In order to monitor the status of critical processes to define the availability of the service we have decided to use SNMP as it is already used to get the other relevant status variables used to scale-in and out the services.
To monitor them it 's enough to include the following lines in /etc/snmp/snmpd.conf:
proc 'pm2-God-Daemon' 1 1
For the specific case of pm2 which is the daemon which launches all the Node.js processes we need to define a environment variable to set the process name to pm2-God-Daemon (it is important not to include whitespaces as it is not supported by snmp):
Environment=PM2_DAEMON_TITLE=pm2-God-Daemon
To check the status of the process it is necessary to check the value of this variable:
$ snmpwalk -v 3 -l authPriv -u "quobis" -a sha -A "asterisk" -x aes -X "asterisk" "192.168.56.102" .1.3.6.1.4.1.2021.2.1.100.1
iso.3.6.1.4.1.2021.2.1.100.1 = INTEGER: 0
If the value is different from 0 the service is down.
In order to monitor the status of critical processes to define the availability of the service we have decided to use SNMP as it is already used to get the other relevant status variables used to scale-in and out the services. To monitor them it 's enough to include the following lines in
/etc/snmp/snmpd.conf:proc 'pm2-God-Daemon' 1 1For the specific case of pm2 which is the daemon which launches all the Node.js processes we need to define a environment variable to set the process name to pm2-God-Daemon (it is important not to include whitespaces as it is not supported by snmp):
Environment=PM2_DAEMON_TITLE=pm2-God-DaemonTo check the status of the process it is necessary to check the value of this variable: $ snmpwalk -v 3 -l authPriv -u "quobis" -a sha -A "asterisk" -x aes -X "asterisk" "192.168.56.102" .1.3.6.1.4.1.2021.2.1.100.1 iso.3.6.1.4.1.2021.2.1.100.1 = INTEGER: 0
If the value is different from 0 the service is down.