Code

a331537814b077f6657ecfd08b21e492ab32a1e1
[pkg-rrdtool.git] / doc / rrdcreate.html
1 <?xml version="1.0" ?>
2 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
3 <html xmlns="http://www.w3.org/1999/xhtml">
4 <head>
5 <title>rrdcreate</title>
6 <meta http-equiv="content-type" content="text/html; charset=utf-8" />
7 <link rev="made" href="mailto:root@localhost" />
8 </head>
10 <body style="background-color: white">
12 <p><a name="__index__"></a></p>
13 <!-- INDEX BEGIN -->
14 <!--
16 <ul>
18         <li><a href="#name">NAME</a></li>
19         <li><a href="#synopsis">SYNOPSIS</a></li>
20         <li><a href="#description">DESCRIPTION</a></li>
21         <li><a href="#aberrant_behavior_detection_with_holtwinters_forecasting">Aberrant Behavior Detection with Holt-Winters Forecasting</a></li>
22         <li><a href="#the_heartbeat_and_the_step">The HEARTBEAT and the STEP</a></li>
23         <li><a href="#how_to_measure">HOW TO MEASURE</a></li>
24         <li><a href="#example">EXAMPLE</a></li>
25         <li><a href="#example_2">EXAMPLE 2</a></li>
26         <li><a href="#example_3">EXAMPLE 3</a></li>
27         <li><a href="#author">AUTHOR</a></li>
28 </ul>
29 -->
30 <!-- INDEX END -->
32 <p>
33 </p>
34 <h1><a name="name">NAME</a></h1>
35 <p>rrdcreate - Set up a new Round Robin Database</p>
36 <p>
37 </p>
38 <hr />
39 <h1><a name="synopsis">SYNOPSIS</a></h1>
40 <p><strong>rrdtool</strong> <strong>create</strong> <em>filename</em>
41 [<strong>--start</strong>|<strong>-b</strong>&nbsp;<em>start&nbsp;time</em>]
42 [<strong>--step</strong>|<strong>-s</strong>&nbsp;<em>step</em>]
43 [<strong>DS:</strong><em>ds-name</em><strong>:</strong><em>DST</em><strong>:</strong><em>dst&nbsp;arguments</em>]
44 [<strong>RRA:</strong><em>CF</em><strong>:</strong><em>cf&nbsp;arguments</em>]</p>
45 <p>
46 </p>
47 <hr />
48 <h1><a name="description">DESCRIPTION</a></h1>
49 <p>The create function of RRDtool lets you set up new Round Robin
50 Database (<strong>RRD</strong>) files.  The file is created at its final, full size
51 and filled with <em>*UNKNOWN*</em> data.</p>
52 <dl>
53 <dt><strong><a name="item_filename"><em>filename</em></a></strong>
55 <dd>
56 <p>The name of the <strong>RRD</strong> you want to create. <strong>RRD</strong> files should end
57 with the extension <em>.rrd</em>. However, <strong>RRDtool</strong> will accept any
58 filename.</p>
59 </dd>
60 </li>
61 <dt><strong><a name="item_time"><strong>--start</strong>|<strong>-b</strong> <em>start time</em> (default: now - 10s)</a></strong>
63 <dd>
64 <p>Specifies the time in seconds since 1970-01-01 UTC when the first
65 value should be added to the <strong>RRD</strong>. <strong>RRDtool</strong> will not accept
66 any data timed before or at the time specified.</p>
67 </dd>
68 <dd>
69 <p>See also AT-STYLE TIME SPECIFICATION section in the
70 <em>rrdfetch</em> documentation for other ways to specify time.</p>
71 </dd>
72 </li>
73 <dt><strong><a name="item_step"><strong>--step</strong>|<strong>-s</strong> <em>step</em> (default: 300 seconds)</a></strong>
75 <dd>
76 <p>Specifies the base interval in seconds with which data will be fed
77 into the <strong>RRD</strong>.</p>
78 </dd>
79 </li>
80 <dt><strong><a name="item_ds_3ads_2dname_3adst_3adst_arguments"><strong>DS:</strong><em>ds-name</em><strong>:</strong><em>DST</em><strong>:</strong><em>dst arguments</em></a></strong>
82 <dd>
83 <p>A single <strong>RRD</strong> can accept input from several data sources (<strong>DS</strong>),
84 for example incoming and outgoing traffic on a specific communication
85 line. With the <strong>DS</strong> configuration option you must define some basic
86 properties of each data source you want to store in the <strong>RRD</strong>.</p>
87 </dd>
88 <dd>
89 <p><em>ds-name</em> is the name you will use to reference this particular data
90 source from an <strong>RRD</strong>. A <em>ds-name</em> must be 1 to 19 characters long in
91 the characters [a-zA-Z0-9_].</p>
92 </dd>
93 <dd>
94 <p><em>DST</em> defines the Data Source Type. The remaining arguments of a
95 data source entry depend on the data source type. For GAUGE, COUNTER,
96 DERIVE, and ABSOLUTE the format for a data source entry is:</p>
97 </dd>
98 <dd>
99 <p><strong>DS:</strong><em>ds-name</em><strong>:</strong><em>GAUGE | COUNTER | DERIVE | ABSOLUTE</em><strong>:</strong><em>heartbeat</em><strong>:</strong><em>min</em><strong>:</strong><em>max</em></p>
100 </dd>
101 <dd>
102 <p>For COMPUTE data sources, the format is:</p>
103 </dd>
104 <dd>
105 <p><strong>DS:</strong><em>ds-name</em><strong>:</strong><em>COMPUTE</em><strong>:</strong><em>rpn-expression</em></p>
106 </dd>
107 <dd>
108 <p>In order to decide which data source type to use, review the
109 definitions that follow. Also consult the section on ``HOW TO MEASURE''
110 for further insight.</p>
111 </dd>
112 <dl>
113 <dt><strong><a name="item_gauge"><strong>GAUGE</strong></a></strong>
115 <dd>
116 <p>is for things like temperatures or number of people in a room or the
117 value of a RedHat share.</p>
118 </dd>
119 </li>
120 <dt><strong><a name="item_counter"><strong>COUNTER</strong></a></strong>
122 <dd>
123 <p>is for continuous incrementing counters like the ifInOctets counter in
124 a router. The <strong>COUNTER</strong> data source assumes that the counter never
125 decreases, except when a counter overflows.  The update function takes
126 the overflow into account.  The counter is stored as a per-second
127 rate. When the counter overflows, RRDtool checks if the overflow
128 happened at the 32bit or 64bit border and acts accordingly by adding
129 an appropriate value to the result.</p>
130 </dd>
131 </li>
132 <dt><strong><a name="item_derive"><strong>DERIVE</strong></a></strong>
134 <dd>
135 <p>will store the derivative of the line going from the last to the
136 current value of the data source. This can be useful for gauges, for
137 example, to measure the rate of people entering or leaving a
138 room. Internally, derive works exactly like COUNTER but without
139 overflow checks. So if your counter does not reset at 32 or 64 bit you
140 might want to use DERIVE and combine it with a MIN value of 0.</p>
141 </dd>
142 <dd>
143 <p>NOTE on COUNTER vs DERIVE</p>
144 </dd>
145 <dd>
146 <p>by Don Baarda &lt;<a href="mailto:don.baarda@baesystems.com">don.baarda@baesystems.com</a>&gt;</p>
147 </dd>
148 <dd>
149 <p>If you cannot tolerate ever mistaking the occasional counter reset for a
150 legitimate counter wrap, and would prefer ``Unknowns'' for all legitimate
151 counter wraps and resets, always use DERIVE with min=0. Otherwise, using
152 COUNTER with a suitable max will return correct values for all legitimate
153 counter wraps, mark some counter resets as ``Unknown'', but can mistake some
154 counter resets for a legitimate counter wrap.</p>
155 </dd>
156 <dd>
157 <p>For a 5 minute step and 32-bit counter, the probability of mistaking a
158 counter reset for a legitimate wrap is arguably about 0.8% per 1Mbps of
159 maximum bandwidth. Note that this equates to 80% for 100Mbps interfaces, so
160 for high bandwidth interfaces and a 32bit counter, DERIVE with min=0 is
161 probably preferable. If you are using a 64bit counter, just about any max
162 setting will eliminate the possibility of mistaking a reset for a counter
163 wrap.</p>
164 </dd>
165 </li>
166 <dt><strong><a name="item_absolute"><strong>ABSOLUTE</strong></a></strong>
168 <dd>
169 <p>is for counters which get reset upon reading. This is used for fast counters
170 which tend to overflow. So instead of reading them normally you reset them
171 after every read to make sure you have a maximum time available before the
172 next overflow. Another usage is for things you count like number of messages
173 since the last update.</p>
174 </dd>
175 </li>
176 <dt><strong><a name="item_compute"><strong>COMPUTE</strong></a></strong>
178 <dd>
179 <p>is for storing the result of a formula applied to other data sources
180 in the <strong>RRD</strong>. This data source is not supplied a value on update, but
181 rather its Primary Data Points (PDPs) are computed from the PDPs of
182 the data sources according to the rpn-expression that defines the
183 formula. Consolidation functions are then applied normally to the PDPs
184 of the COMPUTE data source (that is the rpn-expression is only applied
185 to generate PDPs). In database software, such data sets are referred
186 to as ``virtual'' or ``computed'' columns.</p>
187 </dd>
188 </li>
189 </dl>
190 <p><em>heartbeat</em> defines the maximum number of seconds that may pass
191 between two updates of this data source before the value of the
192 data source is assumed to be <em>*UNKNOWN*</em>.</p>
193 <p><em>min</em> and <em>max</em> define the expected range values for data supplied by a
194 data source. If <em>min</em> and/or <em>max</em> any value outside the defined range
195 will be regarded as <em>*UNKNOWN*</em>. If you do not know or care about min and
196 max, set them to U for unknown. Note that min and max always refer to the
197 processed values of the DS. For a traffic-<strong>COUNTER</strong> type DS this would be
198 the maximum and minimum data-rate expected from the device.</p>
199 <p><em>If information on minimal/maximal expected values is available,
200 always set the min and/or max properties. This will help RRDtool in
201 doing a simple sanity check on the data supplied when running update.</em></p>
202 <p><em>rpn-expression</em> defines the formula used to compute the PDPs of a
203 COMPUTE data source from other data sources in the same &lt;RRD&gt;. It is
204 similar to defining a <strong>CDEF</strong> argument for the graph command. Please
205 refer to that manual page for a list and description of RPN operations
206 supported. For COMPUTE data sources, the following RPN operations are
207 not supported: COUNT, PREV, TIME, and LTIME. In addition, in defining
208 the RPN expression, the COMPUTE data source may only refer to the
209 names of data source listed previously in the create command. This is
210 similar to the restriction that <strong>CDEF</strong>s must refer only to <strong>DEF</strong>s
211 and <strong>CDEF</strong>s previously defined in the same graph command.</p>
212 <dt><strong><a name="item_rra_3acf_3acf_arguments"><strong>RRA:</strong><em>CF</em><strong>:</strong><em>cf arguments</em></a></strong>
214 <dd>
215 <p>The purpose of an <strong>RRD</strong> is to store data in the round robin archives
216 (<strong>RRA</strong>). An archive consists of a number of data values or statistics for
217 each of the defined data-sources (<strong>DS</strong>) and is defined with an <strong>RRA</strong> line.</p>
218 </dd>
219 <dd>
220 <p>When data is entered into an <strong>RRD</strong>, it is first fit into time slots
221 of the length defined with the <strong>-s</strong> option, thus becoming a <em>primary
222 data point</em>.</p>
223 </dd>
224 <dd>
225 <p>The data is also processed with the consolidation function (<em>CF</em>) of
226 the archive. There are several consolidation functions that
227 consolidate primary data points via an aggregate function: <strong>AVERAGE</strong>,
228 <strong>MIN</strong>, <strong>MAX</strong>, <strong>LAST</strong>. The format of <strong>RRA</strong> line for these
229 consolidation functions is:</p>
230 </dd>
231 <dd>
232 <p><strong>RRA:</strong><em>AVERAGE | MIN | MAX | LAST</em><strong>:</strong><em>xff</em><strong>:</strong><em>steps</em><strong>:</strong><em>rows</em></p>
233 </dd>
234 <dd>
235 <p><em>xff</em> The xfiles factor defines what part of a consolidation interval may
236 be made up from <em>*UNKNOWN*</em> data while the consolidated value is still
237 regarded as known. It is given as the ratio of allowed <em>*UNKNOWN*</em> PDPs
238 to the number of PDPs in the interval. Thus, it ranges from 0 to 1 (exclusive).</p>
239 </dd>
240 <dd>
241 <p><em>steps</em> defines how many of these <em>primary data points</em> are used to build
242 a <em>consolidated data point</em> which then goes into the archive.</p>
243 </dd>
244 <dd>
245 <p><em>rows</em> defines how many generations of data values are kept in an <strong>RRA</strong>.</p>
246 </dd>
247 </li>
248 </dl>
249 <p>
250 </p>
251 <hr />
252 <h1><a name="aberrant_behavior_detection_with_holtwinters_forecasting">Aberrant Behavior Detection with Holt-Winters Forecasting</a></h1>
253 <p>In addition to the aggregate functions, there are a set of specialized
254 functions that enable <strong>RRDtool</strong> to provide data smoothing (via the
255 Holt-Winters forecasting algorithm), confidence bands, and the
256 flagging aberrant behavior in the data source time series:</p>
257 <ul>
258 <li>
259 <p><strong>RRA:</strong><em>HWPREDICT</em><strong>:</strong><em>rows</em><strong>:</strong><em>alpha</em><strong>:</strong><em>beta</em><strong>:</strong><em>seasonal period</em>[<strong>:</strong><em>rra-num</em>]</p>
260 </li>
261 <li>
262 <p><strong>RRA:</strong><em>SEASONAL</em><strong>:</strong><em>seasonal period</em><strong>:</strong><em>gamma</em><strong>:</strong><em>rra-num</em></p>
263 </li>
264 <li>
265 <p><strong>RRA:</strong><em>DEVSEASONAL</em><strong>:</strong><em>seasonal period</em><strong>:</strong><em>gamma</em><strong>:</strong><em>rra-num</em></p>
266 </li>
267 <li>
268 <p><strong>RRA:</strong><em>DEVPREDICT</em><strong>:</strong><em>rows</em><strong>:</strong><em>rra-num</em></p>
269 </li>
270 <li>
271 <p><strong>RRA:</strong><em>FAILURES</em><strong>:</strong><em>rows</em><strong>:</strong><em>threshold</em><strong>:</strong><em>window length</em><strong>:</strong><em>rra-num</em></p>
272 </li>
273 </ul>
274 <p>These <strong>RRAs</strong> differ from the true consolidation functions in several ways.
275 First, each of the <strong>RRA</strong>s is updated once for every primary data point.
276 Second, these <strong>RRAs</strong> are interdependent. To generate real-time confidence
277 bounds, a matched set of HWPREDICT, SEASONAL, DEVSEASONAL, and
278 DEVPREDICT must exist. Generating smoothed values of the primary data points
279 requires both a HWPREDICT <strong>RRA</strong> and SEASONAL <strong>RRA</strong>. Aberrant behavior
280 detection requires FAILURES, HWPREDICT, DEVSEASONAL, and SEASONAL.</p>
281 <p>The actual predicted, or smoothed, values are stored in the HWPREDICT
282 <strong>RRA</strong>. The predicted deviations are stored in DEVPREDICT (think a standard
283 deviation which can be scaled to yield a confidence band). The FAILURES
284 <strong>RRA</strong> stores binary indicators. A 1 marks the indexed observation as
285 failure; that is, the number of confidence bounds violations in the
286 preceding window of observations met or exceeded a specified threshold. An
287 example of using these <strong>RRAs</strong> to graph confidence bounds and failures
288 appears in <a href="././rrdgraph.html">the rrdgraph manpage</a>.</p>
289 <p>The SEASONAL and DEVSEASONAL <strong>RRAs</strong> store the seasonal coefficients for the
290 Holt-Winters forecasting algorithm and the seasonal deviations, respectively.
291 There is one entry per observation time point in the seasonal cycle. For
292 example, if primary data points are generated every five minutes and the
293 seasonal cycle is 1 day, both SEASONAL and DEVSEASONAL will have 288 rows.</p>
294 <p>In order to simplify the creation for the novice user, in addition to
295 supporting explicit creation of the HWPREDICT, SEASONAL, DEVPREDICT,
296 DEVSEASONAL, and FAILURES <strong>RRAs</strong>, the <strong>RRDtool</strong> create command supports
297 implicit creation of the other four when HWPREDICT is specified alone and
298 the final argument <em>rra-num</em> is omitted.</p>
299 <p><em>rows</em> specifies the length of the <strong>RRA</strong> prior to wrap around. Remember
300 that there is a one-to-one correspondence between primary data points and
301 entries in these RRAs. For the HWPREDICT CF, <em>rows</em> should be larger than
302 the <em>seasonal period</em>. If the DEVPREDICT <strong>RRA</strong> is implicitly created, the
303 default number of rows is the same as the HWPREDICT <em>rows</em> argument. If the
304 FAILURES <strong>RRA</strong> is implicitly created, <em>rows</em> will be set to the <em>seasonal
305 period</em> argument of the HWPREDICT <strong>RRA</strong>. Of course, the <strong>RRDtool</strong>
306 <em>resize</em> command is available if these defaults are not sufficient and the
307 creator wishes to avoid explicit creations of the other specialized function
308 <strong>RRAs</strong>.</p>
309 <p><em>seasonal period</em> specifies the number of primary data points in a seasonal
310 cycle. If SEASONAL and DEVSEASONAL are implicitly created, this argument for
311 those <strong>RRAs</strong> is set automatically to the value specified by HWPREDICT. If
312 they are explicitly created, the creator should verify that all three
313 <em>seasonal period</em> arguments agree.</p>
314 <p><em>alpha</em> is the adaption parameter of the intercept (or baseline)
315 coefficient in the Holt-Winters forecasting algorithm. See <a href="././rrdtool.html">the rrdtool manpage</a> for a
316 description of this algorithm. <em>alpha</em> must lie between 0 and 1. A value
317 closer to 1 means that more recent observations carry greater weight in
318 predicting the baseline component of the forecast. A value closer to 0 means
319 that past history carries greater weight in predicting the baseline
320 component.</p>
321 <p><em>beta</em> is the adaption parameter of the slope (or linear trend) coefficient
322 in the Holt-Winters forecasting algorithm. <em>beta</em> must lie between 0 and 1
323 and plays the same role as <em>alpha</em> with respect to the predicted linear
324 trend.</p>
325 <p><em>gamma</em> is the adaption parameter of the seasonal coefficients in the
326 Holt-Winters forecasting algorithm (HWPREDICT) or the adaption parameter in
327 the exponential smoothing update of the seasonal deviations. It must lie
328 between 0 and 1. If the SEASONAL and DEVSEASONAL <strong>RRAs</strong> are created
329 implicitly, they will both have the same value for <em>gamma</em>: the value
330 specified for the HWPREDICT <em>alpha</em> argument. Note that because there is
331 one seasonal coefficient (or deviation) for each time point during the
332 seasonal cycle, the adaptation rate is much slower than the baseline. Each
333 seasonal coefficient is only updated (or adapts) when the observed value
334 occurs at the offset in the seasonal cycle corresponding to that
335 coefficient.</p>
336 <p>If SEASONAL and DEVSEASONAL <strong>RRAs</strong> are created explicitly, <em>gamma</em> need not
337 be the same for both. Note that <em>gamma</em> can also be changed via the
338 <strong>RRDtool</strong> <em>tune</em> command.</p>
339 <p><em>rra-num</em> provides the links between related <strong>RRAs</strong>. If HWPREDICT is
340 specified alone and the other <strong>RRAs</strong> are created implicitly, then
341 there is no need to worry about this argument. If <strong>RRAs</strong> are created
342 explicitly, then carefully pay attention to this argument. For each
343 <strong>RRA</strong> which includes this argument, there is a dependency between
344 that <strong>RRA</strong> and another <strong>RRA</strong>. The <em>rra-num</em> argument is the 1-based
345 index in the order of <strong>RRA</strong> creation (that is, the order they appear
346 in the <em>create</em> command). The dependent <strong>RRA</strong> for each <strong>RRA</strong>
347 requiring the <em>rra-num</em> argument is listed here:</p>
348 <ul>
349 <li>
350 <p>HWPREDICT <em>rra-num</em> is the index of the SEASONAL <strong>RRA</strong>.</p>
351 </li>
352 <li>
353 <p>SEASONAL <em>rra-num</em> is the index of the HWPREDICT <strong>RRA</strong>.</p>
354 </li>
355 <li>
356 <p>DEVPREDICT <em>rra-num</em> is the index of the DEVSEASONAL <strong>RRA</strong>.</p>
357 </li>
358 <li>
359 <p>DEVSEASONAL <em>rra-num</em> is the index of the HWPREDICT <strong>RRA</strong>.</p>
360 </li>
361 <li>
362 <p>FAILURES <em>rra-num</em> is the index of the DEVSEASONAL <strong>RRA</strong>.</p>
363 </li>
364 </ul>
365 <p><em>threshold</em> is the minimum number of violations (observed values outside
366 the confidence bounds) within a window that constitutes a failure. If the
367 FAILURES <strong>RRA</strong> is implicitly created, the default value is 7.</p>
368 <p><em>window length</em> is the number of time points in the window. Specify an
369 integer greater than or equal to the threshold and less than or equal to 28.
370 The time interval this window represents depends on the interval between
371 primary data points. If the FAILURES <strong>RRA</strong> is implicitly created, the
372 default value is 9.</p>
373 <p>
374 </p>
375 <hr />
376 <h1><a name="the_heartbeat_and_the_step">The HEARTBEAT and the STEP</a></h1>
377 <p>Here is an explanation by Don Baarda on the inner workings of RRDtool.
378 It may help you to sort out why all this *UNKNOWN* data is popping
379 up in your databases:</p>
380 <p>RRDtool gets fed samples at arbitrary times. From these it builds Primary
381 Data Points (PDPs) at exact times on every ``step'' interval. The PDPs are
382 then accumulated into RRAs.</p>
383 <p>The ``heartbeat'' defines the maximum acceptable interval between
384 samples. If the interval between samples is less than ``heartbeat'',
385 then an average rate is calculated and applied for that interval. If
386 the interval between samples is longer than ``heartbeat'', then that
387 entire interval is considered ``unknown''. Note that there are other
388 things that can make a sample interval ``unknown'', such as the rate
389 exceeding limits, or even an ``unknown'' input sample.</p>
390 <p>The known rates during a PDP's ``step'' interval are used to calculate
391 an average rate for that PDP. Also, if the total ``unknown'' time during
392 the ``step'' interval exceeds the ``heartbeat'', the entire PDP is marked
393 as ``unknown''. This means that a mixture of known and ``unknown'' sample
394 times in a single PDP ``step'' may or may not add up to enough ``unknown''
395 time to exceed ``heartbeat'' and hence mark the whole PDP ``unknown''. So
396 ``heartbeat'' is not only the maximum acceptable interval between
397 samples, but also the maximum acceptable amount of ``unknown'' time per
398 PDP (obviously this is only significant if you have ``heartbeat'' less
399 than ``step'').</p>
400 <p>The ``heartbeat'' can be short (unusual) or long (typical) relative to
401 the ``step'' interval between PDPs. A short ``heartbeat'' means you
402 require multiple samples per PDP, and if you don't get them mark the
403 PDP unknown. A long heartbeat can span multiple ``steps'', which means
404 it is acceptable to have multiple PDPs calculated from a single
405 sample. An extreme example of this might be a ``step'' of 5 minutes and a
406 ``heartbeat'' of one day, in which case a single sample every day will
407 result in all the PDPs for that entire day period being set to the
408 same average rate. <em>-- Don Baarda &lt;<a href="mailto:don.baarda@baesystems.com">don.baarda@baesystems.com</a>&gt;</em></p>
409 <pre>
410        time|
411        axis|
412  begin__|00|
413         |01|
414        u|02|----* sample1, restart &quot;hb&quot;-timer
415        u|03|   /
416        u|04|  /
417        u|05| /
418        u|06|/     &quot;hbt&quot; expired
419        u|07|
420         |08|----* sample2, restart &quot;hb&quot; 
421         |09|   / 
422         |10|  /
423        u|11|----* sample3, restart &quot;hb&quot;
424        u|12|   /
425        u|13|  /
426  step1_u|14| /
427        u|15|/     &quot;swt&quot; expired
428        u|16|
429         |17|----* sample4, restart &quot;hb&quot;, create &quot;pdp&quot; for step1 = 
430         |18|   /  = unknown due to 10 &quot;u&quot; labled secs &gt; &quot;hb&quot;
431         |19|  /
432         |20| /
433         |21|----* sample5, restart &quot;hb&quot;
434         |22|   /
435         |23|  /
436         |24|----* sample6, restart &quot;hb&quot;
437         |25|   /
438         |26|  /
439         |27|----* sample7, restart &quot;hb&quot;
440  step2__|28|   /
441         |22|  /
442         |23|----* sample8, restart &quot;hb&quot;, create &quot;pdp&quot; for step1, create &quot;cdp&quot; 
443         |24|   /
444         |25|  /</pre>
445 <p>graphics by <em><a href="mailto:vladimir.lavrov@desy.de">vladimir.lavrov@desy.de</a></em>.</p>
446 <p>
447 </p>
448 <hr />
449 <h1><a name="how_to_measure">HOW TO MEASURE</a></h1>
450 <p>Here are a few hints on how to measure:</p>
451 <dl>
452 <dt><strong><a name="item_temperature">Temperature</a></strong>
454 <dd>
455 <p>Usually you have some type of meter you can read to get the temperature.
456 The temperature is not really connected with a time. The only connection is
457 that the temperature reading happened at a certain time. You can use the
458 <strong>GAUGE</strong> data source type for this. RRDtool will then record your reading
459 together with the time.</p>
460 </dd>
461 </li>
462 <dt><strong><a name="item_mail_messages">Mail Messages</a></strong>
464 <dd>
465 <p>Assume you have a method to count the number of messages transported by
466 your mailserver in a certain amount of time, giving you data like '5
467 messages in the last 65 seconds'. If you look at the count of 5 like an
468 <strong>ABSOLUTE</strong> data type you can simply update the RRD with the number 5 and the
469 end time of your monitoring period. RRDtool will then record the number of
470 messages per second. If at some later stage you want to know the number of
471 messages transported in a day, you can get the average messages per second
472 from RRDtool for the day in question and multiply this number with the
473 number of seconds in a day. Because all math is run with Doubles, the
474 precision should be acceptable.</p>
475 </dd>
476 </li>
477 <dt><strong><a name="item_it_27s_always_a_rate">It's always a Rate</a></strong>
479 <dd>
480 <p>RRDtool stores rates in amount/second for COUNTER, DERIVE and ABSOLUTE
481 data.  When you plot the data, you will get on the y axis
482 amount/second which you might be tempted to convert to an absolute
483 amount by multiplying by the delta-time between the points. RRDtool
484 plots continuous data, and as such is not appropriate for plotting
485 absolute amounts as for example ``total bytes'' sent and received in a
486 router. What you probably want is plot rates that you can scale to
487 bytes/hour, for example, or plot absolute amounts with another tool
488 that draws bar-plots, where the delta-time is clear on the plot for
489 each point (such that when you read the graph you see for example GB
490 on the y axis, days on the x axis and one bar for each day).</p>
491 </dd>
492 </li>
493 </dl>
494 <p>
495 </p>
496 <hr />
497 <h1><a name="example">EXAMPLE</a></h1>
498 <pre>
499  rrdtool create temperature.rrd --step 300 \
500   DS:temp:GAUGE:600:-273:5000 \
501   RRA:AVERAGE:0.5:1:1200 \
502   RRA:MIN:0.5:12:2400 \
503   RRA:MAX:0.5:12:2400 \
504   RRA:AVERAGE:0.5:12:2400</pre>
505 <p>This sets up an <strong>RRD</strong> called <em>temperature.rrd</em> which accepts one
506 temperature value every 300 seconds. If no new data is supplied for
507 more than 600 seconds, the temperature becomes <em>*UNKNOWN*</em>.  The
508 minimum acceptable value is -273 and the maximum is 5'000.</p>
509 <p>A few archive areas are also defined. The first stores the
510 temperatures supplied for 100 hours (1'200 * 300 seconds = 100
511 hours). The second RRA stores the minimum temperature recorded over
512 every hour (12 * 300 seconds = 1 hour), for 100 days (2'400 hours). The
513 third and the fourth RRA's do the same for the maximum and
514 average temperature, respectively.</p>
515 <p>
516 </p>
517 <hr />
518 <h1><a name="example_2">EXAMPLE 2</a></h1>
519 <pre>
520  rrdtool create monitor.rrd --step 300        \
521    DS:ifOutOctets:COUNTER:1800:0:4294967295   \
522    RRA:AVERAGE:0.5:1:2016                     \
523    RRA:HWPREDICT:1440:0.1:0.0035:288</pre>
524 <p>This example is a monitor of a router interface. The first <strong>RRA</strong> tracks the
525 traffic flow in octets; the second <strong>RRA</strong> generates the specialized
526 functions <strong>RRAs</strong> for aberrant behavior detection. Note that the <em>rra-num</em>
527 argument of HWPREDICT is missing, so the other <strong>RRAs</strong> will implicitly be
528 created with default parameter values. In this example, the forecasting
529 algorithm baseline adapts quickly; in fact the most recent one hour of
530 observations (each at 5 minute intervals) accounts for 75% of the baseline
531 prediction. The linear trend forecast adapts much more slowly. Observations
532 made during the last day (at 288 observations per day) account for only
533 65% of the predicted linear trend. Note: these computations rely on an
534 exponential smoothing formula described in the LISA 2000 paper.</p>
535 <p>The seasonal cycle is one day (288 data points at 300 second intervals), and
536 the seasonal adaption parameter will be set to 0.1. The RRD file will store 5
537 days (1'440 data points) of forecasts and deviation predictions before wrap
538 around. The file will store 1 day (a seasonal cycle) of 0-1 indicators in
539 the FAILURES <strong>RRA</strong>.</p>
540 <p>The same RRD file and <strong>RRAs</strong> are created with the following command,
541 which explicitly creates all specialized function <strong>RRAs</strong>.</p>
542 <pre>
543  rrdtool create monitor.rrd --step 300 \
544    DS:ifOutOctets:COUNTER:1800:0:4294967295 \
545    RRA:AVERAGE:0.5:1:2016 \
546    RRA:HWPREDICT:1440:0.1:0.0035:288:3 \
547    RRA:SEASONAL:288:0.1:2 \
548    RRA:DEVPREDICT:1440:5 \
549    RRA:DEVSEASONAL:288:0.1:2 \
550    RRA:FAILURES:288:7:9:5</pre>
551 <p>Of course, explicit creation need not replicate implicit create, a
552 number of arguments could be changed.</p>
553 <p>
554 </p>
555 <hr />
556 <h1><a name="example_3">EXAMPLE 3</a></h1>
557 <pre>
558  rrdtool create proxy.rrd --step 300 \
559    DS:Total:DERIVE:1800:0:U  \
560    DS:Duration:DERIVE:1800:0:U  \
561    DS:AvgReqDur:COMPUTE:Duration,Requests,0,EQ,1,Requests,IF,/ \
562    RRA:AVERAGE:0.5:1:2016</pre>
563 <p>This example is monitoring the average request duration during each 300 sec
564 interval for requests processed by a web proxy during the interval.
565 In this case, the proxy exposes two counters, the number of requests
566 processed since boot and the total cumulative duration of all processed
567 requests. Clearly these counters both have some rollover point, but using the
568 DERIVE data source also handles the reset that occurs when the web proxy is
569 stopped and restarted.</p>
570 <p>In the <strong>RRD</strong>, the first data source stores the requests per second rate
571 during the interval. The second data source stores the total duration of all
572 requests processed during the interval divided by 300. The COMPUTE data source
573 divides each PDP of the AccumDuration by the corresponding PDP of
574 TotalRequests and stores the average request duration. The remainder of the
575 RPN expression handles the divide by zero case.</p>
576 <p>
577 </p>
578 <hr />
579 <h1><a name="author">AUTHOR</a></h1>
580 <p>Tobias Oetiker &lt;<a href="mailto:tobi@oetiker.ch">tobi@oetiker.ch</a>&gt;</p>
582 </body>
584 </html>