Blob

Date:
Message:
add blog post in README.html
Actions:
History | Blame | Raw File
1 <!DOCTYPE html>

2 <html lang="en" class="black bg-white sans-serif lh-copy">

3 <title>RRDtool looks nicer... - High5!</title>

4 <meta charset="UTF-8">

5 <body>

6 <h1 id="RRDtool%20looks%20nicer...">RRDtool looks nicer...</h1>

8 <p><a href="https://openbsd.amsterdam/">OpenBSD Amsterdam</a> was in search of a lightweight toolset to keep track of resource usage, at a minimum the CPU load generated by the <a href="https://man.openbsd.org/vmm.4">vmm(4)</a>/<a href="https://man.openbsd.org/vmd.8">vmd(8)</a> hosts and the traffic from and to the hosts. A couple of weeks ago we ended up with a workable [MRTG setup]. While it worked, it didn&#39;t look very pretty.</p>

10 <p>In a moment of clarity, we thought about using <a href="https://oss.oetiker.ch/rrdtool/">RRDtool</a>. Heck, why shouldn&#39;t we give it a try? From the previous tooling, we already had some required building blocks in place to make <a href="https://oss.oetiker.ch/mrtg/">MRTG</a> understand the CPU Cores and uptime from <a href="https://openbsd.org/">OpenBSD</a>.</p>

12 <p>We decided to split the collection of the different OIDs (SNMP Object Identifiers) into three different scripts, which <a href="https://man.openbsd.org/cron.1">cron(1)</a> calls, from a wrapper script.</p>

14 <ul>

15 <li>uptime.sh</li>

16 <li>cpu_load.sh</li>

17 <li>interface.sh</li>

18 </ul>

20 <h3 id="uptime.sh">uptime.sh</h3>

22 <pre><code>#!/bin/sh

23 test -n &quot;$1&quot; || exit 1

24 HOST=&quot;$1&quot;

25 COMMUNITY=&quot;public&quot;

26 UPTIMEINFO=&quot;/tmp/${HOST}-uptime.txt&quot;

27 TICKS=$(snmpctl snmp get ${HOST} community ${COMMUNITY} oid hrSystemUptime.0 | cut -d= -f2)

28 DAYS=$(echo &quot;${TICKS}/8640000&quot; | bc -l)

29 HOURS=$(echo &quot;0.${DAYS##*.} * 24&quot; | bc -l)

30 MINUTES=$(echo &quot;0.${HOURS##*.} * 60&quot; | bc -l)

31 SECS=$(echo &quot;0.${MINUTES##*.} * 60&quot; | bc -l)

32 test -n &quot;$DAYS&quot; &amp;&amp; printf &#39;%s days, &#39; &quot;${DAYS%.*}&quot; &gt; ${UPTIMEINFO}

33 printf &#39;%02d\\:%02d\\:%02d\n&#39; &quot;${HOURS%.*}&quot; &quot;${MINUTES%.*}&quot; &quot;${SECS%.*}&quot; &gt;&gt; ${UPTIMEINFO}

34 </code></pre>

36 <p>This is a seperate script, due to the uptime usage of both hosts in both graphs. </p>

38 <p>The origins for this script can be found detailled in our <a href="https://chargen.one/obsdams/using-mrtg-again">MRTG Setup</a>.</p>

40 <h3 id="cpu_load.sh">cpu_load.sh</h3>

42 <pre><code>test -n &quot;$1&quot; || exit 1

43 HOST=&quot;$1&quot;

44 COMMUNITY=&quot;public&quot;

45 RRDFILES=&quot;/var/rrdtool&quot;

46 IMAGES=&quot;/var/www/htdocs&quot;

47 WATERMARK=&quot;OpenBSD Amsterdam - https://obsda.ms&quot;

48 RRDTOOL=&quot;/usr/local/bin/rrdtool&quot;

49 CPUINFO=&quot;/tmp/${HOST}-cpu.txt&quot;

50 UPTIME=$(cat /tmp/${HOST}-uptime.txt)

51 NOW=$(date &quot;+%Y-%m-%d %H:%M:%S %Z&quot; | sed &#39;s/:/\\:/g&#39;)

53 if ! test -f &quot;${RRDFILES}/${HOST}-cpu.rrd&quot;

54 then

55 echo &quot;Creating ${RRDFILES}/${HOST}-cpu.rrd&quot;

56 ${RRDTOOL} create ${RRDFILES}/${HOST}-cpu.rrd \

57     --step 300 \

58     DS:ds0:GAUGE:600:U:U \

59     RRA:MAX:0.5:1:20000

60 fi

62 snmpctl snmp walk ${HOST} community ${COMMUNITY} oid hrProcessorLoad | cut -d= -f2 &gt; ${CPUINFO}

63 CORES=$(grep -cv &quot;^0$&quot; ${CPUINFO})

64 CPU_LOAD_SUM=$(awk &#39;{sum += $1} END {print sum}&#39; ${CPUINFO})

65 CPU_LOAD=$(echo &quot;scale=2; ${CPU_LOAD_SUM}/${CORES}&quot; | bc -l)

67 ${RRDTOOL} update ${RRDFILES}/${HOST}-cpu.rrd N:${CPU_LOAD}

69 ${RRDTOOL} graph ${IMAGES}/${HOST}-cpu.png \

70     --start -43200 \

71     --title &quot;${HOST} - CPU&quot; \

72     --vertical-label &quot;% CPU Used&quot; \

73     --watermark &quot;${WATERMARK}&quot; \

74     DEF:CPU=${RRDFILES}/${HOST}-cpu.rrd:ds0:AVERAGE \

75     AREA:CPU#FFCC00 \

76     LINE2:CPU#CC0033:&quot;CPU&quot; \

77     GPRINT:CPU:MAX:&quot;Max\:%2.2lf %s&quot; \

78     GPRINT:CPU:AVERAGE:&quot;Average\:%2.2lf %s&quot; \

79     GPRINT:CPU:LAST:&quot; Current\:%2.2lf %s\n&quot; \

80     COMMENT:&quot;\\n&quot; \

81     COMMENT:&quot;  SUM CPU Load / Active Cores = % CPU Used\n&quot; \

82     COMMENT:&quot;  Up for ${UPTIME} at ${NOW}&quot;

83 </code></pre>

85 <p>On the first run, <a href="https://oss.oetiker.ch/rrdtool/">RRDtool</a> will create the .rrd file. On every subsequent run, it will update the file with the collected values and update the graph.</p>

87 <p>The origins for this script can be found detailed in our <a href="https://chargen.one/obsdams/using-mrtg-again">MRTG Setup</a>.</p>

89 <h3 id="interface.sh">interface.sh</h3>

91 <pre><code>test -n &quot;$1&quot; || exit 1                                                                             

92 test -n &quot;$2&quot; || exit 1                                                                             

93 HOST=&quot;$1&quot;                                                                                          

94 INTERFACE=&quot;$2&quot;                                                                                     

95 COMMUNITY=&quot;public&quot;                                                                                 

96 RRDFILES=&quot;/var/rrdtool&quot;

97 IMAGES=&quot;/var/www/htdocs&quot;

98 WATERMARK=&quot;OpenBSD Amsterdam - https://obsda.ms&quot;

99 RRDTOOL=&quot;/usr/local/bin/rrdtool&quot;

100 UPTIME=$(cat /tmp/${HOST}-uptime.txt)

101 NOW=$(date &quot;+%Y-%m-%d %H:%M:%S %Z&quot; | sed &#39;s/:/\\:/g&#39;)                                              

103 if ! test -f &quot;${RRDFILES}/${HOST}-${INTERFACE}.rrd&quot;                                                

104 then

105 echo &quot;Creating ${RRDFILES}/${HOST}-${INTERFACE}.rrd&quot;                                               

106 ${RRDTOOL} create ${RRDFILES}/${HOST}-${INTERFACE}.rrd \                                           

107     --step 300 \

108     DS:ds0:COUNTER:600:0:1250000000 \

109     DS:ds1:COUNTER:600:0:1250000000  \

110     RRA:AVERAGE:0.5:1:600 \

111     RRA:AVERAGE:0.5:6:700 \

112     RRA:AVERAGE:0.5:24:775 \

113     RRA:AVERAGE:0.5:288:797 \

114     RRA:MAX:0.5:1:600 \

115     RRA:MAX:0.5:6:700 \

116     RRA:MAX:0.5:24:775 \

117     RRA:MAX:0.5:288:797

118 fi

120 IN=$(snmpctl snmp get ${HOST} community ${COMMUNITY} oid ifInOctets.${INTERFACE} | cut -d= -f2)    

121 OUT=$(snmpctl snmp get ${HOST} community ${COMMUNITY} oid ifOutOctets.${INTERFACE} | cut -d= -f2)  

122 DESCR=$(snmpctl snmp get ${HOST} community ${COMMUNITY} oid ifDescr.${INTERFACE} | cut -d= -f2 | tr

123 -d &#39;&quot;&#39;)

125 ${RRDTOOL} update ${RRDFILES}/${HOST}-${INTERFACE}.rrd N:${IN}:${OUT}                              

127 ${RRDTOOL} graph ${IMAGES}/${HOST}-${INTERFACE}.png \                                              

128     --start -43200 \

129     --title &quot;${HOST} - ${DESCR}&quot; \

130     --vertical-label &quot;Bits per Second&quot; \

131     --watermark &quot;${WATERMARK}&quot; \

132     DEF:IN=${RRDFILES}/${HOST}-${INTERFACE}.rrd:ds0:AVERAGE \                                  

133     DEF:OUT=${RRDFILES}/${HOST}-${INTERFACE}.rrd:ds1:AVERAGE \                                 

134     CDEF:IN_CDEF=&quot;IN,8,*&quot; \

135     CDEF:OUT_CDEF=&quot;OUT,8,*&quot; \

136     AREA:IN_CDEF#00FF00:&quot;In &quot; \

137     GPRINT:IN_CDEF:MAX:&quot;Max\:%5.2lf %s&quot; \

138     GPRINT:IN_CDEF:AVERAGE:&quot;Average\:%5.2lf %s&quot; \                                              

139     GPRINT:IN_CDEF:LAST:&quot; Current\:%5.2lf %s\n&quot; \                                              

140     LINE2:OUT_CDEF#0000FF:&quot;Out&quot; \

141     GPRINT:OUT_CDEF:MAX:&quot;Max\:%5.2lf %s&quot; \

142     GPRINT:OUT_CDEF:AVERAGE:&quot;Average\:%5.2lf %s&quot; \                                             

143     GPRINT:OUT_CDEF:LAST:&quot; Current\:%5.2lf %s\n&quot; \                                             

144     COMMENT:&quot;\\n&quot; \

145     COMMENT:&quot;  Up for ${UPTIME} at ${NOW}&quot;

146 </code></pre>

148 <p>To pinpoint the network interface you want to measure the bandwith for, this command prints the available interfaces:</p>

150 <pre><code>snmpctl snmp walk  community  oid ifDescr

151 </code></pre>

153 <p>This will output a list like:</p>

155 <pre><code>ifDescr.1=&quot;em0&quot;

156 ifDescr.2=&quot;em1&quot;

157 ifDescr.3=&quot;enc0&quot;

158 ifDescr.4=&quot;lo0&quot;

159 ifDescr.5=&quot;bridge880&quot;

160 ifDescr.6=&quot;vlan880&quot;

161 ifDescr.13=&quot;pflog0&quot;

162 ifDescr.669=&quot;tap0&quot;

163 ifDescr.670=&quot;tap1&quot;

164 </code></pre>

166 <p>The number behind <code>ifDescr</code> is the one that you need to feed to <strong>interface.sh</strong>, for example:</p>

168 <pre><code># interface.sh  5

169 </code></pre>

171 <p>Finally the <strong>wrapper.sh</strong> script calls all the aforementioned scripts:</p>

173 <pre><code>#!/bin/sh

174 SCRIPTS=&quot;/var/rrdtool&quot;

175 for i in $(jot 2 1); do ${SCRIPTS}/uptime.sh host${i}.domain.tld; done

176 for i in $(jot 2 1); do ${SCRIPTS}/cpu_load.sh host${i}.domain.tld; done

177 ${SCRIPTS}/interface.sh host1.domain.tld 12

178 ${SCRIPTS}/interface.sh host2.domain.tld 11

179 </code></pre>

181 <p>The resulting graphs:</p>

183 <p><img src="https://ops.high5.nl/obsda.ms/s1.obsda.ms-cpu.png" alt="" />

184 <img src="https://ops.high5.nl/obsda.ms/s1.obsda.ms-12.png" alt="" /></p>

186 <p>You can follow us on <a href="https://twitter.com/@OpenBSDAms">Twitter</a> and <a href="https://bsd.network/@OpenBSDAms">Mastodon</a>.</p>

187 </body>

188 </html>