Need to monitor pool members to see if they are out of the farm

I take it this is not for a typical pool-based health monitor thing, but more for reporting?

Hi Kevin, You're right, it really is more for reporting and not for pool-based health monitors. I'm trying to be able to pump this sort of into some sort of monitoring - either Cacti, Nagios or something similar.

There's really several ways to do this:

1. user_alert.conf script and/or iCall to capture, in real time, when nodes go up and down. From there you run a script that sends an alert, syslog message, email, SNMP trap, whatever, per event.

2. Run a script that periodically checks all of the pool members for their state (up or down) and session (monitor-enabled or user-disabled). The following is a shell of what that could look like:

   !/bin/bash
   
    define pool name
   pool="local-pool";
   
   arr=($(tmsh list ltm pool local-pool |perl -ne 'if (m/([a-zA-Z0-9\.-]+:[^\s]+)\s{/) { print "$1\n"}'));
   
   for x in "${arr[@]}"
   do
       state=`tmsh list ltm pool $pool members { $x { state } } |grep state |awk -F" " '{ print $2 }'`
       session=`tmsh list ltm pool $pool members { $x { session } } |grep session |awk -F" " '{ print $2 }'`
       echo "$x:"
       echo "   state: $state"
       echo "   session: $session"
   done
   

Certainly. Replace the TMSH commands with cURL iControl REST calls and reformat the resulting data as required. I'm sure there's probably a much easier version of this, but here's an idea of what the first request might look like:

curl -sk -u 'admin:admin' -H "Content-Type: application/json" -X GET https://x.x.x.x/mgmt/tm/ltm/pool/local-pool?expandSubcollections=true |sed 's/,/\n/g' |grep "\"name\":" |awk -F"\"name\":" '{ print $2 }' |perl -ne 'if (m/([a-zA-Z0-9\.-]+:[^\s]+)/) { print "$1\n"}' |sed 's/"//g'