@ScarfAntennae said in Gather CPU utilization of host as variable for prometheus exporter:

I was about to link Mike's work here, but wanted to say how we desperately need a metrics system that would be exposed from a centralized location. There's no way in hell I'm giving a docker host credential access to the hypervisor it resides on (and all the other hypervisors) - on the other hand, if the metrics were exposed only via XO and limited in terms of where these metrics could be accessed from, now that would be a sold implementation.

Looks like this wish has been granted with the latest Xen-Orchestra release.

I was able to successfully scrape the metrics from XO using Prometheus. I was wondering if there was any links or json file examples of a good Grafana dashboard to show these metrics? I saw example screenshots in the blog post here but nothing on grafana.com, here in the forums, or on Github had the config. I saw PR #9333 with documentation about the feature, and great rules for alerting, but also no dashboard configs. It has some beginning parts for variables and 2 visualisations, but a copy of the example in the blog would be wonderful as a starting point.

@manilx said in Hundreds of tasks stuck for months:

xo-cli rest del tasks

Whew, thank you! Clearly my search skills failed me since I did not find that post myself

@manilx Mine go to Oct 2023 in case its a contest haha.

@tsrats Meant to add, the XOA has been restarted multiple times during this issue, as have all XCP-ng connected hosts (as well as upgrades, all hosts and XOA are current)

On my instance of Xen-Orchestra, there is upwards of 880 tasks stuck and constantly refreshing/changing places in the list. There is not just one, but many:

700+: "API call: sr.getAllUnhealthyVdiChainsLength"
76: XO user authentication
28: importing vms NNN
21: xoa.licenses.getAll
3: API call: session.getUser

Some of these are recent, some like the importing vms are from last year. Is there some log or setting on XOA that retains these which can be cleared?