cos-agent relation fails after grafana reboot
This bug report will be marked for expiration in 20 days if no further activity occurs. (find out why)
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Ceph Monitor Charm |
Incomplete
|
Medium
|
Unassigned |
Bug Description
Hi team,
My cloud has a ceph-mon:cos-agent - grafana-agent relation. And I can see all the dashboards until grafana restart. Each time it happens, the dashboards disappear from Grafana even though the relation is still present. I'm able to reproduce this with:
juju ssh -m cos --container grafana grafana/0 /charm/bin/pebble restart grafana
Workaround:
juju remove-relation ceph-mon:cos-agent grafana-agent
juju relate ceph-mon:cos-agent grafana-agent
Openstack Yoga Jammy, Ceph Quincy, Juju 3.4.2
COS versions:
$ juju status -m microk8s-
Model Controller Cloud/Region Version SLA Timestamp
cos microk8s-controller microk8s_
App Version Status Scale Charm Channel Rev Address Exposed Message
alertmanager 0.27.0 waiting 1 alertmanager-k8s latest/stable 119 10.152.183.218 no installing agent
catalogue active 1 catalogue-k8s latest/stable 43 10.152.183.33 no
grafana 9.5.3 active 1 grafana-k8s latest/stable 114 10.152.183.86 no
loki 2.9.6 active 1 loki-k8s latest/stable 151 10.152.183.87 no
prometheus 2.50.1 active 1 prometheus-k8s latest/stable 189 10.152.183.121 no
scrape-
scrape-
traefik v2.11.0 active 1 traefik-k8s latest/stable 194 10.152.183.36 no Serving at 172.27.80.50
Changed in charm-ceph-mon: | |
importance: | Undecided → Medium |
status: | New → Triaged |
Hey Natalia, I'm having issues reproducing this here. Would you be able to provide a minimal bundle for reproduction? I'd also note that in my testing there was no update / hook firing on the cos-agent relation for grafana-agent / ceph-mon so ceph-mon would not be aware of any grafana restarts.