extended snmp and system load

[Vincent_Bacher]

Hi mates,

i am thinking about doing some monitoring using extended snmp configuration.

Started with a simple script to see if it works in general:

welt.sh

#!/bin/bash
echo "Hallo Welt!"

and extended the nmpd config in /etc/snmp/userDefinedSettings.conf

extend welt /home/****/welt.sh

and after restart of snmpd it works well

snmpwalk -v3 -l authPriv -u **** -a MD5 -A **** -x DES -X **** 127.0.0.1 nsExtendOutput2
NET-SNMP-EXTEND-MIB::nsExtendOutLine."welt".1 = STRING: Hallo Welt!

So far so good.
Now i want to continue to do some checks with other commands (and of course filtering out relevant kpi), like "pdp broker status" or "fw ctl pstat" or whatever.
And i want to use that on classic appliances, open servers and virtual systems on VSX. (but not on SP systems).

Now my question: Does anybody have already monitoring systems that way using several checks and can tell his/her experiences regarding system load or anything related which could affect performance or stability or anything else of productive gateways?

Are there any best practices what to do, what strictly not to do?

any ideas or hints would be appreciated

best regards

Vincent