Hey Community,
We have a couple of devices that have all their ports appear in "Deleted" list:
Because of that, RRD’s and graphs aren’t created for these ports.
Device deleting/re-adding, ports purging and DB cleanup didn’t help at all.
Will appreciate any advices/suggestions on this as it’s really annoying and unusual behavior.
Seems like all ports are skipped by the badif settings… check your settings. Run discovery and check the output.
Discovery log doesn’t contain anything abnormal related to these ports - they just being discovered and put to the DB with no cautions. Are there any specific DB entries need to be cleared to flush that “Deleted” list?
Device’s ports are not included in bad_if settings or something like that, and the interesting point here is that it happened to just couple of devices, not all of them. All our devices have the same port naming convention, so the bad_if or other regex filters couldn’t cause a problem here.
Yes, I’m only guessing, because you haven’t shown any information that can indicate what your problem is. Such as the discovery debug output or specific port examples.
Sorry, was unable to use pastebin for that so here’s link to alternative service:
https://paste.ofcode.org/3AcDUmZZYmxECnFqican8L
Also here is validate.php output: https://paste.ofcode.org/7W8vtHJP8uCyPHUZJkJjzc
rrdcached test failed because it runs on the other machine
Port issue: “ignored: empty ifDescr, ifAlias and ifName”
Also, you seem to be using distributed polling, but rrdcached is not set correctly
Also, your install is not updating
Any way it can be cleaned up so I can run discovery/polling again and collect all the port parameters?
The issue appears with 3 hosts only so it doesn’t seem as a common issue for the whole inventory.
Your device is returning all three of those datapoints as empty, so LibreNMS does not have any way to label the port. So it doesn’t add it.
This means your device seems to be supplying bad data that LibreNMS cannot fix.