Question Grafana/Monitoring reported boot partition usage warning today, but no usage change lately

[Bitpalast]

Server operating system version

Alma 8 and 9

Plesk version and microupdate number

18.0.77 #2 and 18.0.78 latest

This morning on all our Alma 8 and Alma 9 servers, Grafana Monitoring tool reported that the threshold of boot partition disk usage was exceeded. However, when I skip backwards to previous days, nothing has changed on the boot partitions. This should only leave the conclusion that the Grafana threshold value was changed so that all servers suddenly reported an overuse.

Is this a known issue/behavior?

As I have not visited this report for a very long while (because nothing should really change on the boot partition unless a crash dump is created or a new kernel installed) I am not sure what the threshold was before. I did check whether new files are shown in the boot partition, but the last changes were in May when a kernel update was mandatory due to the Alma CVEs. Nothing changed after that update, yet today was the first time the Grafana Monitoring warnings came in.

 

[Bitpalast]

I also saw that the warning was reported twice: Once at night (e.g. 5.09 am) when obviously something was checked and then again when I opened the panel to check what's going on (at around 9 am). Normally, a disk usage warning only appears once.

 

[Sebahat.hadzhi]

Hi, Peter. There was a recent Grafana upgrade and according to our team's test the thresholds should have been migrated successfully (if legacy alerting was used). I will double-check with our team if that could potentially be the reason for the issue you are noticing.

 

[Kaspar]

For what it's worth, I had almost the same happening on one Alma 9 server server, for me the threshold notifications went off every 4 hours (4 AM, 7 AM, 11 AM) for everything (Memory usage, Disk partition, Services Memory usage), except for anything CPU related. The monitoring dashboard also showed data couldn't be loaded, but I forget what the exact error message was. Oddly enough, the graphs on the Home Dashboard where still working and didn't show any resource usages spikes. I could not be bothered to investigate and re-installed the Monitoring and Grafana extensions instead. Which did resolve the issue for me.