So today, my friend and I were working on this interesting case:
There was an alarm for Network Up-link redundancy lost.
Now the funny thing about this alarm, it is triggered when any vSwitch loses it's network redundancy.
Say, for example. vSwitch 1 has two NICs in active/standby state > No alarms are triggered
When one of the NIC goes down on this vSwitch, then the "Network Uplink redundancy lost" alarm is triggered.
So to notify, we configured email notification for this alarm and configured the mail settings on vCenter as well.
All went good, we received the email stating network redundancy was gone.
Then, we added the NIC back to this vSwitch, and the alert was gone. However, this time, the network uplink restored email was not sent.
We spent a good 30 minutes checking the configuration of the alarm.
1. We checked the trigger conditions:
2. Lost network redundancy was set to Alert Status
3. Restored uplink redundancy to portgroups was set to Normal Status
Actions were set to:
1. Send Notification email.
2. Email Settings
3. All the conditions set to send an alert once.
Yet, we received email for Redundancy down and not redundancy restored.
So, out of curiosity, my colleague said, "Hey! Why not try setting up a similar alarm and copy these triggers over?"
Why not? Off we went with this!
1. We created a new alarm with the alarm name "Network NIC redundancy lost"
2. Monitor for Host
3. Monitor for specific events
4. Enable Alarm
1. Lost Network Redundancy - Alert
2. Restored uplink redundancy to portgroups - Normal
3. Lost Network Redundancy on DVPorts - Alert
4. Restored Network Redundancy to DVPorts - Normal
1. Configured Send notification email and the email address
2. All triggers set to once
3. Disabled the pre-defined alarm which was not working
Simulated this issue again.
We received an email notification when the redundancy was lost
And when the redundancy was restored, an email notification was sent and the alert was cleared automatically.
Something indeed fishy with the pre-defined alarm here!!