I'm using table metrics extensively for collecting data which changes frequently, like running processes or some kind information on user sessions on a terminal server (instead of using instance discovery which I use for metrics which stay unchanged most of the time).
I have an issue with table thresholds and I'm not sure if this is expected behavior:
- Let'say, I'm monitoring the expiration date of TLS certificates with a table metric.
- At some point the expiration date reaches the threshold and an event is generated.
- On the system, the certificate is replaced with a new one and the old (expired) one is removed. This means, the table metric will drop the row for the expired certificate.
- The issue is: The table metric still displays that the threshold has been reached, even if this row has disappeared.
I understand that this is sometimes useful, but in this case the "row" for the expired certificate will never appear again and the threshold on the table metric is reached indefinitely. Even clearing the collected data won't reset this state.
Is there some workaroud for this situation? If no, I would request a new option for table metrics to trigger the threshold "Deactivation event" for disappearing rows.