Folks,
We have two F760s in a cluster running ONTAP 5.3.4, and so far have configured SNMP on there to send traps to a "traphost". I wrote a Perl program that runs on the traphost as a daemon , which intercepts the traps and logs the data to a file. I am interested in having only certain MIB-II parameters sent to the traphost (e.g. cpuUpTime, cpuIdleTime, ...) on a periodic basis (e.g. 60 sec), rather than having the traphost directly query the NetApp periodically.
Is it possible to configure the NetApp to do this? If so, how (please provide an example)? I've read through the sys. admin manual and man pages on SNMP, but they don't provide sufficient info.
Thanks in advance!
--john
* johnj.sasso@ps.ge.com (johnj.sasso@ps.ge.com) done spit this rhetoric:
Folks,
We have two F760s in a cluster running ONTAP 5.3.4, and so far have configured SNMP on there to send traps to a "traphost". I wrote a Perl program that runs on the traphost as a daemon , which intercepts the traps and logs the data to a file. I am interested in having only certain MIB-II parameters sent to the traphost (e.g. cpuUpTime, cpuIdleTime, ...) on a periodic basis (e.g. 60 sec), rather than having the traphost directly query the NetApp periodically.
Is it possible to configure the NetApp to do this? If so, how (please provide an example)? I've read through the sys. admin manual and man pages on SNMP, but they don't provide sufficient info.
It's possible to do this in a round about way. To set up for the cpuUpTime, you could:
snmp traps cpuUpTime.var snmp.1.3.6.1.4.1.789.1.2.1.1.0 snmp traps cpuUpTime.interval 60 snmp traps cpuUpTime.trigger level-trigger snmp traps cpuUpTime.edge-1 0 snmp traps cpuUptime on
Setting the edge to 0 should cause it trigger every 60 seconds, but it may not if the CPU is at 0% (but then again, checking the variable should set it to at least 1%...)