How To Setup Monitoring For Your Droplets

Step 1 - Setup A Monitoring Server

Spin up a CentOS 6.3 x64 droplet that will be used to connect to all your other servers.

For our example, this monitoring server's IP is 198.211.117.251. All the future config snippets will reference this IP, so make sure to change it to your own value.

Step 2 - Install Packages on Hosts to be Monitored

yum -y install net-snmp && chkconfig snmpd on && :> /etc/snmp/snmpd.conf

Add the following lines to /etc/snmp/snmpd.conf:

com2sec mynetwork 198.211.117.251   DigitalOcean
group MyROGroup v1      mynetwork
view all   included .1  80
access MyROGroup ""      any       noauth    exact all    none  none

Our client servers are cloudmail.tk (198.211.107.218) and emailocean.tk (198.211.112.99).

If you have firewall rules on your servers, make sure to allow UDP connections to port 161 from your monitoring droplet, and drop everyone else:

iptables -I INPUT -s 198.211.117.251 -p udp --dport 161 -j ACCEPT
iptables -A INPUT -s 0/0 -p udp --dport 161 -j DROP

Now you can start SNMP daemon on all of your servers that you wish to monitor:

service snmpd start

Step 3 - Configure Monitoring Server

yum -y install net-snmp-utils rrdtool httpd && chkconfig httpd on

Create RRD database files for each client. You only need to run this once:

rrdtool create /root/cloudmail.tk.rrd -s 60 DS:in:COUNTER:300:U:U DS:out:COUNTER:300:U:U RRA:AVERAGE:0.5:1:10080
rrdtool create /root/emailocean.tk.rrd -s 60 DS:in:COUNTER:300:U:U DS:out:COUNTER:300:U:U RRA:AVERAGE:0.5:1:10080

Create a script that would update your graphs. The first set of commands creates and updates the data file, while second one plots this data into a PNG file that will be saved into Apache's DocumentRoot path.

Save this file as /root/rrdupdate.sh and make sure you do not have any spaces after '\' since this is all a single line that was broken into multiple:

#!/bin/bash
rrdupdate /root/cloudmail.tk.rrd N:`snmpget -v 1 -c DigitalOcean -Oqv 198.211.107.218 \
IF-MIB::ifInOctets.1`:`snmpget -v 1 -c DigitalOcean -Oqv 198.211.107.218 IF-MIB::ifOutOctets.1`

rrdtool graph /var/www/html/cloudmail.tk.png \
-a PNG -h 125 -s -43200 --title "cloudmail.tk (198.211.107.218)" \
-v "Data Throughput" \
'DEF:in=/root/cloudmail.tk.rrd:in:AVERAGE' \
'DEF:out=/root/cloudmail.tk.rrd:out:AVERAGE' \
'CDEF:kbin=in,1024,/' \
'CDEF:kbout=out,1024,/' \
'AREA:in#00FF00:Bandwidth In' \
'LINE1:out#0000FF:Bandwidth Out\j' \
'GPRINT:kbin:LAST:Last Bandwidth In\:    %3.2lf KBps' \
'GPRINT:kbout:LAST:Last Bandwidth Out\:   %3.2lf KBps\j' \
'GPRINT:kbin:AVERAGE:Average Bandwidth In\: %3.2lf KBps' \
'GPRINT:kbout:AVERAGE:Average Bandwidth Out\:%3.2lf KBps\j'

rrdupdate /root/emailocean.tk.rrd N:`snmpget -v 1 -c DigitalOcean -Oqv 198.211.112.99 \
IF-MIB::ifInOctets.1`:`snmpget -v 1 -c DigitalOcean -Oqv 198.211.112.99 IF-MIB::ifOutOctets.1`

rrdtool graph /var/www/html/emailocean.tk.png \
-a PNG -h 125 -s -43200 --title "emailocean.tk (198.211.112.99)" \
-v "Data Throughput" \
'DEF:in=/root/emailocean.tk.rrd:in:AVERAGE' \
'DEF:out=/root/emailocean.tk.rrd:out:AVERAGE' \
'CDEF:kbin=in,1024,/' \
'CDEF:kbout=out,1024,/' \
'AREA:in#00FF00:Bandwidth In' \
'LINE1:out#0000FF:Bandwidth Out\j' \
'GPRINT:kbin:LAST:Last Bandwidth In\:    %3.2lf KBps' \
'GPRINT:kbout:LAST:Last Bandwidth Out\:   %3.2lf KBps\j' \
'GPRINT:kbin:AVERAGE:Average Bandwidth In\: %3.2lf KBps' \
'GPRINT:kbout:AVERAGE:Average Bandwidth Out\:%3.2lf KBps\j'

Run your script to make sure there are no errors and that graphs are generated:

bash /root/rrdupdate.sh
ls -lah /var/www/html/*.png

Now just add a crontab entry that would update the graphs every 5 minutes, and start Apache

echo "*/5 * * * * bash /root/rrdupdate.sh" > /var/spool/cron/root
service httpd start

You can include those graphs into your own website, or add a new /var/www/html/index.html

Allow the data to be populated for a few hours and navigate over to your monitoring droplet's IP address:

And you are all done!