Repository navigation
new dashboard cannot update monitoring yaml #1550
Description
Describe the bug
New dashboard cannot update monitoring yaml.
Environment
- Cozystack v37.2
- Provider: on-prem
To Reproduce
Steps to reproduce the behavior:
- Go to dashboard tenant-root monitoring
- Open YAML tab
- Edit value
- Click update button
Expected behaviour
The changes should be applied to YAML
Actual behaviour
The error pop up "Could not find object"
Logs
I don't have logs
Screenshots
Detaild screenshot attached
Additional context
Tried on two indepedent clusters.
Checklist
- [x ] I have checked the documentation
- [x ] I have searched for similar issues
- [x ] I have included all required information
- [x ] I have provided clear steps to reproduce
- [x
] I have included relevant logs
Activity
dosubot commentedon Oct 24, 2025
Hi yviadzishchau! I'm Dosu and I’m helping the cozystack team.
The error "Could not find object" when updating the monitoring YAML in the dashboard is most likely caused by the backend API failing to locate the monitoring resource or its Custom Resource Definition (CRD) in your cluster. This can happen if the resource type is missing, the CRD isn't installed, or there was a recent regression in resource lookup logic—especially around HelmRelease or monitoring resource handling. Similar issues have been reported after changes to resource templates, where reverting to an older version resolved the problem (example).
To troubleshoot, check that the monitoring resource and its CRD exist and are accessible in your cluster (for example, with
kubectl get crdandkubectl get monitoring -n tenant-root). Also, verify that the monitoring HelmRelease is healthy (kubectl get hr -n cozy-monitoring monitoring). If the resource is missing, reinstalling or reconciling the relevant HelmRelease may help. If the resource is present, review recent changes to resource templates or CRDs, as a regression may have affected resource lookup (details).If you have access to backend or dashboard logs, check for more detailed error messages about missing resources or failed API calls. This will help pinpoint whether the issue is with the dashboard, API server, or cluster resource configuration.
To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
yviadzishchau commentedon Oct 26, 2025
Hi
# (@dosu)
$ kubectl get monitoring -n tenant-root
NAME READY AGE VERSION
monitoring True 233d 0.37.2
$ kubectl get hr -n tenant-root monitoring
NAME AGE READY STATUS
monitoring 233d True Helm upgrade succeeded for release tenant-root/monitoring.v80 with chart [email protected]
dosubot commentedon Oct 26, 2025
Thanks for sharing the kubectl output—both the monitoring resource and its HelmRelease are present and healthy, so the dashboard error isn't caused by missing resources. Based on recent code and issue analysis, this error usually happens when the dashboard/backend can't match the monitoring resource to its underlying HelmRelease due to strict criteria like chart name, sourceRef, or a naming prefix mismatch—even if the resource exists and is ready in the cluster. This kind of regression has appeared after changes to resource templates or backend logic, especially in recent releases (details).
If this persists across clusters, it's likely a bug or config drift in the dashboard/backend, not a cluster-specific problem. Reviewing dashboard/backend logs for resource lookup failures or mismatches can help pinpoint the cause. If you recently upgraded or changed templates, reverting to a previous working version or double-checking the chart/sourceRef/prefix config for monitoring may resolve it.
To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
kvaps commentedon Oct 31, 2025
Hey thanks for report, this should be fixed by #1564
this fix will be backported to v0.37
[dashboard] Update openapi-ui v1.0.3 + fixes (#1564)
[dashboard] Update openapi-ui v1.0.3 + fixes (cozystack#1564)