If you like my content please do check out my new blog at thirdpartytools.net !
As I mentioned some time ago we use SCOM to monitor our Netscaler Load Balancers. We ran into an issue where an alert would be raised based on a SNMP trap sent by the Netscaler. Some background info:
– We have an etherchannel (Nortel Avaya SMLT if people are interested) like setup where we use 2 ethernet interfaces on the Netscalers connected to our core switches and we have all our VLAN’s trunked on those ports (including the VLAN where the NSIP’s reside).
– The other ports are not connected – but we had two interfaces enabled so we could use those to connect to the appliances if the etherchannel config got screwded up
– HA monitoring is only enabled on the channel not on any of the individual ethernet interfaces.
What happened was that traps were sent out saying that both nodes missed HA heartbeats – but when we logged into the Netscaler GUI the HA status was fine. When taking a closer look at the snmp trap data it appeared that no HA heartbeats were seen on the two ethernet ports that were enabled but not connected – even though HA monitoring was disabled on these ports.
We couldn’t really override this because the SNMP trap could only be enabled or disabled for all interfaces, so we disabled the interfaces. The only drawback is that we’ll need to use the console port if we can’t reach the Netscalers through the “etherchanneled” interfaces.
Multiple Choice System Engineer 