1RRD-BEGINNERS(1) rrdtool RRD-BEGINNERS(1)
2
6 rrd-beginners - RRDtool Beginners' Guide
7
9 Helping new RRDtool users to understand the basics of RRDtool
10
12 This manual is an attempt to assist beginners in understanding the
13 concepts of RRDtool. It sheds a light on differences between RRDtool
14 and other databases. With help of an example, it explains the structure
15 of RRDtool database. This is followed by an overview of the "graph"
16 feature of RRDtool. At the end, it has sample scripts that illustrate
17 the usage/wrapping of RRDtool within Shell or Perl scripts.
18 What makes RRDtool so special?
20 RRDtool is GNU licensed software developed by Tobias Oetiker, a system
21 manager at the Swiss Federal Institute of Technology. Though it is a
22 database, there are distinct differences between RRDtool databases and
23 other databases as listed below:
24 · RRDtool stores data; that makes it a back-end tool. The RRDtool
26 command set allows one to create graphs; that makes it a front-end
27 tool as well. Other databases just store data and can not create
28 graphs.
29 · In case of linear databases, new data gets appended at the bottom
31 of the database table. Thus its size keeps on increasing, whereas
32 the size of an RRDtool database is determined at creation time.
33 Imagine an RRDtool database as the perimeter of a circle. Data is
34 added along the perimeter. When new data reaches the starting
35 point, it overwrites existing data. This way, the size of an
36 RRDtool database always remains constant. The name "Round Robin"
37 stems from this behavior.
38 · Other databases store the values as supplied. RRDtool can be
40 configured to calculate the rate of change from the previous to the
41 current value and store this information instead.
42 · Other databases get updated when values are supplied. The RRDtool
44 database is structured in such a way that it needs data at
45 predefined time intervals. If it does not get a new value during
46 the interval, it stores an UNKNOWN value for that interval. So,
47 when using the RRDtool database, it is imperative to use scripts
48 that run at regular intervals to ensure a constant data flow to
49 update the RRDtool database.
50 RRDtool is designed to store time series of data. With every data
52 update, an associated time stamp is stored. Time is always expressed in
53 seconds passed since epoch (01-01-1970). RRDtool can be installed on
54 Unix as well as Windows. It comes with a command set to carry out
55 various operations on RRD databases. This command set can be accessed
56 from the command line, as well as from Shell or Perl scripts. The
57 scripts act as wrappers for accessing data stored in RRDtool databases.
58 Understanding by an example
60 The structure of an RRD database is different than other linear
61 databases. Other databases define tables with columns, and many other
62 parameters. These definitions sometimes are very complex, especially in
63 large databases. RRDtool databases are primarily used for monitoring
64 purposes and hence are very simple in structure. The parameters that
65 need to be defined are variables that hold values and archives of those
66 values. Being time sensitive, a couple of time related parameters are
67 also defined. Because of its structure, the definition of an RRDtool
68 database also includes a provision to specify specific actions to take
69 in the absence of update values. Data Source (DS), heartbeat, Date
70 Source Type (DST), Round Robin Archive (RRA), and Consolidation
71 Function (CF) are some of the terminologies related to RRDtool
72 databases.
73 The structure of a database and the terminology associated with it can
75 be best explained with an example.
76 rrdtool create target.rrd \
78 --start 1023654125 \
79 --step 300 \
80 DS:mem:GAUGE:600:0:671744 \
81 RRA:AVERAGE:0.5:12:24 \
82 RRA:AVERAGE:0.5:288:31
83 This example creates a database named target.rrd. Start time
85 (1'023'654'125) is specified in total number of seconds since epoch
86 (time in seconds since 01-01-1970). While updating the database, the
87 update time is also specified. This update time MUST be larger (later)
88 than start time and MUST be in seconds since epoch.
89 The step of 300 seconds indicates that the database expects new values
91 every 300 seconds. The wrapper script should be scheduled to run every
92 step seconds so that it updates the database every step seconds.
93 DS (Data Source) is the actual variable which relates to the parameter
95 on the device that is monitored. Its syntax is
96 DS:variable_name:DST:heartbeat:min:max
98 DS is a key word. "variable_name" is a name under which the parameter
100 is saved in the database. There can be as many DSs in a database as
101 needed. After every step interval, a new value of DS is supplied to
102 update the database. This value is also called Primary Data Point
103 (PDP). In our example mentioned above, a new PDP is generated every 300
104 seconds.
105 Note, that if you do NOT supply new data points exactly every 300
107 seconds, this is not a problem, RRDtool will interpolate the data
108 accordingly.
109 DST (Data Source Type) defines the type of the DS. It can be COUNTER,
111 DERIVE, ABSOLUTE, GAUGE. A DS declared as COUNTER will save the rate of
112 change of the value over a step period. This assumes that the value is
113 always increasing (the difference between the current and the previous
114 value is greater than 0). Traffic counters on a router are an ideal
115 candidate for using COUNTER as DST. DERIVE is the same as COUNTER, but
116 it allows negative values as well. If you want to see the rate of
117 change in free disk space on your server, then you might want to use
118 the DERIVE data type. ABSOLUTE also saves the rate of change, but it
119 assumes that the previous value is set to 0. The difference between the
120 current and the previous value is always equal to the current value.
121 Thus it just stores the current value divided by the step interval (300
122 seconds in our example). GAUGE does not save the rate of change. It
123 saves the actual value itself. There are no divisions or calculations.
124 Memory consumption in a server is a typical example of gauge. The
125 difference between the different types DSTs can be explained better
126 with the following example:
127 Values = 300, 600, 900, 1200
129 Step = 300 seconds
130 COUNTER DS = 1, 1, 1, 1
131 DERIVE DS = 1, 1, 1, 1
132 ABSOLUTE DS = 1, 2, 3, 4
133 GAUGE DS = 300, 600, 900, 1200
134 The next parameter is heartbeat. In our example, heartbeat is 600
136 seconds. If the database does not get a new PDP within 300 seconds, it
137 will wait for another 300 seconds (total 600 seconds). If it doesn't
138 receive any PDP within 600 seconds, it will save an UNKNOWN value into
139 the database. This UNKNOWN value is a special feature of RRDtool - it
140 is much better than to assume a missing value was 0 (zero) or any other
141 number which might also be a valid data value. For example, the
142 traffic flow counter on a router keeps increasing. Lets say, a value is
143 missed for an interval and 0 is stored instead of UNKNOWN. Now when the
144 next value becomes available, it will calculate the difference between
145 the current value and the previous value (0) which is not correct. So,
146 inserting the value UNKNOWN makes much more sense here.
147 The next two parameters are the minimum and maximum value,
149 respectively. If the variable to be stored has predictable maximum and
150 minimum values, this should be specified here. Any update value falling
151 out of this range will be stored as UNKNOWN.
152 The next line declares a round robin archive (RRA). The syntax for
154 declaring an RRA is
155 RRA:CF:xff:step:rows
157 RRA is the keyword to declare RRAs. The consolidation function (CF) can
159 be AVERAGE, MINIMUM, MAXIMUM, and LAST. The concept of the consolidated
160 data point (CDP) comes into the picture here. A CDP is CFed (averaged,
161 maximum/minimum value or last value) from step number of PDPs. This RRA
162 will hold rows CDPs.
163 Lets have a look at the example above. For the first RRA, 12 (steps)
165 PDPs (DS variables) are AVERAGEed (CF) to form one CDP. 24 (rows) of
166 theses CDPs are archived. Each PDP occurs at 300 seconds. 12 PDPs
167 represent 12 times 300 seconds which is 1 hour. It means 1 CDP (which
168 is equal to 12 PDPs) represents data worth 1 hour. 24 such CDPs
169 represent 1 day (1 hour times 24 CDPs). This means, this RRA is an
170 archive for one day. After 24 CDPs, CDP number 25 will replace the 1st
171 CDP. The second RRA saves 31 CDPs; each CPD represents an AVERAGE value
172 for a day (288 PDPs, each covering 300 seconds = 24 hours). Therefore
173 this RRA is an archive for one month. A single database can have many
174 RRAs. If there are multiple DSs, each individual RRA will save data for
175 all the DSs in the database. For example, if a database has 3 DSs and
176 daily, weekly, monthly, and yearly RRAs are declared, then each RRA
177 will hold data from all 3 data sources.
178 Graphical Magic
180 Another important feature of RRDtool is its ability to create graphs.
181 The "graph" command uses the "fetch" command internally to retrieve
182 values from the database. With the retrieved values it draws graphs as
183 defined by the parameters supplied on the command line. A single graph
184 can show different DS (Data Sources) from a database. It is also
185 possible to show the values from more than one database in a single
186 graph. Often, it is necessary to perform some math on the values
187 retrieved from the database before plotting them. For example, in SNMP
188 replies, memory consumption values are usually specified in KBytes and
189 traffic flow on interfaces is specified in Bytes. Graphs for these
190 values will be more meaningful if values are represented in MBytes and
191 mbps. The RRDtool graph command allows one to define such conversions.
192 Apart from mathematical calculations, it is also possible to perform
193 logical operations such as greater than, less than, and if/then/else.
194 If a database contains more than one RRA archive, then a question may
195 arise - how does RRDtool decide which RRA archive to use for retrieving
196 the values? RRDtool looks at several things when making its choice.
197 First it makes sure that the RRA covers as much of the graphing time
198 frame as possible. Second it looks at the resolution of the RRA
199 compared to the resolution of the graph. It tries to find one which has
200 the same or higher better resolution. With the "-r" option you can
201 force RRDtool to assume a different resolution than the one calculated
202 from the pixel width of the graph.
203 Values of different variables can be presented in 5 different shapes in
205 a graph - AREA, LINE1, LINE2, LINE3, and STACK. AREA is represented by
206 a solid colored area with values as the boundary of this area.
207 LINE1/2/3 (increasing width) are just plain lines representing the
208 values. STACK is also an area but it is "stack"ed on top AREA or
209 LINE1/2/3. Another important thing to note is that variables are
210 plotted in the order they are defined in the graph command. Therefore
211 care must be taken to define STACK only after defining AREA/LINE. It is
212 also possible to put formatted comments within the graph. Detailed
213 instructions can be found in the graph manual.
214 Wrapping RRDtool within Shell/Perl script
216 After understanding RRDtool it is now a time to actually use RRDtool in
217 scripts. Tasks involved in network management are data collection, data
218 storage, and data retrieval. In the following example, the previously
219 created target.rrd database is used. Data collection and data storage
220 is done using Shell scripts. Data retrieval and report generation is
221 done using Perl scripts. These scripts are shown below:
222 Shell script (collects data, updates database)
224 #!/bin/sh
226 a=0
227 while [ "$a" == 0 ]; do
228 snmpwalk -c public 192.168.1.250 hrSWRunPerfMem > snmp_reply
229 total_mem=`awk 'BEGIN {tot_mem=0}
230 { if ($NF == "KBytes")
231 {tot_mem=tot_mem+$(NF-1)}
232 }
233 END {print tot_mem}' snmp_reply`
234 # I can use N as a replacement for the current time
235 rrdtool update target.rrd N:$total_mem
236 # sleep until the next 300 seconds are full
237 perl -e 'sleep 300 - time % 300'
238 done # end of while loop
239 Perl script (retrieves data from database and generates graphs and
241 statistics)
242 #!/usr/bin/perl -w
244 # This script fetches data from target.rrd, creates a graph of memory
245 # consumption on the target (Dual P3 Processor 1 GHz, 656 MB RAM)
246 # call the RRD perl module
248 use lib qw( /usr/local/rrdtool-1.0.41/lib/perl ../lib/perl );
249 use RRDs;
250 my $cur_time = time(); # set current time
251 my $end_time = $cur_time - 86400; # set end time to 24 hours ago
252 my $start_time = $end_time - 2592000; # set start 30 days in the past
253 # fetch average values from the RRD database between start and end time
255 my ($start,$step,$ds_names,$data) =
256 RRDs::fetch("target.rrd", "AVERAGE",
257 "-r", "600", "-s", "$start_time", "-e", "$end_time");
258 # save fetched values in a 2-dimensional array
259 my $rows = 0;
260 my $columns = 0;
261 my $time_variable = $start;
262 foreach $line (@$data) {
263 $vals[$rows][$columns] = $time_variable;
264 $time_variable = $time_variable + $step;
265 foreach $val (@$line) {
266 $vals[$rows][++$columns] = $val;}
267 $rows++;
268 $columns = 0;
269 }
270 my $tot_time = 0;
271 my $count = 0;
272 # save the values from the 2-dimensional into a 1-dimensional array
273 for $i ( 0 .. $#vals ) {
274 $tot_mem[$count] = $vals[$i][1];
275 $count++;
276 }
277 my $tot_mem_sum = 0;
278 # calculate the total of all values
279 for $i ( 0 .. ($count-1) ) {
280 $tot_mem_sum = $tot_mem_sum + $tot_mem[$i];
281 }
282 # calculate the average of the array
283 my $tot_mem_ave = $tot_mem_sum/($count);
284 # create the graph
285 RRDs::graph ("/images/mem_$count.png",
286 "--title= Memory Usage",
287 "--vertical-label=Memory Consumption (MB)",
288 "--start=$start_time",
289 "--end=$end_time",
290 "--color=BACK#CCCCCC",
291 "--color=CANVAS#CCFFFF",
292 "--color=SHADEB#9999CC",
293 "--height=125",
294 "--upper-limit=656",
295 "--lower-limit=0",
296 "--rigid",
297 "--base=1024",
298 "DEF:tot_mem=target.rrd:mem:AVERAGE",
299 "CDEF:tot_mem_cor=tot_mem,0,671744,LIMIT,UN,0,tot_mem,IF,1024,/",
300 "CDEF:machine_mem=tot_mem,656,+,tot_mem,-",
301 "COMMENT:Memory Consumption between $start_time",
302 "COMMENT: and $end_time ",
303 "HRULE:656#000000:Maximum Available Memory - 656 MB",
304 "AREA:machine_mem#CCFFFF:Memory Unused",
305 "AREA:tot_mem_cor#6699CC:Total memory consumed in MB");
306 my $err=RRDs::error;
307 if ($err) {print "problem generating the graph: $err\n";}
308 # print the output
309 print "Average memory consumption is ";
310 printf "%5.2f",$tot_mem_ave/1024;
311 print " MB. Graphical representation can be found at /images/mem_$count.png.";
312
314 Ketan Patel <k2pattu@yahoo.com>
3151.7.1 2019-02-04 RRD-BEGINNERS(1)