Hi everyone,
I couldn’t find exactly what I was looking for in the existing topics.
I’m currently working with a LibreNMS instance that monitors around 1,500 devices. In the long term, with 8 pollers and 2 RRDs, we expect to reach about 15,000 devices. Our network changes frequently and generates a lot of alerts. For example, some ports go “down” when we modify their configuration. This triggers an alert in the corresponding port notification (which is expected), but the issue is that the alert remains visible in the notifications window even after it has been acknowledged.
Is there a way, directly from the WEBUI, to add a checkbox and a menu option to acknowledge alerts in bulk, or to delete certain alerts from the database, without having to access the database backend directly?
Thank you,
Guillaume
Bonjour Guillaume,
In LibreNMS logic, an alert is generated from a state. Basically an SQL that triggers the alert. Removing an alert does not make sense, cause as long as the condition is there, it will come back next poll.
You need to define your alerts in a way that only significant alerts are generated. Not always easy.
I personnaly use the description of ports/interfaces to “tag” ports that are backbone, customer, accesspoints, etc etc. And that allows better rules and better alerting.
Hello, thank you for your feedback. I understand what you mean, but my alarms are configured more or less correctly. However, changes in the network topology cause some alarms to appear and never clear, so they remain in the alarm list. Having the ability to acknowledge alarms in bulk would already be a great improvement!
Best regards,
Guillaume