network plugin: Make the receive thread even faster.
On very busy systems, the thread apparently may not be scheduled often enough.
So the receive buffer fills up quickly and data may be lost.
This patch changes the `mutex_lock' to a `mutex_trylock' and data is only
appended to the global receive-queue if the lock can be obtained without
blocking.
If the lock cannot be obtained without blocking, the data is instead appended
to a private queue and that queue is appended to the global queue when the lock
can next be taken.
On very busy systems, the thread apparently may not be scheduled often enough.
So the receive buffer fills up quickly and data may be lost.
This patch changes the `mutex_lock' to a `mutex_trylock' and data is only
appended to the global receive-queue if the lock can be obtained without
blocking.
If the lock cannot be obtained without blocking, the data is instead appended
to a private queue and that queue is appended to the global queue when the lock
can next be taken.
Merge branch 'dm/t_option'
collectd: Don't *abort* on the first read-error with the `-T' option.
The `-T' option used to basically quit the daemon right away when the
first read function of a plugin failed. This patch changes the behavior,
so that:
- All read-functions are tried. If one or more fail, the daemon will
exit with a non-zero exit status, but all read-functions will be
tried.
- Don't quit if one read-function failed without calling all the
shutdown-functions first. This will clean up the UNIX socket of the
unixsock plugin and stuff like that.
The `-T' option used to basically quit the daemon right away when the
first read function of a plugin failed. This patch changes the behavior,
so that:
- All read-functions are tried. If one or more fail, the daemon will
exit with a non-zero exit status, but all read-functions will be
tried.
- Don't quit if one read-function failed without calling all the
shutdown-functions first. This will clean up the UNIX socket of the
unixsock plugin and stuff like that.
collectd: Add the -T option.
The -T switch:
- implies -f
- does not start any read threads
- invokes callback() foreach list_read and will exit 1 on failure, 0
otherwise
Makes for a nicer dryrun than -f + ^C, also makes it possible to wrap
with a test harness such as Test.pm
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
The -T switch:
- implies -f
- does not start any read threads
- invokes callback() foreach list_read and will exit 1 on failure, 0
otherwise
Makes for a nicer dryrun than -f + ^C, also makes it possible to wrap
with a test harness such as Test.pm
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
src/collectd.h: #undef macros before poisoning them.
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
csv plugin: Make it possible to write values to STDOUT.
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@leeloo.lan.home.verplant.org>
openvpn plugin: Added support for LZO compression.
There are four counters for pre/post (de)compression. The code compares
the pre/post counters and submits the current compression ratio based on
the counter change rate.
The type `compression_ratio' (gauge) is used to store the data.
There are four counters for pre/post (de)compression. The code compares
the pre/post counters and submits the current compression ratio based on
the counter change rate.
The type `compression_ratio' (gauge) is used to store the data.
src/common.[ch]: Add `counter_diff', a function to calculate the difference of two counters.
It's not just a simple (old - new), because wrap-around is handled, too.
It's not just a simple (old - new), because wrap-around is handled, too.
processes plugin: Collect VM and stack size stats on Linux.
On other operating systems those values will currently always be zero since I
do not have appropriate test systems available.
The new types "ps_stacksize" and "ps_vm" have been added to types.db.
On other operating systems those values will currently always be zero since I
do not have appropriate test systems available.
The new types "ps_stacksize" and "ps_vm" have been added to types.db.
.gitignore: Added libcollectclient stuff.
libcollectdclient: Added LCC_{BEGIN,END}_DECLS macros.
These macros define an extern "C" block when using C++. This allows to use the
headers from C++ as well.
These macros define an extern "C" block when using C++. This allows to use the
headers from C++ as well.
libcollectdclient: Added header lcc_features.h.
This header provides various macros and functions that may be used to
determine the version of libcollectdclient (i.e. collectd). The file will be
auto-created from lcc_features.h.in by configure.
The macro LCC_VERSION (previously defined in client.h) has been moved to this
header as well and has been renamed to LCC_API_VERSION.
This allows for fine grained version checks at compile- and runtime.
This header provides various macros and functions that may be used to
determine the version of libcollectdclient (i.e. collectd). The file will be
auto-created from lcc_features.h.in by configure.
The macro LCC_VERSION (previously defined in client.h) has been moved to this
header as well and has been renamed to LCC_API_VERSION.
This allows for fine grained version checks at compile- and runtime.
openvpn plugin: Some small cleanups.
value match: Add the `DataSource' and `Satisfy' configuration options.
This adds quite some flexibility to the match. collectd.conf(5) is kept
up to date, too.
This adds quite some flexibility to the match. collectd.conf(5) is kept
up to date, too.
Merge branch 'collectd-4.5'
Conflicts:
src/types.db.pod
Conflicts:
src/types.db.pod
Bumped version to 4.5.2; Updated ChangeLog.
Merge branch 'collectd-4.4' into collectd-4.5
Conflicts:
ChangeLog
version-gen.sh
Conflicts:
ChangeLog
version-gen.sh
Bumped version to 4.4.5; Updated ChangeLog.
contrib/README: Add the solaris-smf subdirectory.
contrib/solaris-smf: Add Solaris SMF information to contrib/.
See README or <http://phosowicz.jogger.pl/2008/12/21/smf-izing-collectd/> for
details.
Signed-off-by: Florian Forster <octo@huhu.verplant.org>
See README or <http://phosowicz.jogger.pl/2008/12/21/smf-izing-collectd/> for
details.
Signed-off-by: Florian Forster <octo@huhu.verplant.org>
Merge branch 'dm/openvpn'
collectd.conf(5): Fix two typos.
collectd.conf(5): Added information about the openvpn plugin.
openvpn plugin: Add a plugin to read OpenVPN statistics.
Simple but useful module for monitoring per-client openvpn traffic..
The plugin reads a statistics file maintained by OpenVPN. The location of the
statistics file is configurable.
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@huhu.verplant.org>
Simple but useful module for monitoring per-client openvpn traffic..
The plugin reads a statistics file maintained by OpenVPN. The location of the
statistics file is configurable.
Signed-off-by: Doug MacEachern <dougm@hyperic.com>
Signed-off-by: Florian Forster <octo@huhu.verplant.org>
Merge branch 'collectd-4.4' into collectd-4.5
Conflicts:
src/network.c
Conflicts:
src/network.c
network plugin: Fix an invalid size of buffer being used.
When specifying the amount of data to copy, we used `sizeof (buffer)' where
`buffer' is a pointer, giving 4 or 8 bytes, depending on the architecture (and
not depending on the actual buffer size). This results in the `type' being sent
much more often than necessary and sometimes not sending a new type when is was
actually necessary. The only prominent case in the default configuration(s) was
`cpufreq' being used instead of `cpu', though.
While in the process, the global `type' buffer was replaced, because an
appropriate buffer is in `value_list_t' now.
Much thanks to Bruno Prémont for reporting and debugging this issue :)
References: #37
When specifying the amount of data to copy, we used `sizeof (buffer)' where
`buffer' is a pointer, giving 4 or 8 bytes, depending on the architecture (and
not depending on the actual buffer size). This results in the `type' being sent
much more often than necessary and sometimes not sending a new type when is was
actually necessary. The only prominent case in the default configuration(s) was
`cpufreq' being used instead of `cpu', though.
While in the process, the global `type' buffer was replaced, because an
appropriate buffer is in `value_list_t' now.
Much thanks to Bruno Prémont for reporting and debugging this issue :)
References: #37
network plugin: A fix for the buffer problem for version 4.4.
This patch fixes the same problem as 6605ff1a, the insufficient copying of
values to the `type' buffer.
Thanks again to Bruno Prémont for reporting and debugging this :)
Resolves: #37
This patch fixes the same problem as 6605ff1a, the insufficient copying of
values to the `type' buffer.
Thanks again to Bruno Prémont for reporting and debugging this :)
Resolves: #37
libcollectdclient: Don't use `PF_UNIX', it's broken on Mac OS X.
Under Mac OS X (10.4, possibly others), PF_UNIX is defined as PF_LOCAL, which
in turn is defined as AF_LOCAL. AF_LOCAL, however, is only defined if
POSIX_C_SOURCE is not.
Under Mac OS X (10.4, possibly others), PF_UNIX is defined as PF_LOCAL, which
in turn is defined as AF_LOCAL. AF_LOCAL, however, is only defined if
POSIX_C_SOURCE is not.
network plugin: Fix an invalid size of buffer being used.
When specifying the amount of data to copy, we used `sizeof (buffer)' where
`buffer' is a pointer, giving 4 or 8 bytes, depending on the architecture (and
not depending on the actual buffer size). This results in the `type' being sent
much more often than necessary and sometimes not sending a new type when is was
actually necessary. The only prominent case in the default configuration(s) was
`cpufreq' being used instead of `cpu', though.
While in the process, the global `type' buffer was replaced, because an
appropriate buffer is in `value_list_t' now.
Much thanks to Bruno Prémont for reporting and debugging this issue :)
References: #37
When specifying the amount of data to copy, we used `sizeof (buffer)' where
`buffer' is a pointer, giving 4 or 8 bytes, depending on the architecture (and
not depending on the actual buffer size). This results in the `type' being sent
much more often than necessary and sometimes not sending a new type when is was
actually necessary. The only prominent case in the default configuration(s) was
`cpufreq' being used instead of `cpu', though.
While in the process, the global `type' buffer was replaced, because an
appropriate buffer is in `value_list_t' now.
Much thanks to Bruno Prémont for reporting and debugging this issue :)
References: #37
src/utils_subst.c: Print the warning of the subst_string function when appropriate.
A leftover from the previous setting, a hard-coded maximum, was forgotten.
A leftover from the previous setting, a hard-coded maximum, was forgotten.
notification target: Add a target that dispatches notifications.
Not tested very well yet, but it works essentially.
Not tested very well yet, but it works essentially.
src/common.c: Change `sstrdup' to not call `strdup' anymore.
src/utils_subst.[ch]: Implement `subst_string'.
Works like `subst', but instead of specifying start and end offsets you
specify `needle', the string that is to be replaced. If `needle' is found
in `string' (using strstr(3)), the offset is calculated and `subst' is
called with the determined parameters.
Works like `subst', but instead of specifying start and end offsets you
specify `needle', the string that is to be replaced. If `needle' is found
in `string' (using strstr(3)), the offset is calculated and `subst' is
called with the determined parameters.
types.db(5): Fix a typo.
replace target: Add a target to replace part of an identifier..
..using regular expressions.
Possible future improvements: Use submatches in the replacement string,
similar to sed's \1, \2, ... back-references.
..using regular expressions.
Possible future improvements: Use submatches in the replacement string,
similar to sed's \1, \2, ... back-references.
types.db(5): Explain in more detail how custom types should be added.
Apparently there was quite some confusion over this.
Apparently there was quite some confusion over this.
Merge branch 'collectd-4.5'
Merge branch 'collectd-4.4' into collectd-4.5
Conflicts:
src/utils_dns.c
Conflicts:
src/utils_dns.c
Do not use *printf() to report errors / debugging messages.
The only exception to that is during configuration (which is done before
daemonization) to let the user know about problems immediately.
The only exception to that is during configuration (which is done before
daemonization) to let the user know about problems immediately.
src/Makefile: Do not print the command when checking for POD ERRORS.
Else, the build output would include the word "error" in a cryptically looking
paragraph which might confuse users.
Else, the build output would include the word "error" in a cryptically looking
paragraph which might confuse users.
collectd.conf(5): Updated the section about the "mysql" plugin.
* Do not talk about what RRD files things end up in - there is more than
"rrdtool" by now.
* Updated the section number of the "Server Status Variables" in the MySQL
reference manual.
* Mention that the MySQL user does not need any special privileges.
* Do not talk about what RRD files things end up in - there is more than
"rrdtool" by now.
* Updated the section number of the "Server Status Variables" in the MySQL
reference manual.
* Mention that the MySQL user does not need any special privileges.
collectd: Abort with an appropriate status in exit_usage().
In case of an error, the exit status code now equals 1.
In case of an error, the exit status code now equals 1.
collectd: Abort if any non-options are left over after command-line parsing.
This happens if any arguments not starting with a dash are passed to collectd.
Any of those are invalid, so let the user know about that.
Thanks to Kris (gruntruk) for (unknowingly) pointing that out ;-)
This happens if any arguments not starting with a dash are passed to collectd.
Any of those are invalid, so let the user know about that.
Thanks to Kris (gruntruk) for (unknowingly) pointing that out ;-)
collectd-nagios: Be more verbose before bailing out with usage().
collectd-nagios: Added "percentage" to the help output.
collectd-nagios: Generalized the "percentage" consolidation function.
Calculation of the percentage is not limited to two values any more but an
arbitrary number may be used now. This allows a more flexible usage.
Also, the documentation has been updated.
Calculation of the percentage is not limited to two values any more but an
arbitrary number may be used now. This allows a more flexible usage.
Also, the documentation has been updated.
collectd-nagios: add new "percentage" aggregate function.
the percentage aggregate function takes exactly two values. it
computes the percentage of the first value versus the sum of the first
and the second value. very useful for checking df values like:
sudo collectd-nagios -s /var/run/collectd-unixsock -H localhost -n df/df-root -d used -d free -g percentage -w 20: -c 10:
which will warn if there is less than 20% free space.
Signed-off-by: Sebastian Harl <sh@tokkee.org>
the percentage aggregate function takes exactly two values. it
computes the percentage of the first value versus the sum of the first
and the second value. very useful for checking df values like:
sudo collectd-nagios -s /var/run/collectd-unixsock -H localhost -n df/df-root -d used -d free -g percentage -w 20: -c 10:
which will warn if there is less than 20% free space.
Signed-off-by: Sebastian Harl <sh@tokkee.org>
collectd-nagios: If DSs are specified, sort the values accordingly.
If a user specifies DSs on the command line, the order or the values might
matter (e.g. when using the upcoming "percentage" consolidation function). So,
in this case we now order the values according to the order of the specified
DSs.
This patch replaces the function ignore_ds() with filter_ds() which is called
once in do_check(). This removes the need to check for ignored values in all
consolidation function implementations.
If a user specifies DSs on the command line, the order or the values might
matter (e.g. when using the upcoming "percentage" consolidation function). So,
in this case we now order the values according to the order of the specified
DSs.
This patch replaces the function ignore_ds() with filter_ds() which is called
once in do_check(). This removes the need to check for ignored values in all
consolidation function implementations.
libcollectdclient: Build the library using -Wall -Werror.
Fixed two minor issues detected by those flags.
Fixed two minor issues detected by those flags.
libcollectdclient/client.c: Let COLLECT_DEBUG decide about debugging.
The file now includes config.h and enables debugging based on COLLECT_DEBUG
instead of some hard-coded value.
The file now includes config.h and enables debugging based on COLLECT_DEBUG
instead of some hard-coded value.
perl plugin, utils_complain: Fixed errors identified by -Wformat-security.
A few "format not a string literal and no format arguments" errors have been
fixed by using "%s" as the format argument.
Thanks to Kevin (onebinary) for reporting this.
A few "format not a string literal and no format arguments" errors have been
fixed by using "%s" as the format argument.
Thanks to Kevin (onebinary) for reporting this.
collectd.conf.in: Fixed a wrong type used in the "tail" example.
collectd.conf.pod: Consistently use spaces instead of tabs in examples.
processes plugin: Get complete command line for processes on Linux as well.
On Linux, /proc/<pid>/cmdline is now used to determine the complete command
line of a process which may then be used with the "ProcessMatch" configuration
option.
On Linux, /proc/<pid>/cmdline is now used to determine the complete command
line of a process which may then be used with the "ProcessMatch" configuration
option.
processes plugin: Do not silently ignore configuration errors.
collectd.conf(5): Documented the "ProcessMatch" option.
configfile.c: Ignore registered undefined config keys.
If a registered config key equals NULL, it is now ignored. Before, this would
cause a segfault. As this has happened a few times in the past, we not handle
it sanely.
If a registered config key equals NULL, it is now ignored. Before, this would
cause a segfault. As this has happened a few times in the past, we not handle
it sanely.
processes plugin: Fixed a segfault when handling unknown config options.
When using STATIC_ARRAY_SIZE() to determine the number of config keys, the
list must not include NULL. Else, the configfile module will handle that as
another config key and dereference it in case no previous key matched, i.e.
when handling unknown config options.
When using STATIC_ARRAY_SIZE() to determine the number of config keys, the
list must not include NULL. Else, the configfile module will handle that as
another config key and dereference it in case no previous key matched, i.e.
when handling unknown config options.
unixsock plugin: Make sure the initialization function is run only once.
If called multiple times, the global `sock_fd' variable will be re-set,
breaking the plugin.
If called multiple times, the global `sock_fd' variable will be re-set,
breaking the plugin.
Merge branch 'collectd-4.5'
configure.in: Check for `kvm_openfiles'.
Check for both, `kvm_nlist' and `kvm_openfiles' before enabling the
tcpconns plugin with libkvm: Solaris has a libkvm, too, but doesn't
provide the necessary functions to read the connections with it. At
least not in the same way.
Check for both, `kvm_nlist' and `kvm_openfiles' before enabling the
tcpconns plugin with libkvm: Solaris has a libkvm, too, but doesn't
provide the necessary functions to read the connections with it. At
least not in the same way.
Merge branch 'collectd-4.4' into collectd-4.5
configure.in: Some improvements for Solaris, expecially the swap plugin.
If Solaris is detected, "-D_POSIX_PTHREAD_SEMANTICS" will be added to
CPPFLAGS automatically now.
If sys/swap.h is not found, print a message suggesting to build a 64bit
binary. (sys/swap.h cannot be included when a 32bit binary is being built,
but it's most likely present under Solaris.)
Enable the swap plugin only if sys/swap.h has been found (when under
Solaris).
If Solaris is detected, "-D_POSIX_PTHREAD_SEMANTICS" will be added to
CPPFLAGS automatically now.
If sys/swap.h is not found, print a message suggesting to build a 64bit
binary. (sys/swap.h cannot be included when a 32bit binary is being built,
but it's most likely present under Solaris.)
Enable the swap plugin only if sys/swap.h has been found (when under
Solaris).
src/collectd-nagios.c: Include "config.h" before <stdlib.h>.
Otherwise _FILE_OFFSET_BITS may be redefined.
Otherwise _FILE_OFFSET_BITS may be redefined.
Merge branch 'collectd-4.5'
src/plugin.[ch]: Don't use unnamed unions.
They're non-standard and cause a lot of trouble.
They're non-standard and cause a lot of trouble.
libcollectdclient/client.c: Define NI_MAXHOST to 1025.
Apparently this define has been obsoleted by RFC 3493, so SunOS 5.11 no
longer defines it. This probably is a problem somewhere else, too.
Apparently this define has been obsoleted by RFC 3493, so SunOS 5.11 no
longer defines it. This probably is a problem somewhere else, too.
Merge branch 'collectd-4.5'
Merge branch 'collectd-4.4' into collectd-4.5
multimeter plugin: Check the return value of `write'.
This resolves warnings when compiling on GNU with:
CPPFLAGS="-D_FORTIFY_SOURCE"
This resolves warnings when compiling on GNU with:
CPPFLAGS="-D_FORTIFY_SOURCE"
collectd.conf(5): Fix a POD error.
value match: Add a match to check the actual value(s) of data-sources.
src/filter_chain.[ch]: Implement the built-in `return' target.
The `stop' target now aborts all processing of the value completely, `return'
only stops the current chain and continues processing the parent chain.
The collectd.conf(5) manual page has been updated and the `FC_ACTION_*' defines
have been renamed to `FC_TARGET_*'.
The `stop' target now aborts all processing of the value completely, `return'
only stops the current chain and continues processing the parent chain.
The collectd.conf(5) manual page has been updated and the `FC_ACTION_*' defines
have been renamed to `FC_TARGET_*'.
collectd.conf(5): Document the regex match and the set target.
regex match: Require *all* regular expressions to match.
Before, when there were multiple regular expressions for the same field,
for example the host, only of of those regular expressions needed to
match for a successful overall match.
The new behavior is not that *all* regular expressions need to match for
a successful overall match. This makes the behavior much easier to
understand.
Before, when there were multiple regular expressions for the same field,
for example the host, only of of those regular expressions needed to
match for a successful overall match.
The new behavior is not that *all* regular expressions need to match for
a successful overall match. This makes the behavior much easier to
understand.
set target: Don't allow setting of the `type' field.
set target: Implement a target to set fields in a value list.
It's currently possible to re-set the `type'. This is most likely not
good and will result in unpredictable behavior. I'll remove that,
probably.
It's currently possible to re-set the `type'. This is most likely not
good and will result in unpredictable behavior. I'll remove that,
probably.
regex match: Add myself to the list of copyright holders.
Sebastian insisted on it ;)
Sebastian insisted on it ;)
src/filter_chain.c: Beatify debug and warning messages.
src/filter_chain.c: Actually abort default targets when one signals `stop'.
It should not behave as documented.
It should not behave as documented.
src/filter_chain.c: Fix a rather unimportant typo.
collectd.conf.in: Add something for the filtering stuff.
Mostly a pointer to the manual page for now.
Mostly a pointer to the manual page for now.
collectd.conf(5): Add an example to the filter documentation.
src/filter_chain.c: Implement the default behavior.
If no <Chain> is defined, send all values to all write plugins.
If no <Chain> is defined, send all values to all write plugins.
src/meta_data.[ch]: Added a framework for meta data.
This framework is planned to be used in at least three places:
- Extend the global cache to allow attaching a meta-data object to each
entry. The meta-data can then be used by other plugins to attach
miscellaneous information to an entry. Possibly uses include
calculation of floating averages or storing the state for a hysteresis
threshold.
- Add a meta-data object to every `value_list_t' in
`plugin_dispatch_values'. The meta-data object can then be used by
matches and targets to pass information about a value-list to one
another.
- Replace the notification meta-data stuff in src/plugin.[ch] with such
an object.
This framework is planned to be used in at least three places:
- Extend the global cache to allow attaching a meta-data object to each
entry. The meta-data can then be used by other plugins to attach
miscellaneous information to an entry. Possibly uses include
calculation of floating averages or storing the state for a hysteresis
threshold.
- Add a meta-data object to every `value_list_t' in
`plugin_dispatch_values'. The meta-data object can then be used by
matches and targets to pass information about a value-list to one
another.
- Replace the notification meta-data stuff in src/plugin.[ch] with such
an object.
Merge branch 'ff/libcollectdclient'
Merge branch 'ff/filter'
src/filter_chain.c, src/match_regex.c: Add some debugging messages.
src/plubin.c: Call `fc_process' instead of calling the write plugins directly.
src/filter_chain.c: Allow the create-function of mathces to be NULL.
Not all targets actually need to be created, for example
the `stop' target.
Not all targets actually need to be created, for example
the `stop' target.
src/filter_chain.c: Add `Target' blocks to `Chain' blocks.
I simply forgot them :/
I simply forgot them :/
match_regex plugin: Fix a typo.
src/configfile.c: Dispatch the `Chain' blocks to fc_configure.
src/plugin.c: Use the write-list in plugin_write.
Not the flush-list, doh!
Not the flush-list, doh!
match_regex plugin: Renamed `filter_pcre' to `match_regex'.
In order to fit into the new match/target schema, the substitute part of the
plugin has been removed for now and will be put in a target plugin in the
future.
The match_regex now registeres a match with the new infrastructure and uses
regular expressions to match certain values based on their identifier.
In order to fit into the new match/target schema, the substitute part of the
plugin has been removed for now and will be put in a target plugin in the
future.
The match_regex now registeres a match with the new infrastructure and uses
regular expressions to match certain values based on their identifier.
src/filter_chain.[ch]: Implement an advanced filtering framework.
The concept separates `filter' plugins into `match' and `target' plugins. Those
two parts can be combined in wild variation, allowing *very* flexible
mechanisms.
The concept is explained in more detail in the `FILTER CONFIGURATION' section
of collectd.conf(5), also added with this commit.
The concept separates `filter' plugins into `match' and `target' plugins. Those
two parts can be combined in wild variation, allowing *very* flexible
mechanisms.
The concept is explained in more detail in the `FILTER CONFIGURATION' section
of collectd.conf(5), also added with this commit.
src/plugin.[ch]: Implement `plugin_write', a function to pass values to specific write plugins.
This function does none of the magic `plugin_dispatch_value' does. It simiply
takes a value and passes it to the requested plugin.
This function is intended to be used with the upcoming `filter chain' code,
which will allow to pass values to specific plugins only.
This function does none of the magic `plugin_dispatch_value' does. It simiply
takes a value and passes it to the requested plugin.
This function is intended to be used with the upcoming `filter chain' code,
which will allow to pass values to specific plugins only.
configure.in: Include <stdint.h> and <inttypes.h> when checking for double storage layout.
Some systems, for example Solaris 9, don't have <stdint.h> and define
`uint32_t' via <inttypes.h> instead. Be more permissive by including
both header files when checking how doubles are stored in memory.
Some systems, for example Solaris 9, don't have <stdint.h> and define
`uint32_t' via <inttypes.h> instead. Be more permissive by including
both header files when checking how doubles are stored in memory.
contrib/collection.cgi: Add meta-graph definitions for the vmem plugin.
contrib/collection.cgi: Add graph definitions for the vmem plugin.