1 =for changes please consult me first. Thanks, Alex
4 rrdtutorial - Alex van den Bogaerdt's RRDtool tutorial
8 RRDtool is written by Tobias Oetiker E<lt>tobi@oetiker.chE<gt> with
9 contributions from many people all around the world. This document is
10 written by Alex van den Bogaerdt E<lt>alex@vandenbogaerdt.nlE<gt> to help you
11 understand what RRDtool is and what it can do for you.
13 The documentation provided with RRDtool can be too technical for some
14 people. This tutorial is here to help you understand the basics of
15 RRDtool. It should prepare you to read the documentation yourself.
16 It also explains the general things about statistics with a focus on
23 Please don't skip ahead in this document! The first part of this
24 document explains the basics and may be boring. But if you don't
25 understand the basics, the examples will not be as meaningful to you.
27 Sometimes things change. This example used to provide numbers like
28 "0.04" in stead of "4.00000e-02". Those are really the same numbers,
29 just written down differently. Don't be alarmed if a future version
30 of rrdtool displays a slightly different form of output. The examples
31 in this document are correct for version 1.2.0 of RRDtool.
33 Also, sometimes bugs do occur. They may also influence the outcome of
34 the examples. Example speed4.png was suffering from this (the handling
35 of unknown data in an if-statement was wrong). Normal data will be
36 just fine (a bug in rrdtool wouldn't last long) but special cases like
37 NaN, INF and so on may last a bit longer. Try another version if you
38 can, or just live with it.
40 I fixed the speed4.png example (and added a note). There may be other
41 examples which suffer from the same or a similar bug. Try to fix it
42 yourself, which is a great excercise. But please do not submit your
43 result as a fix to the source of this document. Discuss it on the
44 user's list, or write to me.
46 =head2 What is RRDtool?
48 RRDtool refers to Round Robin Database tool.
49 Round robin is a technique that works with a fixed amount of data, and a
50 pointer to the current element. Think of a circle with some dots plotted
51 on the edge. These dots are the places where data can be stored. Draw an
52 arrow from the center of the circle to one of the dots; this is the pointer.
53 When the current data is read or written, the pointer moves to the next
54 element. As we are on a circle there is neither a beginning nor an end, you can
55 go on and on and on. After a while, all the available places will be used and
56 the process automatically reuses old locations. This way, the dataset
57 will not grow in size and therefore requires no maintenance.
58 RRDtool works with with Round Robin Databases (RRDs). It stores and retrieves
61 =head2 What data can be put into an RRD?
63 You name it, it will probably fit as long as it is some sort of
64 time-series data. This means you have to be able to measure some value
65 at several points in time and provide this information to RRDtool. If
66 you can do this, RRDtool will be able to store it. The values must be
67 numerical but don't have to be integers, as is the case with MRTG (the
68 next section will give more details on this more specialized application).
70 Many examples below talk about SNMP which is an acronym for Simple Network
71 Management Protocol. "Simple" refers to the protocol. It does not
72 mean it is simple to manage or monitor a network. After working your
73 way through this document, you should know enough to be able to
74 understand what people are talking about. For now, just realize that
75 SNMP can be used to query devices for the values of counters they keep. It
76 is the value from those counters that we want to store in the RRD.
78 =head2 What can I do with this tool?
80 RRDtool originated from MRTG (Multi Router Traffic Grapher). MRTG
81 started as a tiny little script for graphing the use of a university's
82 connection to the Internet. MRTG was later (ab-)used as a tool for
83 graphing other data sources including temperature, speed, voltage,
84 number of printouts and the like.
86 Most likely you will start to use RRDtool to store and process data
87 collected via SNMP. The data will most likely be bytes (or bits)
88 transfered from and to a network or a computer. But it can also be
89 used to display tidal waves, solar radiation, power consumption,
90 number of visitors at an exhibition, noise levels near an airport,
91 temperature on your favorite holiday location, temperature in the
92 fridge and whatever you imagination can come up with.
94 You only need a sensor to measure the data and be able to feed the
95 numbers into RRDtool. RRDtool then lets you create a database, store
96 data in it, retrieve that data and create graphs in PNG format for
97 display on a web browser. Those PNG images are dependent on the data
98 you collected and could be, for instance, an overview of the average
99 network usage, or the peaks that occurred.
101 =head2 What if I still have problems after reading this document?
103 First of all: read it again! You may have missed something.
104 If you are unable to compile the sources and you have a fairly common
105 OS, it will probably not be the fault of RRDtool. There may be pre-compiled
106 versions around on the Internet. If they come from trusted sources, get
109 If on the other hand the program works but does not give you the
110 expected results, it will be a problem with configuring it. Review
111 your configuration and compare it with the examples that follow.
113 There is a mailing list and an archive of it. Read the list for a few
114 weeks and search the archive. It is considered rude to just ask
115 a question without searching the archives: your problem may already have been
116 solved for somebody else! This is true for most, if not all, mailing lists
117 and not only for this particular one. Look in the documentation that
118 came with RRDtool for the location and usage of the list.
120 I suggest you take a moment to subscribe to the mailing list right now
121 by sending an email to E<lt>rrd-users-request@lists.oetiker.chE<gt> with a
122 subject of "subscribe". If you ever want to leave this list, just write
123 an email to the same address but now with a subject of "unsubscribe".
125 =head2 How will you help me?
127 By giving you some detailed descriptions with detailed examples.
128 I assume that following the instructions in the order presented
129 will give you enough knowledge of RRDtool to experiment for yourself.
130 If it doesn't work the first time, don't give up. Reread the stuff that
131 you did understand, you may have missed something.
133 By following the examples you get some hands-on experience and, even
134 more important, some background information of how it works.
136 You will need to know something about hexadecimal numbers. If you don't
137 then start with reading L<bin_dec_hex> before you continue here.
139 =head2 Your first Round Robin Database
141 In my opinion the best way to learn something is to actually do it.
142 Why not start right now? We will create a database, put some values
143 in it and extract this data again. Your output should be the same
144 as the output that is included in this document.
146 We will start with some easy stuff and compare a car with a router,
147 or compare kilometers (miles if you wish) with bits and bytes. It's
148 all the same: some number over some time.
150 Assume we have a device that transfers bytes to and from the Internet.
151 This device keeps a counter that starts at zero when it is turned on,
152 increasing with every byte that is transfered. This counter will probably have
153 a maximum value. If this value is reached and an extra byte is counted,
154 the counter starts over at zero. This is the same as many counters
155 in the world such as the mileage counter in a car.
157 Most discussions about networking talk about bits per second so lets
158 get used to that right away. Assume a byte is eight bits and start to
159 think in bits not bytes. The counter, however, still counts bytes!
160 In the SNMP world most of the counters are 32 bits. That means they are
161 counting from 0 to 4294967295. We will use these values in the examples.
162 The device, when asked, returns the current value of the counter. We
163 know the time that has passes since we last asked so we now know how
164 many bytes have been transfered ***on average*** per second. This is
165 not very hard to calculate. First in words, then in calculations:
171 Take the current counter, subtract the previous value from it.
175 Do the same with the current time and the previous time (in seconds).
179 Divide the outcome of (1) by the outcome of (2), the result is
180 the amount of bytes per second. Multiply by eight to get the
181 number of bits per second (bps).
185 bps = (counter_now - counter_before) / (time_now - time_before) * 8
187 For some people it may help to translate this to an automobile example.
188 Do not try this example, and if you do, don't blame me for the results!
190 People who are not used to think in kilometers per hour can translate
191 most into miles per hour by dividing km by 1.6 (close enough).
192 I will use the following abbreviations:
195 km: kilometer (= 1000 meters).
198 km/h: kilometers per hour
199 m/s: meters per second
201 You are driving a car. At 12:05 you read the counter in the dashboard
202 and it tells you that the car has moved 12345 km until that moment.
203 At 12:10 you look again, it reads 12357 km. This means you have
204 traveled 12 km in five minutes. A scientist would translate that
205 into meters per second and this makes a nice comparison toward the
206 problem of (bytes per five minutes) versus (bits per second).
208 We traveled 12 kilometers which is 12000 meters. We did that in five
209 minutes or 300 seconds. Our speed is 12000m / 300s or 40 m/s.
211 We could also calculate the speed in km/h: 12 times 5 minutes
212 is an hour, so we have to multiply 12 km by 12 to get 144 km/h.
213 For our native English speaking friends: that's 90 mph so don't
214 try this example at home or where I live :)
216 Remember: these numbers are averages only. There is no way to figure out
217 from the numbers, if you drove at a constant speed. There is an example
218 later on in this tutorial that explains this.
220 I hope you understand that there is no difference in calculating m/s or
221 bps; only the way we collect the data is different. Even the k from kilo
222 is the same as in networking terms k also means 1000.
224 We will now create a database where we can keep all these interesting
225 numbers. The method used to start the program may differ slightly from
226 OS to OS, but I assume you can figure it out if it works different on
227 your's. Make sure you do not overwrite any file on your system when
228 executing the following command and type the whole line as one long
229 line (I had to split it for readability)
230 and skip all of the '\' characters.
232 rrdtool create test.rrd \
234 DS:speed:COUNTER:600:U:U \
235 RRA:AVERAGE:0.5:1:24 \
238 (So enter: C<rrdtool create test.rrd --start 920804400 DS ...>)
240 =head2 What has been created?
242 We created the round robin database called test (test.rrd) which starts at
243 noon the day I started writing this document, 7th of March, 1999 (this date
244 translates to 920804400 seconds as explained below). Our database holds
245 one data source (DS) named "speed" that represents a counter. This counter
246 is read every five minutes (this is the default therefore you don't have to
247 put C<--step=300>). In the same database two round robin archives (RRAs)
248 are kept, one averages the data every time it is read (e.g., there's nothing
249 to average) and keeps 24 samples (24 times 5 minutes is 2 hours). The other
250 averages 6 values (half hour) and contains 10 such averages (e.g. 5 hours).
252 RRDtool works with special time stamps coming from the UNIX world.
253 This time stamp is the number of seconds that passed since January
254 1st 1970 UTC. The time stamp value is translated into local time and
255 it will therefore look different for different time zones.
257 Chances are that you are not in the same part of the world as I am.
258 This means your time zone is different. In all examples where I talk
259 about time, the hours may be wrong for you. This has little effect on
260 the results of the examples, just correct the hours while reading.
261 As an example: where I will see "12:05" the UK folks will see "11:05".
263 We now have to fill our database with some numbers. We'll pretend to
264 have read the following numbers:
282 We fill the database as follows:
284 rrdtool update test.rrd 920804700:12345 920805000:12357 920805300:12363
285 rrdtool update test.rrd 920805600:12363 920805900:12363 920806200:12373
286 rrdtool update test.rrd 920806500:12383 920806800:12393 920807100:12399
287 rrdtool update test.rrd 920807400:12405 920807700:12411 920808000:12415
288 rrdtool update test.rrd 920808300:12420 920808600:12422 920808900:12423
290 This reads: update our test database with the following numbers
292 time 920804700, value 12345
293 time 920805000, value 12357
297 As you can see, it is possible to feed more than one value into the
298 database in one command. I had to stop at three for readability but
299 the real maximum per line is OS dependent.
301 We can now retrieve the data from our database using "rrdtool fetch":
303 rrdtool fetch test.rrd AVERAGE --start 920804400 --end 920809200
305 It should return the following output:
310 920805000: 4.0000000000e-02
311 920805300: 2.0000000000e-02
312 920805600: 0.0000000000e+00
313 920805900: 0.0000000000e+00
314 920806200: 3.3333333333e-02
315 920806500: 3.3333333333e-02
316 920806800: 3.3333333333e-02
317 920807100: 2.0000000000e-02
318 920807400: 2.0000000000e-02
319 920807700: 2.0000000000e-02
320 920808000: 1.3333333333e-02
321 920808300: 1.6666666667e-02
322 920808600: 6.6666666667e-03
323 920808900: 3.3333333333e-03
326 If it doesn't, something may be wrong. Perhaps your OS will print
327 "NaN" in a different form. "NaN" stands for "Not A Number". If your OS
328 writes "U" or "UNKN" or something similar that's okay. If something
329 else is wrong, it will probably be due to an error you made (assuming
330 that my tutorial is correct of course :-). In that case: delete the
331 database and try again.
334 The meaning of the above output will become clear below.
336 =head2 Time to create some graphics
338 Try the following command:
340 rrdtool graph speed.png \
341 --start 920804400 --end 920808000 \
342 DEF:myspeed=test.rrd:speed:AVERAGE \
345 This will create speed.png which starts at 12:00 and ends at 13:00.
346 There is a definition of a variable called myspeed, using the data from RRA
347 "speed" out of database "test.rrd". The line drawn is 2 pixels high
348 and represents the variable myspeed. The color is red (specified by
349 its rgb-representation, see below).
351 You'll notice that the start of the graph is not at 12:00 but at 12:05.
352 This is because we have insufficient data to tell the average before
353 that time. This will only happen when you miss some samples, this will
354 not happen a lot, hopefully.
356 If this has worked: congratulations! If not, check what went wrong.
359 The colors are built up from red, green and blue. For each of the
360 components, you specify how much to use in hexadecimal where 00 means
361 not included and FF means fully included.
362 The "color" white is a mixture of red, green and blue: FFFFFF
363 The "color" black is all colors off: 000000
368 magenta #FF00FF (mixed red with blue)
369 gray #555555 (one third of all components)
371 Additionally you can (with a recent RRDtool) add an alpha channel
372 (transparency). The default will be "FF" which means non-transparent.
374 The PNG you just created can be displayed using your favorite image
375 viewer. Web browsers will display the PNG via the URL
376 "file:///the/path/to/speed.png"
378 =head2 Graphics with some math
380 When looking at the image, you notice that the horizontal axis is labeled
381 12:10, 12:20, 12:30, 12:40 and 12:50. Sometimes a label doesn't fit (12:00
382 and 13:00 would be likely candidates) so they are skipped.
384 The vertical axis displays the range we entered. We provided
385 kilometers and when divided by 300 seconds, we get very small
386 numbers. To be exact, the first value was 12 (12357-12345) and divided
387 by 300 this makes 0.04, which is displayed by RRDtool as "40 m"
388 meaning "40/1000". The "m" (milli) has nothing to do with meters (also m),
389 kilometers or millimeters! RRDtool doesn't know about the physical
390 units of our data, it just works with dimensionless numbers.
392 If we had measured our distances in meters, this would have been
393 (12357000-12345000)/300 = 12000/300 = 40.
395 As most people have a better feel for numbers in this range, we'll
396 correct that. We could recreate our database and store the correct
397 data, but there is a better way: we do some calculations while creating
400 rrdtool graph speed2.png \
401 --start 920804400 --end 920808000 \
402 --vertical-label m/s \
403 DEF:myspeed=test.rrd:speed:AVERAGE \
404 CDEF:realspeed=myspeed,1000,\* \
405 LINE2:realspeed#FF0000
407 Note: I need to escape the multiplication operator * with a backslash.
408 If I don't, the operating system may interpret it and use it for file
409 name expansion. You could also place the line within quotation marks
412 "CDEF:realspeed=myspeed,1000,*" \
414 It boils down to: it is RRDtool which should see *, not your shell.
415 And it is your shell interpreting \, not RRDtool. You may need to
416 adjust examples accordingly if you happen to use an operating
417 system or shell which behaves differently.
419 After viewing this PNG, you notice the "m" (milli) has
420 disappeared. This it what the correct result would be. Also, a label
421 has been added to the image. Apart from the things mentioned above,
422 the PNG should look the same.
424 The calculations are specified in the CDEF part above and are in
425 Reverse Polish Notation ("RPN"). What we requested RRDtool to do is:
426 "take the data source myspeed and the number 1000; multiply
427 those". Don't bother with RPN yet, it will be explained later on in
428 more detail. Also, you may want to read my tutorial on CDEFs and Steve
429 Rader's tutorial on RPN. But first finish this tutorial.
431 Hang on! If we can multiply values with 1000, it should also be possible
432 to display kilometers per hour from the same data!
434 To change a value that is measured in meters per second:
436 Calculate meters per hour: value * 3600
437 Calculate kilometers per hour: value / 1000
438 Together this makes: value * (3600/1000) or value * 3.6
440 In our example database we made a mistake and we need to compensate for
441 this by multiplying with 1000. Applying that correction:
443 value * 3.6 * 1000 == value * 3600
445 Now let's create this PNG, and add some more magic ...
447 rrdtool graph speed3.png \
448 --start 920804400 --end 920808000 \
449 --vertical-label km/h \
450 DEF:myspeed=test.rrd:speed:AVERAGE \
451 "CDEF:kmh=myspeed,3600,*" \
452 CDEF:fast=kmh,100,GT,kmh,0,IF \
453 CDEF:good=kmh,100,GT,0,kmh,IF \
454 HRULE:100#0000FF:"Maximum allowed" \
455 AREA:good#00FF00:"Good speed" \
456 AREA:fast#FF0000:"Too fast"
458 Note: here we use another means to escape the * operator by enclosing
459 the whole string in double quotes.
461 This graph looks much better. Speed is shown in km/h and there is even
462 an extra line with the maximum allowed speed (on the road I travel
463 on). I also changed the colors used to display speed and changed it
464 from a line into an area.
466 The calculations are more complex now. For speed measurements within
467 the speed limit they are:
469 Check if kmh is greater than 100 ( kmh,100 ) GT
470 If so, return 0, else kmh ((( kmh,100 ) GT ), 0, kmh) IF
472 For values above the speed limit:
474 Check if kmh is greater than 100 ( kmh,100 ) GT
475 If so, return kmh, else return 0 ((( kmh,100) GT ), kmh, 0) IF
477 =head2 Graphics Magic
479 I like to believe there are virtually no limits to how RRDtool graph
480 can manipulate data. I will not explain how it works, but look at the
483 rrdtool graph speed4.png \
484 --start 920804400 --end 920808000 \
485 --vertical-label km/h \
486 DEF:myspeed=test.rrd:speed:AVERAGE \
487 CDEF:nonans=myspeed,UN,0,myspeed,IF \
488 CDEF:kmh=nonans,3600,* \
489 CDEF:fast=kmh,100,GT,100,0,IF \
490 CDEF:over=kmh,100,GT,kmh,100,-,0,IF \
491 CDEF:good=kmh,100,GT,0,kmh,IF \
492 HRULE:100#0000FF:"Maximum allowed" \
493 AREA:good#00FF00:"Good speed" \
494 AREA:fast#550000:"Too fast" \
495 STACK:over#FF0000:"Over speed"
497 Remember the note in the beginning? I had to remove unknown data from
498 this example. The 'nonans' CDEF is new, and the 6th line (which used to
499 be the 5th line) used to read 'CDEF:kmh=myspeed,3600,*'
501 Let's create a quick and dirty HTML page to view the three PNGs:
503 <HTML><HEAD><TITLE>Speed</TITLE></HEAD><BODY>
504 <IMG src="speed2.png" alt="Speed in meters per second">
506 <IMG src="speed3.png" alt="Speed in kilometers per hour">
508 <IMG src="speed4.png" alt="Traveled too fast?">
511 Name the file "speed.html" or similar, and look at it in your web browser.
513 Now, all you have to do is measure the values regularly and update the
514 database. When you want to view the data, recreate the PNGs and make
515 sure to refresh them in your browser. (Note: just clicking reload may
516 not be enough, especially when proxies are involved. Try shift-reload
519 =head2 Updates in Reality
521 We've already used the C<update> command: it took one or more
522 parameters in the form of "<time>:<value>". You'll be glad to know
523 that you can specify the current time by filling in a "N" as the time.
524 Or you could use the "time" function in Perl (the shortest example in
527 perl -e 'print time, "\n" '
529 How to run a program on regular intervals is OS specific. But here is
530 an example in pseudo code:
532 - Get the value and put it in variable "$speed"
533 - rrdtool update speed.rrd N:$speed
535 (do not try this with our test database, we'll use it in further examples)
537 This is all. Run the above script every five minutes. When you need to know
538 what the graphs look like, run the examples above. You could put them
539 in a script as well. After running that script, view the page
540 index.html we created above.
542 =head2 Some words on SNMP
544 I can imagine very few people that will be able to get real data from
545 their car every five minutes. All other people will have to settle for
546 some other kind of counter. You could measure the number of pages
547 printed by a printer, for example, the cups of coffee made by the
548 coffee machine, a device that counts the electricity used,
549 whatever. Any incrementing counter can be monitored and graphed using
550 the stuff you learned so far. Later on we will also be able to monitor
551 other types of values like temperature.
553 Many people interested in RRDtool will use the counter that keeps track
554 of octets (bytes) transfered by a network device. So let's do just
555 that next. We will start with a description of how to collect data.
557 Some people will make a remark that there are tools which can do this data
558 collection for you. They are right! However, I feel it is important that
559 you understand they are not necessary. When you have to determine why
560 things went wrong you need to know how they work.
562 One tool used in the example has been talked about very briefly in the
563 beginning of this document, it is called SNMP. It is a way of talking
564 to networked equipment. The tool I use below is called "snmpget" and
565 this is how it works:
567 snmpget device password OID
571 snmpget -v[version] -c[password] device OID
573 For device you substitute the name, or the IP address, of your device.
574 For password you use the "community read string" as it is called in the
575 SNMP world. For some devices the default of "public" might work, however
576 this can be disabled, altered or protected for privacy and security
577 reasons. Read the documentation that comes with your device or program.
579 Then there is this parameter, called OID, which means "object identifier".
581 When you start to learn about SNMP it looks very confusing. It isn't
582 all that difficult when you look at the Management Information Base
583 ("MIB"). It is an upside-down tree that describes data, with a single node
584 as the root and from there a number of branches. These branches end
585 up in another node, they branch out, etc. All the branches have a name
586 and they form the path that we follow all the way down. The branches
587 that we follow are named: iso, org, dod, internet, mgmt and mib-2.
588 These names can also be written down as numbers and are 1 3 6 1 2 1.
590 iso.org.dod.internet.mgmt.mib-2 (1.3.6.1.2.1)
592 There is a lot of confusion about the leading dot that some programs
593 use. There is *no* leading dot in an OID. However, some programs
594 can use the above part of OIDs as a default. To indicate the difference
595 between abbreviated OIDs and full OIDs they need a leading dot when
596 you specify the complete OID. Often those programs will leave out
597 the default portion when returning the data to you. To make things
598 worse, they have several default prefixes ...
600 Ok, lets continue to the start of our OID: we had 1.3.6.1.2.1
601 From there, we are especially interested in the branch "interfaces"
602 which has number 2 (e.g., 1.3.6.1.2.1.2 or 1.3.6.1.2.1.interfaces).
604 First, we have to get some SNMP program. First look if there is a
605 pre-compiled package available for your OS. This is the preferred way.
606 If not, you will have to get the sources yourself and compile those.
607 The Internet is full of sources, programs etc. Find information using
608 a search engine or whatever you prefer.
610 Assume you got the program. First try to collect some data that is
611 available on most systems. Remember: there is a short name for the
612 part of the tree that interests us most in the world we live in!
614 I will give an example which can be used on Fedora Core 3. If it
615 doesn't work for you, work your way through the manual of snmp and
616 adapt the example to make it work.
618 snmpget -v2c -c public myrouter system.sysDescr.0
620 The device should answer with a description of itself, perhaps an
621 empty one. Until you got a valid answer from a device, perhaps using a
622 different "password", or a different device, there is no point in
625 snmpget -v2c -c public myrouter interfaces.ifNumber.0
627 Hopefully you get a number as a result, the number of interfaces.
628 If so, you can carry on and try a different program called "snmpwalk".
630 snmpwalk -v2c -c public myrouter interfaces.ifTable.ifEntry.ifDescr
632 If it returns with a list of interfaces, you're almost there.
634 [user@host /home/alex]$ snmpwalk -v2c -c public cisco 2.2.1.2
636 interfaces.ifTable.ifEntry.ifDescr.1 = "BRI0: B-Channel 1"
637 interfaces.ifTable.ifEntry.ifDescr.2 = "BRI0: B-Channel 2"
638 interfaces.ifTable.ifEntry.ifDescr.3 = "BRI0" Hex: 42 52 49 30
639 interfaces.ifTable.ifEntry.ifDescr.4 = "Ethernet0"
640 interfaces.ifTable.ifEntry.ifDescr.5 = "Loopback0"
642 On this cisco equipment, I would like to monitor the "Ethernet0"
643 interface and from the above output I see that it is number four. I try:
645 [user@host /home/alex]$ snmpget -v2c -c public cisco 2.2.1.10.4 2.2.1.16.4
647 interfaces.ifTable.ifEntry.ifInOctets.4 = 2290729126
648 interfaces.ifTable.ifEntry.ifOutOctets.4 = 1256486519
650 So now I have two OIDs to monitor and they are (in full, this time):
658 both with an interface number of 4.
660 Don't get fooled, this wasn't my first try. It took some time for me too
661 to understand what all these numbers mean. It does help a lot when they
662 get translated into descriptive text... At least, when people are talking
663 about MIBs and OIDs you know what it's all about.
664 Do not forget the interface number (0 if it is not interface dependent)
665 and try snmpwalk if you don't get an answer from snmpget.
667 If you understand the above section and get numbers from your device, continue
668 on with this tutorial. If not, then go back and re-read this part.
670 =head2 A Real World Example
672 Let the fun begin. First, create a new database. It contains data from
673 two counters, called input and output. The data is put into archives
674 that average it. They take 1, 6, 24 or 288 samples at a time.
675 They also go into archives that keep the maximum numbers. This will be
676 explained later on. The time in-between samples is 300 seconds, a good
677 starting point, which is the same as five minutes.
679 1 sample "averaged" stays 1 period of 5 minutes
680 6 samples averaged become one average on 30 minutes
681 24 samples averaged become one average on 2 hours
682 288 samples averaged become one average on 1 day
684 Lets try to be compatible with MRTG which stores about the following
687 600 5-minute samples: 2 days and 2 hours
688 600 30-minute samples: 12.5 days
689 600 2-hour samples: 50 days
690 732 1-day samples: 732 days
692 These ranges are appended, so the total amount of data stored in the
693 database is approximately 797 days. RRDtool stores the data
694 differently, it doesn't start the "weekly" archive where the "daily"
695 archive stopped. For both archives the most recent data will be near
696 "now" and therefore we will need to keep more data than MRTG does!
700 600 samples of 5 minutes (2 days and 2 hours)
701 700 samples of 30 minutes (2 days and 2 hours, plus 12.5 days)
702 775 samples of 2 hours (above + 50 days)
703 797 samples of 1 day (above + 732 days, rounded up to 797)
705 rrdtool create myrouter.rrd \
706 DS:input:COUNTER:600:U:U \
707 DS:output:COUNTER:600:U:U \
708 RRA:AVERAGE:0.5:1:600 \
709 RRA:AVERAGE:0.5:6:700 \
710 RRA:AVERAGE:0.5:24:775 \
711 RRA:AVERAGE:0.5:288:797 \
717 Next thing to do is to collect data and store it. Here is an example.
718 It is written partially in pseudo code, you will have to find out what
719 to do exactly on your OS to make it work.
721 while not the end of the universe
724 snmpget router community 2.2.1.10.4
727 snmpget router community 2.2.1.16.4
730 rrdtool update myrouter.rrd N:$in:$out
735 Then, after collecting data for a day, try to create an image using:
737 rrdtool graph myrouter-day.png --start -86400 \
738 DEF:inoctets=myrouter.rrd:input:AVERAGE \
739 DEF:outoctets=myrouter.rrd:output:AVERAGE \
740 AREA:inoctets#00FF00:"In traffic" \
741 LINE1:outoctets#0000FF:"Out traffic"
743 This should produce a picture with one day worth of traffic.
744 One day is 24 hours of 60 minutes of 60 seconds: 24*60*60=86400, we
745 start at now minus 86400 seconds. We define (with DEFs) inoctets and
746 outoctets as the average values from the database myrouter.rrd and draw
747 an area for the "in" traffic and a line for the "out" traffic.
749 View the image and keep logging data for a few more days.
750 If you like, you could try the examples from the test database and
751 see if you can get various options and calculations to work.
753 Suggestion: Display in bytes per second and in bits per second. Make
754 the Ethernet graphics go red if they are over four megabits per
757 =head2 Consolidation Functions
759 A few paragraphs back I mentioned the possibility of keeping
760 the maximum values instead of the average values. Let's go
761 into this a bit more.
763 Recall all the stuff about the speed of the car. Suppose we drove at 144
764 km/h during 5 minutes and then were stopped by the police for 25 minutes.
765 At the end of the lecture we would take our laptop and create and view the
766 image taken from the database. If we look at the second RRA we did
767 create, we would have the average from 6 samples. The samples measured
768 would be 144+0+0+0+0+0=144, divided by 30 minutes, corrected for the
769 error by 1000, translated into km/h, with a result of 24 km/h.
770 I would still get a ticket but not for speeding anymore :)
772 Obviously, in this case we shouldn't look at the averages. In some
773 cases they are handy. If you want to know how many km you had traveled,
774 the averaged picture would be the right one to look at. On the other hand, for
775 the speed that we traveled at, the maximum numbers seen is much more
776 interesting. Later we will see more types.
778 It is the same for data. If you want to know the amount, look at the
779 averages. If you want to know the rate, look at the maximum.
780 Over time, they will grow apart more and more. In the last database
781 we have created, there are two archives that keep data per day. The
782 archive that keeps averages will show low numbers, the archive that
783 shows maxima will have higher numbers.
785 For my car this would translate in averages per day of 96/24=4 km/h
786 (as I travel about 94 kilometers on a day) during working days, and
787 maxima of 120 km/h (my top speed that I reach every day).
789 Big difference. Do not look at the second graph to estimate the
790 distances that I travel and do not look at the first graph to
791 estimate my speed. This will work if the samples are close together,
792 as they are in five minutes, but not if you average.
794 On some days, I go for a long ride. If I go across Europe and travel
795 for 12 hours, the first graph will rise to about 60 km/h. The second
796 one will show 180 km/h. This means that I traveled a distance of 60
797 km/h times 24 h = 1440 km. I did this with a higher speed and a
798 maximum around 180 km/h. However, it probably doesn't mean that I
799 traveled for 8 hours at a constant speed of 180 km/h!
801 This is a real example: go with the flow through Germany (fast!) and stop
802 a few times for gas and coffee. Drive slowly through Austria and the
803 Netherlands. Be careful in the mountains and villages. If you would
804 look at the graphs created from the five-minute averages you would
805 get a totally different picture. You would see the same values on the
806 average and maximum graphs (provided I measured every 300 seconds).
807 You would be able to see when I stopped, when I was in top gear, when
808 I drove over fast highways etc. The granularity of the data is much
809 higher, so you can see more. However, this takes 12 samples per hour,
810 or 288 values per day, so it would be a lot of data over a longer
811 period of time. Therefore we average it, eventually to one value per
812 day. From this one value, we cannot see much detail, of course.
814 Make sure you understand the last few paragraphs. There is no value
815 in only a line and a few axis, you need to know what they mean and
816 interpret the data in an appropriate way. This is true for all data.
818 The biggest mistake you can make is to use the collected data for
819 something that it is not suitable for. You would be better off if
820 you didn't have the graph at all.
823 =head2 Let's review what you now should know
825 You know how to create a database and can put data in it. You can get
826 the numbers out again by creating an image, do math on the data from
827 the database and view the result instead of the raw data. You know
828 about the difference between averages and maximum, and when to use
829 which (or at least you should have an idea).
831 RRDtool can do more than what we have learned up to now. Before you
832 continue with the rest of this doc, I recommend that you reread from
833 the start and try some modifications on the examples. Make sure you
834 fully understand everything. It will be worth the effort and helps
835 you not only with the rest of this tutorial, but also in your day to day
836 monitoring long after you read this introduction.
838 =head2 Data Source Types
840 All right, you feel like continuing. Welcome back and get ready
841 for an increased speed in the examples and explanations.
843 You know that in order to view a counter over time, you have to
844 take two numbers and divide the difference of them between the
845 time lapsed. This makes sense for the examples I gave you but there
846 are other possibilities. For instance, I'm able to retrieve the
847 temperature from my router in three places namely the inlet, the
848 so called hot-spot and the exhaust. These values are not counters.
849 If I take the difference of the two samples and divide that by
850 300 seconds I would be asking for the temperature change per second.
851 Hopefully this is zero! If not, the computer room is probably on fire :)
853 So, what can we do? We can tell RRDtool to store the values we measure
854 directly as they are (this is not entirely true but close enough). The
855 graphs we make will look much better, they will show a rather constant
856 value. I know when the router is busy (it
857 works -> it uses more electricity -> it generates more heat -> the
858 temperature rises). I know when the doors are left open (the room is
859 air conditioned) -> the warm air from the rest of the building flows into the
860 computer room -> the inlet temperature rises). Etc. The data type we
861 use when creating the database before was counter, we now have a
862 different data type and thus a different name for it. It is called
863 GAUGE. There are more such data types:
865 - COUNTER we already know this one
866 - GAUGE we just learned this one
870 The two additional types are DERIVE and ABSOLUTE. Absolute can be used like
871 counter with one difference: RRDtool assumes the counter is reset when
872 it's read. That is: its delta is known without calculation by RRDtool
873 whereas RRDtool needs to calculate it for the counter type.
874 Example: our first example (12345, 12357, 12363, 12363) would read:
875 unknown, 12, 6, 0. The rest of the calculations stay the same.
876 The other one, derive, is like counter. Unlike counter, it can also
877 decrease so it can have a negative delta. Again, the rest of the
878 calculations stay the same.
882 rrdtool create all.rrd --start 978300900 \
883 DS:a:COUNTER:600:U:U \
885 DS:c:DERIVE:600:U:U \
886 DS:d:ABSOLUTE:600:U:U \
888 rrdtool update all.rrd \
889 978301200:300:1:600:300 \
890 978301500:600:3:1200:600 \
891 978301800:900:5:1800:900 \
892 978302100:1200:3:2400:1200 \
893 978302400:1500:1:2400:1500 \
894 978302700:1800:2:1800:1800 \
895 978303000:2100:4:0:2100 \
896 978303300:2400:6:600:2400 \
897 978303600:2700:4:600:2700 \
898 978303900:3000:2:1200:3000
899 rrdtool graph all1.png -s 978300600 -e 978304200 -h 400 \
900 DEF:linea=all.rrd:a:AVERAGE LINE3:linea#FF0000:"Line A" \
901 DEF:lineb=all.rrd:b:AVERAGE LINE3:lineb#00FF00:"Line B" \
902 DEF:linec=all.rrd:c:AVERAGE LINE3:linec#0000FF:"Line C" \
903 DEF:lined=all.rrd:d:AVERAGE LINE3:lined#000000:"Line D"
905 =head2 RRDtool under the Microscope
911 Line A is a COUNTER type, so it should continuously increment and RRDtool
912 must calculate the differences. Also, RRDtool needs to divide the
913 difference by the amount of time lapsed. This should end up as a
914 straight line at 1 (the deltas are 300, the time is 300).
918 Line B is of type GAUGE. These are "real" values so they should match
919 what we put in: a sort of a wave.
923 Line C is of type DERIVE. It should be a counter that can decrease. It does
924 so between 2400 and 0, with 1800 in-between.
928 Line D is of type ABSOLUTE. This is like counter but it works on
929 values without calculating the difference. The numbers are the same
930 and as you can see (hopefully) this has a different result.
934 This translates in the following values, starting at 23:10 and ending
935 at 00:10 the next day (where "u" means unknown/unplotted):
937 - Line A: u u 1 1 1 1 1 1 1 1 1 u
938 - Line B: u 1 3 5 3 1 2 4 6 4 2 u
939 - Line C: u u 2 2 2 0 -2 -6 2 0 2 u
940 - Line D: u 1 2 3 4 5 6 7 8 9 10 u
942 If your PNG shows all this, you know you have entered the data correctly,
943 the RRDtool executable is working properly, your viewer doesn't fool you,
944 and you successfully entered the year 2000 :)
946 You could try the same example four times, each time with only one of
949 Let's go over the data again:
955 Line A: 300,600,900 and so on. The counter delta is a constant 300 and
956 so is the time delta. A number divided by itself is always 1 (except
957 when dividing by zero which is undefined/illegal).
959 Why is it that the first point is unknown? We do know what we put into
960 the database, right? True, But we didn't have a value to calculate the delta
961 from, so we don't know where we started. It would be wrong to assume we
962 started at zero so we don't!
966 Line B: There is nothing to calculate. The numbers are as they are.
970 Line C: Again, the start-out value is unknown. The same story is holds
971 as for line A. In this case the deltas are not constant, therefore the line
972 is not either. If we would put the same numbers in the database as we did for
973 line A, we would have gotten the same line. Unlike type counter,
974 this type can decrease and I hope to show you later on why
975 this makes a difference.
979 Line D: Here the device calculates the deltas. Therefore we DO know the
980 first delta and it is plotted. We had the same input as with line A, but
981 the meaning of this input is different and thus the line is different.
982 In this case the deltas increase each time with 300. The time delta
983 stays at a constant 300 and therefore the division of the two gives
990 There are a few more basics to show. Some important options are still to
991 be covered and we haven't look at counter wraps yet. First the counter wrap:
992 In our car we notice that the counter shows 999987. We travel 20 km and
993 the counter should go to 1000007. Unfortunately, there are only six digits
994 on our counter so it really shows 000007. If we would plot that on a type
995 DERIVE, it would mean that the counter was set back 999980 km. It wasn't,
996 and there has to be some protection for this. This protection is only
997 available for type COUNTER which should be used for this kind of counter
998 anyways. How does it work? Type counter should never decrease and
999 therefore RRDtool must assume it wrapped if it does decrease!
1000 If the delta is negative, this can be compensated for by adding the
1001 maximum value of the counter + 1. For our car this would be:
1003 Delta = 7 - 999987 = -999980 (instead of 1000007-999987=20)
1005 Real delta = -999980 + 999999 + 1 = 20
1007 At the time of writing this document, RRDtool knows of counters that
1008 are either 32 bits or 64 bits of size. These counters can handle the
1009 following different values:
1011 - 32 bits: 0 .. 4294967295
1012 - 64 bits: 0 .. 18446744073709551615
1014 If these numbers look strange to you, you can view them in
1015 their hexadecimal form:
1017 - 32 bits: 0 .. FFFFFFFF
1018 - 64 bits: 0 .. FFFFFFFFFFFFFFFF
1020 RRDtool handles both counters the same. If an overflow occurs and
1021 the delta would be negative, RRDtool first adds the maximum of a small
1022 counter + 1 to the delta. If the delta is still negative, it had to be
1023 the large counter that wrapped. Add the maximum possible value of the
1024 large counter + 1 and subtract the erroneously added small value.
1026 There is a risk in this: suppose the large counter wrapped while adding
1027 a huge delta, it could happen, theoretically, that adding the smaller value
1028 would make the delta positive. In this unlikely case the results would
1029 not be correct. The increase should be nearly as high as the maximum
1030 counter value for that to happen, so chances are you would have several
1031 other problems as well and this particular problem would not even be
1032 worth thinking about. Even though, I did include an example, so you
1033 can judge for yourself.
1035 The next section gives you some numerical examples for counter-wraps.
1036 Try to do the calculations yourself or just believe me if your calculator
1037 can't handle the numbers :)
1041 - 32 bits: (4294967295 + 1) = 4294967296
1042 - 64 bits: (18446744073709551615 + 1)
1043 - correction1 = 18446744069414584320
1047 Should become: 4294967300
1050 Correction1: -4294967196 + 4294967296 = 100
1052 Before: 18446744073709551000
1054 Should become: 18446744073709551800
1056 Delta: -18446744073709550816
1057 Correction1: -18446744073709550816
1058 + 4294967296 = -18446744069414583520
1059 Correction2: -18446744069414583520
1060 + 18446744069414584320 = 800
1062 Before: 18446744073709551615 ( maximum value )
1063 Increase: 18446744069414584320 ( absurd increase, minimum for
1064 Should become: 36893488143124135935 this example to work )
1065 But really is: 18446744069414584319
1067 Correction1: -4294967296 + 4294967296 = 0
1068 (not negative -> no correction2)
1070 Before: 18446744073709551615 ( maximum value )
1071 Increase: 18446744069414584319 ( one less increase )
1072 Should become: 36893488143124135934
1073 But really is: 18446744069414584318
1075 Correction1: -4294967297 + 4294967296 = -1
1076 Correction2: -1 + 18446744069414584320 = 18446744069414584319
1078 As you can see from the last two examples, you need strange numbers
1079 for RRDtool to fail (provided it's bug free of course), so this should
1080 not happen. However, SNMP or whatever method you choose to collect the
1081 data, might also report wrong numbers occasionally. We can't prevent all
1082 errors, but there are some things we can do. The RRDtool "create" command
1083 takes two special parameters for this. They define
1084 the minimum and maximum allowed values. Until now, we used "U", meaning
1085 "unknown". If you provide values for one or both of them and if RRDtool
1086 receives data points that are outside these limits, it will ignore those
1087 values. For a thermometer in degrees Celsius, the absolute minimum is
1088 just under -273. For my router, I can assume this minimum is much higher
1089 so I would set it to 10, where as the maximum temperature I would
1090 set to 80. Any higher and the device would be out of order.
1092 For the speed of my car, I would never expect negative numbers and
1093 also I would not expect a speed higher than 230. Anything else,
1094 and there must have been an error. Remember: the opposite is not true,
1095 if the numbers pass this check, it doesn't mean that they are
1096 correct. Always judge the graph with a healthy dose of suspicion if it
1099 =head2 Data Resampling
1101 One important feature of RRDtool has not been explained yet: it is
1102 virtually impossible to collect data and feed it into RRDtool on exact
1103 intervals. RRDtool therefore interpolates the data, so they are stored
1104 on exact intervals. If you do not know what this means or how it
1105 works, then here's the help you seek:
1107 Suppose a counter increases by exactly one for every second. You want
1108 to measure it in 300 seconds intervals. You should retrieve values
1109 that are exactly 300 apart. However, due to various circumstances you
1110 are a few seconds late and the interval is 303. The delta will also be
1111 303 in that case. Obviously, RRDtool should not put 303 in the database
1112 and make you believe that the counter increased by 303 in 300 seconds.
1113 This is where RRDtool interpolates: it alters the 303 value as if it
1114 would have been stored earlier and it will be 300 in 300 seconds.
1115 Next time you are at exactly the right time. This means that the current
1116 interval is 297 seconds and also the counter increased by 297. Again,
1117 RRDtool interpolates and stores 300 as it should be.
1119 in the RRD in reality
1121 time+000: 0 delta="U" time+000: 0 delta="U"
1122 time+300: 300 delta=300 time+300: 300 delta=300
1123 time+600: 600 delta=300 time+603: 603 delta=303
1124 time+900: 900 delta=300 time+900: 900 delta=297
1126 Let's create two identical databases. I've chosen the time range 920805000
1127 to 920805900 as this goes very well with the example numbers.
1129 rrdtool create seconds1.rrd \
1131 DS:seconds:COUNTER:600:U:U \
1132 RRA:AVERAGE:0.5:1:24
1136 for Unix: cp seconds1.rrd seconds2.rrd
1137 for Dos: copy seconds1.rrd seconds2.rrd
1138 for vms: how would I know :)
1142 rrdtool update seconds1.rrd \
1143 920805000:000 920805300:300 920805600:600 920805900:900
1144 rrdtool update seconds2.rrd \
1145 920805000:000 920805300:300 920805603:603 920805900:900
1149 rrdtool graph seconds1.png \
1150 --start 920804700 --end 920806200 \
1152 --upper-limit 1.05 --lower-limit 0.95 --rigid \
1153 DEF:seconds=seconds1.rrd:seconds:AVERAGE \
1154 CDEF:unknown=seconds,UN \
1155 LINE2:seconds#0000FF \
1157 rrdtool graph seconds2.png \
1158 --start 920804700 --end 920806200 \
1160 --upper-limit 1.05 --lower-limit 0.95 --rigid \
1161 DEF:seconds=seconds2.rrd:seconds:AVERAGE \
1162 CDEF:unknown=seconds,UN \
1163 LINE2:seconds#0000FF \
1166 View both images together (add them to your index.html file)
1167 and compare. Both graphs should show the same, despite the
1168 input being different.
1172 It's time now to wrap up this tutorial. We covered all the basics for
1173 you to be able to work with RRDtool and to read the additional
1174 documentation available. There is plenty more to discover about
1175 RRDtool and you will find more and more uses for this package. You can
1176 easily create graphs using just the examples provided and using only
1177 RRDtool. You can also use one of the front ends to RRDtool that are
1182 Remember to subscribe to the RRDtool mailing list. Even if you are not
1183 answering to mails that come by, it helps both you and the rest of the
1184 users. A lot of the stuff that I know about MRTG (and therefore about
1185 RRDtool) I've learned while just reading the list without posting to
1186 it. I did not need to ask the basic questions as they are answered in
1187 the FAQ (read it!) and in various mails by other users. With
1188 thousands of users all over the world, there will always be people who
1189 ask questions that you can answer because you read this and other
1190 documentation and they didn't.
1194 The RRDtool manpages
1198 I hope you enjoyed the examples and their descriptions. If you do, help
1199 other people by pointing them to this document when they are asking
1200 basic questions. They will not only get their answers, but at the same
1201 time learn a whole lot more.
1203 Alex van den Bogaerdt
1204 E<lt>alex@vandenbogaerdt.nlE<gt>
projects / rrdtool.git / blob