------------------------------------------------------------------------ r1857 | oetiker | 2009-06-08 07:12:26 +0200 | 2 lines make config.h a symlink to rrd_config.h so that intl/Makefile is happy ------------------------------------------------------------------------ r1856 | oetiker | 2009-06-08 06:56:56 +0200 | 4 lines Relax automake requirements and make 1.11 features optional -- kevin brintnall ------------------------------------------------------------------------ r1855 | oetiker | 2009-06-07 22:48:39 +0200 | 2 lines fix some missing rrd_flushcached instances -- kevin ------------------------------------------------------------------------ r1854 | oetiker | 2009-06-07 16:46:21 +0200 | 2 lines switch to the rrd_flushcached interface ------------------------------------------------------------------------ r1853 | oetiker | 2009-06-07 16:45:55 +0200 | 2 lines consider fgets return value ------------------------------------------------------------------------ r1852 | oetiker | 2009-06-07 16:33:06 +0200 | 2 lines do not try to mess with autopoints files ------------------------------------------------------------------------ r1851 | oetiker | 2009-06-07 16:27:52 +0200 | 2 lines check return value from write ------------------------------------------------------------------------ r1850 | oetiker | 2009-06-07 16:19:04 +0200 | 2 lines check return values of dup and ftruncate ------------------------------------------------------------------------ r1849 | oetiker | 2009-06-07 16:12:28 +0200 | 2 lines fix some warnings ------------------------------------------------------------------------ r1848 | oetiker | 2009-06-07 15:57:17 +0200 | 2 lines added a copy to svn since it is changed from the original ------------------------------------------------------------------------ r1847 | oetiker | 2009-06-07 15:53:10 +0200 | 2 lines changed from original version, so include it ... ------------------------------------------------------------------------ r1846 | oetiker | 2009-06-07 15:42:39 +0200 | 2 lines include libtool with the distro ------------------------------------------------------------------------ r1845 | oetiker | 2009-06-07 13:36:20 +0200 | 2 lines another ID tag ------------------------------------------------------------------------ r1844 | oetiker | 2009-06-07 13:35:52 +0200 | 2 lines more id tags added ------------------------------------------------------------------------ r1843 | oetiker | 2009-06-07 13:34:39 +0200 | 2 lines updated for release ------------------------------------------------------------------------ r1842 | oetiker | 2009-06-07 13:34:22 +0200 | 2 lines updated for release ------------------------------------------------------------------------ r1841 | oetiker | 2009-06-07 13:20:15 +0200 | 2 lines added $Id:$ ------------------------------------------------------------------------ r1840 | oetiker | 2009-06-07 13:16:07 +0200 | 2 lines initial news ------------------------------------------------------------------------ r1839 | oetiker | 2009-06-02 15:55:40 +0200 | 2 lines Mention that negative time values should be separated by '--' in rrdupdate.pod -- Sebastian Harl ------------------------------------------------------------------------ r1837 | oetiker | 2009-06-01 16:01:51 +0200 | 2 lines Make rrdtool compile on GNU/kFreeBSD. Fix for Debian bug 494874. Patch by Petr Salinger. ------------------------------------------------------------------------ r1836 | oetiker | 2009-06-01 15:58:58 +0200 | 4 lines make negative update times work as diescribed in the documentation (-5 is NOW-5seconds) this takes care of debian bug 332766 ------------------------------------------------------------------------ r1834 | oetiker | 2009-05-28 07:25:32 +0200 | 2 lines madvise hints may not be ORed since they are enumerated ... #224 ------------------------------------------------------------------------ r1833 | oetiker | 2009-05-27 18:23:40 +0200 | 1 line back to 0.14.4 ------------------------------------------------------------------------ r1830 | oetiker | 2009-05-27 18:21:15 +0200 | 2 lines get the latest gettext ------------------------------------------------------------------------ r1826 | oetiker | 2009-05-27 18:03:01 +0200 | 2 lines added gettext.h to dist ------------------------------------------------------------------------ r1825 | oetiker | 2009-05-27 17:00:04 +0200 | 2 lines added missing gettext.h ------------------------------------------------------------------------ r1824 | oetiker | 2009-05-27 16:59:47 +0200 | 2 lines we don't have any translations ... so do not pretend ------------------------------------------------------------------------ r1823 | oetiker | 2009-05-27 16:31:31 +0200 | 2 lines reduce gettext version requirements ------------------------------------------------------------------------ r1820 | oetiker | 2009-05-27 16:14:10 +0200 | 2 lines fix path ------------------------------------------------------------------------ r1819 | oetiker | 2009-05-27 15:39:44 +0200 | 9 lines the get ready for 1.4rc patch * updated to automake 1.11 with silet build, with this warnings stand out much better in the build process. use make V=1 to see all as before * switched from intltoolize to autopoint for gettext support * moved m4 bits into their own subdirectory --tobi ------------------------------------------------------------------------ r1818 | oetiker | 2009-05-27 15:11:22 +0200 | 2 lines Only include libgen.h when we have it. use pbbasename if we do not. Also fixes #223. ------------------------------------------------------------------------ r1817 | oetiker | 2009-05-26 17:25:41 +0200 | 2 lines update flush to flushcached ------------------------------------------------------------------------ r1816 | oetiker | 2009-05-26 17:24:26 +0200 | 2 lines bindings should invoke new rrd_flushcached (instead of rrd_cmd_flush) -- kevin ------------------------------------------------------------------------ r1815 | oetiker | 2009-05-26 17:20:05 +0200 | 2 lines "ulong" is not portable. must be unsigned long -- kevin ------------------------------------------------------------------------ r1814 | oetiker | 2009-05-26 10:55:50 +0200 | 2 lines update flush document name ------------------------------------------------------------------------ r1813 | oetiker | 2009-05-26 09:14:58 +0200 | 2 lines rename rrd_flush.c to rrd_flushcached.c in Makefile.am ------------------------------------------------------------------------ r1812 | oetiker | 2009-05-26 09:13:52 +0200 | 11 lines resolve the "flush inconsistency": * rrd_flush (a leftover from when rrdtool was using stream based io) is turned into a no-op but still exported to preserve library compatibility. * rrd_cmd_flush (the implementation of "rrdtool flush") is renamed to rrd_flushcached along with its command line invocation which is now rrdtool flushcached --tobi ------------------------------------------------------------------------ r1811 | oetiker | 2009-05-26 00:23:06 +0200 | 16 lines While restoring, rrd_restore uses between 30 and 40 times as much memory as the size of the rrd file to be restored. This is due to the use of libxml2 for parsing the the whole file prior to turning it into rrd structs. This patch switches rrd_restore to using the xmlTextReader interface for incremental consumtion of xml files. This reduces the memory requirements to about twice the size of the rrd file which should be quite manageable for most cases. The parsing is about 25% less efficient than with the original libxml2 version of rrd_restore. This could be in part due to the more liberal use of malloced memory. Optimizing this is for another day though. --tobi ------------------------------------------------------------------------ r1809 | oetiker | 2009-05-25 08:05:55 +0200 | 2 lines only include utime.h when necessary ------------------------------------------------------------------------ r1808 | oetiker | 2009-05-24 12:31:44 +0200 | 11 lines * in 1.3: Changed types used in rrd_file_t from off_t to size_t to prevent: warning: comparison between signed and unsigned * Misc other off_t -> size_t changes to prevent signed warning * in 1.3: Added rrd_dump_opt_r declaration to prevent undef warning * Explicit cast to time_t of check in rrd_fetch to prevent signed warning * Added vasprintf path to sprintf_alloc to prevent excessive mallocing * in 1.3: fixed STRERROR_R_CHAR_P undef warning patch by Steven Hartland killing multiplay.co.uk ------------------------------------------------------------------------ r1806 | oetiker | 2009-05-22 16:54:21 +0200 | 2 lines reset errno right befor the call I want to test ------------------------------------------------------------------------ r1805 | oetiker | 2009-05-22 15:55:46 +0200 | 2 lines inline is something for the compiler to figure ... ------------------------------------------------------------------------ r1804 | oetiker | 2009-05-22 13:32:21 +0200 | 2 lines fix error messages on the road to fix #222 ------------------------------------------------------------------------ r1803 | oetiker | 2009-05-21 07:20:50 +0200 | 2 lines fix text organisation ------------------------------------------------------------------------ r1800 | oetiker | 2009-05-18 16:23:23 +0200 | 2 lines integrated fix for debian bug #529291 as supplied by Anders Hammarquist ------------------------------------------------------------------------ r1799 | oetiker | 2009-05-11 23:25:57 +0200 | 6 lines fix rrd_getops use of external variables (optarg and friends) ... most prominently is kills a segfault when using rrdtool perl bindings on solaris ... thanks to Ihsan Dogan for helping with tracking this down and providing access to sparc solaris boxes. ------------------------------------------------------------------------ r1797 | oetiker | 2009-04-29 08:52:25 +0200 | 2 lines added some remarks on CDEF versus VDEF to the docs -- Pablo Sanchez ------------------------------------------------------------------------ r1796 | oetiker | 2009-04-20 08:48:01 +0200 | 9 lines This patch includes utility functions to support dynamically sized arrays. This simplifies the code that manages some of the dynamic structures inside rrdcached. A few data types have been changed to size_t. -- kevin ------------------------------------------------------------------------ r1795 | oetiker | 2009-04-20 08:46:47 +0200 | 3 lines introduced header_property in info output -- Daniel.Pocock barclayscapital.com ------------------------------------------------------------------------ r1794 | oetiker | 2009-04-17 11:53:57 +0200 | 2 lines add documentation for graph-render-mode ------------------------------------------------------------------------ r1793 | oetiker | 2009-04-15 09:05:58 +0200 | 2 lines fix comments and error messages in rrd_restore patch ------------------------------------------------------------------------ r1792 | oetiker | 2009-04-15 09:02:04 +0200 | 4 lines Fix casting issues in rrd_restore. The bug manifested when compiled on sparc v9 as explained in #218. Patch based in input provided by poster of #218. ------------------------------------------------------------------------ r1791 | oetiker | 2009-04-14 15:55:29 +0200 | 3 lines make ruby bindings compile with newer ruby versions ... fix for #217 patch provided by rrd trac user. ------------------------------------------------------------------------ r1790 | oetiker | 2009-04-14 09:51:20 +0200 | 3 lines Generate an error message when using RRDp with graph - this can not work reliably. (Debian Bug#251701) -- Sebastian Harl ------------------------------------------------------------------------ r1789 | oetiker | 2009-04-14 08:10:21 +0200 | 7 lines rrd_random() is a wrapper around random() that ensures the PRNG is seeded exactly ONCE per process. rrd_utils.c is introduced for functions that do not have a better home. --kevin ------------------------------------------------------------------------ r1788 | oetiker | 2009-04-10 10:52:34 +0200 | 13 lines This patch consolidates several aspects of the command handling in rrdcached: * all commands, help, and context information is reduced to a single data structure. * all checks for command validity (context and permission) are handled in the main dispatch routine This removes duplicate help/syntax text, and long if..elseif..else chains. It facilitates the addition of new commands. -- kevin ------------------------------------------------------------------------ r1787 | oetiker | 2009-04-09 17:07:05 +0200 | 3 lines The bookkeeping of all threads is not necessary, since we cannot pthread_join() them anyway. We only need to know when the last one has exited. -- kevin ------------------------------------------------------------------------ r1786 | oetiker | 2009-04-09 17:05:41 +0200 | 2 lines added missing docs -- kevin ------------------------------------------------------------------------ r1785 | oetiker | 2009-04-09 17:04:27 +0200 | 2 lines added missing newline -- by sebastian harl ------------------------------------------------------------------------ r1778 | oetiker | 2009-04-07 09:23:20 +0200 | 2 lines make sure an empty m4 is in the checkout ------------------------------------------------------------------------ r1773 | oetiker | 2009-04-07 08:37:17 +0200 | 6 lines fix for #213 restore old behaviour of --lazy. Even in lazy mode all the data calculation elements are run. The main time saving comes from the fact that the graph is not being drawn and the image does not have to be compressed. ------------------------------------------------------------------------ r1772 | oetiker | 2009-04-07 08:03:49 +0200 | 2 lines make sure rrd_dump outputs numbers with C locale so that . does not suddenly become , this fixes #212 ------------------------------------------------------------------------ r1771 | oetiker | 2009-04-06 16:49:03 +0200 | 3 lines I thought that it would be nice to be able to pass parameters to python bindings as a list (array) of strings. -- Vytautas Zdanavicius vytaszd at yahoo.com ------------------------------------------------------------------------ r1770 | oetiker | 2009-04-06 16:14:31 +0200 | 2 lines use %lld with (long long) conversion when printing time. ------------------------------------------------------------------------ r1769 | oetiker | 2009-04-06 16:12:21 +0200 | 4 lines printf for time_t with %lli can easily cause a segfault since on many platforms time_t is only 32bit ... by adding (long long) to the time value this gets fixed. ------------------------------------------------------------------------ r1768 | oetiker | 2009-04-06 16:08:16 +0200 | 2 lines lets have an error message when creating the graph fails ------------------------------------------------------------------------ r1764 | oetiker | 2009-03-21 11:45:40 +0100 | 2 lines remove excess flush calls ... ------------------------------------------------------------------------ r1762 | oetiker | 2009-03-21 10:56:43 +0100 | 2 lines remove flush symbol ------------------------------------------------------------------------ r1760 | oetiker | 2009-03-21 10:53:30 +0100 | 5 lines the call to rrd_flush has appeared when rrd_open was fleshed out. It used to be an fflush call which is something entirely different than the current rrd_flush implementation with fdatasync was ... we can safely drop this and gain performance for holt winters in the process ... -- tobi ------------------------------------------------------------------------ r1759 | oetiker | 2009-03-18 18:12:09 +0100 | 2 lines New rrd_graph functions for legend placement control added -- Patch by Melchior Rabe rrdtool mrab.de ------------------------------------------------------------------------ r1758 | oetiker | 2009-03-08 12:12:41 +0100 | 3 lines rrd_flush.c: Provide a more useful error message. by Sebastian Harl ------------------------------------------------------------------------ r1757 | oetiker | 2009-03-08 09:02:38 +0100 | 6 lines Renaming ntmake.pl to ntmake.PL (r1742) had unforseen side effects. At least on freebsd ntmake.PL gets called automatically during build. This is not desired. I renamed the file to ntmake-build sinc it gets called explicilty on windows anyway. Reported by Kevin modified by Tobi ------------------------------------------------------------------------ r1756 | oetiker | 2009-03-07 11:31:51 +0100 | 4 lines This 2-patch series exposes 'flush' methods to the various language bindings. ------------------------------------------------------------------------ r1754 | oetiker | 2009-03-06 06:51:34 +0100 | 2 lines there were two lines of debug info printed by rrd_xport, messing up the xml code. Fix by Alejandro Galue. ------------------------------------------------------------------------ r1752 | oetiker | 2009-03-06 06:39:37 +0100 | 18 lines Even though POSIX/XSI requires "strerror_r" to return an "int", some systems (e.g. the GNU libc) return a "char *" _and_ ignore the second argument (user provided buffer). The configure script now checks for that behavior using AC_FUNC_STRERROR_R. rrd_strerror() in rrd_thread_safe.c has been updated to (hopefully) handle all possible cases. Previously, rrd_strerror() would have returned "strerror_r failed. sorry!" in mostly any cases when using glibc, since "if (strerror_r())" had been used to check for errors which evaluates to true if a (non-NULL) pointer was returned. Now, we, at least, return the error number in case anything else fails. Thanks to Alessandro Iurlano for reporting this issue after spotting it in collectd . patch by Sebastian Harl ------------------------------------------------------------------------ r1751 | oetiker | 2009-03-06 06:30:08 +0100 | 2 lines Make sure we use correct sizes for vsnprintf (sizeof) and strncpy (sizeof-1). -- kevin ------------------------------------------------------------------------ r1749 | oetiker | 2009-02-21 10:43:58 +0100 | 2 lines Misc documentation fixes by Alex van den Bogaerdt ------------------------------------------------------------------------ r1748 | oetiker | 2009-02-21 10:23:04 +0100 | 2 lines Documentation fixes by Bernhard Reutner-Fischer ------------------------------------------------------------------------ r1745 | oetiker | 2009-01-25 23:39:34 +0100 | 2 lines added doc on new graphv graph_start und graph_end info variables ------------------------------------------------------------------------ r1744 | oetiker | 2009-01-25 23:02:57 +0100 | 2 lines fix type conversion error (on windows) unsigned/signed affecting timestamps on rrd_dump. ------------------------------------------------------------------------ r1742 | oetiker | 2009-01-25 22:52:25 +0100 | 2 lines rename ntmake.pl to ntmake.PL so that it does not get installed ------------------------------------------------------------------------ r1741 | oetiker | 2009-01-25 19:38:34 +0100 | 2 lines define cwd where we need it and not outside ------------------------------------------------------------------------ r1733 | oetiker | 2009-01-14 09:14:31 +0100 | 5 lines In rrd_graph: the width/height values are written into rrd_infoval_t's u_cnt but read out of u_int (u_long vs. int). This breaks on LP64BE. Patch by Rafal Boni rafal pobox.com ------------------------------------------------------------------------ r1732 | oetiker | 2009-01-14 08:22:09 +0100 | 2 lines added percentnan (VDEF PERCENT variant that ignores NAN) -- patch by Martin Sperl ------------------------------------------------------------------------ r1726 | oetiker | 2009-01-06 17:32:16 +0100 | 2 lines with MMAP enabled rrd resize GROW was broken ... this fixes it ... ------------------------------------------------------------------------ r1725 | oetiker | 2009-01-06 16:45:53 +0100 | 2 lines TICK with negative numbers should 'start from the top of the graph' looking down ... now it does. ------------------------------------------------------------------------ r1724 | oetiker | 2009-01-06 15:54:41 +0100 | 2 lines fix for segfault in rrd_cgi: caused by freeing a invalid address when printing an error message. ------------------------------------------------------------------------ r1723 | oetiker | 2008-12-26 09:05:03 +0100 | 2 lines Include libgen.h to get prototype for basename in a POSIX conformant way. Fix for #202. ------------------------------------------------------------------------ r1722 | oetiker | 2008-12-26 09:03:06 +0100 | 2 lines improvements for build docs by Zorzoli, Pablo ------------------------------------------------------------------------ r1721 | oetiker | 2008-12-22 22:23:49 +0100 | 2 lines should free memory after a successful fetch too ... ------------------------------------------------------------------------ r1720 | oetiker | 2008-12-22 22:23:04 +0100 | 2 lines documentation fixes ------------------------------------------------------------------------ r1718 | oetiker | 2008-12-22 08:21:46 +0100 | 2 lines Fix for POD error by Sebasian Harl ------------------------------------------------------------------------ r1717 | oetiker | 2008-12-22 08:20:17 +0100 | 2 lines Fixes some typos by Sebastian Harl. ------------------------------------------------------------------------ r1716 | oetiker | 2008-12-19 15:26:47 +0100 | 8 lines win32 fixes by Stefan . Ludewig . exitgames . com * added portable plbasename since basename is not available on stock win32 * rrd_update was using %d to show a timestamp ... %lli is better since it also will show 64bit time_t correctly. ------------------------------------------------------------------------ r1715 | oetiker | 2008-12-19 00:56:52 +0100 | 2 lines fix extra dist to include missing win32 files ------------------------------------------------------------------------ r1708 | oetiker | 2008-12-11 15:31:16 +0100 | 2 lines fixed another seek issue ... ------------------------------------------------------------------------ r1707 | oetiker | 2008-12-11 11:01:33 +0100 | 5 lines The spec file for building the rrdtool as an RPM that comes with the release is not "friendly" to older versions of RHEL and derivatives. -- Martin Sperl ------------------------------------------------------------------------ r1706 | oetiker | 2008-12-11 10:59:10 +0100 | 2 lines include luapod into distro ------------------------------------------------------------------------ r1705 | oetiker | 2008-12-10 14:18:10 +0100 | 4 lines On some architectures, strtod does not turn "NaN" into 'Not A Number (NAN)' se we better provide our own code here. ------------------------------------------------------------------------ r1703 | oetiker | 2008-12-10 13:42:46 +0100 | 4 lines rrdtool dump is now backward compatible with rrdtool 1.3 dump again. Default is --header=dtd and --no-header is an alias for --header=none ------------------------------------------------------------------------ r1702 | oetiker | 2008-12-10 09:29:38 +0100 | 2 lines minichange ------------------------------------------------------------------------ r1701 | oetiker | 2008-12-09 19:30:43 +0100 | 4 lines * forward ported right axis patch from 1.2 branche * fixed --full-size-mode sizing issue when used in combination with --title the work in the axis patch has been sponsored by VoltWerk. ------------------------------------------------------------------------ r1700 | oetiker | 2008-12-08 17:07:27 +0100 | 2 lines Fix for last method in ruby bindings -- Hiroyuki Ikezoe ------------------------------------------------------------------------ r1699 | oetiker | 2008-12-05 16:39:55 +0100 | 2 lines integrate osx build hints as indicated in #192 ------------------------------------------------------------------------ r1697 | oetiker | 2008-12-05 16:23:45 +0100 | 2 lines deleted a bit much from rrd_open ... merged back in ------------------------------------------------------------------------ r1696 | oetiker | 2008-12-05 16:18:11 +0100 | 4 lines check for broken implementations of msync / MS_ASYNC where the mtime does not get updated. warn and explicitly call utime in rrd_open when a file is opened rw. ------------------------------------------------------------------------ r1695 | oetiker | 2008-12-05 11:23:46 +0100 | 2 lines added rrdcached examples to extra_dist to make sure they get distributed ------------------------------------------------------------------------ r1694 | oetiker | 2008-12-04 17:01:37 +0100 | 2 lines Legends should NOT be placed in reverse order when using --full-size-mode ... this fixed #198. ------------------------------------------------------------------------ r1693 | oetiker | 2008-12-04 16:49:11 +0100 | 2 lines add de-referencing to step in fetch check ------------------------------------------------------------------------ r1692 | oetiker | 2008-12-04 16:48:39 +0100 | 2 lines fix merg error with win32 port ------------------------------------------------------------------------ r1691 | oetiker | 2008-12-04 10:03:48 +0100 | 2 lines Catch CDEF:tot= and raise and error to fix #199. Thanks to Jeron Massar for reporting this. ------------------------------------------------------------------------ r1690 | oetiker | 2008-12-04 08:02:56 +0100 | 4 lines Update to WIN32-BUILD-TIPS.txt and VC project files so that the rrdtool.exe can be built too. -- Stefan Ludewig Stefan.Ludewig exitgames.com ------------------------------------------------------------------------ r1689 | oetiker | 2008-12-03 12:08:48 +0100 | 4 lines When fetching data from 'past' an rra, rrd_fetch was seeking past the rrd file. This did cause seek errors in some setups. Fixed. Thanks to Daniel Pocock for finding the problem. ------------------------------------------------------------------------ r1686 | oetiker | 2008-12-03 10:54:09 +0100 | 2 lines integrated rrdcached examples into build process ------------------------------------------------------------------------ r1685 | oetiker | 2008-12-03 09:47:56 +0100 | 2 lines add rrdcached examples -- kevin ------------------------------------------------------------------------ r1683 | oetiker | 2008-12-02 13:35:02 +0100 | 2 lines fixed solaris build instructions ------------------------------------------------------------------------ r1682 | oetiker | 2008-12-01 12:05:52 +0100 | 2 lines added libdbi doc to extradist ------------------------------------------------------------------------ r1678 | oetiker | 2008-11-20 14:04:38 +0100 | 27 lines Attached a patch for the rrd-tool LIBDBI integration with the following improvements: a) correct error handling in case of libdbi being unable to load the driver was producing segmentation faults. b) better parsing of datasources * until now timestamp fields had to be integer and had to contain a unix timestamp - now you can now also use DateTime fields (you still need to specify it, as the time-range needs to be defined correctly) * data fields are now no longer limited to (var)char or DOUBLE fields - FLOAT, INTEGER,... are now also supported. c) there is a bug with at least LIBDBI 0.8.1 in conjunction with mysql that can result in segmentation faults when BINARY/BLOB fields are accessed - rrdtool will now tell you about this fact before dying ;) d) also the value of rrdderivemaxstep only gets applied if derive has been selected correctly. e) "GROUP BY timestamp" has been removed from SQL statement. f) "ORDER BY timestamp" will be added only in the case of fetching "derived" data. -- Martin Sperl ------------------------------------------------------------------------ r1677 | oetiker | 2008-11-18 18:19:17 +0100 | 5 lines win32 portability patch and win32/rrdlib.vcproj file for the source -- christof wegmann Christof.Wegmann with exitgames.com some fixes to make the result work on unix again ... ------------------------------------------------------------------------ r1674 | oetiker | 2008-11-18 08:14:10 +0100 | 2 lines missing , added to help string ------------------------------------------------------------------------ r1672 | oetiker | 2008-11-17 17:21:41 +0100 | 2 lines added build instruction for libdbi docu ------------------------------------------------------------------------ r1670 | oetiker | 2008-11-16 23:24:44 +0100 | 2 lines missing pod files ------------------------------------------------------------------------ r1669 | oetiker | 2008-11-15 15:35:51 +0100 | 2 lines switch tu using AS_HELP_STRING for formatting ... ------------------------------------------------------------------------ r1668 | oetiker | 2008-11-15 15:15:38 +0100 | 3 lines fix typo in handle_request_help --kevin ------------------------------------------------------------------------ r1667 | oetiker | 2008-11-15 15:14:19 +0100 | 8 lines doc fixups for rrdcacehd * documented the -p option * added nbsp between option and argument * layout one option per line --kevin ------------------------------------------------------------------------ r1666 | oetiker | 2008-11-11 17:36:45 +0100 | 2 lines make basename work ------------------------------------------------------------------------ r1665 | oetiker | 2008-11-11 17:36:13 +0100 | 14 lines This patch allows for multiple RRD writer threads to service the queue. Now, rrdcached can have more simultaneous I/O requests into the kernel. This may allow the kernel to re-order disk writes, resulting in better disk throughput. In practice, throughput is increased by 2-3x. The flush (-f) timer maintenance has been moved from queue_thread_main into its own thread. Be more careful about when to use pthread_cond_signal vs _broadcast, since multiple threads may be waiting on &queue_cond. --kevin ------------------------------------------------------------------------ r1663 | oetiker | 2008-11-09 20:19:48 +0100 | 2 lines be even more liberal with NOT using isfinite on solaris2.8 ------------------------------------------------------------------------ r1662 | oetiker | 2008-11-09 01:00:33 +0100 | 2 lines use basename to determine filename for imginfo as suggested by sebastian ------------------------------------------------------------------------ r1661 | oetiker | 2008-11-08 23:57:38 +0100 | 3 lines rrdcached: move queue length decrement into remove_from_queue -- kevin ------------------------------------------------------------------------ r1660 | oetiker | 2008-11-08 23:57:18 +0100 | 3 lines rrdcached: examine the current queue with the "QUEUE" command -- kevin ------------------------------------------------------------------------ r1659 | oetiker | 2008-11-08 23:50:51 +0100 | 4 lines This helps with portability on platforms where realloc doesn't handle NULL. -- kevin ------------------------------------------------------------------------ r1658 | oetiker | 2008-11-08 19:01:33 +0100 | 3 lines document the "QUIT" command -- kevin ------------------------------------------------------------------------ r1657 | oetiker | 2008-11-08 19:01:01 +0100 | 4 lines rrdcached: journal_replay default failure reason, just in case avoids a "may be used uninitialized" warning -- kevin ------------------------------------------------------------------------ r1656 | oetiker | 2008-11-08 19:00:21 +0100 | 3 lines added "quit" command -- kevin ------------------------------------------------------------------------ r1655 | oetiker | 2008-11-08 18:57:32 +0100 | 9 lines rrdcached now frees all of its resources correctly. This facilitates memory debugging. g_tree now knows how to free the nodes when it removes them. Also, use g_tree_replace instead of g_tree_insert. This fixes a bug triggered when the same file was simultaneously inserted by two clients. --kevin ------------------------------------------------------------------------ r1654 | oetiker | 2008-11-08 18:56:48 +0100 | 2 lines do not quit when libdbi is missing ------------------------------------------------------------------------ r1653 | oetiker | 2008-11-07 23:46:40 +0100 | 2 lines fixed --font TITLE:12: problem -- tobi ------------------------------------------------------------------------ r1652 | oetiker | 2008-11-07 23:23:16 +0100 | 2 lines fix scaling issue -- Martin ------------------------------------------------------------------------ r1651 | oetiker | 2008-11-07 15:09:29 +0100 | 2 lines extra file to make libdbi patch work ------------------------------------------------------------------------ r1650 | oetiker | 2008-11-07 15:07:53 +0100 | 38 lines As some of you may know that I have created a patch for rrdtool 1.2 a few years ago, so that a database could be queried for values for graphing. The patch has been mostly rewritten and the following changes have been made: * high dependency on mysql has been reduced by avoiding the temporary tables (which was bad for mysql replication) * The number of executed SQL-Statements for one CDEF has been reduced to 1 compared to 11 SQLs (including CREATE TEMPORARY TABLE) - for patch against version 1.2 * All consolidation is done in rrdtool itself (MIN,MAX,AVERAGE) * Additional consolidation functions are COUNT and SIGMA, which give information on statistics on a per "time-bin" basis. * All these consolidation values are always returned as separate columns, that are returned by RRD and the consolidation function given as Argument is ignored. Main reason is that this way there is only one call to rrd_fetcht and thus the database even if we need to fetch for example min, avg and max. Compare this to 3 calls in case of different consolidation functions - and if you want to get SIGMA and COUNT as well it is still only one call to the backend and the database. * Some previous existing features have been taken out at the moment to allow for this reduced set of SQL queries. o prediction using the values from the last X days at the same time o the corresponding sigma calculation * The idea is to create generic CDEF's that will do the same thing, but that is also available when using RRD-files (similar to TREND, but with another scope) This will get posted as a separate patch. * Overall performance should be much better and the patch as a whole simpler. * The patch also includes modifications to the configuration infrastructure, to make libdbi support optional. -- Martin Sperl ------------------------------------------------------------------------ r1649 | oetiker | 2008-11-07 15:02:09 +0100 | 3 lines new CDEF functions for predictions PREDICT and PREDICTSIGMA -- Martin Sperl ------------------------------------------------------------------------ r1648 | oetiker | 2008-11-07 14:51:24 +0100 | 10 lines Much simpler handling of timestamp errors. Return an error to the user when any of the time stamp values are invalid. This is similar to RRDTool's normal behavior. Removed the complex logic previously used to return error codes to the user. This solves a bug where non-advancing timestamps could have produced incorrect error output during "BATCH" mode. The bug was cause by using the sock->wbuf pointer for the error output. -- kevin brintnall ------------------------------------------------------------------------ r1647 | oetiker | 2008-11-07 14:45:34 +0100 | 4 lines rrdcached: pull in rrd_config.h so we can use its defines ... particularly HAVE_VSNPRINTF -- kevin brintnall ------------------------------------------------------------------------ r1645 | oetiker | 2008-11-04 08:12:46 +0100 | 15 lines I realize now that the problem is the line test -f lua/Makefile && cd lua && $(MAKE) install || true in the target "install-data-local", in bindings/Makefile.am. It forces execution of bindings/lua/Makefile independently of lua being found or not. I added that line in my first patch, following perl, python and ruby build style, but it's not needed after I switched to automake. The make recursion is controlled by SUBDIRS, which will only contain "lua" if BUILD_LUA is true. -- Fidelis Assis ------------------------------------------------------------------------ r1644 | oetiker | 2008-11-02 18:25:36 +0100 | 2 lines fix build process for systems without lua ------------------------------------------------------------------------ r1643 | oetiker | 2008-11-02 18:09:36 +0100 | 3 lines PTHREAD_COND_INITIALIZER may only be used to initialize static cond variables. -- Sebastian Harl ------------------------------------------------------------------------ r1642 | oetiker | 2008-11-02 16:25:48 +0100 | 2 lines reverting non build makefile patch ... we need to lua makefile for building the ditribution ------------------------------------------------------------------------ r1641 | oetiker | 2008-11-01 10:09:07 +0100 | 2 lines do not create the lua makefile if lua is not to be compiled. ------------------------------------------------------------------------ r1638 | oetiker | 2008-10-30 15:32:09 +0100 | 2 lines fix from alex ------------------------------------------------------------------------ r1637 | oetiker | 2008-10-28 09:57:13 +0100 | 5 lines - remove the spaccing between the elements - add xsd support to dump output - change the argument "[--no-header|-n]" to [--header|-h {xsd,dtd}] -- tobias.lindenmann 1und1.de ------------------------------------------------------------------------ r1632 | oetiker | 2008-10-26 16:18:33 +0100 | 2 lines make compilation on older OSX versios work by removing the # from the librrd.sym file ------------------------------------------------------------------------ r1628 | oetiker | 2008-10-24 08:20:21 +0200 | 2 lines fixed typo ... GAUGE not GUAGE ------------------------------------------------------------------------ r1627 | oetiker | 2008-10-22 22:41:59 +0200 | 4 lines The previous code relied on the assumption that pthread_cond_init(&cond) was equivalent to memset(&cond,0). This may not be true on all platforms. -- kevin ------------------------------------------------------------------------ r1626 | oetiker | 2008-10-22 08:02:23 +0200 | 15 lines remove_cache_item() did not check whether a file was in queue before modifying the cache head/tail pointers. Therefore, the process of flushing old files may perturb the cache_queue_head pointer. This caused some nodes with CI_FLAGS_IN_QUEUE to be un-linked from the queue list. Thereafter, they would not be flushed by any periodic process (although they could be revived with FLUSH or UPDATE). This caused a slow memory leak for files that are no longer updated. Pending updates for these "abandoned" files would remain in memory ad infinitum. With this patch, remove_from_queue() will check that the item is queued before modifying the head/tail pointers. This restores the intended behavior. --kevin ------------------------------------------------------------------------ r1625 | oetiker | 2008-10-22 07:08:47 +0200 | 2 lines this was replaced by Makefile.am ------------------------------------------------------------------------ r1624 | oetiker | 2008-10-21 08:10:32 +0200 | 2 lines guess one needs that in svn too for luabindings to work ------------------------------------------------------------------------ r1623 | oetiker | 2008-10-21 07:42:50 +0200 | 13 lines * Open all listen sockets in daemonize(), while we still have stderr. Changed open_listen_socket_* routines to complain to stderr. Now, any errors in binding to the listen sockets are much more obvious. * Simplified exit of parent after fork() * PID file will be correctly cleaned up if there is a failure in daemonize(). * unlink the unix socket before trying to bind() (after we're sure we have the PID file) -- kevin ------------------------------------------------------------------------ r1622 | oetiker | 2008-10-20 13:46:08 +0200 | 9 lines rrd_notify_row patch: - Delegate choice of starting row for newly created RRD files to the rrd_open.c API. - Introduce the rrd_notify_row() function so that an implementation can choose to align the rows of new RRDs with existing RRDs, if desirable. - Maintain the existing behaviour (random starting row) by default. -- Daniel.Pocock ------------------------------------------------------------------------ r1621 | oetiker | 2008-10-20 08:06:16 +0200 | 4 lines - Lua module build and install process rewritten using automake+libtool - improved check for compat-5.1 availability -- Fidelis Assis ------------------------------------------------------------------------ r1620 | oetiker | 2008-10-19 00:32:19 +0200 | 6 lines rrd_open should not create files with restrictive masks Removed unnecessary "mode" varaible. The mode is only used when O_CREAT is specified, where we want to use 0666 (as rrd_create_fn did r<=1612). --kevin ------------------------------------------------------------------------ r1619 | oetiker | 2008-10-18 17:50:07 +0200 | 7 lines - encapsulate fd and mmap related variables within a private data structure - rrd_file_t keeps a pointer to the private data structure of type void*, so that other block storage implementations can store their internal data with rrd_file_t -- Daniel.Pocock ------------------------------------------------------------------------ r1618 | oetiker | 2008-10-17 07:50:22 +0200 | 3 lines rrdc_flush_if_daemon should not clobber error message -- kevin (as suggested by Sebastian) ------------------------------------------------------------------------ r1617 | oetiker | 2008-10-17 07:44:09 +0200 | 2 lines use double for legend positioning to fight wiggle ------------------------------------------------------------------------ r1616 | oetiker | 2008-10-16 23:33:08 +0200 | 5 lines fix version numbers in compile instructions tobi ------------------------------------------------------------------------ r1614 | oetiker | 2008-10-16 23:30:17 +0200 | 4 lines In almost all cases where rrdc_flush can fail, it will leave a more descriptive error message anyway.. better not overwrite it. -- kevin ------------------------------------------------------------------------ r1613 | oetiker | 2008-10-16 23:12:27 +0200 | 14 lines - rrd_open() calculates file size for new files and calls mmap once for the whole file - rrd_resize() cleaned up, no longer passing a size through the cookie argument - rrd_init(&my_rrd) must be called before rrd_open() - if people are calling rrd_open directly from application code, this might be troublesome. Alternative solutions: creating an additional function, rrd_open_create(), or adding an extra argument to rrd_open() for setting the file size -- Daniel Pocock ------------------------------------------------------------------------ r1612 | oetiker | 2008-10-16 23:05:51 +0200 | 2 lines move rrd_lock into rrd_open where the general rrd_file ops are located. -- Daniel Pocock ------------------------------------------------------------------------ r1610 | oetiker | 2008-10-16 08:11:54 +0200 | 3 lines error reporting caused crash in journal replay ------------------------------------------------------------------------ r1609 | oetiker | 2008-10-16 08:11:44 +0200 | 2 lines better build docs ------------------------------------------------------------------------ r1607 | oetiker | 2008-10-15 07:29:41 +0200 | 2 lines reverted r1601 and r1606 since r1601 introduced a non portable mremap and r1606 has a dependency on r1601. ------------------------------------------------------------------------ r1606 | oetiker | 2008-10-14 22:14:35 +0200 | 15 lines This moves selection of the initial RRA row into the rrd_open.c API The current implementation (random row) is used by default. However, it now provides an opportunity for alternative implementations to integrate with rrdtool in a single place. Maybe there are other places in rrdtool where I should insert calls to the function rrd_notify_row()? This has been tested with rrdtool create and rrdtool info to verify that random rows are selected by default (existing behaviour preserved). -- Daniel Pocock ------------------------------------------------------------------------ r1605 | oetiker | 2008-10-14 21:23:24 +0200 | 8 lines Under most circumstances, rrdcached can detect a stale pid file. If the process in the pid file does not exist, or cannot be signalled by the rrdcached owner, then rrdcached will replace the pid file and start normally. Otherwise, it will complain verbosely to STDERR. --kevin ------------------------------------------------------------------------ r1604 | oetiker | 2008-10-14 21:08:56 +0200 | 12 lines * this preserves principle of least surprise when dealing with files that are reachable via many path strings. i.e. when $PWD=/base/dir the following files are the same: /base/dir/x.rrd x.rrd ../dir/x.rrd * for performance, absolute paths (starting with '/') are not resolved. this reduces the number of stat(2) system calls. ------------------------------------------------------------------------ r1603 | oetiker | 2008-10-14 21:08:36 +0200 | 15 lines Now, the daemon will check that a base directory is NOT reached via symbolic link. Documentation added to illustrate the restriction. This allows several simplifying (and performance-enhancing) assumptions to be made elsewhere in the code: * it ensures that paths resolved in the client via realpath() will match our data structure keys * it's possible to generate the correct absolute path when given a relative path by simply prepending the base directory * it's not necessary to resolve paths that begin with '/' -- kevin ------------------------------------------------------------------------ r1602 | oetiker | 2008-10-14 21:08:00 +0200 | 2 lines rrdcached treats relative and absolute paths the same -- kevin ------------------------------------------------------------------------ r1601 | oetiker | 2008-10-14 13:33:37 +0200 | 13 lines Some observations I made while implementing this: - In rrd_open(), the call to rrd_init() clobbers values already populated in rrd_create.c - The logic for newfile_size in rrd_open() wasn't really able to cope with the logic for a file starting with size 0 I've tried to deal with these issues, but maybe there is a more elegant solution. -- Daniel Pocock ------------------------------------------------------------------------ r1600 | oetiker | 2008-10-14 08:37:19 +0200 | 3 lines clearer way of advancing the flush time -- kevin ------------------------------------------------------------------------ r1599 | oetiker | 2008-10-14 00:07:14 +0200 | 18 lines This patch reduces the number of time()/gettimeofday() system calls when doing high volume processing. This enables about 25% speed increase during journal replay and "BATCH" processing. (this is a function of syscall overhead). * note when "BATCH" processing or journal replay starts, use that timestamp for all commands * use the batch start time to detect when we're in batch mode. no longer need a separate boolean. * pass the time_t into handle_request * pass the time_t through to the commands that need it -- kevin brintnall ------------------------------------------------------------------------ r1598 | oetiker | 2008-10-13 20:41:12 +0200 | 3 lines fixed bad folding marker (on handle_request_flush) -- kevin ------------------------------------------------------------------------ r1597 | oetiker | 2008-10-11 11:53:40 +0200 | 4 lines this ensures that the response is protocol-compliant even if additional info has been added to the write buffer during processing. -- kevin brintnall ------------------------------------------------------------------------ r1596 | oetiker | 2008-10-11 11:37:53 +0200 | 10 lines This patch introduces a feature whereby rrdcached will disallow updates that do not advance the update time. This prevents the updates from being discarded later by rrd_update_r. This patch attempts to make the most of the protocol's limited ability to return error text when using a -1 return code. -- kevin brintnall ------------------------------------------------------------------------ r1595 | oetiker | 2008-10-10 07:21:19 +0200 | 5 lines This patch ensures that when rrdcached is stopped, it cleans up the pid file. Apparently this is necessary if RRDCACHED_USER is not the default "rrdcached". -- Bernard Li ------------------------------------------------------------------------ r1594 | oetiker | 2008-10-10 07:21:01 +0200 | 4 lines The patch I submitted for rrdtool.spec introduced a bug where there are two ldconfig calls in the %postun section. This patch fixes that. -- Bernard Li ------------------------------------------------------------------------ r1593 | oetiker | 2008-10-10 07:19:06 +0200 | 11 lines This patch does two things: - remove the --pidfile check, for compatibility with older RHEL/Centos systems - update the username used by the init script, to remain consistent with Bernard's changes to the spec file -- Daniel.Pocock ------------------------------------------------------------------------ r1592 | oetiker | 2008-10-08 22:47:04 +0200 | 2 lines Split-out rrd-cached subpackage -- Bernard Li ------------------------------------------------------------------------ r1591 | oetiker | 2008-10-08 07:49:24 +0200 | 5 lines This patch updates the spec file and includes the librrd.pc file in the -devel subpackage so that you can build the RPM again. -- Bernard ------------------------------------------------------------------------ r1590 | oetiker | 2008-10-08 07:47:39 +0200 | 2 lines fix crash reported by Bernard Li - connection_thread_main: avoid double calls to close_connection -- kevin ------------------------------------------------------------------------ r1589 | oetiker | 2008-10-08 07:46:21 +0200 | 1 line fixed typo ------------------------------------------------------------------------ r1588 | oetiker | 2008-10-07 23:10:54 +0200 | 7 lines This patch introduces two new commands for cache management: PENDING: shows any un-written updates for a file FORGET : remove a file completely from cache -- kevin brintnall ------------------------------------------------------------------------ r1587 | oetiker | 2008-10-07 23:08:30 +0200 | 15 lines This patch moves the permission handling code around a bit. * moved privilege checks into the command handler functions (possible now that we pass the sock data structures around) * on UPDATE, delay journal_write until after check_file_access(). previously, it was possible for a high-priv socket to introduce commands into the journal that could be replayed if they were still in the journal at next startup. * moved has_privilege() further up in the file to avoid need for prototype. -- kevin brintnall ------------------------------------------------------------------------ r1586 | oetiker | 2008-10-07 18:28:24 +0200 | 24 lines This patch introduces some extra safety checks in journal processing, and cleans up the code a little bit. * moved journal initialization to its own function; main() is cleaner * any time we process a file, log the results (previous code only loggded if there was a valid entry) * After reading journals at startup, only trigger full flush out to disk if the user specified -F. Avoids unnecessary IO on startup unless the user also wants unnecessary IO on shutdown. * journal_replay is much more careful about files it will open * must be a regular file * must be owned by daemon user * must not be group/other writable * Ensure that the journal gets created with the right permissions. ... even when the daemon is invoked with a permissive umask. equivalent to "chmod a-x,go-w" -- kevin ------------------------------------------------------------------------ r1584 | oetiker | 2008-10-07 17:37:34 +0200 | 4 lines aniel Pocock reported that the argument may be NULL in low-diskspace situations, so check for that here to prevent a segmentation fault. -- Florian Forster ------------------------------------------------------------------------ r1583 | oetiker | 2008-10-06 21:14:17 +0200 | 2 lines Makefile.lua is now generated ------------------------------------------------------------------------ r1582 | oetiker | 2008-10-06 21:13:30 +0200 | 2 lines added missing file from lua 5.0 integration patch ------------------------------------------------------------------------ r1581 | oetiker | 2008-10-06 21:05:47 +0200 | 12 lines This patch introduces "BATCH" mode. In this mode, a client can feed multiple commands to rrdcached without waiting for acknowledgement. This permits multiple commands to be sent for each read()/write(). This can dramatically increase the command throughput by increasing the amount of work done per system call. It enables over 100k updates/second with no CPU utilization due to the reduced system calls. -- kevin brintnall ------------------------------------------------------------------------ r1580 | oetiker | 2008-10-06 21:04:48 +0200 | 15 lines This patch introduces buffered I/O to rrdcached. Now, rrdcached can interpret as many commands as arrive in a single read(), and it will use fewer write()s when there are multiple output lines. All routines now pass around listen_socket_t objects instead of file descriptors. All I/O is now contained in two routines. It's no longer necessary to specify the line count in multi-line outputs, since that is calculated automatically. This is the foundation for accepting batched commands. -- kevin brintnall ------------------------------------------------------------------------ r1579 | oetiker | 2008-10-06 07:05:19 +0200 | 2 lines Support to Lua 5.0 -- Fidelis Assis ------------------------------------------------------------------------ r1575 | oetiker | 2008-10-04 18:05:11 +0200 | 3 lines initial librrd file added ... this will need more love since it does not pick up any -rpath settings ------------------------------------------------------------------------ r1571 | oetiker | 2008-10-04 15:34:54 +0200 | 3 lines * rrd_resize fixed so that it does NOT alter the original file (mmap side effect) * rrd_resize can can deal with version 4 files, so let it ------------------------------------------------------------------------ r1570 | oetiker | 2008-10-04 00:53:28 +0200 | 11 lines The patch fixs rrdtools behaviour when presented with trunkated rrd files. * Make sure that we do not extend past the end of the file when reading the header. * Makes sure that the data portion of the file is large enough, based on the ds_cnt and the total number of RRA rows -- kevin brintnall ------------------------------------------------------------------------ r1567 | oetiker | 2008-10-02 21:17:18 +0200 | 2 lines folding fix ------------------------------------------------------------------------ r1566 | oetiker | 2008-10-02 21:08:30 +0200 | 3 lines missing spaces in extconf.rb caused compile failure -- kevin ------------------------------------------------------------------------ r1565 | oetiker | 2008-10-02 19:17:58 +0200 | 2 lines permission fix for rrdcached socket -- Daniel.Pocock ------------------------------------------------------------------------ r1559 | oetiker | 2008-10-01 22:22:57 +0200 | 3 lines since rrdcached uses pthread functions, use the threadsafe version of librrd as well. This will also reasolve build problems on boxes there the ptherad functions must be linked explicitly. ------------------------------------------------------------------------ r1558 | oetiker | 2008-10-01 22:04:47 +0200 | 2 lines Clearer explanation on which paths are permitted when -B is supplied -- kevin ------------------------------------------------------------------------ r1557 | oetiker | 2008-10-01 22:01:43 +0200 | 26 lines Fixes for the following compiler warnings: - unused variable - unused parameter - assignment / argument discards qualifiers from pointer target type - comparison between signed and unsigned - too many arguments to function - assignment makes pointer from integer without a cast - incompatible pointer type - differ in signedness - implicit declaration of function - enumeration value not handled in switch - value computed is not used Most notably, a possible segfault in the Rrd_Lastupdate() code of the TCL bindings has been fixed. Also, -Wundef (warn if an undefined identifier is evaluated in an #if directive) has been removed from CFLAGS. I don't see any problem with letting undefined identifiers evaluate to "false" in rrdtool. Keeping that option would produce a lot of (imho unnecessary) errors which would need to be fixed using ugly preprocessor statements like '#if defined(FOO) && FOO'. -- Sebastian Harl ------------------------------------------------------------------------ r1556 | oetiker | 2008-10-01 21:48:15 +0200 | 15 lines I've adapted an init script for rrdcached, and also incorporated it into the spec file so that it is deployed with the RPM. There are also some other changes to the spec file so that I could build an RPM successfully from trunk. I'm happy to tidy up the spec file some more if no one else wants to mandate the best way to do it. By default, rrdcached runs as nobody. I've tested this on a server running Ganglia gmetad. Regards, Daniel ------------------------------------------------------------------------ r1555 | oetiker | 2008-10-01 21:44:36 +0200 | 11 lines Now, moving a value to the head of the queue is O(1). Before it was O(queue size). This improves performance of individual flushes when there is a large number of files in the queue. As a result, we don't hold the cache_lock as much. Revamped enqueue_cache_item to take advantage of the new structure. Renamed _wipe_ci_values to look nicer with other code. --kevin ------------------------------------------------------------------------ r1552 | oetiker | 2008-10-01 18:53:56 +0200 | 2 lines lcd is actually gcd! ------------------------------------------------------------------------ r1551 | oetiker | 2008-09-30 20:44:03 +0200 | 4 lines When -B is specified, the daemon will only operate on files within the base directory. Symlink detection is omitted for performance reasons (if a user can create a symlink, they can probably overwrite the RRDs anyway). -- kevin ------------------------------------------------------------------------ r1550 | oetiker | 2008-09-30 20:43:15 +0200 | 2 lines check permissions before writing to journal -- kevin ------------------------------------------------------------------------ r1549 | oetiker | 2008-09-30 20:42:34 +0200 | 3 lines The daemon should behave the same way w/r/t files whether we "-g" or not. -- kevin ------------------------------------------------------------------------ r1548 | oetiker | 2008-09-29 22:00:34 +0200 | 6 lines This bug caused the last line in each journal file to be processed a second time. Since it had been modified due to tokenizing, it failed syntax check. The daemon would always record one failed line at end-of-journal as a result. No data loss incurred by this bug. -- kevin ------------------------------------------------------------------------ r1547 | oetiker | 2008-09-29 21:58:34 +0200 | 13 lines This patch introduces the concept of socket privilege levels. "UPDATE" and "FLUSHALL" commands are restricted to high-privilege sockets. "FLUSH" commands can be executed on any socket. This is ideal for multi-user installations where only certain users need write access to the RRD files. Now, nearly all socket information is passed around the daemon in listen_socket_t data structures. In case there is other per-socket state (i.e. if we add authentication) we can put it there. Also, I created a new "open_listen_socket_network" and removed the network setup from "open_listen_socket". -- kevin ------------------------------------------------------------------------ r1546 | oetiker | 2008-09-29 21:56:10 +0200 | 6 lines This patch provides better error messages to the client when something goes wrong with the daemon. When possible, the daemon error message is passed through to rrd_set_error() on the client. Prior to this patch, most error conditions would result in "Internal error", which is not very helpful. -- kevin brintnall ------------------------------------------------------------------------ r1543 | oetiker | 2008-09-28 23:37:46 +0200 | 2 lines spelling fix -- kevin ------------------------------------------------------------------------ r1542 | oetiker | 2008-09-28 23:36:46 +0200 | 4 lines This patch removes an extra "SIGNALS" section in the rrdcached.pod and merges "[BUG] fixed hang in flush_file() introduced by per-file flush condition". -- kevin brintnall ------------------------------------------------------------------------ r1541 | oetiker | 2008-09-28 21:27:45 +0200 | 3 lines a few missing {{{ folding }}} markers added -- kevin brintnall ------------------------------------------------------------------------ r1540 | oetiker | 2008-09-28 21:26:39 +0200 | 3 lines This patch introduces "fast shutdown" mode and two new signals. -- kevin brintnall ------------------------------------------------------------------------ r1539 | oetiker | 2008-09-28 21:25:40 +0200 | 4 lines Moved signal handler setup out of daemonize(). Coalesced common code in preparation for new signals. Documented behavior of existing signals. -- kevin brintnall ------------------------------------------------------------------------ r1537 | oetiker | 2008-09-28 21:23:23 +0200 | 9 lines Attached is a patch to lower the version requirements of libtool and automake. I have tested this on CentOS 4.x with the specified versions of libtool and automake and was able to build RRDTool fine. I did *not* test building with PHP, tcl, ruby or Python though. I also abstracted the version numbers of all the dependencies such that editing them in the future will be easier. -- Bernard Li ------------------------------------------------------------------------ r1535 | oetiker | 2008-09-28 21:12:22 +0200 | 2 lines added support for FLUSHALL command -- kevin brintnall ------------------------------------------------------------------------ r1534 | oetiker | 2008-09-28 21:10:57 +0200 | 1 line move cache broadcast into enqueue_cache_item -- kevin brintnall ------------------------------------------------------------------------ r1533 | oetiker | 2008-09-28 21:10:15 +0200 | 6 lines When -z is specified, some updates may be timestamped up to seconcds in the future. Therefore, a timeout of now+1 may not be sufficient. Set abs_timeout past the point where any updates are currently specified. -- kevin brintnall ------------------------------------------------------------------------ r1532 | oetiker | 2008-09-28 21:09:34 +0200 | 15 lines The PID file is created with open() in the parent process, while we still have STDERR open. If it cannot be created, it complains verbosely to stderr. The PID file is written in the child process. The only way the fdopen() will fail on a fd that is already open is if you're completely out of memory. As in other places in the code, I didn't consider this a case that required a very verbose message. (Search for "strdup failed"). If you still think a more verbose message is called for, please suggest one. The attached patch corrects the error message to complain about fdopen() vs fopen(). I hadn't noticed that until you brought it up. -- kevin brintnall ------------------------------------------------------------------------ r1531 | oetiker | 2008-09-28 21:08:41 +0200 | 2 lines rrd.h, librrd.sym: Define and export rrd_info_r() -- Sebastian Harl ------------------------------------------------------------------------ r1530 | oetiker | 2008-09-28 17:01:43 +0200 | 7 lines vdef calc was using end_orig to determine for which range it should do its calculations which is odd, since orig is only the requested range as invocation time and not the data range deliverd by fetch. It does fall completely flat when shifting since shifting does not affect the original data. Bug #177 reported by hokiel ------------------------------------------------------------------------ r1529 | oetiker | 2008-09-27 13:44:58 +0200 | 2 lines rrdxport was completely broken for exporting datasources that did not have a uniform step size. Thanks to Peter Valdemar Mørch for finding this. ------------------------------------------------------------------------ r1528 | oetiker | 2008-09-26 07:11:32 +0200 | 2 lines create the pid file before forking, so we can complain to stderr -- kevin brintnall ------------------------------------------------------------------------ r1527 | oetiker | 2008-09-26 07:10:25 +0200 | 8 lines This patch ensures that the "FLUSH" command will write the updates out to RRD before returning to the user. Before, it returned when the update was "dequeued"; updates were not necessarily on disk. Also, for new nodes, the cache_lock is not held while we are setting up the new node. We don't want to be holding the lock if the stat() blocks. -- kevin brintnal ------------------------------------------------------------------------ r1526 | oetiker | 2008-09-25 23:07:35 +0200 | 2 lines Use the same IPv6/IPv4 as for the client as Florian did for the server -- kevin brintnall ------------------------------------------------------------------------ r1525 | oetiker | 2008-09-25 22:31:29 +0200 | 10 lines This fixes a couple problems when exiting due to signal: * connection threads exit without closing client socket (so client blocks waiting for response that never comes) * listen_queue_thread blocks on poll() until a new connection comes in (because no poll timeout specified) -- kevin brintnall ------------------------------------------------------------------------ r1524 | oetiker | 2008-09-25 22:11:43 +0200 | 2 lines set SO_REUSEADDR on the listen socket -- kevin brintnall ------------------------------------------------------------------------ r1523 | oetiker | 2008-09-25 22:10:17 +0200 | 17 lines Support for IPv6 has been broken with revision 1522: Because IPv6-addresses contain colons, simply checking for a colon and using everything after it does destroy correctly formatted IPv6-addresses. This patch checks for dots '.' in the address. If the address contains at least one dot, it is considered to be a hostname or an IPv4-address and a simple search for a colon is done. If no dot is found, the code will check for an opening square bracket '[' at the beginning of the address. If one if found, the format [address]:port is assumed. If neither applies, the default port will be used. -- Florian Forster ------------------------------------------------------------------------ r1522 | oetiker | 2008-09-25 17:25:51 +0200 | 4 lines This adds support for
: in the rrd client library. Obviously this is required to take advantage of the server's ability to bind to a non-standard port -- kevin brintnall ------------------------------------------------------------------------ r1521 | oetiker | 2008-09-25 00:21:21 +0200 | 2 lines Allow ports to be specified with -l addr:port also. -- kevin brintnall ------------------------------------------------------------------------ r1520 | oetiker | 2008-09-25 00:07:33 +0200 | 2 lines the rrdlua.pod is in the doc directory ------------------------------------------------------------------------ r1519 | oetiker | 2008-09-24 08:04:26 +0200 | 2 lines added section on error reporting -- Florian Forster ------------------------------------------------------------------------ r1517 | oetiker | 2008-09-22 08:35:26 +0200 | 3 lines I finally finished the first version of the patch (attached) -- Fidelis Assis fidelis pobox.com (this does not seem to quite work yet at least not in my hardy setup) ------------------------------------------------------------------------ r1515 | oetiker | 2008-09-22 07:38:28 +0200 | 2 lines use default prefix for ruby install if no user defined prefix is available ------------------------------------------------------------------------ r1513 | oetiker | 2008-09-19 22:34:42 +0200 | 2 lines Bsd lirary path switch added to perl bindings makefile -- as suggested by Kevin Brintnall ------------------------------------------------------------------------ r1512 | oetiker | 2008-09-18 15:30:48 +0200 | 3 lines Fix for debian bug 498183 ... never *advise after the end of the file. On sparc this causes a segfault. Thanks to Sebastian Harl and Jurij Smakov for finding this. ------------------------------------------------------------------------ r1507 | oetiker | 2008-09-14 17:30:43 +0200 | 2 lines fix RRDCacheD documentation -- Florian Forster ------------------------------------------------------------------------ r1506 | oetiker | 2008-09-14 17:28:34 +0200 | 13 lines The previous code was broken: The response was read using `read(2)'. If the server wasn't sending fast enough, the client would stop reading before the entire message had been read. This patch changes the communication code to use the (line based) `fgets' function rather than the lower level `read' function. After reading the first line (which contains the total number of line to be expected), this precise number of lines is read - blocking if necessary. Also, the missing four new statistic values have been added to `rrdc_stats_get'. --Folorian Forester ------------------------------------------------------------------------ r1505 | oetiker | 2008-09-14 12:35:51 +0200 | 2 lines did not pick up all the changes for rrdcached in the first round ... so here is the second batch. ------------------------------------------------------------------------ r1504 | oetiker | 2008-09-14 11:49:03 +0200 | 4 lines RRDcached patch. This implements an infrastructure, where rrd updates can be sent to a daemon which caches them prior to bulk-updateing rrd files. See the rrdcached manual page. -- Created by Florian Forster with some help from Kevin Brintnall. ------------------------------------------------------------------------ r1502 | oetiker | 2008-09-13 13:42:08 +0200 | 7 lines regarding #183: * move part of rrd_config.h out to a separate file, so that autoconf is not confused by the more complex expressions. * fix configure variable name to allow caching * another attemt at fixing solaris 2.8 compile so that it works out of the box. ------------------------------------------------------------------------ r1501 | oetiker | 2008-09-13 12:56:40 +0200 | 2 lines fix graph docs ------------------------------------------------------------------------ r1500 | oetiker | 2008-09-11 22:27:47 +0200 | 3 lines fixed processing of custom fonts René GARCIA ------------------------------------------------------------------------ r1499 | oetiker | 2008-09-11 09:01:30 +0200 | 2 lines no , on the last member in C ------------------------------------------------------------------------ r1488 | oetiker | 2008-09-07 09:19:03 +0200 | 5 lines The attached patch eliminates the many places where the rra_current pointer was maintained; the existing write/seek mechanisms update rrd_file->pos correctly. Also, the seek at the front of process_arg is not necessary; write_to_rras will now take care to correct the file position at every update cycle. This eliminates unnecessary rrd_seek() calls when MMAP is not enabled. -- kevin brintnall kbrint rufus.net ------------------------------------------------------------------------ r1484 | oetiker | 2008-09-06 16:24:43 +0200 | 1 line prepare for the release of rrdtool-1.3.2 ------------------------------------------------------------------------ r1483 | oetiker | 2008-09-06 13:46:41 +0200 | 2 lines we need rrd_format.h here for things to compile properly ------------------------------------------------------------------------ r1482 | oetiker | 2008-09-06 13:43:10 +0200 | 2 lines updated opensolaris build instructions ------------------------------------------------------------------------ r1481 | oetiker | 2008-09-06 12:43:07 +0200 | 4 lines only link build and use rrd_getopt if there is no getopt_long provided by by the os ... this should also fix problems for solaris 2.8 and other non getopt_long oses who were not taken care of anymore after the intriduction of librrd.sym ------------------------------------------------------------------------ r1480 | oetiker | 2008-09-06 12:39:29 +0200 | 1 line fix for multi update - data corruption bug as reported (and patched) in #178 by kevin brintnall ------------------------------------------------------------------------ r1479 | oetiker | 2008-09-06 12:32:57 +0200 | 2 lines make pod checker happy ------------------------------------------------------------------------ r1478 | oetiker | 2008-09-06 10:58:18 +0200 | 4 lines inline functions are not supported in c99 ... lets remove this. the compiler will foture it out if they are important. ------------------------------------------------------------------------ r1477 | oetiker | 2008-09-04 23:45:44 +0200 | 2 lines introduced a persistant fontmap ... this makes our pango rendering 6 times faster starting from the second graph we draw. ------------------------------------------------------------------------ r1476 | oetiker | 2008-09-01 01:42:02 +0200 | 2 lines now the results actually look as they did with the old code only things are FASTER ------------------------------------------------------------------------ r1475 | oetiker | 2008-09-01 00:53:49 +0200 | 3 lines Improve performance by keeping multiple pango font descriptions. For the first invocation we are about 20% times as fast for the second one about 4-5 times faster ... ------------------------------------------------------------------------ r1474 | oetiker | 2008-08-31 12:55:24 +0200 | 3 lines trying to improve the pango performance I added facilities to cache layout,context and more in between invocations of the text function, unfortunately it does not seem to help all that much. ------------------------------------------------------------------------ r1473 | oetiker | 2008-08-31 12:52:49 +0200 | 1 line use the standart info free function ------------------------------------------------------------------------ r1471 | oetiker | 2008-08-28 07:49:02 +0200 | 2 lines fix handling of error conditions in rrd_tool.c ... errno is not for detection, only for information AFTER detection. ------------------------------------------------------------------------ r1470 | oetiker | 2008-08-28 06:13:39 +0200 | 2 lines free the linebuffer if you don't return it! ------------------------------------------------------------------------ r1469 | oetiker | 2008-08-28 00:06:43 +0200 | 2 lines some more leak plugging ------------------------------------------------------------------------ r1468 | oetiker | 2008-08-27 23:48:50 +0200 | 2 lines drop the dummy surface prior to creating the real one ------------------------------------------------------------------------ r1467 | oetiker | 2008-08-27 23:46:01 +0200 | 2 lines added some freeing for error conditions ------------------------------------------------------------------------ r1466 | oetiker | 2008-08-27 23:25:43 +0200 | 2 lines fixed indentation ... gnu indent results are not realy beautifl. I might switch to uncrustify. ------------------------------------------------------------------------ r1465 | oetiker | 2008-08-27 23:25:04 +0200 | 2 lines Many rrd_create memory leaks patched ... based on valgrind analysis by Sven Engelhardt. Thanks! ------------------------------------------------------------------------ r1464 | oetiker | 2008-08-27 23:18:46 +0200 | 2 lines fix indentation ------------------------------------------------------------------------ r1463 | oetiker | 2008-08-27 08:52:33 +0200 | 3 lines added sparc-sun-solaris2.8 to list of broken isfinite implementations ------------------------------------------------------------------------ r1462 | oetiker | 2008-08-23 00:04:37 +0200 | 2 lines it is rb_rrd_infocall not rrd_infocall ... ------------------------------------------------------------------------ r1461 | oetiker | 2008-08-16 09:13:01 +0200 | 5 lines When in lazy mode, and the graph does not get regenerated, then there are no results from PRINT either. Fix for #163. ------------------------------------------------------------------------ r1460 | oetiker | 2008-08-16 09:11:46 +0200 | 1 line documentation updates ------------------------------------------------------------------------ r1459 | oetiker | 2008-08-12 08:48:37 +0200 | 2 lines make sure imageinfo only reports the filename and not the whole path. This is a regression from 1.2.x where this was implemented properly. ------------------------------------------------------------------------ r1457 | oetiker | 2008-08-04 22:35:45 +0200 | 12 lines The attached patch converts the given string to UTF-8 using glib's g_locale_to_utf8() function. This function expects the string to be encoded as specified by the locale settings, so this will fail if e.g. you're using some script which is encoded in some way incompatible to the locale settings. In this case, the patch falls back to use the original string as provided by the user as there is no way to tell how to interpret that string. So, in this case, you will still get the old behavior (that pango warning and messed up strings in the graph). However, imho this is perfectly fine since it should be up to the user to take care of a sane environment. -- Sebastian Harl ------------------------------------------------------------------------ r1456 | oetiker | 2008-07-29 16:42:08 +0200 | 2 lines newer perlapi docu suggests to use SV * instead of void for managed returns ------------------------------------------------------------------------ r1455 | oetiker | 2008-07-29 07:08:46 +0200 | 2 lines fixed some c++ style lines -- David Masterson ------------------------------------------------------------------------ r1454 | oetiker | 2008-07-28 09:23:15 +0200 | 2 lines spelling fix ------------------------------------------------------------------------ r1453 | oetiker | 2008-07-23 17:03:43 +0200 | 3 lines forward ported patch for rra cur_row randomization in rrd_restore from 1.2.28 to 1.3.1 ------------------------------------------------------------------------ r1447 | oetiker | 2008-07-23 15:02:26 +0200 | 1 line prepare for the release of rrdtool-1.3.1 ------------------------------------------------------------------------ r1446 | oetiker | 2008-07-23 14:45:52 +0200 | 3 lines added support for restoring rrdtool 1.0.x files ... removed support for old HW patched rrd format created by an early patched version of 1.0.x ------------------------------------------------------------------------ r1445 | oetiker | 2008-07-22 17:15:44 +0200 | 2 lines Fix for #167 - rrdcreate is arguably missing a check for 'step>=1' for RRAs with AVERAGE as a CF. I ran into this when specifying 0.6 as a step value, assuming floats were also accepted. Patch attached here. -- Bogdan ------------------------------------------------------------------------ r1444 | oetiker | 2008-07-01 08:35:54 +0200 | 1 line header-fix ------------------------------------------------------------------------ r1443 | oetiker | 2008-07-01 08:34:24 +0200 | 2 lines improve layout ------------------------------------------------------------------------ r1441 | oetiker | 2008-06-29 18:23:50 +0200 | 2 lines added rrd filename to illegal update message -- he at uninett.no ------------------------------------------------------------------------ r1440 | oetiker | 2008-06-29 16:08:51 +0200 | 2 lines rrdtool.spec fix by Anton Blanchard to build without php patch ------------------------------------------------------------------------ r1439 | oetiker | 2008-06-29 15:57:00 +0200 | 2 lines make sure image-with and image-height get returned even when rrdtool is called with --lazy as reported in bug #163 ------------------------------------------------------------------------ r1436 | oetiker | 2008-06-19 15:18:04 +0200 | 3 lines rrdtool.fetch is returning an extra, bogus, data point. Attached is a patch to correct this. - Gilad Raphaelli. Fix for #161 ------------------------------------------------------------------------ r1435 | oetiker | 2008-06-12 15:54:23 +0200 | 2 lines added note on %S ------------------------------------------------------------------------ r1434 | oetiker | 2008-06-11 17:49:24 +0200 | 2 lines made news the same as the announcement ------------------------------------------------------------------------ r1432 | oetiker | 2008-06-11 01:12:55 +0200 | 1 line prepare for the release of rrdtool-1.3.0 ------------------------------------------------------------------------ r1428 | oetiker | 2008-06-10 18:38:10 +0200 | 2 lines oops it is 1.3.0 not 1.4.0 yet ------------------------------------------------------------------------ r1427 | oetiker | 2008-06-10 18:37:39 +0200 | 2 lines preparing for 1.3 ------------------------------------------------------------------------ r1426 | oetiker | 2008-06-10 18:32:34 +0200 | 4 lines document the 'internal' functions in rrd.h but mark them as deprecated, so that anyone using them will see that they should not be doing it. ------------------------------------------------------------------------ r1425 | oetiker | 2008-06-10 13:16:40 +0200 | 2 lines flip check for row_count with setting for better code logic ------------------------------------------------------------------------ r1424 | oetiker | 2008-06-10 11:35:03 +0200 | 2 lines check row count in RRAs to be > 0 ... by Sebastian Harl ------------------------------------------------------------------------ r1423 | oetiker | 2008-06-10 09:18:02 +0200 | 2 lines for windows we must open files in O_BINARY mode ------------------------------------------------------------------------ r1421 | oetiker | 2008-06-09 18:10:39 +0200 | 6 lines do NOT export lowlevel functions, since I have no intention supporting these interfaces across versions. The symbols are in the library, so if someone WANTS to code to them, they are free todo so, but they will have to use our internal header files. ------------------------------------------------------------------------ r1420 | oetiker | 2008-06-09 17:49:54 +0200 | 2 lines add all lowlevel file access functions from the rrd_open interface to the export list. ------------------------------------------------------------------------ r1419 | oetiker | 2008-06-09 07:12:27 +0200 | 2 lines fixed indenting ------------------------------------------------------------------------ r1418 | oetiker | 2008-06-09 07:11:39 +0200 | 1 line update to rpm spec from Jarod Wilson ------------------------------------------------------------------------ r1417 | oetiker | 2008-06-08 23:15:32 +0200 | 2 lines export some more symbols from librrd ------------------------------------------------------------------------ r1416 | oetiker | 2008-06-08 23:06:35 +0200 | 2 lines windows should use a windows font ------------------------------------------------------------------------ r1415 | oetiker | 2008-06-08 23:05:50 +0200 | 2 lines drop special case windows font code ... it makes no sense with the new setup ------------------------------------------------------------------------ r1413 | oetiker | 2008-06-08 19:08:47 +0200 | 1 line prepare for the release of rrdtool-1.3rc9 ------------------------------------------------------------------------ r1412 | oetiker | 2008-06-08 19:08:11 +0200 | 2 lines fixed pango markup switch ... we still want the text to show even if no markup is enabled ------------------------------------------------------------------------ r1411 | oetiker | 2008-06-08 18:47:22 +0200 | 2 lines rrd_freemem has to be exported as well ... ------------------------------------------------------------------------ r1409 | oetiker | 2008-06-08 18:23:39 +0200 | 2 lines fix for ticket #121 jmpbuf on aix is a problem it seems, maybe this makes it better. ------------------------------------------------------------------------ r1407 | oetiker | 2008-06-08 17:59:20 +0200 | 1 line prepare for the release of rrdtool-1.3rc8 ------------------------------------------------------------------------ r1406 | oetiker | 2008-06-08 17:55:54 +0200 | 1 line prepare for the release of rrdtool-1.3rc8 ------------------------------------------------------------------------ r1405 | oetiker | 2008-06-08 17:55:09 +0200 | 62 lines From: Sebastian Harl sh tokkee.org Date: Sun, 8 Jun 2008 15:01:05 +0200 Subject: [PATCH] Cleanup the symbols exported by librrd. Up to know librrd exported a lot of symbols, most of which are to be regarded as private symbols. This somewhat pollutes the API as the symbols could, in theory, be used by external software and, more importantly, makes symbol based dependencies (as recently introduced in e.g. Debian) somewhat harder to implement. This patch does a somewhat large-scale cleanup of the exported symbols: * Introduced a librrd.sym file which contains all symbols that are to be exported. This file is then passed to libtool using the -export-symbols option which tells the linker to export the given symbols only (note: according to the libtool manual, this has no effect on some architectures - however, I assume that most architectures in use today do support it). librrd.sym contains all symbols originally defined in rrd.h sans LockRRD() (which has been moved to rrd_tool.h). The following functions have been added to rrd.h and the list of exported symbols (some of them have been renamed, see below): - rrd_info() - rrd_info_free() - rrd_info_print() - rrd_info_push() - rrd_lastupdate() - rrd_update_v() - rrd_strerror() * Prefixed all public functions and types with "rrd_" to avoid name clashes with other libraries. Also, while I was at it, I introduced typedefs for all custom types and prefixed the time names with "_t" to improve consistency: - enum info_type -> rrd_info_type_t - enum timetype -> rrd_timetype_t - union infoval -> rrd_infoval_t - struct info_t -> rrd_info_t - struct rrd_context -> rrd_context_t - struct rrd_time_value -> rrd_time_value_t - info_free() -> rrd_info_free() - info_free() -> rrd_info_free() - info_print() -> rrd_info_print() - info_push() -> rrd_info_push() - LockRRD() -> rrd_lock() (not public though) - parsetime() -> rrd_parsetime() (and: src/parsetime.c -> src/rrd_parsetime.c) - proc_start_end() -> rrd_proc_start_end() - set_to_DINF() -> rrd_set_to_DINF() - set_to_DNAN() -> rrd_set_to_DNAN() * Moved readfile() from rrd_open.c to rrd_cgi.c and declared it static. This function is used in rrd_cgi.c only. * rrd_lock() (f.k.a. LockRRD()) now accepts a rrd_file_t pointer instead of an integer to increase encapsulation. ------------------------------------------------------------------------ r1404 | oetiker | 2008-06-07 16:35:03 +0200 | 2 lines don't bother with don't need when there is neither madvise not fadvise calls ------------------------------------------------------------------------ r1403 | oetiker | 2008-06-06 18:38:25 +0200 | 2 lines fix for --font bug (font names wer not being picked up at all). Bug reported by Dmitry B. Bigunayk ------------------------------------------------------------------------ r1402 | oetiker | 2008-06-05 22:56:24 +0200 | 2 lines added pango markup name ------------------------------------------------------------------------ r1401 | oetiker | 2008-06-05 21:59:18 +0200 | 2 lines fixed indenting ------------------------------------------------------------------------ r1400 | oetiker | 2008-06-05 21:59:01 +0200 | 1 line disable pango markup by default. added --pango-markup option for explicit enablement. ------------------------------------------------------------------------ r1399 | oetiker | 2008-06-05 21:58:51 +0200 | 1 line disable pango markup by default. added --pango-markup option for explicit enablement. ------------------------------------------------------------------------ r1398 | oetiker | 2008-06-05 21:57:25 +0200 | 1 line fixed indenting ------------------------------------------------------------------------ r1397 | oetiker | 2008-06-04 18:18:42 +0200 | 2 lines enable new building for newer tcl versions ------------------------------------------------------------------------ r1396 | oetiker | 2008-06-02 08:46:57 +0200 | 2 lines updated documentation of library versioning system ------------------------------------------------------------------------ r1395 | oetiker | 2008-06-02 08:43:43 +0200 | 3 lines Had to reset the interface version number due to the change in the rrd_context implementation. ------------------------------------------------------------------------ r1394 | oetiker | 2008-06-02 08:21:42 +0200 | 2 lines use DNAN instead of NAN since there seem to be issues with gcc on solaris and the NAN experession (and we use DNAN everywhere else anyway). ------------------------------------------------------------------------ r1393 | oetiker | 2008-06-01 23:12:26 +0200 | 2 lines integrated gentoo bits and spell fixes by Peter Volkov ------------------------------------------------------------------------ r1392 | oetiker | 2008-06-01 22:44:01 +0200 | 3 lines Patch for rrd_context system by sebastian harl, making it simpler and less error prone. Removing one buffer overflow in the process. ------------------------------------------------------------------------ r1389 | oetiker | 2008-05-31 17:58:45 +0200 | 1 line prepare for the release of rrdtool-1.3rc7 ------------------------------------------------------------------------ r1388 | oetiker | 2008-05-31 17:57:39 +0200 | 1 line prepare for the release of rrdtool-1.3rc7 ------------------------------------------------------------------------ r1387 | oetiker | 2008-05-31 17:56:37 +0200 | 2 lines added pkgconfig build instructions ------------------------------------------------------------------------ r1386 | oetiker | 2008-05-31 14:28:11 +0200 | 2 lines make sure vdef calcs only calc as far as originally requested. ------------------------------------------------------------------------ r1385 | oetiker | 2008-05-28 00:24:32 +0200 | 3 lines fixed maxlength for sprintf_alloc 50 was not enough ... 1024+strlen(fmt) is much better. This makes imginfo work even when there are long image paths. ------------------------------------------------------------------------ r1382 | oetiker | 2008-05-26 13:45:15 +0200 | 2 lines updated todo list ------------------------------------------------------------------------ r1380 | oetiker | 2008-05-26 10:56:58 +0200 | 1 line prepare for the release of rrdtool-1.3rc6 ------------------------------------------------------------------------ r1379 | oetiker | 2008-05-26 09:46:35 +0200 | 2 lines added notes about new grid lines ------------------------------------------------------------------------ r1378 | oetiker | 2008-05-26 09:45:19 +0200 | 5 lines * make y-axis labels NOT run into each other (classic and alt) * space horizontal grid lines and lables a bit further apart for tight layouts. * let horizontal lines and y axis labels run a little closer for classic grid ------------------------------------------------------------------------ r1377 | oetiker | 2008-05-25 23:17:30 +0200 | 2 lines fixed calculation of space at the bottom of the graph ------------------------------------------------------------------------ r1375 | oetiker | 2008-05-25 22:02:23 +0200 | 1 line prepare for the release of rrdtool-1.3rc5 ------------------------------------------------------------------------ r1374 | oetiker | 2008-05-25 22:00:45 +0200 | 2 lines fix update compatibility with rrd-1.0.x files ------------------------------------------------------------------------ r1373 | oetiker | 2008-05-21 00:22:27 +0200 | 2 lines droped rrd_extra.h from the distro with the python update ... must remove it from the Makefile too ------------------------------------------------------------------------ r1372 | oetiker | 2008-05-19 23:14:36 +0200 | 5 lines fixed python bindings: * info call isnnow aequivalent to rrdtool info (the same as in the perl and ruby bindings) * graphv and updatev calls added ------------------------------------------------------------------------ r1371 | oetiker | 2008-05-19 00:11:56 +0200 | 3 lines * document tabwidth option * fix \t -> tab expansion ------------------------------------------------------------------------ r1370 | oetiker | 2008-05-18 17:39:41 +0200 | 2 lines ytr boundery checking ... this is from the GD days ... we don't need this anymore ------------------------------------------------------------------------ r1369 | oetiker | 2008-05-18 17:11:52 +0200 | 2 lines fix indenting ------------------------------------------------------------------------ r1368 | oetiker | 2008-05-18 17:11:42 +0200 | 2 lines pdf, svg and ps formats were not being created ------------------------------------------------------------------------ r1366 | oetiker | 2008-05-18 15:06:44 +0200 | 1 line prepare for the release of rrdtool-1.3rc4 ------------------------------------------------------------------------ r1365 | oetiker | 2008-05-18 15:04:36 +0200 | 2 lines fix year ------------------------------------------------------------------------ r1364 | oetiker | 2008-05-18 14:53:01 +0200 | 2 lines PRINT returns from rrd_graph did not work due to an off by one error ------------------------------------------------------------------------ r1363 | oetiker | 2008-05-18 10:42:18 +0200 | 2 lines updated build instructions to latest libraries ------------------------------------------------------------------------ r1362 | oetiker | 2008-05-18 08:37:43 +0200 | 2 lines improved handling of libintl issues ------------------------------------------------------------------------ r1360 | oetiker | 2008-05-16 14:29:42 +0200 | 1 line prepare for the release of rrdtool-1.3rc3 ------------------------------------------------------------------------ r1359 | oetiker | 2008-05-16 14:28:37 +0200 | 2 lines added --no-header to documentation and fixed implementation ------------------------------------------------------------------------ r1358 | oetiker | 2008-05-16 14:20:05 +0200 | 2 lines fixed dump_opt implementation ------------------------------------------------------------------------ r1357 | oetiker | 2008-05-16 14:18:05 +0200 | 2 lines added no-header option to rrd_dump ------------------------------------------------------------------------ r1356 | oetiker | 2008-05-16 14:07:58 +0200 | 2 lines fix handling of values <= zero for logarithmic display ------------------------------------------------------------------------ r1354 | oetiker | 2008-05-16 00:39:55 +0200 | 1 line prepare for the release of rrdtool-1.3rc2 ------------------------------------------------------------------------ r1353 | oetiker | 2008-05-16 00:39:09 +0200 | 1 line prepare for the release of rrdtool-1.3rc2 ------------------------------------------------------------------------ r1352 | oetiker | 2008-05-16 00:37:26 +0200 | 2 lines fix portability to mac osx and freebsd -- tobi ------------------------------------------------------------------------ r1351 | oetiker | 2008-05-13 01:10:59 +0200 | 2 lines add solaris pod2man location last ------------------------------------------------------------------------ r1350 | oetiker | 2008-05-13 01:03:04 +0200 | 2 lines some changes to make things work on opensolaris 2008.05 ------------------------------------------------------------------------ r1349 | oetiker | 2008-05-13 00:42:25 +0200 | 2 lines for some reason tclrrd.c does not get picked up using $< at least on opensolaris ------------------------------------------------------------------------ r1348 | oetiker | 2008-05-13 00:23:19 +0200 | 3 lines allow rrd_rpncalc to build without cairo dependencies ... fix warnings ------------------------------------------------------------------------ r1346 | oetiker | 2008-05-12 23:14:12 +0200 | 2 lines add intltool*in files to dist ------------------------------------------------------------------------ r1345 | oetiker | 2008-05-12 18:15:25 +0200 | 2 lines fixed indenting ------------------------------------------------------------------------ r1344 | oetiker | 2008-05-12 18:15:02 +0200 | 2 lines removed c++ comments // ------------------------------------------------------------------------ r1343 | oetiker | 2008-05-12 18:09:02 +0200 | 2 lines added notes about rewrites in this release ------------------------------------------------------------------------ r1342 | oetiker | 2008-05-12 18:08:40 +0200 | 2 lines improved handling of library versioning ------------------------------------------------------------------------ r1341 | oetiker | 2008-05-12 17:35:40 +0200 | 2 lines should not leave debug enabled ------------------------------------------------------------------------ r1340 | oetiker | 2008-05-12 17:33:31 +0200 | 10 lines Fixed handling of unknown data at PDP build time. There was a long standing (even documented) missfeature in rrdtool which caused uknown-data to be accepted as long as it was less than the mrhb. This was never the intended behaviour and had interesting side effects (http://oss.oetiker.ch/rrdtool-trac/ticket/125): If you have a 60 Second step with 59s unknown data and 1 second of known data the whole become known data as long as the mrhb was > step. The intended behaviour was that a step should become unknown as soon as it contains more than 50% of unknown data. The patch fixes both the documentation and the code. ------------------------------------------------------------------------ r1339 | oetiker | 2008-05-12 15:22:26 +0200 | 2 lines the last_ds value fuer unknown is 'U' and not 'UNKN' ------------------------------------------------------------------------ r1338 | oetiker | 2008-05-04 20:59:20 +0200 | 2 lines fix for sun sudio 11 incompatibility http://www.netbsd.org/cgi-bin/query-pr-single.pl?number=38555 ------------------------------------------------------------------------ r1335 | oetiker | 2008-05-02 01:23:35 +0200 | 1 line indenting fixes ------------------------------------------------------------------------ r1334 | oetiker | 2008-05-02 01:23:25 +0200 | 1 line make sure properties are shown via graph_v even in --graph-only mode ------------------------------------------------------------------------ r1333 | oetiker | 2008-05-01 08:56:32 +0200 | 2 lines Make the MakeMakefile script a bit more robust ------------------------------------------------------------------------ r1332 | oetiker | 2008-04-22 23:40:45 +0200 | 2 lines fix for #148 ------------------------------------------------------------------------ r1331 | oetiker | 2008-04-21 00:46:39 +0200 | 7 lines Introduced a alternated interface to rrd_graph using rrd_info style return values: rrd_graph_v The graph_v interface returnes additional information about the graph including the location of the graphing area within the image. When called with '-' as filename it will even return the image data. The new interface is supported in the rrdtool command line, RRDs perl and ruby bindings. ------------------------------------------------------------------------ r1329 | oetiker | 2008-04-17 07:35:16 +0200 | 2 lines indent fix ------------------------------------------------------------------------ r1328 | oetiker | 2008-04-17 07:35:01 +0200 | 2 lines do not reset the last_ds store unless the data source type is actually changed. ------------------------------------------------------------------------ r1327 | oetiker | 2008-04-16 17:50:59 +0200 | 2 lines fixed 2. x-grid example ... since the lable is valid for the whole day, it must be 86400 wide. --tobi ------------------------------------------------------------------------ r1324 | oetiker | 2008-04-12 11:29:10 +0200 | 2 lines alter order of header inclusion for FreeBSD 4.8 compatibility ------------------------------------------------------------------------ r1323 | oetiker | 2008-04-12 11:21:31 +0200 | 3 lines do not mess with dst status after running localtime, the system gets confused otherwhise when it comes to times during the switch over periode as DST changes to non DST. -- tobi ------------------------------------------------------------------------ r1322 | oetiker | 2008-03-26 00:02:32 +0100 | 2 lines added missing " escapes to xml header in rrd_tool.c ------------------------------------------------------------------------ r1321 | oetiker | 2008-03-25 23:59:49 +0100 | 2 lines added Florian octo Forster to contrib list ------------------------------------------------------------------------ r1320 | oetiker | 2008-03-25 23:59:30 +0100 | 2 lines added missing =back for propper pod syntax ------------------------------------------------------------------------ r1319 | oetiker | 2008-03-25 23:58:16 +0100 | 2 lines fix indent ------------------------------------------------------------------------ r1318 | oetiker | 2008-03-25 23:58:08 +0100 | 2 lines fix broken header comment section -- Florian octo Forster ------------------------------------------------------------------------ r1317 | oetiker | 2008-03-25 23:57:42 +0100 | 2 lines cast size_t to int to avoid warning -- Florian octo Forster ------------------------------------------------------------------------ r1316 | oetiker | 2008-03-25 23:56:41 +0100 | 2 lines do not declare DEBUG functions unless compiled in DEBUG mode -- Florian octo Forster ------------------------------------------------------------------------ r1315 | oetiker | 2008-03-25 23:55:56 +0100 | 2 lines make helptexts read only memory ------------------------------------------------------------------------ r1314 | oetiker | 2008-03-25 23:55:13 +0100 | 2 lines stop complaints about unused variables if not running DEBUG ------------------------------------------------------------------------ r1313 | oetiker | 2008-03-25 23:49:47 +0100 | 5 lines Added check for intltool version. Use Makefile for cleanup. -- Florian octo Forster ------------------------------------------------------------------------ r1311 | oetiker | 2008-03-23 14:43:22 +0100 | 2 lines added pointer to florian forsters dtd to the rrd dumpt ------------------------------------------------------------------------ r1310 | oetiker | 2008-03-20 00:36:26 +0100 | 2 lines fix pkgconfig file name ------------------------------------------------------------------------ r1306 | oetiker | 2008-03-15 11:39:48 +0100 | 3 lines reindent aded id keyword substitution ------------------------------------------------------------------------ r1305 | oetiker | 2008-03-15 11:32:10 +0100 | 2 lines reat nan as FALSE in an IF CDEF ------------------------------------------------------------------------ r1303 | oetiker | 2008-03-07 09:57:01 +0100 | 10 lines a nan-safe add operator (ADDNAN) into rrd. I used it to add several incomplete graphs. NaN + NaN => NaN x + NaN => x NaN + y => y x + y => x + y -- Timo Stripf tstripf gmx.de ------------------------------------------------------------------------ r1301 | oetiker | 2008-03-02 16:38:05 +0100 | 2 lines Fix error "cairo_restore without matching cairo_save" when zero-time point is out of graph ------------------------------------------------------------------------ r1300 | oetiker | 2008-02-25 08:11:00 +0100 | 2 lines Ignore RRA-end when comparing. coverage. ------------------------------------------------------------------------ r1298 | oetiker | 2008-02-24 15:27:34 +0100 | 2 lines fix indents gone out of kileter by patching ... ------------------------------------------------------------------------ r1297 | oetiker | 2008-02-24 15:26:46 +0100 | 2 lines get rid of a few type cast warnings ------------------------------------------------------------------------ r1296 | oetiker | 2008-02-24 15:26:16 +0100 | 2 lines fix one segfault while trying to restore an invalid xml file. -- tobi ------------------------------------------------------------------------ r1294 | oetiker | 2008-02-21 08:20:49 +0100 | 2 lines make ruby obey destdir ------------------------------------------------------------------------ r1293 | oetiker | 2008-02-21 08:18:58 +0100 | 2 lines initial ------------------------------------------------------------------------ r1292 | oetiker | 2008-02-21 08:12:27 +0100 | 2 lines make inbternationalized version actually build ------------------------------------------------------------------------ r1291 | oetiker | 2008-02-21 07:59:19 +0100 | 2 lines added missing bits from internationalization ------------------------------------------------------------------------ r1290 | oetiker | 2008-02-19 13:56:44 +0100 | 12 lines Generate a random cur_row for each RRA during create/restore operations. This effectively randomizes the block crossings among RRDs created around the same time. Previously, RRDs that were created/restored en masse would cross block boundaries simultaneously, which is sub-optimal. Also, this patch enables the user to see the RRA's cur_row pointer via rrdinfo. This was useful during debugging. -- kevin brintnall kbrint qwest.net ------------------------------------------------------------------------ r1289 | oetiker | 2008-02-18 16:04:42 +0100 | 2 lines Added I18N support for messages printed by rrd_tool.c -- http://oss.oetiker.ch/rrdtool-trac/ticket/144 ------------------------------------------------------------------------ r1285 | oetiker | 2008-02-15 09:49:07 +0100 | 2 lines check for pdf as well ------------------------------------------------------------------------ r1284 | oetiker | 2008-02-15 09:44:04 +0100 | 2 lines check for cairo-ps and cairo-svg explicitly ------------------------------------------------------------------------ r1280 | oetiker | 2008-02-06 01:12:27 +0100 | 2 lines ready for beta 4 ------------------------------------------------------------------------ r1279 | oetiker | 2008-02-06 01:10:33 +0100 | 4 lines * replaced strtok with strtok_r for thread safety * fixed im initialization broken after the introduction of dashes ... ------------------------------------------------------------------------ r1278 | oetiker | 2008-02-04 23:34:56 +0100 | 2 lines in rrd_create we do not use mmaping and thus need to free rrd struct members in any case ... ------------------------------------------------------------------------ r1277 | oetiker | 2008-02-04 23:16:14 +0100 | 2 lines this should help with the memory leak ------------------------------------------------------------------------ r1276 | oetiker | 2008-02-04 19:18:07 +0100 | 2 lines added zpl 2.1 to floss exception ------------------------------------------------------------------------ r1274 | oetiker | 2008-02-01 07:52:56 +0100 | 2 lines fixed parsing of fontnames with embeded spaces ------------------------------------------------------------------------ r1272 | oetiker | 2008-01-14 18:11:50 +0100 | 2 lines remove c++ warnings flag .. this is not c++ after all ------------------------------------------------------------------------ r1271 | oetiker | 2008-01-14 17:52:49 +0100 | 2 lines fixed indenting ------------------------------------------------------------------------ r1270 | oetiker | 2008-01-14 17:51:42 +0100 | 2 lines fix indents ------------------------------------------------------------------------ r1269 | oetiker | 2008-01-14 17:50:51 +0100 | 5 lines * fixed madvise/fadvise detection * fixed many compiler warnings -- Peter Breitenlohner peb mppmu.mpg.de ------------------------------------------------------------------------ r1266 | oetiker | 2008-01-12 19:37:10 +0100 | 1 line weekday and time are too tight ------------------------------------------------------------------------ r1264 | oetiker | 2008-01-04 23:53:46 +0100 | 2 lines fontmap resolution was not matching scaled_font resolution ... this killed layouting ... much better now ------------------------------------------------------------------------ r1263 | oetiker | 2008-01-03 22:30:59 +0100 | 2 lines get rid of DASHED, only use dashes syntax. less redundancy -- thomas.gutzler gmail.com ------------------------------------------------------------------------ r1262 | oetiker | 2008-01-02 23:11:26 +0100 | 2 lines support for dashed lines in graphs ------------------------------------------------------------------------ r1261 | oetiker | 2008-01-02 23:06:24 +0100 | 2 lines fix indenting ------------------------------------------------------------------------ r1260 | oetiker | 2008-01-02 23:06:16 +0100 | 2 lines ignore errors with txt doc building ------------------------------------------------------------------------ r1255 | oetiker | 2007-12-12 00:41:21 +0100 | 2 lines building the docs depends on the presence of pod2man ... which does not seem to be installed necessarily ------------------------------------------------------------------------ r1253 | oetiker | 2007-12-08 23:15:22 +0100 | 2 lines added shared library locations into the docs ------------------------------------------------------------------------ r1252 | oetiker | 2007-12-08 22:35:51 +0100 | 2 lines include stdlib since putenv is in there on solaris ------------------------------------------------------------------------ r1251 | oetiker | 2007-12-08 17:57:04 +0100 | 6 lines fixing solaris portability * isinf and isnan definitions fixed * check if rt must be linked * ordering must be read write in getopt * updated build instructions ------------------------------------------------------------------------ r1247 | oetiker | 2007-11-26 21:12:11 +0100 | 2 lines fix libdir for python build ------------------------------------------------------------------------ r1244 | oetiker | 2007-11-21 11:51:48 +0100 | 2 lines fixed NetBSD spelling ------------------------------------------------------------------------ r1241 | oetiker | 2007-11-21 07:56:19 +0100 | 2 lines define bsd source to get access to chroot ------------------------------------------------------------------------ r1240 | oetiker | 2007-11-21 07:47:17 +0100 | 2 lines make perlbuild more portable ------------------------------------------------------------------------ r1238 | oetiker | 2007-11-20 23:22:31 +0100 | 2 lines enable madvise defines in netbsd ------------------------------------------------------------------------ r1234 | oetiker | 2007-11-20 01:08:20 +0100 | 3 lines remove last traces of rrd_nan_inf.h fix handling of min==max where min < 0 ------------------------------------------------------------------------ r1228 | oetiker | 2007-11-15 15:39:56 +0100 | 2 lines re-indented files that have gone out of indent-style over the last few weeks ------------------------------------------------------------------------ r1227 | oetiker | 2007-11-15 15:36:39 +0100 | 2 lines integrate rrd_nan_inf header ------------------------------------------------------------------------ r1226 | oetiker | 2007-11-15 15:35:15 +0100 | 2 lines Improve layout of documentation. -- Bernhard Fischer ------------------------------------------------------------------------ r1225 | oetiker | 2007-11-15 15:34:14 +0100 | 5 lines * rrd_update(): Unify error path. * parse_template(): Likewise. * allocate_data_structures(): Make error strings use consistent punctuation. -- bernhard fischer ------------------------------------------------------------------------ r1224 | oetiker | 2007-11-15 15:32:37 +0100 | 2 lines make sure this compiles with aix ------------------------------------------------------------------------ r1213 | oetiker | 2007-11-08 11:16:27 +0100 | 2 lines fixed start end time in ruby fetch and added step size -- Mike Perham mperham gmail ------------------------------------------------------------------------ r1211 | oetiker | 2007-11-06 22:27:52 +0100 | 2 lines updated build instructions ------------------------------------------------------------------------ r1209 | oetiker | 2007-11-04 22:56:19 +0100 | 2 lines handle errors properly ... in RRDp read call ------------------------------------------------------------------------ r1204 | oetiker | 2007-09-16 17:35:11 +0200 | 2 lines no reason to die when posixfadvise is not sucessful ------------------------------------------------------------------------ r1203 | oetiker | 2007-09-11 08:28:15 +0200 | 4 lines Allow to the the smoothing window size other thatn the default 5%. -- Evan Miller emiller imvu.com ------------------------------------------------------------------------ r1202 | oetiker | 2007-09-11 08:16:52 +0200 | 2 lines added msync before unmap ------------------------------------------------------------------------ r1201 | oetiker | 2007-09-08 07:23:23 +0200 | 3 lines added STDEV aggregation function for VDEF. -- Patrick J Cherry patrick bytemark.co.uk ------------------------------------------------------------------------ r1200 | oetiker | 2007-09-08 00:53:41 +0200 | 2 lines a few leanups -- Bernhard Fischer ------------------------------------------------------------------------ r1199 | oetiker | 2007-09-06 11:06:04 +0200 | 2 lines added missing ; ------------------------------------------------------------------------ r1198 | oetiker | 2007-09-06 10:42:22 +0200 | 3 lines * mimic write() and read() even better -- Bernhard Fischer ------------------------------------------------------------------------ r1197 | oetiker | 2007-09-06 10:40:36 +0200 | 5 lines rrd_open.c (rrd_read): Mimicing read() behaviour for EOF case and NULL buffer case. -- Bernhard Fischer ------------------------------------------------------------------------ r1196 | oetiker | 2007-09-02 18:55:01 +0200 | 3 lines fixed speling of horizon ------------------------------------------------------------------------ r1194 | oetiker | 2007-08-29 08:43:16 +0200 | 2 lines a better performance tester ------------------------------------------------------------------------ r1193 | oetiker | 2007-08-16 09:31:53 +0200 | 2 lines added holt winters example back into the examples documentation ------------------------------------------------------------------------ r1192 | oetiker | 2007-08-14 23:59:31 +0200 | 18 lines Fix for HoltWinters phase-shift bug described below. When one or more primary data point times were missed, the SEASONAL and DEVSEASONAL archives were marked as being up-to-date, so that they would not be written to. It was correct not to write to these archives, but the code failed to advance the pointers within the SEASONAL and DEVSEASONAL archives so that future updates would go to the correct location in the archives. Rather than mark these archives as up-to-date (by setting rra_step_cnt[rra_idx] = 0), my patch allocates a new "skip_update" array that is set to 1 for SEASONAL and DEVSEASONAL archives that have missed one or more primary data points. When an RRA is written to, the cur_row pointer advancement happens for all archives, but the skip_update array is checked just before actually writing out the changes. Please give it a whirl! -- Evan Miller emiller imvu.com ------------------------------------------------------------------------ r1191 | oetiker | 2007-08-13 22:06:10 +0200 | 3 lines Refactored rrd_update code in preparation of finding the HW update problem -- Evan Miller ------------------------------------------------------------------------ r1190 | oetiker | 2007-08-07 17:02:08 +0200 | 2 lines updated build instructions for 1.3 ------------------------------------------------------------------------ r1185 | oetiker | 2007-08-03 21:46:17 +0200 | 1 line prepare for the release of rrdtool-1.2.99907080300 ------------------------------------------------------------------------ r1184 | oetiker | 2007-08-03 21:45:01 +0200 | 2 lines remove some things we do not need anymore in 1.3 ------------------------------------------------------------------------ r1183 | oetiker | 2007-08-03 21:43:47 +0200 | 1 line prepare for the release of rrdtool-1.2.99907080300 ------------------------------------------------------------------------ r1182 | oetiker | 2007-08-03 21:43:21 +0200 | 2 lines prepare for 1.3 beta 1 ------------------------------------------------------------------------ r1181 | oetiker | 2007-08-03 21:26:41 +0200 | 2 lines continue with the revision numbering ------------------------------------------------------------------------ r1180 | oetiker | 2007-08-03 21:17:13 +0200 | 2 lines added notes on pango inline formatting ------------------------------------------------------------------------ r1178 | oetiker | 2007-08-01 17:29:09 +0200 | 2 lines only release cairo stuff if we ever initialized it. ------------------------------------------------------------------------ r1177 | oetiker | 2007-08-01 17:28:03 +0200 | 2 lines fix for 64bit portability problem ... unsigned long is not int ------------------------------------------------------------------------ r1175 | oetiker | 2007-07-31 06:53:04 +0200 | 2 lines reindented ------------------------------------------------------------------------ r1173 | oetiker | 2007-07-25 22:38:37 +0200 | 2 lines fixed indenting ------------------------------------------------------------------------ r1172 | oetiker | 2007-07-25 22:38:26 +0200 | 2 lines make sure we check input even when the previous update was a 'U' ... and some indenting fixes ------------------------------------------------------------------------ r1169 | oetiker | 2007-07-22 17:51:39 +0200 | 2 lines new consolidation functions must be added last or this will break binary compatibility ------------------------------------------------------------------------ r1168 | oetiker | 2007-07-22 17:50:21 +0200 | 2 lines don't force data out ... let cache management do this ------------------------------------------------------------------------ r1167 | oetiker | 2007-07-22 11:47:53 +0200 | 1 line Only create version 4 rrd files if the new holtwinters MHW.. CF is used. ------------------------------------------------------------------------ r1166 | oetiker | 2007-07-21 21:55:00 +0200 | 5 lines * remove some experimental code from rrd_open * rrd_dontneed will release all data except for header blocks and RRA hot blocks which are going to be updated withing 10 minutes. ------------------------------------------------------------------------ r1165 | oetiker | 2007-07-21 21:53:52 +0200 | 3 lines for now, do not release fetched data ... in the future we may call rrd_dontneed and let the users decied with an option if they want to keep the data in cache ... ------------------------------------------------------------------------ r1164 | oetiker | 2007-07-21 21:52:17 +0200 | 2 lines the fonts are now taken from the system. we do not distribute them anympore ------------------------------------------------------------------------ r1163 | oetiker | 2007-07-21 21:37:50 +0200 | 2 lines not on update rrd version ------------------------------------------------------------------------ r1162 | oetiker | 2007-07-21 00:20:18 +0200 | 2 lines we have a new holtwinders aggregation fucntion, so we have a new version ... ------------------------------------------------------------------------ r1161 | oetiker | 2007-07-21 00:19:53 +0200 | 2 lines check for fadvise all the time ------------------------------------------------------------------------ r1160 | oetiker | 2007-07-21 00:16:36 +0200 | 2 lines better docs on the aggregate functions ------------------------------------------------------------------------ r1159 | oetiker | 2007-07-19 00:02:18 +0200 | 2 lines added note on locale ------------------------------------------------------------------------ r1158 | oetiker | 2007-07-18 23:59:10 +0200 | 2 lines some more piechart removed ------------------------------------------------------------------------ r1157 | oetiker | 2007-07-18 23:47:26 +0200 | 1 line make sure all ascii to float parsing uses LC_NUMERIC = C so that we do not stuble on locales confusing the , with a . ------------------------------------------------------------------------ r1156 | oetiker | 2007-07-18 02:30:49 +0200 | 9 lines * rrd_open: rrd_close does not purge file from cache * rrd_open: new function rrd_dontneed for purging un-needed pages from core * rrd_open: in linux at least only fadivse DONTNEED has the power to purge pages from cache, so letst call madvise as well as fadvise * rrd_create: uses open/write/close now, flushes file to disk and keeps only hot pages in core * rrd_update: keeps only hot pages in core * configure enables FADVISE even when mmap is in use ------------------------------------------------------------------------ r1155 | oetiker | 2007-07-17 23:46:00 +0200 | 2 lines added comment on float cookie ------------------------------------------------------------------------ r1154 | oetiker | 2007-07-16 08:37:07 +0200 | 2 lines fix for fetch in ruby bindings and support for info added. ------------------------------------------------------------------------ r1149 | oetiker | 2007-07-12 22:33:47 +0200 | 2 lines improve wording ------------------------------------------------------------------------ r1148 | oetiker | 2007-07-12 22:33:37 +0200 | 2 lines fix tcl build settup ------------------------------------------------------------------------ r1146 | oetiker | 2007-07-12 01:08:47 +0200 | 2 lines usr rrd_free for freeing the rrd structs ... vito caputo ------------------------------------------------------------------------ r1141 | oetiker | 2007-07-05 17:19:27 +0200 | 2 lines prevent potential segfaults on boxes where time_t is not long ------------------------------------------------------------------------ r1134 | oetiker | 2007-06-18 20:25:35 +0200 | 2 lines fix indentation ------------------------------------------------------------------------ r1133 | oetiker | 2007-06-18 18:05:07 +0200 | 8 lines Bernhard Fischer: - move several static struct option out of loops and makes them non-static - moves some functions from old-style definitions into new-style definitions ------------------------------------------------------------------------ r1132 | oetiker | 2007-06-17 01:20:23 +0200 | 2 lines Bernhard is not Benrard ------------------------------------------------------------------------ r1130 | oetiker | 2007-06-15 23:49:52 +0200 | 1 line notes on evans contribs ------------------------------------------------------------------------ r1129 | oetiker | 2007-06-15 23:41:54 +0200 | 3 lines rrdtool.spec from Jarod "redhat" Wilson ------------------------------------------------------------------------ r1127 | oetiker | 2007-06-15 10:09:13 +0200 | 2 lines updated contributors ------------------------------------------------------------------------ r1126 | oetiker | 2007-06-15 10:01:04 +0200 | 2 lines added note on TEXTALIGN ------------------------------------------------------------------------ r1125 | oetiker | 2007-06-15 09:59:01 +0200 | 46 lines There are two popular variants of the Holt-Winters forecasting method; RRDtool supports the "additive" method, which means that seasonal variation is simply added to the baseline. For our application, it would be more appropriate to use the "multiplicative" Holt-Winters method, where seasonal variation is a coefficient multiplied by the baseline. Quick example to illustrate the difference: if the average doubles season-over-season, the additive method would predict the delta between min and max to be constant, whereas the multiplicative method would predict the delta to double as well. Attached is a patch against trunk to support the multiplicative method. I've done this with a new consolidation function, MHWPREDICT, which is essentially interchangeable with HWPREDICT. There is a noticeable improvement in prediction deviations for certain types of functions; the attachments show HWPREDICT and MHWPREDICT predictions for a function with an x*sin(x) component. Because HWPREDICT and MHWPREDICT differ only in their equations, I've factored out their math into rrd_hw_math.c. The appropriate smoothing functions are passed to the update functions in a container of function pointers, which are called where appropriate. Thus the additive and multiplicative methods use the same update functions, and the right equations are evaluated without having flag checks everywhere. This approach, I think, makes the algorithms quite clear, with minimal duplicate code. I have moved update_hwpredict, update_seasonal, update_devpredict, update_devseasonal, and update_failures into a separate file, rrd_hw_update.c, with some slight refactoring related to rrd_hw_math.c. I ran some regression tests against trunk to make sure I didn't break anything with the existing HWPREDICT code. MHWPREDICT uses the same deviation smoothing and failure detection algorithms as HWPREDICT. Some helpful references on the multiplicative Holt-Winters method: http://www.it.iitb.ac.in/~praj/acads/seminar/04329008_ExponentialSmoothing.pdf (a student's quick overview of additive vs. multiplicative HW) http://ideas.repec.org/p/msh/ebswps/1999-1.html (paper on variations to the multiplicative Holt-Winters, including variance calculations; FYI, my implementation uses "Model 1") My employer and the owner of this patch (IMVU, Inc.) is happy to license it under the same terms as RRDtool, i.e. give it to the project. -- Evan Miller emiller imvu.com ------------------------------------------------------------------------ r1124 | oetiker | 2007-06-15 08:43:36 +0200 | 2 lines oops missed to update the in memory output interface here ------------------------------------------------------------------------ r1123 | oetiker | 2007-06-15 08:28:47 +0200 | 3 lines added interface to get rendered image via a pointer to allocated memory ... Evan Miller emmiller gmail.com ------------------------------------------------------------------------ r1122 | oetiker | 2007-06-15 08:12:55 +0200 | 2 lines improve fontnaming ------------------------------------------------------------------------ r1121 | oetiker | 2007-06-14 22:30:13 +0200 | 2 lines fixed indentation ------------------------------------------------------------------------ r1120 | oetiker | 2007-06-14 22:28:44 +0200 | 1 line added TEXTALIGN command to change the default text alignment ------------------------------------------------------------------------ r1119 | oetiker | 2007-06-13 18:53:32 +0200 | 2 lines fixed tcl includes ------------------------------------------------------------------------ r1118 | oetiker | 2007-06-13 18:51:06 +0200 | 2 lines be more helpful when raising rb_eTypeError in string_arr string_arr_new(VALUE rb_strings). This patch indicates which index is in error as well as the type you are erroneously passing. -- anonymous ------------------------------------------------------------------------ r1116 | oetiker | 2007-06-12 22:36:50 +0200 | 6 lines Bernhard Fischer: - fix nulling the updvals - cosmetic change to move the option struct out of the loop. - need to operate on a copy of optarg ------------------------------------------------------------------------ r1115 | oetiker | 2007-06-11 21:45:42 +0200 | 2 lines added TRENDNAN function -- Timo Stripf ------------------------------------------------------------------------ r1114 | oetiker | 2007-06-11 21:45:16 +0200 | 1 line allow output to stdout -- Timo Stripf ------------------------------------------------------------------------ r1113 | oetiker | 2007-06-11 18:49:01 +0200 | 2 lines added more news ------------------------------------------------------------------------ r1112 | oetiker | 2007-06-11 18:34:51 +0200 | 2 lines fixed indentation, added documentation note on gridfitting. ------------------------------------------------------------------------ r1111 | oetiker | 2007-06-11 18:33:59 +0200 | 2 lines fix arrow locations ------------------------------------------------------------------------ r1110 | oetiker | 2007-06-11 18:20:20 +0200 | 2 lines disable gridfitting for vector formats ------------------------------------------------------------------------ r1109 | oetiker | 2007-06-11 18:14:40 +0200 | 2 lines make pdf output work ------------------------------------------------------------------------ r1108 | oetiker | 2007-06-11 17:32:41 +0200 | 1 line added detail ------------------------------------------------------------------------ r1107 | oetiker | 2007-06-11 17:11:10 +0200 | 5 lines --font-render-mode is back, and there is also --graph-render-mode now. It allows to enable mono graphs that look strikingly like rrdtool 1.0.x --tobi ------------------------------------------------------------------------ r1106 | oetiker | 2007-06-11 15:36:40 +0200 | 5 lines Bernhard Fischer: - NEWS mention the new file accessors - rrd_update.c: remove unused some cruft ------------------------------------------------------------------------ r1105 | oetiker | 2007-06-11 12:09:54 +0200 | 7 lines Bernhard Fischer: - fix maintaining our file-pos when we have to wrap the rra_ptr in update(). - no need to update the header-data a second time for the mmap case. ------------------------------------------------------------------------ r1104 | oetiker | 2007-06-11 02:23:51 +0200 | 2 lines updated news ------------------------------------------------------------------------ r1103 | oetiker | 2007-06-11 02:22:23 +0200 | 4 lines * added some documentation * updated version number ------------------------------------------------------------------------ r1102 | oetiker | 2007-06-11 01:55:13 +0200 | 2 lines switched graphics library from arts to cairo+pango ------------------------------------------------------------------------ r1101 | oetiker | 2007-06-04 13:15:11 +0200 | 3 lines Bernhard Fischer: - new configure option --enable-static-programs ------------------------------------------------------------------------ r1100 | oetiker | 2007-06-01 21:11:49 +0200 | 5 lines Bernhard: - implement resize SHRINK for FD (mmap is to be done) - indent ------------------------------------------------------------------------ r1099 | oetiker | 2007-06-01 19:59:25 +0200 | 2 lines --full-size-mode with and without --no-legend should work now -- Matthew Chambers ------------------------------------------------------------------------ r1098 | oetiker | 2007-06-01 19:35:46 +0200 | 4 lines Bernhard - fill the file with DNAN and add the remaining data in after the fact ... ------------------------------------------------------------------------ r1097 | oetiker | 2007-06-01 19:34:15 +0200 | 3 lines Bernhard Fischer - implement resize, growing part. truncation is to be done. ------------------------------------------------------------------------ r1096 | oetiker | 2007-06-01 11:10:33 +0200 | 6 lines Bernhard Fischer - use 'z' length modifier when printing size_t - add a cast to __rrd_read and document why it is there - add RANDOM hint for the header ------------------------------------------------------------------------ r1095 | oetiker | 2007-05-31 20:47:13 +0200 | 7 lines Bernhard: - plug tiny mem-leak in error path of rrd_close where freeing the rrd_file was forgotten. - improve two error messages - use CHECK_MADVISE_OVERLAPS per default ------------------------------------------------------------------------ r1094 | oetiker | 2007-05-30 21:56:34 +0200 | 2 lines fix rrd_write -- Bernhard Fischer ------------------------------------------------------------------------ r1093 | oetiker | 2007-05-30 14:41:39 +0200 | 1 line fix indenting ... again ------------------------------------------------------------------------ r1092 | oetiker | 2007-05-30 14:39:30 +0200 | 6 lines From Bernhard Fischer - rrd_close(): call close on the file and use rrd_close consistently - clean up some error paths. The fadvise error path is leaking memory (see XXX in these spots). ------------------------------------------------------------------------ r1089 | oetiker | 2007-05-30 07:23:07 +0200 | 3 lines switch for rrd_graph to specify the outer-size of the graph and not just the size of the canvas: --full-size-mode -- matthew.chambers vanderbilt.edu ------------------------------------------------------------------------ r1088 | oetiker | 2007-05-29 23:29:17 +0200 | 22 lines More updates from Bernhard Fischer - flag rrd_resize's old file with RRD_COPY - cleanup error-handling pathes in rrd_update and fix a few typos in comments - rrd_close(): implement printing mincore results for the rrd if DEBUG=2 was defined - rrd_open(): madvise start addresses need to be page-aligned; implement alternative path to the fine-grained (i.e. exact) madvise by flagging just the first two pages as needed (see TWO_PAGES). Implement alternative path that records the last madvise()ed area to avoid redundant calls to madvise() on identical areas (due to page-alignment constraints) -- see CHECK_MADVISE_OVERLAPS. Implement path for USE_DIRECT_IO. - configure: add check for O_DIRECT flag to open(2). Add option --enable-direct-io. Add _GNU_SOURCE to CFLAGS to silence warnings about chroot which is marked LEGACY since SUSv2 and is a non POSIX extension. Make checks for posix_fadvise() dependant on --disable-mmap, since we do not need fadvise for the mmap case. ------------------------------------------------------------------------ r1086 | oetiker | 2007-05-25 17:24:32 +0200 | 2 lines tobis old cruft removal program ... ------------------------------------------------------------------------ r1085 | oetiker | 2007-05-25 15:14:35 +0200 | 8 lines updates from Bernhard Fischer rep dot nop gmail com - add some more rrd_sterror calls to failure paths. - make rrdtool updatev work like 1.2.23 (i.e. multiple values are still broken, but at least updating a single value works; see http://oss.oetiker.ch/rrdtool-trac/ticket/8) - indent touched files. ------------------------------------------------------------------------ r1084 | oetiker | 2007-05-25 15:12:07 +0200 | 19 lines Updates from Bernhard Fischer rep dot nop gmail com - fix typo in rrd_fetch where rrd_read result was checked against an incorrect size. - in rrd_fetch, drop rrd_head_size in favour of rrd_file->header_len - in rrd_fetch, make the message "post fetch" unambiguous (now past vs. post). - change usage of param rdwr of rrd_open: allow for RRD_READONLY, RRD_READWRITE, RRD_CREAT, RRD_READAHEAD; adjust callers accordingly: + rrd_resize needs CREAT + rrd_dump may want READAHEAD - implement FD based I/O in rrd_open, rrd_read, rrd_write, rrd_seek. - in rrd_update, unify write_RRA_row(). - sort | uniq the -T in .indent.pro (info_t was duplicated) - add stub of an option to use O_DIRECT to the configury - in Makefile.am, simplify the "indent" invocation of find: My find may not support "-o" resp. "-or" nor braces. Using -name "*.[ch]" works everywhere, AFAIK. ------------------------------------------------------------------------ r1083 | oetiker | 2007-05-24 10:26:39 +0200 | 2 lines use proper preprocessor directives for the config-checks. -- tobi ------------------------------------------------------------------------ r1082 | oetiker | 2007-05-24 09:21:48 +0200 | 2 lines fixed version number for development ------------------------------------------------------------------------ r1081 | oetiker | 2007-05-24 08:16:01 +0200 | 3 lines indent all the rest of the code, and add some typedefs to indent.pro there is now a indent makefile target ------------------------------------------------------------------------ r1080 | oetiker | 2007-05-23 23:40:21 +0200 | 2 lines removed superfluouse code ------------------------------------------------------------------------ r1079 | oetiker | 2007-05-23 23:39:58 +0200 | 2 lines added more indenting rules ------------------------------------------------------------------------ r1078 | oetiker | 2007-05-23 23:33:45 +0200 | 3 lines reindented everything according to .indent.pro ------------------------------------------------------------------------ r1077 | oetiker | 2007-05-23 22:57:43 +0200 | 2 lines profile must be in the source directory to work ------------------------------------------------------------------------ r1076 | oetiker | 2007-05-23 22:57:24 +0200 | 2 lines tabs be gone! ------------------------------------------------------------------------ r1075 | oetiker | 2007-05-23 22:42:13 +0200 | 4 lines * progress in moving all the fileaccess over to a wrapper system that can do fd based and mmap based fileaccess transparently ... * small fixes left and right to improve code quality and stability. -- Bernhard Fischer rep dot nop gmail com ------------------------------------------------------------------------ r1074 | oetiker | 2007-05-23 18:10:23 +0200 | 2 lines we are going to use indent from now on to keep code indenting consistent throughout the project ------------------------------------------------------------------------ r1073 | oetiker | 2007-05-23 18:08:14 +0200 | 2 lines new trunk based on current 1.2 ------------------------------------------------------------------------ r1071 | oetiker | 2007-05-22 22:46:25 +0200 | 2 lines fixed spelling ------------------------------------------------------------------------ r1070 | oetiker | 2007-05-22 22:44:57 +0200 | 4 lines * better tests for madvise, fadvise and fdatasync * fix inclusion of fcntl.h * clenups for configure.ac -- Bernhard Fischer rep dot nop gmail com ------------------------------------------------------------------------ r1067 | oetiker | 2007-05-19 07:47:29 +0200 | 2 lines update defines according to manual pageentry ------------------------------------------------------------------------ r1066 | oetiker | 2007-05-18 18:40:55 +0200 | 2 lines added madvise MADV_RANDOM is rrd_update uses mmap ------------------------------------------------------------------------ r1065 | oetiker | 2007-05-18 13:56:34 +0200 | 5 lines * dropping in rrd_update is probably overkill since we may be dropping too much. * in create we have to flush before dropping. * read fcntl.h after defining the __USE_XOPEN2K * fix drop code in fetch ... ------------------------------------------------------------------------ r1064 | oetiker | 2007-05-17 13:38:54 +0200 | 2 lines now the fadvise code actually compiles ... let the testing begin. ------------------------------------------------------------------------ r1063 | oetiker | 2007-05-17 10:44:26 +0200 | 2 lines new better ? performance test ... work in progress ------------------------------------------------------------------------ r1062 | oetiker | 2007-05-17 10:44:06 +0200 | 7 lines added posix_fadvise support (untested) ... this should help performance by stopping read-ahead and droping buffer cache for all rrd data except the file header portion. Newly created files are fdsynced to disk and released from cache after creation, to soften the blow on buffer cache by creating new rrds. ------------------------------------------------------------------------ r1061 | oetiker | 2007-05-16 22:35:56 +0200 | 2 lines remove tabs and replace with 8 spaces ------------------------------------------------------------------------ r1060 | oetiker | 2007-05-15 10:39:17 +0200 | 2 lines use libdir instead of exec_prefix/lib ... maybe someone wants their libraries in another place ------------------------------------------------------------------------ r1059 | oetiker | 2007-05-15 10:36:52 +0200 | 2 lines yet another attempt to make the python build environment REALY cool ------------------------------------------------------------------------ r1057 | oetiker | 2007-05-09 17:26:57 +0200 | 3 lines make sure the python extension gets the final resting place of the rrdlibrary compiled in ... ------------------------------------------------------------------------ r1056 | oetiker | 2007-05-08 17:10:21 +0200 | 3 lines ruby complained about finding a previous copy of the rrdtool extension ... very odd thing actually ... adding RUBYARCHDIR= seems to help ... ------------------------------------------------------------------------ r1055 | oetiker | 2007-05-06 21:38:22 +0200 | 2 lines don't use round since it is c99 ... use floor(x+0.5) for positive numbers instead. ------------------------------------------------------------------------ r1054 | oetiker | 2007-05-06 10:50:11 +0200 | 2 lines print the name of the HW cf in debug mode -- Helge Oldach ------------------------------------------------------------------------ r1053 | oetiker | 2007-05-03 14:49:44 +0200 | 2 lines added some more rounding for int to float comparisons ------------------------------------------------------------------------ r1052 | oetiker | 2007-05-03 14:43:15 +0200 | 2 lines Testing an double and an integer for equality is bound to produce odd results on times. Don't do it! -- rrdtool@oldach.net ------------------------------------------------------------------------ r1050 | oetiker | 2007-05-02 20:06:05 +0200 | 1 line prepare for the release of rrdtool-1.2.23 ------------------------------------------------------------------------ r1049 | oetiker | 2007-05-02 20:05:37 +0200 | 2 lines fix python install for virgin systems ------------------------------------------------------------------------ r1047 | oetiker | 2007-05-02 17:31:29 +0200 | 1 line prepare for the release of rrdtool-1.2.22 ------------------------------------------------------------------------ r1046 | oetiker | 2007-05-02 17:31:05 +0200 | 2 lines fixed old urls in code ------------------------------------------------------------------------ r1045 | oetiker | 2007-05-02 17:23:34 +0200 | 2 lines fix python for staged installs ------------------------------------------------------------------------ r1044 | oetiker | 2007-05-02 07:08:21 +0200 | 2 lines fixed path to rrd_format.h in tcl bindings ------------------------------------------------------------------------ r1042 | oetiker | 2007-05-01 22:41:59 +0200 | 1 line prepare for the release of rrdtool-1.2.21 ------------------------------------------------------------------------ r1041 | oetiker | 2007-05-01 22:40:15 +0200 | 2 lines if the graph goes 'down' minval must be hanged instead ------------------------------------------------------------------------ r1039 | oetiker | 2007-05-01 18:36:02 +0200 | 1 line prepare for the release of rrdtool-1.2.20 ------------------------------------------------------------------------ r1038 | oetiker | 2007-05-01 18:24:03 +0200 | 2 lines updated win32 and netware Makefiles -- Guenter Knauf ------------------------------------------------------------------------ r1037 | oetiker | 2007-05-01 18:23:19 +0200 | 2 lines added some const casts ------------------------------------------------------------------------ r1036 | oetiker | 2007-05-01 18:19:47 +0200 | 3 lines added --alt-autoscale-min (see --alt-autoscale-max) -- Helge Oldach ------------------------------------------------------------------------ r1032 | oetiker | 2007-04-10 07:43:34 +0200 | 2 lines untabified rrd_graph.c ------------------------------------------------------------------------ r1031 | oetiker | 2007-04-10 07:41:50 +0200 | 2 lines use the shorthand for vidx ... ------------------------------------------------------------------------ r1030 | oetiker | 2007-04-10 07:27:04 +0200 | 2 lines the rrd_graph_check_vname function is not used anywhere in the code ... ------------------------------------------------------------------------ r1029 | oetiker | 2007-04-06 09:28:33 +0200 | 2 lines release font resources after thy have been used ... -- Travis Spencer ------------------------------------------------------------------------ r1028 | oetiker | 2007-04-02 08:21:19 +0200 | 6 lines new function rrd_fetch_r() (and make the strings const char* instead of char*). The only difference between rrd_fetch_r() and rrd_fetch_fn() is that rrd_fetch_r() receives the consolidation function as a string (instead of an enum cf_en) and is thread-safe -- Sam Umbach ------------------------------------------------------------------------ r1027 | oetiker | 2007-04-02 08:18:15 +0200 | 2 lines make xml2rrd more robust on invalid input -- Florian Forster ------------------------------------------------------------------------ r1026 | oetiker | 2007-03-29 21:08:26 +0200 | 2 lines added vardef ------------------------------------------------------------------------ r1021 | oetiker | 2007-03-15 22:10:18 +0100 | 2 lines fix the snprintf call for vtag ... it was using the wrong size -- kili ------------------------------------------------------------------------ r1020 | oetiker | 2007-03-15 22:03:51 +0100 | 2 lines fixed strerror_r usage ... #85 ------------------------------------------------------------------------ r1019 | oetiker | 2007-03-11 14:08:31 +0100 | 2 lines fix netware and win32 makefiles for new getopt names ------------------------------------------------------------------------ r1018 | oetiker | 2007-03-11 14:04:41 +0100 | 2 lines enable ID keyword in all source files ------------------------------------------------------------------------ r1017 | oetiker | 2007-03-11 13:57:38 +0100 | 2 lines do not use gnuisms in Makefile rules ($<) ------------------------------------------------------------------------ r1016 | oetiker | 2007-03-11 13:41:46 +0100 | 2 lines use Guenters latest Netware Makefile ... ------------------------------------------------------------------------ r1014 | oetiker | 2007-03-10 18:44:12 +0100 | 2 lines axe the double quotes on $perlcc in configure to allow for program names with arguments. ------------------------------------------------------------------------ r1013 | oetiker | 2007-03-10 18:36:32 +0100 | 2 lines integers can be negative ... fixed the test for this in rrd_update -- rians cc usu edu ------------------------------------------------------------------------ r1012 | oetiker | 2007-03-03 12:03:38 +0100 | 4 lines * build extensions last since this seems to be most likely to fail * do more testing before building the ruby extension ------------------------------------------------------------------------ r1010 | oetiker | 2007-02-24 18:23:02 +0100 | 2 lines fix memmory problem in rrd_dump.c ------------------------------------------------------------------------ r1008 | oetiker | 2007-02-14 19:54:29 +0100 | 2 lines new operator ABS added -- ------------------------------------------------------------------------ r1007 | oetiker | 2007-02-14 19:48:21 +0100 | 2 lines renamed getopt to rrd_getopt to avoid confusion on some systems ... ------------------------------------------------------------------------ r1006 | oetiker | 2007-02-14 19:47:41 +0100 | 2 lines cosmetic fix for usage message -- günter knauff ------------------------------------------------------------------------ r1005 | oetiker | 2007-02-14 19:46:56 +0100 | 2 lines added new win32 component to the release tarball ------------------------------------------------------------------------ r1004 | oetiker | 2007-02-14 19:46:16 +0100 | 2 lines updates to the windows build system by guenter knauff ------------------------------------------------------------------------ r1003 | oetiker | 2007-02-14 07:53:01 +0100 | 2 lines allow --units-length to reserve label space even when --y-grid=none is in effect [#75], tobi ------------------------------------------------------------------------ r1002 | oetiker | 2007-02-14 07:33:49 +0100 | 3 lines parsing absolute time should not send us 30,000 years into the future -- Tatsuki Makino tatsuki_makino hotmail.com ------------------------------------------------------------------------ r1001 | oetiker | 2007-02-14 07:31:52 +0100 | 3 lines Printstrftime on rrd_cgi.c returns an illegal value. It causes segmentation fault error. -- Tatsuki Makino tatsuki_makino hotmail.com ------------------------------------------------------------------------ r998 | oetiker | 2007-02-09 23:19:48 +0100 | 2 lines more build patches from Günter Knauf ------------------------------------------------------------------------ r997 | oetiker | 2007-02-08 07:01:43 +0100 | 2 lines updates for Win32 and NW Makefiles -- Guenter Knauf ------------------------------------------------------------------------ r996 | oetiker | 2007-02-08 06:54:23 +0100 | 2 lines added get_ver.awk to release ------------------------------------------------------------------------ r995 | oetiker | 2007-02-08 06:51:48 +0100 | 2 lines there is no Makefile.in for python anymore! ------------------------------------------------------------------------ r994 | oetiker | 2007-02-07 22:35:39 +0100 | 2 lines DPRINTF may not be the ideal define ... use something more RRDish ... like RRDPRINTF ------------------------------------------------------------------------ r993 | oetiker | 2007-02-07 22:35:07 +0100 | 2 lines use setup.py for python install and build work instad of trying todo it on our own ------------------------------------------------------------------------ r992 | oetiker | 2007-02-07 22:31:33 +0100 | 2 lines added Makefile targets for python ------------------------------------------------------------------------ r991 | oetiker | 2007-02-07 22:30:18 +0100 | 2 lines python does no get built externally ------------------------------------------------------------------------ r990 | oetiker | 2007-02-02 19:22:34 +0100 | 2 lines added IBM Public License to the list of Exceptions ------------------------------------------------------------------------ r989 | oetiker | 2007-02-02 13:19:12 +0100 | 2 lines increasse portability (no // comment, no ... arguments for cpp) ------------------------------------------------------------------------ r988 | oetiker | 2007-02-01 23:00:23 +0100 | 3 lines we should use the headers from the local rrdtool setup and not the system ones! ------------------------------------------------------------------------ r986 | oetiker | 2007-02-01 06:51:34 +0100 | 1 line prepare for the release of rrdtool-1.2.19 ------------------------------------------------------------------------ r985 | oetiker | 2007-02-01 06:45:03 +0100 | 3 lines libtool builds shared libraries anyway, no need for extra LDFLAGS that bite non GNU ld versions ------------------------------------------------------------------------ r979 | oetiker | 2007-01-29 17:47:28 +0100 | 2 lines resolve snprintf overflow warning by using sizeof -- Anna Bernathova suse cz ------------------------------------------------------------------------ r977 | oetiker | 2007-01-23 21:54:46 +0100 | 1 line prepare for the release of rrdtool-1.2.18 ------------------------------------------------------------------------ r976 | oetiker | 2007-01-23 21:52:35 +0100 | 4 lines the change from R920 to fetch was not helping ... no low res got prefered in non full coverage situations .. ------------------------------------------------------------------------ r975 | oetiker | 2007-01-23 18:30:27 +0100 | 2 lines another control sequence that got lost ... ------------------------------------------------------------------------ r974 | oetiker | 2007-01-23 09:28:56 +0100 | 2 lines read up to full potential length of gdp->rrd ------------------------------------------------------------------------ r972 | oetiker | 2007-01-22 17:34:55 +0100 | 2 lines added missing B formatting instruction ------------------------------------------------------------------------ r971 | oetiker | 2007-01-22 11:49:19 +0100 | 2 lines create non version link ------------------------------------------------------------------------ r969 | oetiker | 2007-01-22 11:33:30 +0100 | 1 line prepare for the release of rrdtool-1.2.17 ------------------------------------------------------------------------ r968 | oetiker | 2007-01-22 11:29:11 +0100 | 1 line prepare for the release of rrdtool-1.2.17 ------------------------------------------------------------------------ r967 | oetiker | 2007-01-22 11:28:39 +0100 | 6 lines introduced \n as an alias for \l since people seem to be using it despite the documentation not talking about this ... so now it is official and I bet there will soon be people asking why this was only working at the end of a line. -- tobi ------------------------------------------------------------------------ r966 | oetiker | 2007-01-21 23:22:30 +0100 | 2 lines removed mentiones of \n ... ------------------------------------------------------------------------ r962 | oetiker | 2007-01-20 23:54:25 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r961 | oetiker | 2007-01-20 23:48:37 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r960 | oetiker | 2007-01-20 23:22:10 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r959 | oetiker | 2007-01-20 21:19:13 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r958 | oetiker | 2007-01-20 21:17:31 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r957 | oetiker | 2007-01-20 21:16:15 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r956 | oetiker | 2007-01-20 21:03:14 +0100 | 2 lines this did not compile after integrating the new fetchlast command -- Andy Riebs ------------------------------------------------------------------------ r955 | oetiker | 2007-01-20 17:41:15 +0100 | 1 line prepare for the release of rrdtool-1.2.16 ------------------------------------------------------------------------ r954 | oetiker | 2007-01-20 17:39:32 +0100 | 2 lines get ready for 1.2.16 release ------------------------------------------------------------------------ r953 | oetiker | 2007-01-19 00:43:26 +0100 | 2 lines aargh another missing bit from lastupdate added ... ------------------------------------------------------------------------ r952 | oetiker | 2007-01-18 00:35:01 +0100 | 2 lines finish integration of lastupdate commmand ... ------------------------------------------------------------------------ r951 | oetiker | 2007-01-17 22:31:23 +0100 | 4 lines New functions lastupdate to efficiently get the last values fed into the rrd ... this also changes that the last values get stored even for ABSOLUTE and GAUGE data sources ... -- andy.riebs hp.com ------------------------------------------------------------------------ r938 | oetiker | 2006-12-16 17:14:43 +0100 | 2 lines add missing \0 to the end of several strncpy strings ------------------------------------------------------------------------ r937 | oetiker | 2006-12-16 17:13:53 +0100 | 5 lines replace malloc/strncpy by strdup add \0 to the end of several strncpy strings add missing free fix error string rendering :... makes no sense here ------------------------------------------------------------------------ r936 | oetiker | 2006-12-16 17:11:46 +0100 | 2 lines move headers around to make more sense ------------------------------------------------------------------------ r935 | oetiker | 2006-12-16 17:10:58 +0100 | 2 lines init the context on first call ------------------------------------------------------------------------ r934 | oetiker | 2006-12-16 17:05:07 +0100 | 2 lines make sure there is ample space in the static strings ------------------------------------------------------------------------ r933 | oetiker | 2006-12-16 17:04:05 +0100 | 2 lines oops 99 is the last item ------------------------------------------------------------------------ r932 | oetiker | 2006-12-16 17:02:43 +0100 | 3 lines it is sufficient to set the last item to 0 ------------------------------------------------------------------------ r931 | oetiker | 2006-12-16 17:02:00 +0100 | 2 lines getopt is required for rrdupdate too (there are platforms where this is not in libc ... eg solaris) ------------------------------------------------------------------------ r930 | oetiker | 2006-12-16 17:01:25 +0100 | 2 lines we have 2006 now ------------------------------------------------------------------------ r929 | oetiker | 2006-12-16 16:56:26 +0100 | 4 lines it makes no sense allocating an array and then asiging it to a fixed string ... this caused stuff to crash badly as long error strings were assigned. ------------------------------------------------------------------------ r927 | oetiker | 2006-12-11 20:11:36 +0100 | 2 lines added note on alternate CFLAGS for Forte ------------------------------------------------------------------------ r925 | oetiker | 2006-12-07 20:51:59 +0100 | 2 lines c is a valid formatting character ------------------------------------------------------------------------ r922 | oetiker | 2006-11-19 12:15:32 +0100 | 2 lines fixed typo ------------------------------------------------------------------------ r920 | oetiker | 2006-11-06 07:53:13 +0100 | 4 lines when fetching, don't pick a higher res rra just because it's coverage at the 'end' of the requested range is better ... -- Sebastian Pachuta seba123 seba123.webd.pl ------------------------------------------------------------------------ r919 | oetiker | 2006-10-27 16:03:22 +0200 | 2 lines use integer time ------------------------------------------------------------------------ r918 | oetiker | 2006-10-27 00:26:31 +0200 | 2 lines added perftest tool ------------------------------------------------------------------------ r917 | oetiker | 2006-10-25 19:19:46 +0200 | 2 lines do not distribute windows files anymore ... ------------------------------------------------------------------------ r916 | oetiker | 2006-10-25 19:18:27 +0200 | 2 lines distribute win32 and netware build files too ------------------------------------------------------------------------ r915 | oetiker | 2006-10-25 19:17:07 +0200 | 2 lines do not remove Makefiles from win32 and netware build dirs ------------------------------------------------------------------------ r914 | oetiker | 2006-10-25 19:12:22 +0200 | 3 lines fix build procedure for ruby (clean target) don't run autoreconf at the end of MakeMakefiles as this has a tendency todo the wrong things ------------------------------------------------------------------------ r913 | oetiker | 2006-10-25 18:39:31 +0200 | 2 lines added path to sepp ------------------------------------------------------------------------ r912 | oetiker | 2006-10-25 18:30:39 +0200 | 2 lines aclocal expects a space after -I ... ------------------------------------------------------------------------ r911 | oetiker | 2006-10-25 18:24:44 +0200 | 2 lines tell aclocal where to find its config ... ------------------------------------------------------------------------ r910 | oetiker | 2006-10-25 18:11:57 +0200 | 2 lines aclocal show know where it's stuff is ... ------------------------------------------------------------------------ r908 | oetiker | 2006-10-22 23:20:45 +0200 | 8 lines make time paring more robust ... it can now deal with time date date time time date + xxx date time + xxx ------------------------------------------------------------------------ r907 | oetiker | 2006-10-22 23:09:27 +0200 | 2 lines check if valid control codes are in use ------------------------------------------------------------------------ r906 | oetiker | 2006-10-22 18:01:42 +0200 | 2 lines fixed ruby demo ------------------------------------------------------------------------ r905 | oetiker | 2006-10-22 18:01:32 +0200 | 2 lines fixed address ------------------------------------------------------------------------ r903 | oetiker | 2006-10-03 09:52:51 +0200 | 2 lines added missing file ------------------------------------------------------------------------ r902 | oetiker | 2006-09-28 14:38:34 +0200 | 2 lines fixed doc regarding site install ------------------------------------------------------------------------ r901 | oetiker | 2006-09-27 23:48:05 +0200 | 2 lines added ruby bindings ... thanks to Loïs LHERBIER lois.lherbier covadis.ch ------------------------------------------------------------------------ r900 | oetiker | 2006-09-27 23:46:23 +0200 | 2 lines make rrdupdate realy light as it was intended in the first place -- Peter Breitenlohner peb mppmu.mpg.de ------------------------------------------------------------------------ r899 | oetiker | 2006-09-27 23:45:47 +0200 | 1 line tod can come after the date as well as before the date ------------------------------------------------------------------------ r898 | oetiker | 2006-09-18 07:45:16 +0200 | 2 lines fix faild netware makefile merge -- Guenter Knauf ------------------------------------------------------------------------ r897 | oetiker | 2006-09-17 23:08:29 +0200 | 2 lines make mvs stop complaining about uninitialized variables ... -- norman wheeler bigpond.com ------------------------------------------------------------------------ r896 | oetiker | 2006-09-17 23:05:18 +0200 | 2 lines fixed spelling -- Peter Breitenlohner ------------------------------------------------------------------------ r895 | oetiker | 2006-09-17 22:46:17 +0200 | 3 lines fix generation of cgi-demo.cgi -- Peter Breitenlohner peb mppmu.mpg.de ------------------------------------------------------------------------ r894 | oetiker | 2006-09-17 22:31:26 +0200 | 4 lines make rrdtool use rrd_config.h instead of config.h since this is just bound to lead to confusion when the wrong config.h gets sourced -- Guenter Knauf fix clean target in bindings directory ------------------------------------------------------------------------ r893 | oetiker | 2006-09-10 21:30:22 +0200 | 2 lines the msvc files need dos file endings ... ------------------------------------------------------------------------ r892 | oetiker | 2006-09-10 21:26:05 +0200 | 2 lines we want an int, so lets tell the compiler to make it not complain ------------------------------------------------------------------------ r891 | oetiker | 2006-09-10 21:20:59 +0200 | 2 lines netware und win32 fixes from guenter ------------------------------------------------------------------------ r890 | oetiker | 2006-09-05 21:55:36 +0200 | 2 lines we do not need an extra echo there ------------------------------------------------------------------------ r889 | oetiker | 2006-09-05 20:34:47 +0200 | 2 lines put all the architecture specific stuff in separate subdirectories ... one for netwara and one for win32 -- Guenter Knauf ------------------------------------------------------------------------ r888 | oetiker | 2006-09-05 20:19:02 +0200 | 3 lines noone needs this anymore ------------------------------------------------------------------------ r887 | oetiker | 2006-09-04 23:10:55 +0200 | 1 line in logarithmic mode minval must not be <= 0 ------------------------------------------------------------------------ r883 | oetiker | 2006-08-24 06:14:54 +0200 | 2 lines add more submission encuragement -- alex ------------------------------------------------------------------------ r882 | oetiker | 2006-08-21 13:41:57 +0200 | 2 lines updated windows build instructions from guenter ------------------------------------------------------------------------ r881 | oetiker | 2006-08-16 08:07:26 +0200 | 2 lines contain a potential problem with log grid painting Bug #54 ------------------------------------------------------------------------ r880 | oetiker | 2006-08-15 07:51:04 +0200 | 1 line do not loose filedescriptors when rrd_first errors out ------------------------------------------------------------------------ r879 | oetiker | 2006-08-15 07:44:54 +0200 | 2 lines catch empty XXXX: commands in graph ------------------------------------------------------------------------ r878 | oetiker | 2006-08-15 07:24:39 +0200 | 2 lines windows is not NT anymore ... lets call it WIN32 ------------------------------------------------------------------------ r877 | oetiker | 2006-08-13 18:41:23 +0200 | 2 lines use unix fileendings here! ------------------------------------------------------------------------ r876 | oetiker | 2006-08-13 18:40:17 +0200 | 2 lines new version added by Guenter Knauf ------------------------------------------------------------------------ r875 | oetiker | 2006-08-13 18:25:00 +0200 | 3 lines added new rrdstrversion function -- Guenter Knauf ------------------------------------------------------------------------ r874 | oetiker | 2006-08-13 18:22:53 +0200 | 3 lines use th OS provided timezone names for %Z this should make things more flexile will only work on windows, but since we are using strftime.c on windows only this should not be a problem -- Guenter Knauf ------------------------------------------------------------------------ r873 | oetiker | 2006-08-13 18:13:13 +0200 | 2 lines include strftime.h for windows builds -- Guenter Knauff ------------------------------------------------------------------------ r872 | oetiker | 2006-08-13 18:10:23 +0200 | 2 lines removed mscv defines to confignt/config.h completely -- Guenter Knauf ------------------------------------------------------------------------ r871 | oetiker | 2006-08-13 18:09:25 +0200 | 2 lines working windows config.h ------------------------------------------------------------------------ r870 | oetiker | 2006-08-13 18:07:36 +0200 | 3 lines can't define new variables in the middle of the code (unless everyone was using c99 which is not the case) ------------------------------------------------------------------------ r869 | oetiker | 2006-08-13 17:21:12 +0200 | 3 lines added special labeling for LONGTERM graphs -- Paul Boven p.boven sara.nl ------------------------------------------------------------------------ r867 | oetiker | 2006-08-10 09:46:22 +0200 | 2 lines it seems some headers have nan predefined ... so lets just use other variable names ------------------------------------------------------------------------ r866 | oetiker | 2006-08-04 17:06:18 +0200 | 2 lines copy of the solaris isnan hack from configure.ac ------------------------------------------------------------------------ r865 | oetiker | 2006-08-04 00:21:37 +0200 | 3 lines solaris 10 has isnan defined as a sun forte builtin ... gcc can not deal with this this will replace isnan with an fpclass expression hopefully working around the problem ------------------------------------------------------------------------ r864 | oetiker | 2006-08-02 17:05:36 +0200 | 3 lines fixing python bindings ... Jarod Wilson jwilson redhat.com ------------------------------------------------------------------------ r863 | oetiker | 2006-08-01 15:08:33 +0200 | 2 lines unknownaszero was not a good idea ... bye bye ------------------------------------------------------------------------ r862 | oetiker | 2006-08-01 00:48:45 +0200 | 2 lines new options --enumds and --unknwonaszero for xport ------------------------------------------------------------------------ r860 | oetiker | 2006-07-14 14:11:16 +0200 | 1 line prepare for the release of rrdtool-1.2.15 ------------------------------------------------------------------------ r859 | oetiker | 2006-07-14 14:06:09 +0200 | 1 line we were leaking directory handles ... ------------------------------------------------------------------------ r858 | oetiker | 2006-07-14 13:34:23 +0200 | 1 line remove an excess =over ------------------------------------------------------------------------ r857 | oetiker | 2006-07-14 13:32:39 +0200 | 2 lines added missing =over ------------------------------------------------------------------------ r855 | oetiker | 2006-07-14 13:12:45 +0200 | 1 line prepare for the release of rrdtool-1.2.14 ------------------------------------------------------------------------ r854 | oetiker | 2006-07-14 11:47:34 +0200 | 2 lines do not reset im->gdes[i].step blindly. We initialize it to im->step at allocation time already ... the fix in r291 was over the top. -- niels weaklogic.com ------------------------------------------------------------------------ r853 | oetiker | 2006-07-14 10:56:28 +0200 | 3 lines added pwd command for remote server reset errno to zero before each loop -- Damien.Stuart usi.net ------------------------------------------------------------------------ r852 | oetiker | 2006-07-13 10:44:47 +0200 | 2 lines fixed makefiles without libcgi ... guenter knauf ------------------------------------------------------------------------ r851 | oetiker | 2006-07-08 00:37:20 +0200 | 2 lines added first function to python bindings ... -- Ulf Lilleengen lulf pvv.ntnu.no ------------------------------------------------------------------------ r850 | oetiker | 2006-07-06 00:00:25 +0200 | 2 lines we have no cgilib anymore .... ------------------------------------------------------------------------ r849 | oetiker | 2006-07-04 23:04:12 +0200 | 2 lines let rrdtool understand headers in rrdtool restore input ------------------------------------------------------------------------ r847 | oetiker | 2006-06-18 23:21:56 +0200 | 3 lines Added AVG function to CDEF language. Martin Sperl martin sperl.org ------------------------------------------------------------------------ r846 | oetiker | 2006-06-16 00:16:04 +0200 | 2 lines fixed libpng link ------------------------------------------------------------------------ r845 | oetiker | 2006-06-15 09:44:22 +0200 | 2 lines added hint for RHEL ------------------------------------------------------------------------ r843 | oetiker | 2006-06-13 23:20:36 +0200 | 2 lines added missing tcl files ------------------------------------------------------------------------ r842 | oetiker | 2006-06-07 23:01:35 +0200 | 2 lines fix calc of fahrenheit ... ------------------------------------------------------------------------ r841 | oetiker | 2006-06-06 15:25:32 +0200 | 2 lines additional AIX hints from zoran.majcenic inet.hr ------------------------------------------------------------------------ r840 | oetiker | 2006-06-06 10:16:05 +0200 | 3 lines new, working rrdtool.spec file from Jarod Wilson jwilson redhat.com ------------------------------------------------------------------------ r839 | oetiker | 2006-05-27 07:44:15 +0200 | 2 lines python uses the name of the module for loding, so this should better match up -- Duncan Webb duncan dwebb ch ------------------------------------------------------------------------ r838 | oetiker | 2006-05-25 17:13:56 +0200 | 2 lines addeed AIX instructions ------------------------------------------------------------------------ r837 | oetiker | 2006-05-23 08:52:22 +0200 | 2 lines adjust label spacing to make sure labels don't overlap for the default font. ------------------------------------------------------------------------ r836 | oetiker | 2006-05-22 00:14:27 +0200 | 2 lines add note about variable name picking as suggested by alex ------------------------------------------------------------------------ r835 | oetiker | 2006-05-22 00:07:02 +0200 | 1 line another Makefile update from guenter ------------------------------------------------------------------------ r834 | oetiker | 2006-05-22 00:06:17 +0200 | 2 lines tiny updates to netware and win32 makefiles -- Guenter Knauf ------------------------------------------------------------------------ r833 | oetiker | 2006-05-21 23:53:57 +0200 | 2 lines fixed type hte->the ------------------------------------------------------------------------ r832 | oetiker | 2006-05-21 23:47:23 +0200 | 3 lines make vdef time part available in the data_calc stage so that the legend printer can decide properly ------------------------------------------------------------------------ r831 | oetiker | 2006-05-21 23:15:39 +0200 | 2 lines new GPRINT option :strftime to print time associated with a VDEF value ------------------------------------------------------------------------ r830 | oetiker | 2006-05-21 22:20:11 +0200 | 2 lines make configure test which flags gcc actually accepts ... ------------------------------------------------------------------------ r829 | oetiker | 2006-05-21 15:08:46 +0200 | 2 lines don't deprecate HRULE anymore ------------------------------------------------------------------------ r828 | oetiker | 2006-05-21 14:37:12 +0200 | 2 lines allow DS names starting with the same letters as operators ------------------------------------------------------------------------ r827 | oetiker | 2006-05-12 15:26:00 +0200 | 2 lines we wan't to dist thread_save_nt too ------------------------------------------------------------------------ r826 | oetiker | 2006-05-11 13:11:04 +0200 | 2 lines make DNAN and DINF faster by caching the result of the first calculation -- pascal.gloor spale.com ------------------------------------------------------------------------ r825 | oetiker | 2006-05-11 09:37:58 +0200 | 2 lines tell Makefile.PL how to store an rpath under aix ------------------------------------------------------------------------ r824 | oetiker | 2006-05-10 22:51:09 +0200 | 2 lines improved scaling for --logarithmic mode ... -- beat.zahnd space.unibe.ch ------------------------------------------------------------------------ r823 | oetiker | 2006-05-09 20:46:14 +0200 | 2 lines don't panic if the user forgets to specify the name of the rrd file ------------------------------------------------------------------------ r822 | oetiker | 2006-05-07 12:46:24 +0200 | 2 lines make legend spacing more reliable. handle the fact that some graph comands do not contribute to the legend -- tobi ------------------------------------------------------------------------ r821 | oetiker | 2006-05-06 15:29:33 +0200 | 3 lines lets have index.html -- Peter Breitenlohner ------------------------------------------------------------------------ r820 | oetiker | 2006-05-06 15:28:57 +0200 | 4 lines Avoid gcc warning (discards qualifier) in tclrrd.c -- Peter Breitenlohner ------------------------------------------------------------------------ r819 | oetiker | 2006-05-06 15:28:16 +0200 | 2 lines Avoid gcc warning ("static not first") -- Peter Breitenlohner ------------------------------------------------------------------------ r818 | oetiker | 2006-05-06 15:24:23 +0200 | 3 lines Avoid gcc warnings about use of "long long" -- Peter Breitenlohner ------------------------------------------------------------------------ r817 | oetiker | 2006-05-06 15:21:42 +0200 | 2 lines Use tclpkgdir instead of pkglibdir in order to avoid automake warning -- Peter Breitenlohner ------------------------------------------------------------------------ r816 | oetiker | 2006-05-06 15:20:26 +0200 | 4 lines The rrdtoolmodule.so should be installed in pyexecdir instead of pythondir (they differ if prefix!=exec_prefix) -- Peter Breitenlohner ------------------------------------------------------------------------ r815 | oetiker | 2006-05-04 22:41:51 +0200 | 3 lines fix for debian bug 359071 ... reporting the long cdef in the error message was too much ... plus some other small cleanups in this context ------------------------------------------------------------------------ r813 | oetiker | 2006-05-04 15:26:21 +0200 | 1 line prepare for the release of rrdtool-1.2.13 ------------------------------------------------------------------------ r812 | oetiker | 2006-05-04 15:24:16 +0200 | 2 lines fix my mail address ------------------------------------------------------------------------ r811 | oetiker | 2006-05-04 15:12:01 +0200 | 2 lines fewer double frees ... -- slamb slamb.org ------------------------------------------------------------------------ r810 | oetiker | 2006-05-04 15:02:26 +0200 | 4 lines * fixed argument parsing for long options * no more doublefree on abort * enable restore of xml dumps created with rrdtool 1.0.x ------------------------------------------------------------------------ r809 | oetiker | 2006-05-04 14:11:09 +0200 | 3 lines if end % step == 0 we should still fetch a full step ... no need for special handling ------------------------------------------------------------------------ r808 | oetiker | 2006-05-04 09:48:04 +0200 | 4 lines updated handling of subsecond resolution in connection with unknown data ... bugreport #38 pointed out that the old handling where the seconds were rounded could lead to cases where the data stored was higher than the data input ... now I always floor the unknown seconds and this will lead to the same or lower data but not higher ... -- tobi ------------------------------------------------------------------------ r806 | oetiker | 2006-05-02 22:52:46 +0200 | 2 lines variables should be defined at the bein of the block ... else we loose portability! ------------------------------------------------------------------------ r805 | oetiker | 2006-04-29 10:57:25 +0200 | 2 lines fixed spelling ------------------------------------------------------------------------ r804 | oetiker | 2006-04-27 08:37:03 +0200 | 2 lines new units=si option -- wim.heirmann elis.urgent.be ------------------------------------------------------------------------ r802 | oetiker | 2006-04-25 23:00:09 +0200 | 2 lines compile without multibyte support ------------------------------------------------------------------------ r801 | oetiker | 2006-04-25 22:59:41 +0200 | 2 lines lets have proper dependencies ------------------------------------------------------------------------ r800 | oetiker | 2006-04-24 17:14:10 +0200 | 2 lines make sure things are initialised ... ------------------------------------------------------------------------ r796 | oetiker | 2006-04-18 07:29:34 +0200 | 2 lines shared should be share ! ------------------------------------------------------------------------ r795 | oetiker | 2006-04-17 22:25:33 +0200 | 3 lines Add a second label to the Y axis if there is only going to be one -- Jason A. Smith smithj4 X bnl.gov ------------------------------------------------------------------------ r790 | oetiker | 2006-04-14 12:17:28 +0200 | 2 lines no more people links! ------------------------------------------------------------------------ r788 | oetiker | 2006-04-14 11:22:00 +0200 | 2 lines remove people references ------------------------------------------------------------------------ r785 | oetiker | 2006-04-07 16:37:52 +0200 | 2 lines add solaris hint to bild procedure ------------------------------------------------------------------------ r784 | oetiker | 2006-04-04 20:46:08 +0200 | 2 lines allow for input lines of arbitrary length in rrdtool pipe mode -- roger.meier terreactive.ch ------------------------------------------------------------------------ r782 | oetiker | 2006-04-01 17:36:05 +0200 | 1 line watermartk feature for rrdgraph by Ronan Mullally ------------------------------------------------------------------------ r781 | oetiker | 2006-03-26 16:36:35 +0200 | 2 lines vnames can be up to MAX_VNAME_LEN long ... not only 30 chars ... -- Ulf Harnhammar metaur telia.com ------------------------------------------------------------------------ r780 | oetiker | 2006-03-26 16:29:53 +0200 | 2 lines potential bufferoverflow throught large windir avoided ------------------------------------------------------------------------ r776 | oetiker | 2006-03-14 15:59:57 +0100 | 2 lines added note on the fact that no OK will be printed in rrdtool pipe mode if an error occures. ------------------------------------------------------------------------ r775 | oetiker | 2006-03-14 08:11:00 +0100 | 2 lines RRDtool in remote mode will either print an ERROR line or an OK line not both. ------------------------------------------------------------------------ r774 | oetiker | 2006-03-14 00:21:27 +0100 | 2 lines in pipe mode, make rrd_tool send errors to STDOUT (most of the time) -- tobi ------------------------------------------------------------------------ r772 | oetiker | 2006-03-06 08:15:45 +0100 | 2 lines at least freebsd will not do utf8 bits properly without this ------------------------------------------------------------------------ r769 | oetiker | 2006-03-04 00:11:55 +0100 | 3 lines * simplify code for STACK function ... there were two code paths (new as well as legacy from 1.0.x) * make TICK work for negative fractions ------------------------------------------------------------------------ r768 | oetiker | 2006-03-04 00:10:46 +0100 | 2 lines remove ~ from my wishlist address ... ------------------------------------------------------------------------ r767 | oetiker | 2006-03-04 00:10:14 +0100 | 2 lines make this compile even when HAVE_MBSTOWCS is not set ------------------------------------------------------------------------ r766 | oetiker | 2006-02-23 08:26:28 +0100 | 3 lines separate matching of full and partial match evaluation for better result stability. The fetch result was dependent on the RRA ordering before. -- Jo Rhett jrhett svcolo.com ------------------------------------------------------------------------ r764 | oetiker | 2006-02-20 07:20:37 +0100 | 2 lines YYYYMMDDhhmm is not supported ... only YYYYMMDD ------------------------------------------------------------------------ r761 | oetiker | 2006-02-08 12:35:10 +0100 | 2 lines OS and COPYRIGHT removed from the rrdcgi INTERNAL function since we do not have this info ------------------------------------------------------------------------ r760 | oetiker | 2006-02-06 20:49:23 +0100 | 1 line new rrdcgi function RRD::INTERNAL for accessing VERSION, COPYRIGHT, COMPILETIME, OS -- Guenter Knauf gk gknw.de ------------------------------------------------------------------------ r759 | oetiker | 2006-02-06 20:46:26 +0100 | 2 lines we have 2006 now so let's show it ------------------------------------------------------------------------ r755 | oetiker | 2006-01-30 00:28:21 +0100 | 1 line more makefile cleanup from Guenter Knauf gk gknw.de ------------------------------------------------------------------------ r752 | oetiker | 2006-01-25 20:09:12 +0100 | 2 lines allow RRDp to catch errors ------------------------------------------------------------------------ r751 | oetiker | 2006-01-22 13:02:47 +0100 | 3 lines clarified the documentation of xff -- Michael Bunk mb computer-leipzig.com ------------------------------------------------------------------------ r750 | oetiker | 2006-01-22 12:00:18 +0100 | 2 lines the operator in the example is * not + -- Nick Sharp njsharp bigpond.net.au ------------------------------------------------------------------------ r748 | oetiker | 2006-01-15 14:31:12 +0100 | 2 lines make sure we pickup the correct config.h ... -- Blair Zajac blair orcaware.com ------------------------------------------------------------------------ r747 | oetiker | 2006-01-15 13:31:03 +0100 | 2 lines make sure the image size only gets printed from rrd_tool when --imageinfo is NOT set. ------------------------------------------------------------------------ r746 | oetiker | 2006-01-15 13:25:36 +0100 | 2 lines make sure gcc does not thing gridind may be undefined ------------------------------------------------------------------------ r745 | oetiker | 2006-01-15 12:12:51 +0100 | 2 lines HWPREDICT without reference number will implicitly create the other Holt-Winters RRAs. ------------------------------------------------------------------------ r744 | oetiker | 2006-01-15 10:03:30 +0100 | 2 lines updated NetWare and Cygwin Builds ------------------------------------------------------------------------ r743 | oetiker | 2006-01-15 09:55:53 +0100 | 2 lines make sure gridind is defined under all conditions ------------------------------------------------------------------------ r742 | oetiker | 2006-01-14 10:10:16 +0100 | 2 lines add perlbindings and optional export filename to rrd_dump -- Nicola Worthington nicolaw arwen.tfb.net ------------------------------------------------------------------------ r740 | oetiker | 2006-01-05 21:39:12 +0100 | 2 lines let the building of the perl extension be more reliable ... ------------------------------------------------------------------------ r739 | oetiker | 2006-01-05 21:07:41 +0100 | 2 lines remove redundant if statement in rrd_gfx pdf code ... -- NormW ------------------------------------------------------------------------ r737 | oetiker | 2006-01-04 23:07:46 +0100 | 2 lines Compile time improvments for NetWare and Win32 -- Guenter Knauf gk.gknw.de ------------------------------------------------------------------------ r736 | oetiker | 2006-01-04 21:12:20 +0100 | 2 lines make sure this actually compiles with HAVE_MBSTOWCS is not defined ------------------------------------------------------------------------ r735 | oetiker | 2005-12-30 15:36:30 +0100 | 3 lines only use getuid if it is actually available ... ------------------------------------------------------------------------ r734 | oetiker | 2005-12-26 17:01:40 +0100 | 3 lines added heartbeat grahics ------------------------------------------------------------------------ r733 | oetiker | 2005-12-19 13:28:55 +0100 | 2 lines fix tcl issues ... build with pre tcl8.4 ... do not try to install into the tcl tree unless --enable-tcl-site is given -- tobi ------------------------------------------------------------------------ r729 | oetiker | 2005-12-18 21:30:31 +0100 | 1 line prepare for the release of rrdtool-1.2.12 ------------------------------------------------------------------------ r728 | oetiker | 2005-12-18 21:29:51 +0100 | 3 lines added FLOSS exception from http://www.mysql.com/company/legal/licensing/foss-exception.html in order to not end up in license incompatibility hell -- tobi ------------------------------------------------------------------------ r725 | oetiker | 2005-11-30 10:19:16 +0100 | 3 lines if called with --lazy we still want to PRINT ... -- Bartek Szady bszx from bszx.eu.org ------------------------------------------------------------------------ r723 | oetiker | 2005-11-28 22:59:15 +0100 | 2 lines we should still consider the mrhb ... ------------------------------------------------------------------------ r722 | oetiker | 2005-11-20 11:16:38 +0100 | 2 lines improve win32 compilation -- Norm ------------------------------------------------------------------------ r721 | oetiker | 2005-11-18 21:38:51 +0100 | 2 lines deal with versions ending in letters ------------------------------------------------------------------------ r718 | oetiker | 2005-11-16 14:14:25 +0100 | 3 lines have indivitudal lines for generating output files, so that cygwin does not choke ------------------------------------------------------------------------ r717 | oetiker | 2005-11-15 00:07:30 +0100 | 2 lines make mmap test a little saver ------------------------------------------------------------------------ r714 | oetiker | 2005-11-14 12:46:31 +0100 | 2 lines cygwin can actually do mmap ... -- larryjadams with comcast.net ------------------------------------------------------------------------ r713 | oetiker | 2005-11-14 08:32:23 +0100 | 2 lines guess the option should still be called template :-) ------------------------------------------------------------------------ r712 | oetiker | 2005-11-13 19:32:10 +0100 | 2 lines improve screen apperance of configure output as suggested by Dale -- tobi ------------------------------------------------------------------------ r711 | oetiker | 2005-11-13 19:31:38 +0100 | 2 lines don't use clean: target ... automake wants this to be clean-local: ------------------------------------------------------------------------ r710 | oetiker | 2005-11-13 00:36:10 +0100 | 3 lines add xmlns declaration to svg output -- R.P. Aditya ------------------------------------------------------------------------ r709 | oetiker | 2005-11-13 00:30:50 +0100 | 2 lines clean target was broken ------------------------------------------------------------------------ r708 | oetiker | 2005-11-13 00:10:48 +0100 | 5 lines * integrate necessary cgi functionality into rrd_cgi.c so that libcgi is not required anymore ... -- D. Walsh * check if we have to include malloc/malloc.h to make malloc work ... -- tobi ------------------------------------------------------------------------ r707 | oetiker | 2005-11-07 20:12:47 +0100 | 3 lines the variable template has to generic a name, it clashes with other things in some environments. ------------------------------------------------------------------------ r706 | oetiker | 2005-10-28 07:55:36 +0200 | 3 lines added openbsd note ------------------------------------------------------------------------ r705 | oetiker | 2005-10-24 22:55:03 +0200 | 2 lines use multithreded version of librrd -- Oleg Derevenetz ------------------------------------------------------------------------ r702 | oetiker | 2005-10-11 19:09:09 +0200 | 2 lines improved error messages for rrd_graph commandline parsing -- Alex ------------------------------------------------------------------------ r701 | oetiker | 2005-10-11 18:51:50 +0200 | 2 lines fix file generation ------------------------------------------------------------------------ r700 | oetiker | 2005-10-11 18:41:15 +0200 | 2 lines die on error! -- Alex ------------------------------------------------------------------------ r699 | oetiker | 2005-10-11 18:37:58 +0200 | 2 lines the images should be written to the current directory ... -- Alex ------------------------------------------------------------------------ r698 | oetiker | 2005-10-11 18:32:09 +0200 | 2 lines do not complain about strict aliasing issues -- Alex ------------------------------------------------------------------------ r697 | oetiker | 2005-10-08 23:58:57 +0200 | 2 lines allow rrd::graph to write directly to a tcl stream ... by Dave Bodenstab ------------------------------------------------------------------------ r696 | oetiker | 2005-10-07 09:48:52 +0200 | 2 lines updates for tcl bindings by -- Dave Bodenstab ------------------------------------------------------------------------ r695 | oetiker | 2005-10-07 09:48:01 +0200 | 2 lines included favicon in distro ------------------------------------------------------------------------ r691 | oetiker | 2005-09-26 22:33:45 +0200 | 2 lines plugged two memmory leaks happening when a requested font is not found. ------------------------------------------------------------------------ r690 | oetiker | 2005-09-24 01:06:12 +0200 | 3 lines hmpf its the gdef step we are talking about not the image step ! ------------------------------------------------------------------------ r689 | oetiker | 2005-09-22 22:18:12 +0200 | 2 lines do not 'reload' data if you already have it! ------------------------------------------------------------------------ r688 | oetiker | 2005-09-22 22:17:42 +0200 | 3 lines the -r option should not expect an argument ------------------------------------------------------------------------ r685 | oetiker | 2005-09-19 09:42:17 +0200 | 2 lines do not free stuff that gets freed later anyway ------------------------------------------------------------------------ r684 | oetiker | 2005-09-18 20:28:19 +0200 | 2 lines Netware Build Updates --- Guenter Knauf ------------------------------------------------------------------------ r683 | oetiker | 2005-09-18 18:59:34 +0200 | 3 lines allocate memory for one more row ... fetch can not return 0 reows ... ------------------------------------------------------------------------ r682 | oetiker | 2005-09-18 17:39:49 +0200 | 2 lines comparing floating point numbers is dangerouse ... don't do this directly! ------------------------------------------------------------------------ r681 | oetiker | 2005-09-01 23:23:34 +0200 | 3 lines since we use firstdir even without CHROOT it should better be defined in any case -- Guenter Knauf ------------------------------------------------------------------------ r680 | oetiker | 2005-08-30 07:52:41 +0200 | 3 lines When selecting the x-axis label take the length of the time shown in the graph into consideration just showing %H:%M in a two day presentation is a problem ... -- LAUKENS Niels ------------------------------------------------------------------------ r679 | oetiker | 2005-08-28 09:33:13 +0200 | 3 lines fixed location of example scripts ------------------------------------------------------------------------ r678 | oetiker | 2005-08-27 23:29:21 +0200 | 7 lines resolve subsecond resolution logging and handling of unknown values by starting off with an unknown value when building a pdp instead of starting out with 0 and then turning it into unknown later on if the number of uknown seconds is equal or larger than the interval ... ------------------------------------------------------------------------ r677 | oetiker | 2005-08-27 22:55:46 +0200 | 2 lines the rrdtool binary one more level up the path ... ------------------------------------------------------------------------ r676 | oetiker | 2005-08-27 22:52:12 +0200 | 2 lines update for new prefix setting ------------------------------------------------------------------------ r675 | oetiker | 2005-08-25 22:43:04 +0200 | 3 lines when resize converts a v1 rrd to v3 it should also change the version number in the head ... ------------------------------------------------------------------------ r674 | oetiker | 2005-08-23 23:18:23 +0200 | 3 lines fixes for compiling rrdtool with win32 mingw and netware -- Guenter Knauf ------------------------------------------------------------------------ r673 | oetiker | 2005-08-16 07:30:29 +0200 | 3 lines added notes about VDEF expressions not being general purpose ------------------------------------------------------------------------ r672 | oetiker | 2005-08-11 21:18:36 +0200 | 2 lines only bomb is there is no chroot and someone tries to use it. ------------------------------------------------------------------------ r669 | oetiker | 2005-08-04 07:08:04 +0200 | 3 lines New VDEF functions for least squares slope, intercept & correlation for simple forecasting -- Trent Burkard ------------------------------------------------------------------------ r668 | oetiker | 2005-08-04 06:59:01 +0200 | 2 lines fix rounding issues that prevented UNKNOWN to work as soon as subsecond resolution was used ------------------------------------------------------------------------ r667 | oetiker | 2005-08-04 06:58:11 +0200 | 2 lines avoid drawing lines where two points are in the same spot. this mai be confusing libart ------------------------------------------------------------------------ r666 | oetiker | 2005-08-02 17:54:32 +0200 | 2 lines make sure clean works even when the perl Makefile is broken ------------------------------------------------------------------------ r665 | oetiker | 2005-07-26 11:13:36 +0200 | 2 lines initialize font to empty string for deault font case. ------------------------------------------------------------------------ r663 | oetiker | 2005-07-25 16:12:58 +0200 | 1 line prepare for the release of rrdtool-1.2.11 ------------------------------------------------------------------------ r662 | oetiker | 2005-07-25 16:07:35 +0200 | 3 lines prepared for release ------------------------------------------------------------------------ r660 | oetiker | 2005-07-22 15:16:48 +0200 | 2 lines do not quit so easily when runing in interactive mode ------------------------------------------------------------------------ r659 | oetiker | 2005-07-22 15:16:18 +0200 | 2 lines init ytr after all the changes have happened ------------------------------------------------------------------------ r658 | oetiker | 2005-07-22 15:10:50 +0200 | 2 lines make --no-gridfit with --only-graph work ------------------------------------------------------------------------ r657 | oetiker | 2005-07-22 00:23:28 +0200 | 2 lines when stacking known on unknown, then we asume unknown was zero ... ------------------------------------------------------------------------ r654 | oetiker | 2005-07-20 09:20:04 +0200 | 2 lines make sure gcc-4.0 does not optimize our tests away :-) ------------------------------------------------------------------------ r653 | oetiker | 2005-07-19 12:00:13 +0200 | 2 lines fixed some corner cases in drawing code to get the same results as before with way more performance ------------------------------------------------------------------------ r652 | oetiker | 2005-07-12 00:49:09 +0200 | 2 lines optimized drawing routines ... up to 6 times faster (in the case of smokeping) ------------------------------------------------------------------------ r651 | oetiker | 2005-07-07 12:27:55 +0200 | 2 lines colors start with # not : ------------------------------------------------------------------------ r650 | oetiker | 2005-07-04 20:49:56 +0200 | 3 lines make setting just the size and not the font work -- Havard Eidnes ------------------------------------------------------------------------ r649 | oetiker | 2005-06-28 23:03:20 +0200 | 3 lines * the makefiles should not depend on gnumake features * make the perl module compilation environement configurable -- Stan Sinyagin ssinyagin on yahoo.com ------------------------------------------------------------------------ r646 | oetiker | 2005-06-25 07:24:52 +0200 | 2 lines added GNUMAKE and EGREP hints ------------------------------------------------------------------------ r645 | oetiker | 2005-06-20 11:15:10 +0200 | 3 lines x-axis labels should be aligned to their baseline alt-y-grid labels without fractionals should have no fractionals when printed ------------------------------------------------------------------------ r642 | oetiker | 2005-06-17 11:05:17 +0200 | 1 line prepare for the release of rrdtool-1.2.10 ------------------------------------------------------------------------ r641 | oetiker | 2005-06-17 00:09:36 +0200 | 2 lines when building the sgridstep in alt-y-grid mode, take the --base into consideration. ------------------------------------------------------------------------ r639 | oetiker | 2005-06-13 23:56:48 +0200 | 2 lines simplify code for generating y axis labels a bit. ------------------------------------------------------------------------ r638 | oetiker | 2005-06-13 15:20:45 +0200 | 2 lines wide chars should better be unsigned ------------------------------------------------------------------------ r637 | oetiker | 2005-06-13 14:56:11 +0200 | 2 lines fall back to assuming latin1 encoding if mbtowchar conversion fails ------------------------------------------------------------------------ r636 | oetiker | 2005-06-13 07:43:44 +0200 | 2 lines allow for VNAMES starting with a number and add documentation -- Alex ------------------------------------------------------------------------ r635 | oetiker | 2005-06-13 07:43:16 +0200 | 2 lines add notes about small numbers for --alt-y-grid ------------------------------------------------------------------------ r634 | oetiker | 2005-06-12 22:01:57 +0200 | 3 lines kill the last stored counter value if the updates comes after mrhb to make sure it does not get taken into account accedenly later on. ------------------------------------------------------------------------ r633 | oetiker | 2005-06-12 22:00:07 +0200 | 3 lines make --alt-y-grid work properly in many more cases than before ------------------------------------------------------------------------ r632 | oetiker | 2005-06-11 21:03:46 +0200 | 2 lines be more careful when checking if a file is an rrdfile ------------------------------------------------------------------------ r631 | oetiker | 2005-06-11 08:17:02 +0200 | 4 lines distribute all generated documentation install in share and not shared drop index.html link ------------------------------------------------------------------------ r630 | oetiker | 2005-06-11 08:14:25 +0200 | 2 lines install examples in an rrdtool specific subdirectory if the prefix does not contain rrd -- Peter Breitenlohner ------------------------------------------------------------------------ r629 | oetiker | 2005-06-11 07:41:31 +0200 | 3 lines avoid problems when building with VPATH set and --disable-perl fix make test for perl-piped -- Peter Breitenlohner ------------------------------------------------------------------------ r628 | oetiker | 2005-06-11 07:40:04 +0200 | 2 lines let 'out-of tree' builds work for everythin but the perl module -- Peter Breitenlohner ------------------------------------------------------------------------ r627 | oetiker | 2005-06-11 07:37:55 +0200 | 2 lines defined DPRINTF in a more compatible way -- Peter Breitenlohner ------------------------------------------------------------------------ r626 | oetiker | 2005-06-11 07:31:22 +0200 | 1 line link libcgi only against rrd_cgi since it is only needed there -- Peter Breitenlohner ------------------------------------------------------------------------ r625 | oetiker | 2005-06-11 07:29:45 +0200 | 1 line Make sure librrd_th properly depends on libpthread. -- Peter Breitenlohner ------------------------------------------------------------------------ r624 | oetiker | 2005-06-11 07:27:08 +0200 | 2 lines avoid gcc (3.4.3) warnings char format, different type arg -- Peter Breitenlohner ------------------------------------------------------------------------ r623 | oetiker | 2005-06-11 07:26:13 +0200 | 2 lines fix various problems with the tcl bindings -- Peter Breitenlohner ------------------------------------------------------------------------ r622 | oetiker | 2005-06-11 07:22:14 +0200 | 2 lines added peter breitenlohner ------------------------------------------------------------------------ r621 | oetiker | 2005-06-11 07:20:59 +0200 | 2 lines remove various gcc (3.4.3) warnings such as discards qualifier all due to 'char **' vs. 'const char **' discrepancies. -- Peter Breitenlohner peb with mppmu.mpg.de ------------------------------------------------------------------------ r620 | oetiker | 2005-06-11 07:15:37 +0200 | 2 lines re-innitializing sort of defeats the purpose ------------------------------------------------------------------------ r619 | oetiker | 2005-06-11 07:03:33 +0200 | 2 lines be more carefull in checking for sucessful open ------------------------------------------------------------------------ r618 | oetiker | 2005-06-10 21:25:36 +0200 | 2 lines vnames MUST start with a letter ------------------------------------------------------------------------ r617 | oetiker | 2005-06-10 21:24:06 +0200 | 2 lines kill another warning ------------------------------------------------------------------------ r616 | oetiker | 2005-06-10 20:44:40 +0200 | 6 lines make --font more flexible: * TITLE:30: just alter the size of the title * DEFAULT:0:/tmp/funfont.ttf alter the font but not the size ------------------------------------------------------------------------ r613 | oetiker | 2005-06-10 10:42:44 +0200 | 2 lines added missing cd and replace one path with a variable ------------------------------------------------------------------------ r612 | oetiker | 2005-06-08 00:10:01 +0200 | 2 lines make pdf/eps/svg formats utf8 aware too -- Peter Speck ------------------------------------------------------------------------ r611 | oetiker | 2005-06-06 07:32:44 +0200 | 2 lines sync up with Dag Wieers version from http://dag.wieers.com/packages/rrdtool/rrdtool.spec ------------------------------------------------------------------------ r610 | oetiker | 2005-06-06 00:23:32 +0200 | 3 lines New functions for CDEF ATAN2 RAD2DEG and DEG2RAD -- Simon Melhuish ------------------------------------------------------------------------ r605 | oetiker | 2005-05-31 08:22:31 +0200 | 2 lines fixed epoch ------------------------------------------------------------------------ r603 | oetiker | 2005-05-28 15:01:43 +0200 | 1 line prepare for the release of rrdtool-1.2.9 ------------------------------------------------------------------------ r602 | oetiker | 2005-05-28 14:53:13 +0200 | 2 lines add notes on building on Mac OS X ------------------------------------------------------------------------ r601 | oetiker | 2005-05-28 14:52:50 +0200 | 3 lines kill a few warnings ------------------------------------------------------------------------ r600 | oetiker | 2005-05-28 14:52:26 +0200 | 2 lines make the bindings work on os x ------------------------------------------------------------------------ r599 | oetiker | 2005-05-27 21:17:17 +0200 | 2 lines add note on variable names ------------------------------------------------------------------------ r598 | oetiker | 2005-05-27 21:12:23 +0200 | 2 lines remove one more warning ------------------------------------------------------------------------ r597 | oetiker | 2005-05-27 21:12:12 +0200 | 2 lines there is realy no reason to have the vnames constraind so much ------------------------------------------------------------------------ r596 | oetiker | 2005-05-27 21:01:53 +0200 | 3 lines only print ttf font errors when compiled with -DDEBUG ------------------------------------------------------------------------ r593 | oetiker | 2005-05-23 00:29:11 +0200 | 2 lines don't go into an endless loop upon finding a bad character in a truetype font ------------------------------------------------------------------------ r591 | oetiker | 2005-05-20 08:20:19 +0200 | 2 lines go into rrdbuild too if we use tcsh ------------------------------------------------------------------------ r590 | oetiker | 2005-05-20 07:48:52 +0200 | 2 lines initial checking ------------------------------------------------------------------------ r589 | oetiker | 2005-05-20 07:46:52 +0200 | 2 lines remove extra whitespace at the end of lines ------------------------------------------------------------------------ r588 | oetiker | 2005-05-20 07:43:27 +0200 | 2 lines introduce builddir ------------------------------------------------------------------------ r586 | oetiker | 2005-05-20 07:14:20 +0200 | 2 lines iconf is not needed so to not check for it ------------------------------------------------------------------------ r585 | oetiker | 2005-05-20 01:21:51 +0200 | 1 line added propper numeric version number to version ------------------------------------------------------------------------ r584 | oetiker | 2005-05-20 01:21:20 +0200 | 4 lines add multibyte to wide char conversion ability. Now every string users input should should be handled propperly by rrdtool, whatever your locale says will work. ------------------------------------------------------------------------ r583 | oetiker | 2005-05-20 01:19:09 +0200 | 1 line windows at least stores null terminated char strings in dummychar1 -- Philippe ------------------------------------------------------------------------ r581 | oetiker | 2005-05-18 22:20:55 +0200 | 2 lines make tcl compile again ... don't know if it will work though ------------------------------------------------------------------------ r580 | oetiker | 2005-05-18 21:43:03 +0200 | 2 lines fix rrd_version prototype ------------------------------------------------------------------------ r579 | oetiker | 2005-05-18 21:41:08 +0200 | 2 lines cleaned up version number code and added new function rrd_version -- partly by Burton Strauss ------------------------------------------------------------------------ r578 | oetiker | 2005-05-18 21:18:59 +0200 | 3 lines help building perl on hpux ------------------------------------------------------------------------ r571 | oetiker | 2005-05-16 22:56:09 +0200 | 1 line prepare for the release of rrdtool-1.2.8 ------------------------------------------------------------------------ r570 | oetiker | 2005-05-16 22:55:29 +0200 | 4 lines redo --disable-* options for python and tcl add some more quotes to protect potentially empty variables in tests which should make configure more robust. ------------------------------------------------------------------------ r569 | oetiker | 2005-05-16 22:04:42 +0200 | 2 lines if python stuff does not work, do not die, just do not compile python ------------------------------------------------------------------------ r568 | oetiker | 2005-05-16 22:02:26 +0200 | 2 lines fix extra argument when testing for PERLCC problem -- Simon Leinen ------------------------------------------------------------------------ r565 | oetiker | 2005-05-16 14:18:16 +0200 | 1 line prepare for the release of rrdtool-1.2.7 ------------------------------------------------------------------------ r564 | oetiker | 2005-05-16 14:14:19 +0200 | 3 lines added note about --alt-y-mrtg ------------------------------------------------------------------------ r563 | oetiker | 2005-05-16 12:12:26 +0200 | 3 lines fix the color square spacing so that it works regardles of the fontsize and with transparend backgrounds. ------------------------------------------------------------------------ r562 | oetiker | 2005-05-16 11:01:01 +0200 | 2 lines added alt-y-mrtg option as a dummy so that old apps who use it do not crash when doing so. ------------------------------------------------------------------------ r561 | oetiker | 2005-05-16 08:33:30 +0200 | 2 lines reintroduced the FRAME color ... somehow it was lost from 1.0.x ------------------------------------------------------------------------ r560 | oetiker | 2005-05-15 16:16:15 +0200 | 1 line make transparent backgrounds work propperly ------------------------------------------------------------------------ r559 | oetiker | 2005-05-14 23:57:46 +0200 | 2 lines it seems that _WIN32 is more commonly defined thatn WIN32 on win32 ... (gifford.hesketh on gmail.com) ------------------------------------------------------------------------ r558 | oetiker | 2005-05-14 19:32:05 +0200 | 3 lines added a note to dump old rrd files with a current rrdtool dump to get the upgraded on the fly. Adding 001 support should not be all that tough though. Any takers ? ------------------------------------------------------------------------ r557 | oetiker | 2005-05-14 17:07:35 +0200 | 1 line check cookie read for success ------------------------------------------------------------------------ r556 | oetiker | 2005-05-14 11:35:02 +0200 | 5 lines if the y value is out of bounds, make it just a wee bit larger or smaller than the bound so that the code can detect this by comparing it with the bound, but make it so small that it is not visible. ------------------------------------------------------------------------ r555 | oetiker | 2005-05-13 00:27:06 +0200 | 1 line one sed is quite enough ------------------------------------------------------------------------ r554 | oetiker | 2005-05-13 00:22:45 +0200 | 2 lines [ must be quoted by [ ... ------------------------------------------------------------------------ r553 | oetiker | 2005-05-13 00:12:56 +0200 | 2 lines add global option ------------------------------------------------------------------------ r552 | oetiker | 2005-05-13 00:06:13 +0200 | 2 lines fox sed line for unsetting header cache ------------------------------------------------------------------------ r551 | oetiker | 2005-05-13 00:03:33 +0200 | 2 lines make sure the header gets tested twice ------------------------------------------------------------------------ r550 | oetiker | 2005-05-13 00:03:01 +0200 | 1 line add notes on python error, and install location ------------------------------------------------------------------------ r549 | oetiker | 2005-05-12 13:32:39 +0200 | 1 line allow 3/4 component color rrdtool graph --color too -- Alex ------------------------------------------------------------------------ r548 | oetiker | 2005-05-12 13:28:13 +0200 | 2 lines make the areas a little bigger still for propper overlapping ------------------------------------------------------------------------ r547 | oetiker | 2005-05-12 13:22:11 +0200 | 3 lines make parsing of DS arguments more robust and give more sensible error messages. -- Alex ------------------------------------------------------------------------ r546 | oetiker | 2005-05-12 10:38:53 +0200 | 4 lines make the areas that makeup the graphs a little larger, so that they overlap a bit and do not exhibit white lines when touching each other. ------------------------------------------------------------------------ r545 | oetiker | 2005-05-12 07:48:56 +0200 | 2 lines some more warning fixes ------------------------------------------------------------------------ r544 | oetiker | 2005-05-12 07:41:09 +0200 | 2 lines add unused to prevent warnings about self ------------------------------------------------------------------------ r543 | oetiker | 2005-05-11 20:31:00 +0200 | 2 lines intial python bindings documentation ------------------------------------------------------------------------ r541 | oetiker | 2005-05-11 14:09:36 +0200 | 2 lines fix wording for min/max ... they are not optional ------------------------------------------------------------------------ r540 | oetiker | 2005-05-10 22:19:55 +0200 | 2 lines second installment with warning fixes from alan ------------------------------------------------------------------------ r539 | oetiker | 2005-05-10 20:52:43 +0200 | 2 lines python bindings add -- Alan Milligan alan from balclutha.org ------------------------------------------------------------------------ r538 | oetiker | 2005-05-10 20:35:55 +0200 | 2 lines fix for character escaping in EPS -- Peter Speck ------------------------------------------------------------------------ r536 | oetiker | 2005-05-10 08:07:41 +0200 | 1 line prepare for the release of rrdtool-1.2.6 ------------------------------------------------------------------------ r535 | oetiker | 2005-05-10 08:06:49 +0200 | 2 lines I guess the pdf entities want to be escaped chars not just control caracters with a prepended backslash ... ------------------------------------------------------------------------ r534 | oetiker | 2005-05-10 07:18:28 +0200 | 2 lines we should not use C++ comments in c files ------------------------------------------------------------------------ r533 | oetiker | 2005-05-10 06:14:17 +0200 | 2 lines we should distribute unused.h if we use it :-) ------------------------------------------------------------------------ r531 | oetiker | 2005-05-09 22:15:38 +0200 | 1 line prepare for the release of rrdtool-1.2.5 ------------------------------------------------------------------------ r530 | oetiker | 2005-05-09 22:13:28 +0200 | 2 lines silence warnings for UNUSED variables without changeing the funtion interfaces -- Alex ------------------------------------------------------------------------ r529 | oetiker | 2005-05-09 21:32:12 +0200 | 2 lines allow decimal positions y axis labels as soon as the max entry is < 10 ------------------------------------------------------------------------ r528 | oetiker | 2005-05-09 21:11:05 +0200 | 2 lines fix rendering of the color spots especially when proportional fonts are in use. ------------------------------------------------------------------------ r526 | oetiker | 2005-05-09 16:01:21 +0200 | 3 lines EPS/SVG/PDF code ... more comments and better font finder including removal of hardcoded SVG font. -- Peter Speck speck from vitality.dk ------------------------------------------------------------------------ r525 | oetiker | 2005-05-09 15:22:23 +0200 | 3 lines * switch to courier if the font is unknown * fix font positioning ------------------------------------------------------------------------ r523 | oetiker | 2005-05-09 07:39:59 +0200 | 2 lines perls $^O is set to linux not Linux ------------------------------------------------------------------------ r522 | oetiker | 2005-05-09 00:02:35 +0200 | 2 lines svg/pdf/eps rendering improvements. There are still issues though ... -- peter speck ------------------------------------------------------------------------ r521 | oetiker | 2005-05-08 23:25:42 +0200 | 2 lines added support for getting ascent/descent size info for ps fonts -- Peter Speck ------------------------------------------------------------------------ r520 | oetiker | 2005-05-08 22:15:18 +0200 | 2 lines we do not actually need these files ... goodbye ------------------------------------------------------------------------ r518 | oetiker | 2005-05-08 18:59:20 +0200 | 1 line prepare for the release of rrdtool-1.2.4 ------------------------------------------------------------------------ r517 | oetiker | 2005-05-08 18:22:25 +0200 | 4 lines * adjust spacing at the left border of the graph. use less white space * increasse minimal grid spacing and label spacing ------------------------------------------------------------------------ r516 | oetiker | 2005-05-08 18:02:28 +0200 | 2 lines increase minimal vertical spacing for y-axis lables ------------------------------------------------------------------------ r515 | oetiker | 2005-05-08 18:01:40 +0200 | 2 lines shorten space for y-axis labels a bit ------------------------------------------------------------------------ r514 | oetiker | 2005-05-08 17:40:28 +0200 | 2 lines remove the premature fox for the unused parameter in xport ------------------------------------------------------------------------ r513 | oetiker | 2005-05-08 17:30:12 +0200 | 2 lines sync with reality ------------------------------------------------------------------------ r512 | oetiker | 2005-05-08 17:19:02 +0200 | 2 lines be more portable with shorter strings ------------------------------------------------------------------------ r511 | oetiker | 2005-05-08 17:18:47 +0200 | 2 lines we do not use the affine functions ------------------------------------------------------------------------ r510 | oetiker | 2005-05-08 17:07:27 +0200 | 2 lines removed extra $ from from PERLCC test ------------------------------------------------------------------------ r509 | oetiker | 2005-05-08 10:20:48 +0200 | 2 lines fix rendering errors by removing single spaces ------------------------------------------------------------------------ r507 | oetiker | 2005-05-08 01:54:52 +0200 | 1 line prepare for the release of rrdtool-1.2.3 ------------------------------------------------------------------------ r506 | oetiker | 2005-05-08 01:26:30 +0200 | 2 lines fix some more y0, y1 complaints ------------------------------------------------------------------------ r505 | oetiker | 2005-05-08 01:26:14 +0200 | 3 lines When encountering an empty legend, do NOT make space for drawing the color spats in front of it. ------------------------------------------------------------------------ r503 | oetiker | 2005-05-07 17:57:55 +0200 | 1 line prep for 1.2.2 release ------------------------------------------------------------------------ r502 | oetiker | 2005-05-07 17:51:47 +0200 | 2 lines increment library revision ------------------------------------------------------------------------ r501 | oetiker | 2005-05-07 17:45:52 +0200 | 2 lines fix these annoying y0 warnings. ------------------------------------------------------------------------ r500 | oetiker | 2005-05-07 17:45:27 +0200 | 1 line add rrdtool attribution string back using a transparent version of the font color ------------------------------------------------------------------------ r499 | oetiker | 2005-05-07 17:23:41 +0200 | 2 lines added build instructions ------------------------------------------------------------------------ r498 | oetiker | 2005-05-07 12:38:34 +0200 | 2 lines fix for rotation code in pdf,eps,svg formats -- Peter Speck ------------------------------------------------------------------------ r497 | oetiker | 2005-05-07 12:34:48 +0200 | 2 lines fix handling of PREV(x) command ------------------------------------------------------------------------ r493 | oetiker | 2005-05-07 01:00:53 +0200 | 5 lines * added RRDTOOL/TOBI OETIKER back in * in riggid mode, cut data and 'min' and 'max' not two above * gibt ytop0 a default value ------------------------------------------------------------------------ r492 | oetiker | 2005-05-07 00:59:22 +0200 | 2 lines fixed documentation strings for rrdgraph ------------------------------------------------------------------------ r491 | oetiker | 2005-05-07 00:03:16 +0200 | 2 lines added doc for new slope-mode ------------------------------------------------------------------------ r490 | oetiker | 2005-05-06 23:52:55 +0200 | 3 lines since the drawing code is now much simpler we do not need to perturbe the points or rewind the graph anymore this saves quite some time ... ------------------------------------------------------------------------ r489 | oetiker | 2005-05-06 17:44:11 +0200 | 6 lines * allow for 3(4) letter color names * better debugging code * fix tick, line, stack parsing -- Alex van den Bogaerdt ------------------------------------------------------------------------ r488 | oetiker | 2005-05-06 17:24:27 +0200 | 2 lines if stacking on nothing we should not try to figure out WHAT we stacked on since this will result in a segfault ... -- Alex van den Bogaerdt ------------------------------------------------------------------------ r485 | oetiker | 2005-05-06 00:15:10 +0200 | 3 lines clarify LINE/AREA syntax -- Alex van den Bogaerdt alex with ergens.op.het.net ------------------------------------------------------------------------ r484 | oetiker | 2005-05-05 11:07:56 +0200 | 2 lines better not to write into argv elements ... Henrik Stoerner ------------------------------------------------------------------------ r483 | oetiker | 2005-05-03 22:23:05 +0200 | 1 line do not die upon wront input when in remote mode ------------------------------------------------------------------------ r482 | oetiker | 2005-05-03 20:10:16 +0200 | 4 lines install documentation into $PREFIX/shared/doc/rrdtool-$VERSION/{txt,html} but drop the rrdtool-$VERSION/ bit if $PREFIX matches on rrd assuming that the tool is being installed into a custom directory anyway. ------------------------------------------------------------------------ r481 | oetiker | 2005-05-03 19:24:50 +0200 | 4 lines new options to tune how fontsmoothing works: -R/--font-render-mode {normal,light,mono} and -B/--font-smoothing-threshold size -- John yffffffff4271ef37 at f4n.org ------------------------------------------------------------------------ r480 | oetiker | 2005-05-03 19:22:54 +0200 | 2 lines note about win32comp ------------------------------------------------------------------------ r478 | oetiker | 2005-05-02 22:27:13 +0200 | 1 line files that help getting that native windows port going -- Philippe.Simonet with swisscom.com ------------------------------------------------------------------------ r477 | oetiker | 2005-05-02 21:47:05 +0200 | 2 lines when walking the bytes use the pitch not the width ------------------------------------------------------------------------ r476 | oetiker | 2005-05-02 21:33:53 +0200 | 2 lines pow is alreay used as a function ... so lets call it power instead ------------------------------------------------------------------------ r475 | oetiker | 2005-05-02 21:32:49 +0200 | 2 lines no infinite line width please ------------------------------------------------------------------------ r472 | oetiker | 2005-05-02 00:07:34 +0200 | 1 line prep for 1.2.1 release ------------------------------------------------------------------------ r471 | oetiker | 2005-05-02 00:04:34 +0200 | 2 lines add some more vertical space above the graph if no title is available ------------------------------------------------------------------------ r470 | oetiker | 2005-05-01 23:24:06 +0200 | 2 lines use the pen to determine string length, do not look at the realy length, or space will get ignored. ------------------------------------------------------------------------ r469 | oetiker | 2005-05-01 23:16:49 +0200 | 3 lines DejaVuSansMono-Roman is a variant of VeraMono with lots of added unicode glyphes ... check out http://dejavu.sourceforge.net ------------------------------------------------------------------------ r468 | oetiker | 2005-05-01 22:49:54 +0200 | 2 lines fix spaceing ... more space for the title and more space the the left of the yaxis lable ------------------------------------------------------------------------ r467 | oetiker | 2005-05-01 15:19:39 +0200 | 2 lines the multibyte support tried did not work ... takeing it out again ------------------------------------------------------------------------ r466 | oetiker | 2005-05-01 15:18:48 +0200 | 2 lines update tested library version ------------------------------------------------------------------------ r465 | oetiker | 2005-05-01 13:02:40 +0200 | 1 line undef jmpbuf to make compile work on AIX 5.1 ------------------------------------------------------------------------ r464 | oetiker | 2005-05-01 13:00:36 +0200 | 2 lines increasse revision of shared library number since the library source code has changed ------------------------------------------------------------------------ r463 | oetiker | 2005-05-01 13:00:10 +0200 | 2 lines add wide character support ------------------------------------------------------------------------ r462 | oetiker | 2005-05-01 12:42:17 +0200 | 3 lines do not make the graph wider if the title does not fit ... this is wagging the dog the graph is not ABOUT the title ------------------------------------------------------------------------ r461 | oetiker | 2005-05-01 12:37:00 +0200 | 1 line fix description of VRULE ------------------------------------------------------------------------ r460 | oetiker | 2005-05-01 12:23:11 +0200 | 2 lines reset optind opterr inside the function calls ... ------------------------------------------------------------------------ r459 | oetiker | 2005-05-01 12:22:35 +0200 | 2 lines fixed documentation regarding floating point line width ------------------------------------------------------------------------ r458 | oetiker | 2005-05-01 12:20:55 +0200 | 4 lines relax parsing of LINE commands LINE: is valid as well as LINE: eg. LINE1.2323: ------------------------------------------------------------------------ r457 | oetiker | 2005-05-01 11:16:38 +0200 | 2 lines allow for characters over 127 ------------------------------------------------------------------------ r456 | oetiker | 2005-04-30 11:56:58 +0200 | 3 lines added notes on escaping colons in COMMENT and put deprecated forms of command to the back and not to the front ------------------------------------------------------------------------ r455 | oetiker | 2005-04-30 11:51:12 +0200 | 2 lines always open with rb/wb when it is a binary file ... unix and windows will work alike with this ... ------------------------------------------------------------------------ r454 | oetiker | 2005-04-30 11:48:49 +0200 | 2 lines this file is not required ... ------------------------------------------------------------------------ r453 | oetiker | 2005-04-30 11:41:14 +0200 | 2 lines even better lets have rb for CYGWIN too ... one never knows it is windows after all. ------------------------------------------------------------------------ r452 | oetiker | 2005-04-30 11:40:24 +0200 | 2 lines oops inverted code here ... windows needs rb unix not ------------------------------------------------------------------------ r451 | oetiker | 2005-04-30 11:38:50 +0200 | 2 lines let rrd_graph honor the font name (RRD_DEFAULT_FONT) defined in confignt/config.h ------------------------------------------------------------------------ r450 | oetiker | 2005-04-27 00:08:22 +0200 | 1 line use pod syntax not html syntax in pod :-) ------------------------------------------------------------------------ r449 | oetiker | 2005-04-27 00:04:25 +0200 | 2 lines more fixes by fritz ------------------------------------------------------------------------ r443 | oetiker | 2005-04-26 07:33:18 +0200 | 2 lines obfuscate email addresses ------------------------------------------------------------------------ r442 | oetiker | 2005-04-26 07:30:22 +0200 | 5 lines * remove jakes old address * add security warning ------------------------------------------------------------------------ r440 | oetiker | 2005-04-26 00:55:15 +0200 | 2 lines remove pdf link ------------------------------------------------------------------------ r436 | oetiker | 2005-04-26 00:18:15 +0200 | 4 lines use time_t to improve portability to 64bit systems -- John R Mocho ------------------------------------------------------------------------ r429 | oetiker | 2005-04-25 23:01:22 +0200 | 2 lines added new state ------------------------------------------------------------------------ r428 | oetiker | 2005-04-25 22:45:00 +0200 | 1 line prep for 1.2.0 release ------------------------------------------------------------------------ r426 | oetiker | 2005-04-25 21:07:18 +0200 | 2 lines resolved all XXX situations ------------------------------------------------------------------------ r425 | oetiker | 2005-04-25 19:31:11 +0200 | 2 lines another batch of fixes from fritz ------------------------------------------------------------------------ r423 | oetiker | 2005-04-25 00:21:34 +0200 | 1 line fixed contributiors listing ------------------------------------------------------------------------ r422 | oetiker | 2005-04-24 19:23:57 +0200 | 2 lines big patch-up for the 1.2. Documents Revisited by Fritz Zaucker ------------------------------------------------------------------------ r420 | oetiker | 2005-04-22 20:35:15 +0200 | 1 line prep for 1.2rc9 release ------------------------------------------------------------------------ r417 | oetiker | 2005-04-19 23:09:27 +0200 | 1 line fix titles ------------------------------------------------------------------------ r416 | oetiker | 2005-04-19 23:05:39 +0200 | 2 lines fixed title ------------------------------------------------------------------------ r414 | oetiker | 2005-04-19 00:39:10 +0200 | 3 lines src moved back to pod. this include thing was a neat idea but it just adds another twist to things which is not necessary. fixed some pod errors ------------------------------------------------------------------------ r413 | oetiker | 2005-04-19 00:13:20 +0200 | 2 lines misc fixes for better display ------------------------------------------------------------------------ r412 | oetiker | 2005-04-18 13:07:02 +0200 | 2 lines fixed format ------------------------------------------------------------------------ r411 | oetiker | 2005-04-18 00:43:11 +0200 | 3 lines fixed many pod bugs while creating the new website ------------------------------------------------------------------------ r410 | oetiker | 2005-04-17 19:29:29 +0200 | 2 lines make sure argument descriptions comply with Bold = fixed and Italics = variable ------------------------------------------------------------------------ r409 | oetiker | 2005-04-17 18:26:18 +0200 | 2 lines fix use of =item commands ------------------------------------------------------------------------ r408 | oetiker | 2005-04-17 18:25:53 +0200 | 2 lines allow to build pods (for website) ------------------------------------------------------------------------ r406 | oetiker | 2005-04-14 00:00:20 +0200 | 4 lines * draw the axis last to have them always on top * use a smaller font for labeling the axis ------------------------------------------------------------------------ r405 | oetiker | 2005-04-13 21:53:41 +0200 | 2 lines have better default compression ------------------------------------------------------------------------ r404 | oetiker | 2005-04-13 21:53:16 +0200 | 1 line tune fontsizes and spacing to be more in line with rrdtool 1.0 ------------------------------------------------------------------------ r402 | oetiker | 2005-04-12 23:52:18 +0200 | 1 line prep for 1.2rc8 release ------------------------------------------------------------------------ r401 | oetiker | 2005-04-12 23:10:39 +0200 | 1 line prep for 1.2rc8 release ------------------------------------------------------------------------ r400 | oetiker | 2005-04-12 22:56:19 +0200 | 5 lines * better fix for units-exponent issue ... now the whole scaling happens only at print time ... this is more stable -- tobi * remove the newline from ctime (%c) in VDEF -- alex ------------------------------------------------------------------------ r398 | oetiker | 2005-04-12 01:23:59 +0200 | 1 line prep for 1.2rc7 release ------------------------------------------------------------------------ r397 | oetiker | 2005-04-12 01:22:57 +0200 | 6 lines * adjusted element spacing within the graph * re-intruduced --units-length option * stabilized --units-exponent option by jumping to alt-y-grid mode once the default autogrid can not cope with the data anymore. ------------------------------------------------------------------------ r396 | oetiker | 2005-04-12 01:20:52 +0200 | 1 line added note about incompatible changes ------------------------------------------------------------------------ r395 | oetiker | 2005-04-11 14:29:32 +0200 | 2 lines started change log ------------------------------------------------------------------------ r394 | oetiker | 2005-04-11 14:17:36 +0200 | 1 line added notes on INF ------------------------------------------------------------------------ r393 | oetiker | 2005-04-11 11:51:22 +0200 | 4 lines * we are now creating true RGBA pngs * removed all memory leeks from rrd_gfx ------------------------------------------------------------------------ r392 | oetiker | 2005-04-11 00:02:10 +0200 | 2 lines started updating for 1.2 release ------------------------------------------------------------------------ r391 | oetiker | 2005-04-10 20:11:58 +0200 | 1 line clarified output -- alex ------------------------------------------------------------------------ r390 | oetiker | 2005-04-10 20:10:56 +0200 | 1 line get the docs ready for the 1.2 release. remove notes about things that never got implemented. -- alex ------------------------------------------------------------------------ r389 | oetiker | 2005-04-10 20:05:51 +0200 | 1 line do not draw AREAS that have no hight (done right). draw an arrow on the positiv y axis. ------------------------------------------------------------------------ r388 | oetiker | 2005-04-10 17:35:28 +0200 | 1 line updated for 1.2 release -- alex ------------------------------------------------------------------------ r387 | oetiker | 2005-04-10 16:00:12 +0200 | 1 line fixed typo -- alex ------------------------------------------------------------------------ r386 | oetiker | 2005-04-10 15:57:34 +0200 | 2 lines G does not exist E does ------------------------------------------------------------------------ r385 | oetiker | 2005-04-10 14:12:21 +0200 | 2 lines fixed spelling and working -- Alex van den Bogaerdt alex at ergens.op.het.net ------------------------------------------------------------------------ r383 | oetiker | 2005-04-10 13:53:01 +0200 | 1 line added extra strftime.[ch] which supports ISO 8601 week numbers, together with instructions in the NT-BUILD-TIPPS file ------------------------------------------------------------------------ r381 | oetiker | 2005-04-10 13:33:10 +0200 | 1 line prep for 1.2rc6 release ------------------------------------------------------------------------ r380 | oetiker | 2005-04-10 13:31:32 +0200 | 2 lines remove excess comment ------------------------------------------------------------------------ r379 | oetiker | 2005-04-10 13:30:14 +0200 | 4 lines make --disable-perl work allow configuration of the default font at configure time ------------------------------------------------------------------------ r378 | oetiker | 2005-04-10 13:29:16 +0200 | 1 line revamped drawing routines to be less complex and more stable. be kind to libart ------------------------------------------------------------------------ r377 | oetiker | 2005-04-10 11:38:38 +0200 | 1 line allow --color arguments with RRGGBB and not only RRGGBBAA ------------------------------------------------------------------------ r376 | oetiker | 2005-04-10 11:16:10 +0200 | 1 line everyone has other names for their shared libraries, so lets drop the dependency ------------------------------------------------------------------------ r375 | oetiker | 2005-04-10 11:15:19 +0200 | 1 line mention updatev in the synopsis ------------------------------------------------------------------------ r374 | oetiker | 2005-04-09 16:21:59 +0200 | 2 lines we do have 95% ------------------------------------------------------------------------ r372 | oetiker | 2005-04-07 23:35:32 +0200 | 1 line prep for 1.2rc5 release ------------------------------------------------------------------------ r371 | oetiker | 2005-04-07 23:31:37 +0200 | 1 line fix implementation of --only-graph ------------------------------------------------------------------------ r370 | oetiker | 2005-04-07 23:31:09 +0200 | 1 line make libart happy by NOT adjusting coordinates and by using the ART_WIND_RULE_NONZERO winding rule ------------------------------------------------------------------------ r369 | oetiker | 2005-04-07 23:30:16 +0200 | 1 line return propper exit codes ------------------------------------------------------------------------ r368 | oetiker | 2005-04-07 22:07:45 +0200 | 2 lines fix bitpatterns of extraflags so that they do not interfear ------------------------------------------------------------------------ r366 | oetiker | 2005-04-07 01:58:41 +0200 | 1 line prep for 1.2rc4 release ------------------------------------------------------------------------ r365 | oetiker | 2005-04-07 01:24:35 +0200 | 1 line prep for 1.2rc4 release ------------------------------------------------------------------------ r364 | oetiker | 2005-04-07 01:18:37 +0200 | 2 lines document RRD_DEFAULT_FONT environment variable ------------------------------------------------------------------------ r363 | oetiker | 2005-04-07 00:40:53 +0200 | 2 lines make sure WIN32 sections do NOT kick in for cygwin compiles ------------------------------------------------------------------------ r362 | oetiker | 2005-04-06 00:49:55 +0200 | 2 lines only unwind the areas, the lines should be fine by themselves! ------------------------------------------------------------------------ r360 | oetiker | 2005-04-05 22:32:39 +0200 | 1 line prep for 1.2rc3 release ------------------------------------------------------------------------ r359 | oetiker | 2005-04-05 22:17:45 +0200 | 4 lines * fix transparency rendering by rewinding the paths propperly * fix x-axis label drawing by considering label precision ------------------------------------------------------------------------ r358 | oetiker | 2005-04-05 21:14:16 +0200 | 1 line new originaly copy ------------------------------------------------------------------------ r357 | oetiker | 2005-04-05 21:12:54 +0200 | 2 lines fixed propperties to have a RAW font ------------------------------------------------------------------------ r353 | oetiker | 2005-04-04 23:56:30 +0200 | 1 line make configure work even when no prefix is specified ------------------------------------------------------------------------ r352 | oetiker | 2005-04-04 23:40:11 +0200 | 1 line prep for 1.2rc2 release ------------------------------------------------------------------------ r351 | oetiker | 2005-04-04 23:22:05 +0200 | 1 line prepared for 1.2rc2 release ------------------------------------------------------------------------ r350 | oetiker | 2005-04-04 23:21:43 +0200 | 1 line prepared for 1.2rc1 release ------------------------------------------------------------------------ r349 | oetiker | 2005-04-04 23:12:21 +0200 | 1 line prepared for 1.2rc1 release ------------------------------------------------------------------------ r348 | oetiker | 2005-04-04 23:07:55 +0200 | 1 line prepared for 1.2rc1 release ------------------------------------------------------------------------ r347 | oetiker | 2005-04-04 23:05:32 +0200 | 1 line prepared for 1.2rc1 release ------------------------------------------------------------------------ r345 | oetiker | 2005-04-04 22:56:49 +0200 | 1 line fixed prefix detection code again ------------------------------------------------------------------------ r344 | oetiker | 2005-04-04 22:53:23 +0200 | 1 line fixed autoprefix ------------------------------------------------------------------------ r342 | oetiker | 2005-04-04 22:48:58 +0200 | 2 lines fixed config script for prefixing ------------------------------------------------------------------------ r340 | oetiker | 2005-04-04 22:37:26 +0200 | 2 lines get ready for 1.2 ------------------------------------------------------------------------ r338 | oetiker | 2005-04-04 21:55:30 +0200 | 1 line branche for 1.2 ------------------------------------------------------------------------ r336 | oetiker | 2005-04-03 14:44:15 +0200 | 2 lines update build instructions and configure output. ------------------------------------------------------------------------ r335 | oetiker | 2005-04-03 13:44:58 +0200 | 2 lines paramters in SHIFT are separated by : like everywhere else ------------------------------------------------------------------------ r334 | oetiker | 2005-04-03 13:36:58 +0200 | 3 lines the piechart code does not produce release ready results ... hide it behind WITH_PIECHART define ------------------------------------------------------------------------ r333 | oetiker | 2005-04-03 13:11:50 +0200 | 2 lines improve png writing speed by lowering the compression ------------------------------------------------------------------------ r332 | oetiker | 2005-04-03 13:11:21 +0200 | 2 lines make sure the font path is shorter than the buffer ------------------------------------------------------------------------ r331 | oetiker | 2005-04-03 12:10:56 +0200 | 3 lines fix graph layout ... especially consider text propperties propperly ------------------------------------------------------------------------ r330 | oetiker | 2005-04-03 11:37:08 +0200 | 3 lines the legend string of a PRINT command gets renderd later by runnning it through printf. ------------------------------------------------------------------------ r329 | oetiker | 2005-04-03 11:35:52 +0200 | 4 lines MMAP support for rrdtool this is suposed to speed-up uptime 4 times. -- Radoslaw Karas ------------------------------------------------------------------------ r328 | oetiker | 2005-03-20 22:17:50 +0100 | 1 line revert to last working version ------------------------------------------------------------------------ r327 | oetiker | 2005-03-20 22:11:17 +0100 | 1 line fix quoting ------------------------------------------------------------------------ r326 | oetiker | 2005-03-20 22:00:13 +0100 | 2 lines quote sed expression ------------------------------------------------------------------------ r325 | oetiker | 2005-03-20 21:59:31 +0100 | 5 lines document tabbing support by adding in the textformatting notes document --tabwidth argument document AXIS color ------------------------------------------------------------------------ r324 | oetiker | 2005-03-20 21:58:28 +0100 | 8 lines * allow configuration of the default tab width * added new configurable color AXIS * made tabbing support work with legend elements * use propper font for the legend: LEGEND * fix apperance by makeing grid more transparent * draw axis over the graph not under it. ------------------------------------------------------------------------ r323 | oetiker | 2005-03-20 21:56:16 +0100 | 1 line added tabbig support ------------------------------------------------------------------------ r322 | oetiker | 2005-03-20 21:55:27 +0100 | 1 line include the thread defs after they have been found not before ------------------------------------------------------------------------ r320 | oetiker | 2005-03-13 17:34:27 +0100 | 2 lines propperly integrate rrd_first ------------------------------------------------------------------------ r319 | oetiker | 2005-03-13 17:13:49 +0100 | 1 line autotools still run after configure this is not good ... run autoreconf at the end of MakeMakefile. Maybe this will help ------------------------------------------------------------------------ r318 | oetiker | 2005-03-13 17:09:21 +0100 | 1 line set install patch to version ------------------------------------------------------------------------ r317 | oetiker | 2005-03-13 17:08:43 +0100 | 1 line re-run autotools to make sure dependent dates are correct ------------------------------------------------------------------------ r316 | oetiker | 2005-03-13 17:00:51 +0100 | 1 line reintroduce final header regen ------------------------------------------------------------------------ r315 | oetiker | 2005-03-13 16:57:37 +0100 | 1 line still trying to get the order of calling autotools right ------------------------------------------------------------------------ r314 | oetiker | 2005-03-13 16:54:06 +0100 | 1 line create aclocal.m4 before libtoolize ------------------------------------------------------------------------ r313 | oetiker | 2005-03-13 16:51:06 +0100 | 1 line make the headers first ------------------------------------------------------------------------ r312 | oetiker | 2005-03-13 16:16:05 +0100 | 1 line add the documentation for first ------------------------------------------------------------------------ r311 | oetiker | 2005-03-13 16:14:31 +0100 | 1 line ntconfig is no more ------------------------------------------------------------------------ r310 | oetiker | 2005-03-13 16:12:44 +0100 | 2 lines missed the actual code for first ------------------------------------------------------------------------ r309 | oetiker | 2005-03-12 18:06:46 +0100 | 1 line rrd_first code contributed by Burton Strauss ------------------------------------------------------------------------ r307 | oetiker | 2005-03-10 18:03:34 +0100 | 17 lines * Updated perl compile system. It now uses Makefile.PL for everything, but gets it to install the perl module in @prefix@/lib/perl such that you can use the same use lib line for whatever platform and perlversion you install into this directory. They all live neatly next to each other. * fixed examples to work with new perl install system. Added a few exec_prefix lines where appropriate * fixed piped demo * fixed VeraMono.ttf file. This somehow got damaged by cvs ... * added thread defines to normal compile run, so that _r functions to not through errors. * tested compilation and installation ------------------------------------------------------------------------ r306 | oetiker | 2005-03-05 00:53:39 +0100 | 8 lines revamped configure system ... lots more stuff is in acinclude now still a bit of a mess, but better than before ... if configure does not find the dependent libraries and header it will not try to get help from pkg-config. If you know where your libraries are it is best to just use CPPFLAGS and LDFLAGS to specify their location. ------------------------------------------------------------------------ r305 | oetiker | 2005-03-04 01:12:37 +0100 | 15 lines Big autotool update: * Upgraded to the latest version of autoconf, automake and libtool * replaced pthread detection code in configure with a more effective one * added a test to see if special solaris defines are required for ctime_r to behave posixly correct. * Fixed lots of warnings from autotools * started adding support for pkg-config ... this will eventually make the whole library finding much simpler ------------------------------------------------------------------------ r300 | oetiker | 2005-03-03 14:34:53 +0100 | 1 line ignorelists are a propperty of an svn directory not a file ------------------------------------------------------------------------ r298 | oetiker | 2005-02-27 23:00:54 +0100 | 2 lines do not complain when setting environemnt variable sucessfully. -- hunter@mimuw.edu.pl ------------------------------------------------------------------------ r297 | oetiker | 2005-02-13 17:13:33 +0100 | 3 lines let rrd_graph return the actual value range it picked ... -- Henrik Stoerner ------------------------------------------------------------------------ r296 | oetiker | 2005-01-28 15:02:17 +0100 | 2 lines fix for memory leak. Rene Gallati ------------------------------------------------------------------------ r295 | oetiker | 2005-01-03 23:25:37 +0100 | 2 lines update todo ------------------------------------------------------------------------ r294 | oetiker | 2004-10-26 00:22:53 +0200 | 2 lines * fore-rules-legend option -- author name missing ------------------------------------------------------------------------ r293 | oetiker | 2004-10-26 00:21:19 +0200 | 2 lines fixed segmentation falt problem in rrd_tool -- Yasuhiro Sumi ------------------------------------------------------------------------ r292 | oetiker | 2004-10-21 21:21:24 +0200 | 2 lines fix return codes in the error case -- Haroon Rafique ------------------------------------------------------------------------ r291 | oetiker | 2004-09-24 23:11:09 +0200 | 3 lines misc fixed and TREND and reduce functionality by -- David M. Grimes ------------------------------------------------------------------------ r290 | oetiker | 2004-08-24 07:26:09 +0200 | 5 lines CDEF operators SHIFT, SQRT, SORT, and REV (reverse). See documentation for what they do. This included removal of redundant code in the rrd_xport path, replaced with a call to rrd_graph_script(). -- David M. Grimes ------------------------------------------------------------------------ r289 | oetiker | 2004-08-09 10:38:55 +0200 | 2 lines added note on escaping : ------------------------------------------------------------------------ r288 | oetiker | 2004-08-08 23:54:07 +0200 | 2 lines port forward of fixes from the stable version of rrd_cgi ------------------------------------------------------------------------ r287 | oetiker | 2004-08-08 01:46:35 +0200 | 3 lines when moving the data pointers we should take into account how fahr off they are. Found by David M. Grimes ------------------------------------------------------------------------ r286 | oetiker | 2004-08-05 23:24:46 +0200 | 2 lines removed nonexisting files from manifest ------------------------------------------------------------------------ r285 | jake | 2004-07-30 02:30:13 +0200 | 4 lines Fix the Win32 build for executable and perl-shared library. See NT-BUILD-TIPS.txt for step by step instructions. Remove VC++ project files not being maintained by anyone. ------------------------------------------------------------------------ r284 | oetiker | 2004-07-14 12:55:25 +0200 | 2 lines added Peter Speck ------------------------------------------------------------------------ r283 | oetiker | 2004-06-26 11:03:50 +0200 | 3 lines allow building rpm from cvs where there is not ChangeLog (this gets generated when the distribution tar is built) -- Tobi and Mike Slifcak ------------------------------------------------------------------------ r282 | oetiker | 2004-06-23 22:36:39 +0200 | 4 lines It's quie enough that the title is placed on the top of the graph, there's no need to emphasize it more. -- Stanislav Sinyagin ------------------------------------------------------------------------ r281 | oetiker | 2004-06-23 10:39:44 +0200 | 2 lines reverted ... $< does not work ------------------------------------------------------------------------ r280 | oetiker | 2004-06-23 00:09:07 +0200 | 4 lines The "$^" variable is GNU make-specific, and fails to compile under FreeBSD 4.10. Please change it to $<, which is more compatible. -- Stanislav Sinyagin ------------------------------------------------------------------------ r279 | oetiker | 2004-06-18 00:09:24 +0200 | 2 lines fixed type for text_prop_conv prototype ------------------------------------------------------------------------ r278 | oetiker | 2004-06-07 21:07:57 +0200 | 3 lines Fix for debian freetype linking -- Peter Hirdina ------------------------------------------------------------------------ r277 | oetiker | 2004-05-27 00:11:12 +0200 | 2 lines reduce compiler warnings. Many small fixes. -- Mike Slifcak ------------------------------------------------------------------------ r276 | oetiker | 2004-05-27 00:10:24 +0200 | 2 lines improve redhat/fedora building -- Mike Slifcak ------------------------------------------------------------------------ r275 | oetiker | 2004-05-27 00:09:58 +0200 | 2 lines update debian build instructions -- Mike Slifcak ------------------------------------------------------------------------ r274 | oetiker | 2004-05-26 00:08:12 +0200 | 3 lines The counter was the only variable which needed to be an integer. -- Mike Slifcak ------------------------------------------------------------------------ r273 | oetiker | 2004-05-26 00:07:30 +0200 | 10 lines The rrdtool RPM package was depending on the rrdtool-perl package. These changes move the examples, which are primarily perl scripts, in with the rrdtool-perl RPM package, thereby de-coupling the rest of the RRDtool from Perl. Perhaps this is silliness ? If there is no need for rrdtool separate from the Perl modules, let me know, then I'll just create an rrdtool and an rrdtool-devel RPM package (that would be a cleaner build and scripting, anyway). -- Mike Slifcak ------------------------------------------------------------------------ r272 | oetiker | 2004-05-26 00:03:04 +0200 | 2 lines added dtd descriptsion s or rrdtool xml output -- Wolfgang Schrimm ------------------------------------------------------------------------ r271 | oetiker | 2004-05-25 23:06:59 +0200 | 2 lines added dump and restore ... Mike Schilli ------------------------------------------------------------------------ r270 | oetiker | 2004-05-25 23:01:50 +0200 | 2 lines added some contributors ------------------------------------------------------------------------ r269 | oetiker | 2004-05-25 23:00:48 +0200 | 2 lines fixed time stamp in example -- Mike Slifcak ------------------------------------------------------------------------ r268 | oetiker | 2004-05-25 22:59:49 +0200 | 2 lines more fixes for rpm spec -- Mike Slifcak ------------------------------------------------------------------------ r267 | oetiker | 2004-05-25 22:58:57 +0200 | 2 lines single word fix -- Mike Slifcak ------------------------------------------------------------------------ r266 | oetiker | 2004-05-25 22:57:58 +0200 | 4 lines Clear out previous legend array contents before freeing legend array. Problem could manifest under extreme resource limits. Not tested. -- Mike Slifcak ------------------------------------------------------------------------ r265 | oetiker | 2004-05-25 22:55:37 +0200 | 2 lines subordinate term was improperly formed -- Mike Slifcak ------------------------------------------------------------------------ r264 | oetiker | 2004-05-25 22:53:21 +0200 | 2 lines prevent small leak when resources are exhausted -- Mike Slifcak ------------------------------------------------------------------------ r263 | oetiker | 2004-05-25 22:52:16 +0200 | 2 lines fix spelling and syntax, especially in messages that are printed -- Mike Slifcak ------------------------------------------------------------------------ r262 | oetiker | 2004-05-25 22:51:49 +0200 | 2 lines Update displayed copyright messages to be consistent. -- Mike Slifcak ------------------------------------------------------------------------ r261 | oetiker | 2004-05-25 22:51:12 +0200 | 2 lines don't process the old rrdgraph single file document -- Mike Slifcak ------------------------------------------------------------------------ r260 | oetiker | 2004-05-25 22:50:36 +0200 | 2 lines fix rpm rules -- Mike Slifcak ------------------------------------------------------------------------ r259 | oetiker | 2004-05-25 22:50:07 +0200 | 2 lines fixed debian dependencies -- Mike Slifcak ------------------------------------------------------------------------ r258 | oetiker | 2004-05-25 18:51:49 +0200 | 3 lines The expression's head was first checking for LT, and then for LTIME, and the latter was never reached. -- Stanislav Sinyagin ------------------------------------------------------------------------ r257 | oetiker | 2004-05-19 07:11:07 +0200 | 2 lines integrated from debian ------------------------------------------------------------------------ r256 | oetiker | 2004-05-18 20:54:50 +0200 | 2 lines its 365 days in a year .... -- Mike Slifcak ------------------------------------------------------------------------ r255 | oetiker | 2004-05-18 20:53:03 +0200 | 2 lines big spell checking patch -- slif@bellsouth.net ------------------------------------------------------------------------ r254 | oetiker | 2004-05-18 20:51:55 +0200 | 2 lines make force option work ------------------------------------------------------------------------ r253 | oetiker | 2004-05-18 20:51:02 +0200 | 2 lines added beginners guide off debian -- slif@bellsouth.net ------------------------------------------------------------------------ r252 | oetiker | 2004-05-18 20:49:50 +0200 | 2 lines add 4chars and fix two examples -- Mike Slifcak ------------------------------------------------------------------------ r251 | oetiker | 2004-05-18 20:40:31 +0200 | 2 lines initial debian build system added -- Mike Slifcak ------------------------------------------------------------------------ r250 | oetiker | 2004-05-18 20:36:05 +0200 | 2 lines fix library numbering -- Mike Slifcak ------------------------------------------------------------------------ r249 | oetiker | 2004-05-18 20:25:45 +0200 | 2 lines better cleaning -- Mike Slifcak ------------------------------------------------------------------------ r248 | oetiker | 2004-05-12 23:57:53 +0200 | 3 lines locate the mkinstalldirs script at top level accept "configure --mandir=" settings -- Mike Slifcak ------------------------------------------------------------------------ r247 | oetiker | 2004-05-04 23:01:29 +0200 | 4 lines Added the ATAN function. This is being used to convert a DS for each vector component of wind direction into a single direction for graphing. CDEF:avdir=yavg,xavg,/,ATAN,57.296,*,xavg,0,LT,180,0,IF,+,DUP,0,LT,360,0,IF,+ -- Daniel Shiels ------------------------------------------------------------------------ r246 | oetiker | 2004-05-04 22:54:47 +0200 | 2 lines updated spec file -- Chris Adams ------------------------------------------------------------------------ r245 | oetiker | 2004-05-04 07:25:19 +0200 | 3 lines die only after all the version checks are done and not for each one -- Alex van den Bogaerdt ------------------------------------------------------------------------ r244 | oetiker | 2004-05-03 16:05:48 +0200 | 6 lines * Allows rrdtool 1.1.x (2004-04-29) to compile on freebsd. * Add configure options --disable-rrdcgi disable building of rrdcgi --disable-pthread disable multithread support -- Stanislav Sinyagin ------------------------------------------------------------------------ r243 | oetiker | 2004-05-02 23:10:59 +0200 | 7 lines See the patch, attached. It changes the default angle to 90, and allows one to change it from configure command line: ./configure RRDGRAPH_YLEGEND_ANGLE=270.0 -- Stanislav Sinyagin ------------------------------------------------------------------------ r242 | oetiker | 2004-04-18 12:21:31 +0200 | 2 lines fixed O_BINARY ------------------------------------------------------------------------ r241 | oetiker | 2004-03-23 22:34:28 +0100 | 2 lines make sure fole is opened binary on restore ... to make things work on windows -- "Girod, Laurent" ------------------------------------------------------------------------ r240 | oetiker | 2004-03-21 12:40:49 +0100 | 2 lines Added --only-graph option -- Piotr Kodzis ------------------------------------------------------------------------ r239 | oetiker | 2004-03-14 21:58:32 +0100 | 3 lines make MakeMaker work on debian and fix some leftovers from the library rip-out transition -- Tobi ------------------------------------------------------------------------ r238 | oetiker | 2004-01-20 00:41:34 +0100 | 2 lines initial ------------------------------------------------------------------------ r237 | oetiker | 2004-01-20 00:39:42 +0100 | 2 lines kill ------------------------------------------------------------------------ r236 | oetiker | 2004-01-20 00:39:01 +0100 | 2 lines remove ------------------------------------------------------------------------ r235 | oetiker | 2004-01-20 00:37:19 +0100 | 2 lines kill it ------------------------------------------------------------------------ r234 | oetiker | 2004-01-20 00:35:21 +0100 | 2 lines killem ------------------------------------------------------------------------ r233 | oetiker | 2004-01-20 00:27:17 +0100 | 2 lines all must go ------------------------------------------------------------------------ r232 | oetiker | 2004-01-20 00:17:00 +0100 | 2 lines misc fixes to get rrdtool working without included libraries. ------------------------------------------------------------------------ r231 | oetiker | 2004-01-20 00:16:03 +0100 | 2 lines modifications for 'no more local libaries' ------------------------------------------------------------------------ r230 | oetiker | 2004-01-15 19:14:02 +0100 | 2 lines addition of RRDs::times -- Christophe Kalt ------------------------------------------------------------------------ r229 | oetiker | 2003-12-26 17:54:55 +0100 | 2 lines using --riggid it is possible to get a case where minval is bigger than maxval ... this breaks the horizontal grid ... ------------------------------------------------------------------------ r228 | oetiker | 2003-11-27 07:31:38 +0100 | 2 lines Stop RRDs for segfaulting on invalid graph input -- Ian Holsman ------------------------------------------------------------------------ r227 | oetiker | 2003-11-27 07:30:36 +0100 | 2 lines make rrd_cgi build again ... Ian Holsman ------------------------------------------------------------------------ r226 | oetiker | 2003-11-23 13:55:37 +0100 | 2 lines New recursive parser for rrdcgi by Arend-Jan Wijtzes ------------------------------------------------------------------------ r225 | oetiker | 2003-11-19 08:06:44 +0100 | 4 lines Adding attributions: * xml fix was by Ian Holsman * font rotation was by Chris Turbeville ------------------------------------------------------------------------ r224 | oetiker | 2003-11-19 07:58:36 +0100 | 2 lines fixed svg generation ------------------------------------------------------------------------ r223 | oetiker | 2003-11-17 12:43:56 +0100 | 2 lines more correct usage added -- Eldad Zack ------------------------------------------------------------------------ r222 | oetiker | 2003-11-15 00:10:51 +0100 | 2 lines handle NULL pointers gracefully ------------------------------------------------------------------------ r221 | oetiker | 2003-11-12 23:14:26 +0100 | 2 lines allow to pass an open filehandle into rrd_graph as an extra argument ------------------------------------------------------------------------ r220 | oetiker | 2003-11-12 23:13:55 +0100 | 2 lines make create much faster .. -- David M. Grimes ------------------------------------------------------------------------ r219 | oetiker | 2003-11-11 22:19:52 +0100 | 3 lines Allow to force overwrite when restoring from xml to rrd. -- Eldad Zack ------------------------------------------------------------------------ r218 | oetiker | 2003-11-11 20:46:21 +0100 | 2 lines replaced time_value with rrd_time_value as MacOS X introduced a struct of that name in their standard headers ------------------------------------------------------------------------ r217 | oetiker | 2003-11-11 20:38:03 +0100 | 3 lines rrd files should NOT change size ever ... bulk update code wa buggy. -- David M. Grimes ------------------------------------------------------------------------ r216 | oetiker | 2003-11-04 23:20:33 +0100 | 2 lines 20 chars max in a ds name -- Kuba Filipowicz ------------------------------------------------------------------------ r215 | oetiker | 2003-10-26 20:29:56 +0100 | 2 lines added rotation ... ------------------------------------------------------------------------ r214 | oetiker | 2003-09-04 15:16:12 +0200 | 2 lines should not assigne but compare ... grrrrr ------------------------------------------------------------------------ r213 | oetiker | 2003-09-02 23:58:35 +0200 | 2 lines be pickier about what we accept in rrd_update. Complain if things do not work out ------------------------------------------------------------------------ r212 | oetiker | 2003-08-28 23:48:53 +0200 | 2 lines make sure we get only 1 flag -- Chris Turbeville ------------------------------------------------------------------------ r211 | oetiker | 2003-07-28 23:59:22 +0200 | 10 lines I was just taking a look at the development RRDtool build (I'm looking forward to SVG support), and had a few problems with an out-of-tree build. I find using a separate build directory makes things much easier to manage, especially when building for multiple architectures. I've attached the minor patches for the progress that I made (I couldn't get the Perl modules to work, but the main binaries seem okay), and would be grateful if you could commit them (or something like them!). -- Joseph Walton ------------------------------------------------------------------------ r210 | jake | 2003-07-24 16:51:46 +0200 | 2 lines Allow hyphen character in DEF_NAM_FMT to match DS_NAM_FMT. ------------------------------------------------------------------------ r209 | jake | 2003-07-17 01:41:37 +0200 | 4 lines Updated MS Visual C++ files from 7.0 to 7.1. Add code to rrd_graph.c: rrd_graph_init() to use %windir% environment variable to find the default font at runtime. ------------------------------------------------------------------------ r208 | jake | 2003-07-15 02:58:38 +0200 | 2 lines Moved OP_COUNT to the end of the enumeration. ------------------------------------------------------------------------ r207 | oetiker | 2003-05-20 22:51:04 +0200 | 5 lines I've made a patch to rrdcgi to add which allows you to re-format start and end-time at-style times using strftime. This allows easy timespans in the graph (e.g. -2weeks) to be formatted into more usual times. -- "Erskine, Thomas" ------------------------------------------------------------------------ r206 | oetiker | 2003-05-03 17:41:41 +0200 | 3 lines small patch to rrd_create.c that corrects problem with the version number of the newly created files. -- Sasha Mikheev ------------------------------------------------------------------------ r205 | oetiker | 2003-04-29 23:56:49 +0200 | 6 lines readline in rrd_open.c reads the file in 8 KB blocks, and calls realloc for each block. realloc is very slow in Mac OS X for huge blocks, e.g. when restoring databases from huge xml files. This patch finds the size of the file, and starts out with malloc'ing the full size. -- Peter Speck ------------------------------------------------------------------------ r204 | oetiker | 2003-04-29 21:37:27 +0200 | 2 lines added comment on iso ------------------------------------------------------------------------ r203 | jake | 2003-04-29 21:14:12 +0200 | 3 lines Change updatev RRA return from index_number to cf_nam, pdp_cnt. Also revert accidental addition of -I to aclocal MakeMakefile. ------------------------------------------------------------------------ r202 | oetiker | 2003-04-26 07:11:27 +0200 | 2 lines only restore V2 and larger. ------------------------------------------------------------------------ r201 | jake | 2003-04-25 20:35:08 +0200 | 2 lines Alternate update interface, updatev. Returns info about CDPs written to disk as result of update. Output format is similar to rrd_info, a hash of key-values. ------------------------------------------------------------------------ r200 | oetiker | 2003-04-24 00:47:00 +0200 | 2 lines fixed version checks to only complain if xml version is > than current RRD version ------------------------------------------------------------------------ r199 | oetiker | 2003-04-11 21:43:44 +0200 | 6 lines New special value COUNT which allows calculations based on the position of a value within a data set. Bug fix in rrd_rpncalc.c. PREV returned erroneus value for the second value. Bug fix in rrd_restore.c. Bug causing seek error when accesing an RRD restored from an xml that holds an RRD version <3. -- Ruben Justo ------------------------------------------------------------------------ r198 | oetiker | 2003-04-04 22:51:21 +0200 | 4 lines I think I found a very small bug in rrd graph PDF output : the is_stream member of pdf_buffer is not initialized, making bad PDF sometimes (I found a case where is_stream was != 0 for the font initialization). -- lhoudard@netcourrier.com ------------------------------------------------------------------------ r197 | jake | 2003-04-02 00:52:23 +0200 | 2 lines Fix Win32 build. VC++ 6.0 and 7.0 now use the thread-safe code. ------------------------------------------------------------------------ r196 | oetiker | 2003-03-31 23:22:12 +0200 | 4 lines enables RRDtool updates with microsecond or in case of windows millisecond precision. This is needed to reduce time measurement error when archive step is small. (<30s) -- Sasha Mikheev ------------------------------------------------------------------------ r195 | oetiker | 2003-03-25 23:29:10 +0100 | 2 lines use the environment variable VCINSTALLDIR -- Ian Holsman ------------------------------------------------------------------------ r194 | oetiker | 2003-03-24 23:08:20 +0100 | 2 lines strerror should not turn us recursive here ... -- Peter Stamfest ------------------------------------------------------------------------ r193 | oetiker | 2003-03-24 23:05:26 +0100 | 2 lines strerror must not become recursive -- Peter Stamfest ------------------------------------------------------------------------ r192 | oetiker | 2003-03-12 21:39:59 +0100 | 2 lines remove ^M ------------------------------------------------------------------------ r191 | oetiker | 2003-03-10 01:30:34 +0100 | 3 lines handle cases with two negative numbers -- Sasha Mikheev ------------------------------------------------------------------------ r190 | oetiker | 2003-03-08 19:44:06 +0100 | 3 lines don't display legends for [HV]RULEs out of graph bounds (rrdgraph) -- Christophe Kalt ------------------------------------------------------------------------ r189 | oetiker | 2003-03-01 23:25:56 +0100 | 2 lines realy suppress size output when talking to stdout. Patch from Mat Zimmerman @ debian ------------------------------------------------------------------------ r188 | oetiker | 2003-02-24 19:26:37 +0100 | 2 lines fix for : in piped demo ------------------------------------------------------------------------ r187 | oetiker | 2003-02-22 22:57:35 +0100 | 3 lines Initial checkin -- Peter Stamfest ------------------------------------------------------------------------ r186 | oetiker | 2003-02-22 22:57:03 +0100 | 3 lines a patch to avoid a memory leak and a Makefile.am patch to distribute all required source files -- Peter Stamfest ------------------------------------------------------------------------ r185 | oetiker | 2003-02-21 23:40:07 +0100 | 2 lines colons must be escaped ------------------------------------------------------------------------ r184 | oetiker | 2003-02-20 22:48:57 +0100 | 3 lines make the threading work and compile, fix linking to libpng and update the timestaps int the files ------------------------------------------------------------------------ r183 | oetiker | 2003-02-20 22:27:48 +0100 | 4 lines updated copyrigh dates removed error.h from threaded variant added compiletime to rrdtool ------------------------------------------------------------------------ r182 | oetiker | 2003-02-16 13:32:24 +0100 | 2 lines if output is - there is not need to deref the handler ------------------------------------------------------------------------ r181 | oetiker | 2003-02-16 13:31:58 +0100 | 2 lines expect seems to exist on some systems in the system inc files ------------------------------------------------------------------------ r180 | oetiker | 2003-02-13 08:05:59 +0100 | 12 lines Find attached the patch I promised to send to you. Please note that there are three new source files (src/rrd_is_thread_safe.h, src/rrd_thread_safe.c and src/rrd_not_thread_safe.c) and the introduction of librrd_th. This library is identical to librrd, but it contains support code for per-thread global variables currently used for error information only. This is similar to how errno per-thread variables are implemented. librrd_th must be linked alongside of libpthred There is also a new file "THREADS", holding some documentation. -- Peter Stamfest ------------------------------------------------------------------------ r179 | oetiker | 2003-02-12 08:06:56 +0100 | 2 lines improved dist target -- Peter Stamfest ------------------------------------------------------------------------ r178 | oetiker | 2003-02-11 11:16:46 +0100 | 2 lines removed lg we do not allow this in the code at th moment ------------------------------------------------------------------------ r177 | oetiker | 2003-02-05 19:16:02 +0100 | 2 lines rrd resize SHRINK was broken :-) -- Scott Mace ------------------------------------------------------------------------ r176 | oetiker | 2003-01-31 07:46:32 +0100 | 2 lines The NaN check should only use lastgdes when following a STACK -- Scott Mace ------------------------------------------------------------------------ r175 | oetiker | 2003-01-30 22:39:21 +0100 | 2 lines fix stak+nan error ... Scott Mace ------------------------------------------------------------------------ r174 | oetiker | 2003-01-29 08:17:23 +0100 | 2 lines fixed the processing of TICK:vname#color:frac:legend -- Scott Mace ------------------------------------------------------------------------ r173 | oetiker | 2003-01-25 23:50:32 +0100 | 3 lines added function (--no-minor) to turn off minor gridlines on graphs -- Travis Brown ------------------------------------------------------------------------ r172 | oetiker | 2003-01-17 00:27:54 +0100 | 3 lines fix border condition in rra selection of rrd_fetch -- Stanislav Sinyagin ------------------------------------------------------------------------ r171 | oetiker | 2003-01-15 20:24:13 +0100 | 2 lines * single letter variants for all graph options -- James Overbeck ------------------------------------------------------------------------ r170 | oetiker | 2002-12-14 23:30:41 +0100 | 3 lines fix rrdtool compile on solaris where CC for perl is "gcc -B/usr/ccs/bin/" -- Russell Van Tassell ------------------------------------------------------------------------ r169 | oetiker | 2002-11-29 12:57:17 +0100 | 2 lines note on info added ------------------------------------------------------------------------ r168 | oetiker | 2002-11-19 23:33:09 +0100 | 2 lines added g as valid string format value ------------------------------------------------------------------------ r167 | oetiker | 2002-11-19 00:37:56 +0100 | 3 lines check for -OPT:IEEE_NaN_inf=ON on SGI C compiler -- Albert Chin-A-Young ------------------------------------------------------------------------ r166 | oetiker | 2002-11-03 16:07:13 +0100 | 2 lines fixed ------------------------------------------------------------------------ r165 | oetiker | 2002-10-24 23:17:51 +0200 | 2 lines added patch for apache 2 compatibility ------------------------------------------------------------------------ r164 | oetiker | 2002-10-07 08:22:35 +0200 | 2 lines keep up with the jhonses ------------------------------------------------------------------------ r163 | oetiker | 2002-10-07 08:21:22 +0200 | 2 lines -z does not take an option -- Tomoyuki Murakami ------------------------------------------------------------------------ r162 | oetiker | 2002-09-02 20:41:19 +0200 | 2 lines fixed leak in VDEF_PERCENT handlin -- Perry Stoll ------------------------------------------------------------------------ r161 | oetiker | 2002-08-01 07:42:27 +0200 | 3 lines fix for segfault condition in print_calc -- Paul Clifford ------------------------------------------------------------------------ r160 | oetiker | 2002-07-31 07:42:48 +0200 | 2 lines improved fetch explanation -- Nenad.Antic@era.ericsson.se ------------------------------------------------------------------------ r159 | oetiker | 2002-07-13 20:35:37 +0200 | 3 lines fixed DEF_NAM_FMT definition added double include protection the rrd_graph.h and rrd_rpncalc.h ------------------------------------------------------------------------ r158 | oetiker | 2002-07-12 13:20:32 +0200 | 2 lines fixed PREV parsing ... Gonzalo Augusto Arana Tagle ------------------------------------------------------------------------ r157 | oetiker | 2002-07-06 17:45:15 +0200 | 2 lines new operators must be added at the END of the sequence ... ------------------------------------------------------------------------ r156 | oetiker | 2002-07-05 20:57:02 +0200 | 2 lines new operand on rrdgraph CDEFs: PREV(xxxx) -- Gonzalo Augusto Arana Tagle ------------------------------------------------------------------------ r155 | oetiker | 2002-07-05 20:47:46 +0200 | 2 lines explain tzset ------------------------------------------------------------------------ r154 | oetiker | 2002-07-02 12:35:23 +0200 | 2 lines fixed DINF for win32 ------------------------------------------------------------------------ r153 | alex | 2002-06-29 17:33:58 +0200 | 2 lines Changed DEF:... ------------------------------------------------------------------------ r152 | alex | 2002-06-29 17:24:26 +0200 | 3 lines Changed parsing again. Added a DEBUG prefix to all grapher commands ------------------------------------------------------------------------ r151 | alex | 2002-06-29 16:55:47 +0200 | 2 lines Not needed anymore ------------------------------------------------------------------------ r150 | alex | 2002-06-27 21:34:25 +0200 | 3 lines DEF now takes "--start" and "--end" LINEx, AREA and such now take parameter "STACK" ------------------------------------------------------------------------ r149 | alex | 2002-06-24 01:13:53 +0200 | 2 lines Updated docs for the "DEF ... :step=nnnn" case ------------------------------------------------------------------------ r148 | alex | 2002-06-24 00:29:40 +0200 | 3 lines Added "step=1800" and such to "DEF" Cleaned some of the signed vs. unsigned problems ------------------------------------------------------------------------ r147 | alex | 2002-06-24 00:28:06 +0200 | 2 lines In stead of a zillion warnings, only one place gives a warning for DNAN ------------------------------------------------------------------------ r146 | jake | 2002-06-20 02:21:03 +0200 | 2 lines More Win32 build changes; thanks to Kerry Calvert. ------------------------------------------------------------------------ r145 | oetiker | 2002-06-14 14:15:12 +0200 | 2 lines better explanations -- "Shipway, Steve" ------------------------------------------------------------------------ r144 | jake | 2002-05-22 17:25:43 +0200 | 2 lines Changed intercept to slope for beta parameter in rrdtune doc. ------------------------------------------------------------------------ r143 | oetiker | 2002-05-22 07:35:09 +0200 | 3 lines VC++ .NET (7.0) project files -- Eric Chamberlain ------------------------------------------------------------------------ r142 | oetiker | 2002-05-16 21:18:35 +0200 | 2 lines added -W ... lets have some more warnings ... with gcc 3.1 985 ------------------------------------------------------------------------ r141 | oetiker | 2002-05-15 08:14:40 +0200 | 2 lines afm fixes -- Peter Speck ------------------------------------------------------------------------ r140 | jake | 2002-05-14 23:52:58 +0200 | 2 lines Fix Win32 Build compatibility. ------------------------------------------------------------------------ r139 | oetiker | 2002-05-14 23:48:05 +0200 | 2 lines double free fixed ------------------------------------------------------------------------ r138 | oetiker | 2002-05-14 07:28:45 +0200 | 3 lines make opendir readdir chdir chroot and thus the serverfunctionality a conditional compile depending on the presence of the necessary function calls. ------------------------------------------------------------------------ r137 | oetiker | 2002-05-11 11:09:57 +0200 | 2 lines make dirent sys/types and sys/stat autoconfable .... ------------------------------------------------------------------------ r136 | oetiker | 2002-05-08 22:34:56 +0200 | 3 lines Add funtionality to rrdtool to run it as a 'server' -- Hartmut.Vogler@epost.de ------------------------------------------------------------------------ r135 | oetiker | 2002-05-07 23:58:33 +0200 | 3 lines new command rrdtool xport integrated -- Wolfgang Schrimm ------------------------------------------------------------------------ r134 | oetiker | 2002-05-02 15:23:56 +0200 | 2 lines typo fixed ------------------------------------------------------------------------ r133 | oetiker | 2002-04-29 19:11:03 +0200 | 2 lines badformat parser fixed ... for good this time I hope ------------------------------------------------------------------------ r132 | oetiker | 2002-04-28 21:13:04 +0200 | 2 lines inserted PRINT checker at the wrong point ... ------------------------------------------------------------------------ r131 | oetiker | 2002-04-28 16:14:48 +0200 | 5 lines It's nice and simple, the error checker for the PRINT stuff which is passed to printf() is not allowing certain valid printf() style operations which make text alignment much more fun. -- Richard A Steenbergen ------------------------------------------------------------------------ r130 | oetiker | 2002-04-24 23:04:46 +0200 | 2 lines fixed setlocale issues ------------------------------------------------------------------------ r129 | oetiker | 2002-04-24 22:54:09 +0200 | 2 lines fixed adress for wishlist ------------------------------------------------------------------------ r128 | oetiker | 2002-04-24 22:49:26 +0200 | 2 lines Patch for supporting PDF as output format ------------------------------------------------------------------------ r127 | oetiker | 2002-04-24 22:49:00 +0200 | 6 lines Modification of rrd_graph.c/h so it uses dashes for gridlines much like rrdtool 1.0 Avoids drawing minor vertical gridlines where a major gridline is to avoid artifacts from having the grey line beneath the red one. Patches src/rrd_graph.c and src/rrd_graph.h -- Peter Speck ------------------------------------------------------------------------ r125 | oetiker | 2002-04-24 22:43:27 +0200 | 2 lines fixed setlocale issues ------------------------------------------------------------------------ r124 | oetiker | 2002-04-16 07:00:07 +0200 | 2 lines doc for gridfit by peter ------------------------------------------------------------------------ r123 | oetiker | 2002-04-15 23:29:24 +0200 | 20 lines Evenly-spaced y-axis gridlines. I had to split horizontal_grid() into calc_horizontal_grid and draw_horizontal_grid as the calculated info is needed in the adjustment code. My previous patch had the problem that it adjusted the y-axis too late, e.g. after the data lines was drawn. The result of the calc is stored in struct ygrid_scale_t which image_desc_t has as a member. --no-gridfit is implemented. The round-to-integer coordinates for png is moved to the libart code in rrd_gfx. The 'close path' code is cleaned up so the node list is left unchanged if you want to save the same graph in multiple formats in one run. The rounding is done on the scaled coordinates (zoom). I have made a simple version for logarithmic y scales as such a scale might have 5 gridlines with 4 difference spacings. This version only uses y = 10^x values for modifying the scale. -- Peter Speck ------------------------------------------------------------------------ r122 | oetiker | 2002-04-09 23:35:49 +0200 | 2 lines remove boguos description of goodfor and refresh ------------------------------------------------------------------------ r121 | oetiker | 2002-04-09 23:34:56 +0200 | 2 lines removed tzset as it is in rrd_graph now ------------------------------------------------------------------------ r120 | oetiker | 2002-04-09 23:34:13 +0200 | 2 lines added tzset and setlocale to rrd_graph ------------------------------------------------------------------------ r119 | oetiker | 2002-04-08 00:07:46 +0200 | 2 lines added missing files from peters patch ------------------------------------------------------------------------ r118 | oetiker | 2002-04-07 22:20:27 +0200 | 18 lines 3 patches, #3 depends on #1 as EPS uses AFM for stringwidth too. src/rrd_afm.[ch], src/rrd_afm_data.[ch] and changes to rrd_gfx.[ch] to use AFM for stringwidth in SVG output. libraries/afm with .afm files for the standard 14 postscript fonts and perl script to create src/rrd_afm_data.c Includes small test program to show examples of calculated stringwidth and actual stringwidth. Adds EPS output support. Modifies rrd_tool.c as rrd_graph() changes the argv pointer, and rrd_tool then always wrote the 200x100 output line to the file (which probably doesn't do anything for png images, but eps files fail having it after the %%EOF marker). -- Peter Speck ------------------------------------------------------------------------ r117 | alex | 2002-04-06 14:40:59 +0200 | 2 lines Different default font when WIN32 is defined ------------------------------------------------------------------------ r116 | alex | 2002-04-06 14:25:29 +0200 | 3 lines Removed references to GIF Changed y0 and such into Y0 and so ------------------------------------------------------------------------ r115 | jake | 2002-04-06 01:51:21 +0200 | 4 lines Updated/added MVSC++ 6.0 project files for compilation of rrd.lib and rrdtool.exe on Win32. Changes for RRDs compilation on Win32. Moved src/ntconfig.h to confignt/config.h. ------------------------------------------------------------------------ r114 | oetiker | 2002-04-03 16:52:15 +0200 | 23 lines As gfx_canvas_t now has excatly the same lifespan as image_desc_t, I've made 'gfx_canvas_t canvas' a member of image_desc_t and dropped it as a separate parameter in all function calls. imgformat, interlaced and zoom are moved to gfx_canvas_t. I have dropped my old fontlib-enum as imgformat contains that information. The gfx_render_xxx switch is moved to rrd_gfx so rrd_graph is (almost) only bothered with image formats in cmd line parsing. gfx_close_path is added. gfx_new_dashed_line is added with 2 new arguments: length of a dash and length between dashes. gfx_new_line is still there for plain lines. Having dash-length == 0 creates normal line. rrd_graph.c is not updated for dashed lines. It's not decided how one should specify which and how lines are dashed. An extension of the color specification? svg is updated for dashes and opacity, but libart code is not. I've fixed indent in SVG, all lines had a space before func decl. etc. Misc small fixes, e.g. sscanf of gfx_color_t, position of x-axis arrow. --- Peter Speck ------------------------------------------------------------------------ r113 | oetiker | 2002-04-03 07:32:31 +0200 | 2 lines i suck. sorry. add ,0666 to the open(). -- Paul Vixie ------------------------------------------------------------------------ r112 | oetiker | 2002-04-02 23:32:15 +0200 | 2 lines added call to tzset to activate TZ settings -- Paul A Vixie ------------------------------------------------------------------------ r111 | oetiker | 2002-04-02 23:31:08 +0200 | 2 lines aded missing fcntl.h ------------------------------------------------------------------------ r110 | oetiker | 2002-04-02 21:37:54 +0200 | 2 lines rrd restore should not burn down existing files -- Paul Vixie ------------------------------------------------------------------------ r109 | oetiker | 2002-04-01 20:32:17 +0200 | 2 lines merged svg update ------------------------------------------------------------------------ r108 | oetiker | 2002-04-01 20:31:22 +0200 | 3 lines "!" takes a higher preference than "||" this means rrd_update N:: would segfault -- Oliver Cook ------------------------------------------------------------------------ r107 | jake | 2002-03-28 18:33:49 +0100 | 2 lines Added comment suggesting use of -I flag with aclocal. ------------------------------------------------------------------------ r106 | oetiker | 2002-03-26 08:02:28 +0100 | 2 lines added SVG support -- Peter Speck ------------------------------------------------------------------------ r105 | alex | 2002-03-23 21:41:48 +0100 | 2 lines Better positioning of the pie when it is alone on the canvas ------------------------------------------------------------------------ r104 | alex | 2002-03-23 21:01:13 +0100 | 2 lines Modified rrd_graph ------------------------------------------------------------------------ r103 | alex | 2002-03-23 20:59:39 +0100 | 2 lines Changes in rrd_graph; see NEWS ------------------------------------------------------------------------ r102 | oetiker | 2002-03-23 10:05:32 +0100 | 2 lines removed reference to GIF ------------------------------------------------------------------------ r101 | oetiker | 2002-03-23 10:01:43 +0100 | 2 lines remove all traces of GIF and make PNG the default ------------------------------------------------------------------------ r100 | oetiker | 2002-03-23 09:53:31 +0100 | 3 lines Move CDEF start pointers if start of cdef is a step ahead of the start of the data -- Ashok Mandala ------------------------------------------------------------------------ r99 | oetiker | 2002-03-21 23:39:03 +0100 | 2 lines fixed color area in legend and fixed color area in 3d border ------------------------------------------------------------------------ r98 | alex | 2002-03-21 13:00:51 +0100 | 3 lines Pie charts didn't have anti-aliasing; building them clockwise seems to solve this problem. ------------------------------------------------------------------------ r97 | oetiker | 2002-03-20 23:48:25 +0100 | 3 lines fixed longstanding bug affection CDEFS where values from rrds with different resolutions got mixed .. ------------------------------------------------------------------------ r96 | alex | 2002-03-17 23:40:18 +0100 | 2 lines Changed the way circle sections are drawn. ------------------------------------------------------------------------ r95 | alex | 2002-03-13 03:58:44 +0100 | 2 lines New, hopefully better, implementation of PART drawing ------------------------------------------------------------------------ r94 | oetiker | 2002-03-12 08:19:09 +0100 | 2 lines realy remove zlibe 1.1.3 dir ------------------------------------------------------------------------ r93 | oetiker | 2002-03-12 08:18:06 +0100 | 2 lines updated ------------------------------------------------------------------------ r92 | oetiker | 2002-03-12 08:17:46 +0100 | 2 lines ignore generated files ------------------------------------------------------------------------ r91 | oetiker | 2002-03-12 08:16:32 +0100 | 2 lines missing bits ------------------------------------------------------------------------ r90 | oetiker | 2002-03-12 07:43:38 +0100 | 2 lines updated to zlib 1.1.4 ------------------------------------------------------------------------ r89 | oetiker | 2002-03-12 07:42:00 +0100 | 2 lines replace zlib 1.1.3 with zlib 1.1.4 ------------------------------------------------------------------------ r88 | alex | 2002-03-11 00:08:37 +0100 | 2 lines Pie chart support added to rrdtool graph ------------------------------------------------------------------------ r87 | oetiker | 2002-03-10 23:49:24 +0100 | 2 lines added quotes to echo ------------------------------------------------------------------------ r86 | alex | 2002-03-10 17:22:28 +0100 | 3 lines Realigned function rpn_calc() Stack checking now done using a macro ------------------------------------------------------------------------ r85 | alex | 2002-03-10 15:58:30 +0100 | 3 lines Fixed some typos/errors Updated for the NE and ISINF operators in RPN ------------------------------------------------------------------------ r84 | alex | 2002-03-10 15:53:04 +0100 | 3 lines Using the font option resulted in a segfault. Needs more care, the current change is just a hack ------------------------------------------------------------------------ r83 | alex | 2002-03-10 15:48:59 +0100 | 2 lines Added NE and ISINF operands to RPN ------------------------------------------------------------------------ r82 | oetiker | 2002-03-10 13:28:38 +0100 | 2 lines better version check ------------------------------------------------------------------------ r81 | alex | 2002-03-08 23:14:22 +0100 | 2 lines Needed to escape a wildcard ------------------------------------------------------------------------ r80 | alex | 2002-03-08 23:11:37 +0100 | 2 lines Fixed some typos ------------------------------------------------------------------------ r79 | oetiker | 2002-02-18 22:52:20 +0100 | 2 lines added autotools version check to MakeMakefiles ------------------------------------------------------------------------ r78 | oetiker | 2002-02-09 07:21:11 +0100 | 2 lines do propper error checking and release memmory when it is not required anymore ------------------------------------------------------------------------ r77 | oetiker | 2002-02-08 19:40:37 +0100 | 2 lines allow rrd_cgi to deal with umlauts -- Alexander Schwartz ------------------------------------------------------------------------ r76 | oetiker | 2002-02-03 09:10:36 +0100 | 2 lines fixed spelling ------------------------------------------------------------------------ r75 | oetiker | 2002-02-02 15:36:14 +0100 | 2 lines fixed data pointer storage ------------------------------------------------------------------------ r74 | oetiker | 2002-02-01 21:34:49 +0100 | 2 lines fixed version number and date/time ------------------------------------------------------------------------ r73 | oetiker | 2002-02-01 21:29:32 +0100 | 2 lines *** empty log message *** ------------------------------------------------------------------------ r72 | oetiker | 2002-01-31 23:44:34 +0100 | 2 lines this file and all the external libraries configurability was contributed by mat zimmermann ------------------------------------------------------------------------ r71 | oetiker | 2002-01-31 23:43:13 +0100 | 2 lines allow use of libraries already installed on the system ------------------------------------------------------------------------ r70 | oetiker | 2002-01-31 12:56:24 +0100 | 2 lines make "18:00 yesterday" work Pavel Mores ------------------------------------------------------------------------ r69 | oetiker | 2002-01-31 08:23:12 +0100 | 2 lines weeknumber is not %V which is more appropriate than %W ------------------------------------------------------------------------ r68 | oetiker | 2002-01-31 07:36:27 +0100 | 2 lines copy ltmain.sh ------------------------------------------------------------------------ r67 | oetiker | 2002-01-17 21:31:31 +0100 | 2 lines this also ------------------------------------------------------------------------ r66 | oetiker | 2002-01-17 21:30:46 +0100 | 2 lines this is not needed anymore ------------------------------------------------------------------------ r65 | oetiker | 2002-01-17 21:28:01 +0100 | 2 lines killing remains ------------------------------------------------------------------------ r64 | oetiker | 2002-01-17 00:11:11 +0100 | 2 lines added places for download ------------------------------------------------------------------------ r63 | oetiker | 2002-01-15 23:58:09 +0100 | 2 lines ignore irrelevant things ------------------------------------------------------------------------ r62 | oetiker | 2002-01-15 23:53:19 +0100 | 2 lines added new .cvsignore files ------------------------------------------------------------------------ r61 | oetiker | 2002-01-15 23:51:07 +0100 | 15 lines The BIG graph update * Replace libgd with libart * Added freetype * Updated zlib and libpng * rrd_gfx.c intrduced as libart wrapper * LINE takes now a float as argument * rrdtool uses truetype for fonts * thanks to libart there is now full alpha transparenc and antialiasing. * the new option --font lets customize the font and size for various graph elements * Updated to -> libtool 1.4.2 automake 2.12 autoconf 2.52 * new --zoom commandline option for zoom ans shrinking -- tobias oetiker ------------------------------------------------------------------------ r60 | alex | 2002-01-04 02:11:35 +0100 | 2 lines Reworked rrd_graph_script() ------------------------------------------------------------------------ r59 | alex | 2001-12-24 07:51:49 +0100 | 21 lines A patch of size 44Kbytes... in short: Found and repaired the off-by-one error in rrd_fetch_fn(). As a result I had to remove the hacks in rrd_fetch_fn(), rrd_tool.c, vdef_calc(), data_calc(), data_proc() and reduce_data(). There may be other places which I didn't find so be careful. Enhanced debugging in rrd_fetch_fn(), it shows the RRA selection process. Added the ability to print VDEF timestamps. At the moment it is a hack, I needed it now to fix the off-by-one error. If the format string is "%c" (and nothing else!), the time will be printed by both ctime() and as a long int. Moved some code around (slightly altering it) from rrd_graph() initializing now in rrd_graph_init() options parsing now in rrd_graph_options() script parsing now in rrd_graph_script() ------------------------------------------------------------------------ r58 | alex | 2001-12-22 03:49:23 +0100 | 3 lines Somehow eight lines were missing from function rrd_graph(). Also fixed minor bug in vdef_calc(). ------------------------------------------------------------------------ r57 | oetiker | 2001-12-17 13:48:43 +0100 | 2 lines fix overflow error ... ------------------------------------------------------------------------ r56 | jake | 2001-12-11 23:55:28 +0100 | 2 lines Fixed uninitialized ptr causing seg fault invoking info for COMPUTE data sources. ------------------------------------------------------------------------ r55 | oetiker | 2001-11-18 09:41:16 +0100 | 2 lines return "" instead of NULL for -- Michael ------------------------------------------------------------------------ r54 | oetiker | 2001-11-17 17:57:55 +0100 | 2 lines fixed path to mkinstalldirs -- Laurent Saehyun Kim?\239?\191?\189 ------------------------------------------------------------------------ r53 | oetiker | 2001-09-08 20:25:17 +0200 | 2 lines spelling updates by Martin Schulze ------------------------------------------------------------------------ r52 | jake | 2001-08-23 00:29:07 +0200 | 13 lines Contents of this patch: (1) Adds/revises documentation for rrd tune in rrd_tool.c and pod files. (2) Moves some initialization code from rrd_create.c to rrd_hw.c. (3) Adds another pass to smoothing for SEASONAL and DEVSEASONAL RRAs. This pass computes the coefficients as deviations from an average; the average is added the baseline coefficient of HWPREDICT. Statistical texts suggest this to preserve algorithm stability. It will not invalidate RRD files created and smoothed with the old code. (4) Adds the aberrant-reset flag to rrd tune. This operation, which is specified for a single data source, causes the holt-winters algorithm to forget everthing it has learned and start over. (5) Fixes a few out-of-date code comments. ------------------------------------------------------------------------ r51 | oetiker | 2001-08-13 20:58:07 +0200 | 2 lines spell fix for si units (only in comments) ------------------------------------------------------------------------ r50 | alex | 2001-07-29 00:34:42 +0200 | 2 lines Removing rrdgraph*.pod from the cvs ------------------------------------------------------------------------ r49 | alex | 2001-07-29 00:21:44 +0200 | 4 lines Edited Makefile.am and the rrdgraph_*.src files. By mistake the previous update was made on the rrdgraph_*.pod files however they are not to be edited. This is now corrected. ------------------------------------------------------------------------ r48 | alex | 2001-07-26 04:27:46 +0200 | 4 lines Supporting functions for rrd_graph. Functions that are used frequently and/or from different places should be added here in stead of repeating them in rrd_graph.c over and over again. ------------------------------------------------------------------------ r47 | alex | 2001-07-26 04:25:38 +0200 | 2 lines Moved most typedefs etc. from rrd_graph.c to this file ------------------------------------------------------------------------ r46 | alex | 2001-07-26 04:22:18 +0200 | 3 lines Added VDEF TOTAL Moved most typedefs etc. to rrd_graph.h ------------------------------------------------------------------------ r45 | alex | 2001-07-26 04:19:48 +0200 | 2 lines Added rrd_graph_helper ------------------------------------------------------------------------ r44 | alex | 2001-07-26 04:15:02 +0200 | 2 lines Updated for VDEF TOTAL ------------------------------------------------------------------------ r43 | alex | 2001-07-26 04:11:46 +0200 | 2 lines Announce of VDEF in rrd_graph ------------------------------------------------------------------------ r42 | alex | 2001-07-21 20:17:44 +0200 | 2 lines Made PRINT and GPRINT aware of VDEF statements ------------------------------------------------------------------------ r41 | oetiker | 2001-07-21 00:34:29 +0200 | 2 lines missing documentation ... -- Alex van den Bogaerdt ------------------------------------------------------------------------ r40 | oetiker | 2001-07-21 00:34:06 +0200 | 2 lines only a comment ------------------------------------------------------------------------ r39 | oetiker | 2001-07-21 00:33:17 +0200 | 2 lines small fixes -- Alex van den Bogaerdt ------------------------------------------------------------------------ r38 | oetiker | 2001-07-19 00:30:29 +0200 | 2 lines VDEF and VRULE are comig along -- Alex van den Bogaerdt ------------------------------------------------------------------------ r37 | oetiker | 2001-06-05 15:42:55 +0200 | 2 lines spell fix ------------------------------------------------------------------------ r36 | oetiker | 2001-05-09 07:31:01 +0200 | 5 lines Bug fix: when update of multiple PDP/CDP RRAs coincided with interpolation of multiple PDPs an incorrect value was stored as the CDP. Especially evident for GAUGE data sources. Minor changes to rrdcreate.pod. -- Jake Brutlag ------------------------------------------------------------------------ r35 | oetiker | 2001-03-31 17:21:59 +0200 | 3 lines added 'all' dependancy to site-perl-install rule as many seem to skip this step when instaling cricket ... probably an incomplete cricket doku. ------------------------------------------------------------------------ r34 | oetiker | 2001-03-15 20:43:04 +0100 | 2 lines added PROJECTS file ------------------------------------------------------------------------ r33 | oetiker | 2001-03-11 13:03:12 +0100 | 2 lines reindented ------------------------------------------------------------------------ r32 | oetiker | 2001-03-11 12:49:28 +0100 | 2 lines reindented the cource for better readability ------------------------------------------------------------------------ r31 | oetiker | 2001-03-11 00:54:41 +0100 | 7 lines Support for COMPUTE data sources (CDEF data sources). Removes the RPN parser and calculator from rrd_graph and puts then in a new file, rrd_rpncalc.c. Changes to core files rrd_create and rrd_update. Some clean-up of aberrant behavior stuff, including a bug fix. Documentation update (rrdcreate.pod, rrdupdate.pod). Change xml format. -- Jake Brutlag ------------------------------------------------------------------------ r30 | oetiker | 2001-03-07 22:21:54 +0100 | 4 lines complete rewrite of rrdgraph documentation. This also includs info on upcomming/planned changes to the rrdgraph interface and functionality -- Alex van den Bogaerdt ------------------------------------------------------------------------ r28 | oetiker | 2001-03-04 14:50:31 +0100 | 2 lines fidex jackes adderss ------------------------------------------------------------------------ r27 | oetiker | 2001-03-04 14:12:44 +0100 | 2 lines added note on counter vs derive -- Don BAARDA ------------------------------------------------------------------------ r26 | oetiker | 2001-03-04 14:01:56 +0100 | 6 lines Aberrant Behavior Detection support. A brief overview added to rrdtool.pod. Major updates to rrd_update.c, rrd_create.c. Minor update to other core files. This is backwards compatible! But new files using the Aberrant stuff are not readable by old rrdtool versions. See http://cricket.sourceforge.net/aberrant/rrd_hw.htm -- Jake Brutlag ------------------------------------------------------------------------ r25 | oetiker | 2001-03-04 14:00:25 +0100 | 2 lines another do not report Makefile and Makefile.in ------------------------------------------------------------------------ r24 | oetiker | 2001-03-04 13:51:55 +0100 | 2 lines this file has no purpose ------------------------------------------------------------------------ r23 | oetiker | 2001-03-04 13:32:07 +0100 | 2 lines rrdupdate is a generated file it has no place in cvs ------------------------------------------------------------------------ r21 | oetiker | 2001-03-04 12:31:21 +0100 | 2 lines repository cleanup ------------------------------------------------------------------------ r20 | oetiker | 2001-03-04 12:14:26 +0100 | 3 lines added at-style-time@value:value syntax to rrd_update -- Dave Bodenstab ------------------------------------------------------------------------ r19 | oetiker | 2001-03-04 11:29:20 +0100 | 3 lines fixed filedescriptor leak -- Mike Franusich ------------------------------------------------------------------------ r18 | oetiker | 2001-03-02 23:48:39 +0100 | 2 lines added test for hpux. if matched, compile zlib with -fpic instead of -fPIC ------------------------------------------------------------------------ r17 | oetiker | 2001-03-01 22:37:54 +0100 | 2 lines added check for the presence of the compiler used to build perl modules ------------------------------------------------------------------------ r16 | oetiker | 2001-02-26 00:18:40 +0100 | 2 lines they do not belong into cvs ad they are autogenerated ------------------------------------------------------------------------ r15 | oetiker | 2001-02-25 23:57:02 +0100 | 2 lines Fixed ln for modules ------------------------------------------------------------------------ r14 | oetiker | 2001-02-25 23:53:17 +0100 | 2 lines fixed compile order ... src before bindings ------------------------------------------------------------------------ r6 | oetiker | 2001-02-25 23:25:02 +0100 | 2 lines Initial revision ------------------------------------------------------------------------