Configuring to send logs if vs,pool,poolmember is down

A pool member going down will create an entry in /var/log/ltm, and will also send a trap as defined in /etc/alertd/

/var/log/ltm:

Mar 31 19:35:53 ltm-1154-217 notice mcpd[7360]: 01070638:5: Pool /Common/bob member /Common/1.2.3.4:80 monitor status down. [ /Common/gateway_icmp: down ]  [ was unchecked for 0hr:0min:16sec ]
Mar 31 19:35:53 ltm-1154-217 err tmm[14221]: 01010028:3: No members available for pool /Common/bob

The action(s) to be taken for each type of log message are defined in the map files in /etc/alertd/*.maps.h . We can identify the actions by searching for a bit of unique text from the error message, in order to get the error name. The error ID is a good choice to search on:

grep 01070638 /etc/alertd/*
/etc/alertd/bigip_mcpd_error_maps.h:0 LOG_NOTICE        01070638 BIGIP_MCPD_MCPDERR_POOL_MEMBER_MON_STATUS Pool %s  member %s:%u monitor status %s

This reveals that the name of the error for a pool member changing state is BIGIP_MCPD_MCPDERR_POOL_MEMBER_MON_STATUS

We then check /etc/alertd/alert.conf for BIGIP_MCPD_MCPDERR_POOL_MEMBER_MON_STATUS :

alert BIGIP_MCPD_MCPDERR_POOL_MEMBER_MON_STATUS { snmptrap OID=".1.3.6.1.4.1.3375.2.4.0.10" }

So the action is to send an SNMP trap with an OID of 1.3.6.1.4.1.3375.2.4.0.10

Similarly, we can see what happens when a pool member goes up (.3375.2.4.0.11), a virtual going down (.3375.2.4.0.136), etc.

It is also possible to define custom behaviours, such as sending an email alert. For more information on this, check out SOL3667: Configuring alerts to send email notifications