So you followed the steps in the previous post about enabling SNMP traps on ESX4. Now you probably want to pick those up by something useful. Opsview can be configured to handle the traps quite easily. Just follow the steps below and your server will be listening to those pesky traps. After that, you’ll need to write a couple of service check handlers in Opsview to make sense of the traps. More on that later. This post is just about picking them up. Continue reading »
Oct 05
Mar 13
This is a potentially controversial patch, so may not get into the Nagios core code.
We’re working on integrating SNMPtraps into Opsview which are passive checks by nature. However, when a service is initially added into Nagios, the CGIs show them in a PENDING state, which looks like an error. We prefer to have a sea of green when things are OK.
A PENDING state is fine for a distributed monitoring setup, because the active check on the slaves will get through to the master soon, but not with other “irregular” passive checks. I couldn’t find a way to distinguish between a passive, “going to get a result from a distributed slave soon”, and a passive, “don’t know when the next result is going to be”.
So this patch will amend the CGIs so that passive checks (or more precisely, checks that are not scheduled to run) are displayed as OK, rather than PENDING.
Opsview is a leading Open Source application and network monitoring suite. Labs is where our engineers discuss new projects, new approaches and new frameworks they’re using.
Recent Comments