Fix grafana dashboards not working #856
Labels
enhancement
New feature or request
Comments
|
Thank you for reporting us your feedback! The internal ticket has been created: https://warthogs.atlassian.net/browse/KF-5488.
|
This was referenced Mar 27, 2024
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 4, 2024
* Rename relation from `grafana-dashboards` to `grafana-dashboard` in order to: * follow what every other CKF charm does. * not diverge from the grafana library default. Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/notebook-operators
that referenced
this issue
Apr 5, 2024
* Remove `uid` from the `datasource` fields. * fix typo * Add tag `ckf` to dashboard. * Increase retry_for_attempts to 10 attempts * unpin prometheus Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/argo-operators
that referenced
this issue
Apr 5, 2024
* Add `ckf` tag to argo-controller's grafana dashboard.
* Fix dashboard panels not working by:
* Replacing unavailable metrics with available ones
* Change 2 minutes instead of 1 in places where rate() is used since
this requires more than one scrape data points.
* Remove rate() from panel that shows total number of log messages.
Part of canonical/bundle-kubeflow#856
Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 5, 2024
* Add `ckf` tag to the grafana dashboard.
* Fix dashboard panels not working by:
* Replacing unavailable metrics with available ones
* Adding 2 minutes instead of 1 in places where rate() is used since
this requires more than one scrape data points.
* Remove rate() from panels that shows percentages.
* Remove labels where the metrics don't provide them.
Part of canonical/bundle-kubeflow#856
Refs canonical/bundle-kubeflow#834
Closes #73
orfeas-k
added a commit
to canonical/seldon-core-operator
that referenced
this issue
Apr 5, 2024
* Fix grafana dashboard by removing `uid` from the `datasource` fields. * Add tags `ckf` and `seldon` to dashboard. Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/minio-operator
that referenced
this issue
Apr 8, 2024
Fix the dashboard: * Remove `job=scrape_jobs` label from dashboard fields * Remove panels using metrics that are not provided by the minio workload. * Reorder panels for filling gaps created by removed panels. On top of that, it adds a `ckf` tag to the dashboard. Ref canonical/bundle-kubeflow#856
orfeas-k
added a commit
to canonical/katib-operators
that referenced
this issue
Apr 8, 2024
* Add `ckf` and `katib` tag to katib-controller's grafana dashboard.
* Fix `datasource` field
* Fix current experiments & trials panel by:
* adding a `legendFormat` field
* converting it to a time series & mapping no value to 0 in order to account
for the fact that katib-controller doesn't output any `current` metrics when
there is no experiment or trial running.
Ref canonical/bundle-kubeflow#856
Ref canonical/bundle-kubeflow#834
|
Fixed in referenced PRs |
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 9, 2024
* Rename relation from `grafana-dashboards` to `grafana-dashboard` in order to: * follow what every other CKF charm does. * not diverge from the grafana library default. Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 9, 2024
* Add `ckf` tag to the grafana dashboard.
* Fix dashboard panels not working by:
* Replacing unavailable metrics with available ones
* Adding 2 minutes instead of 1 in places where rate() is used since
this requires more than one scrape data points.
* Remove rate() from panels that shows percentages.
* Remove labels where the metrics don't provide them.
Part of canonical/bundle-kubeflow#856
Refs canonical/bundle-kubeflow#834
Closes #73
orfeas-k
added a commit
to canonical/notebook-operators
that referenced
this issue
Apr 9, 2024
* Remove `uid` from the `datasource` fields. * fix typo * Add tag `ckf` to dashboard. * Increase retry_for_attempts to 10 attempts * unpin prometheus Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/argo-operators
that referenced
this issue
Apr 9, 2024
* Add `ckf` tag to argo-controller's grafana dashboard.
* Fix dashboard panels not working by:
* Replacing unavailable metrics with available ones
* Change 2 minutes instead of 1 in places where rate() is used since
this requires more than one scrape data points.
* Remove rate() from panel that shows total number of log messages.
Part of canonical/bundle-kubeflow#856
Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/seldon-core-operator
that referenced
this issue
Apr 9, 2024
* Fix grafana dashboard by removing `uid` from the `datasource` fields. * Add tags `ckf` and `seldon` to dashboard. Part of canonical/bundle-kubeflow#856 Refs canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/minio-operator
that referenced
this issue
Apr 9, 2024
Fix the dashboard: * Remove `job=scrape_jobs` label from dashboard fields * Remove panels using metrics that are not provided by the minio workload. * Reorder panels for filling gaps created by removed panels. On top of that, it adds a `ckf` tag to the dashboard. Ref canonical/bundle-kubeflow#856
orfeas-k
added a commit
to canonical/katib-operators
that referenced
this issue
Apr 9, 2024
* Add `ckf` and `katib` tag to katib-controller's grafana dashboard.
* Fix `datasource` field
* Fix current experiments & trials panel by:
* adding a `legendFormat` field
* converting it to a time series & mapping no value to 0 in order to account
for the fact that katib-controller doesn't output any `current` metrics when
there is no experiment or trial running.
Ref canonical/bundle-kubeflow#856
Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 9, 2024
* Add `ckf` tag to the grafana dashboard. * Fix dashboard panels not working by: * Replacing unavailable metrics with available ones * Adding 2 minutes instead of 1 in places where rate() is used since this requires more than one scrape data points. * Remove rate() from panels that shows percentages. * Remove labels where the metrics don't provide them. Part of canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834 Ref #73
orfeas-k
added a commit
to canonical/envoy-operator
that referenced
this issue
Apr 9, 2024
* Rename relation from `grafana-dashboards` to `grafana-dashboard` in order to: * follow what every other CKF charm does. * not diverge from the grafana library default. Ref canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/minio-operator
that referenced
this issue
Apr 9, 2024
Fix the dashboard: * Remove `job=scrape_jobs` label from dashboard fields * Remove panels using metrics that are not provided by the minio workload. * Reorder panels for filling gaps created by removed panels. On top of that, it adds a `ckf` tag to the dashboard. Ref canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/argo-operators
that referenced
this issue
Apr 9, 2024
* Add `ckf` tag to argo-controller's grafana dashboard.
* Fix dashboard panels not working by:
* Replacing unavailable metrics with available ones
* Change 2 minutes instead of 1 in places where rate() is used since
this requires more than one scrape data points.
* Remove rate() from panel that shows total number of log messages.
Ref canonical/bundle-kubeflow#856
Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/seldon-core-operator
that referenced
this issue
Apr 9, 2024
* Fix grafana dashboard by removing `uid` from the `datasource` fields. * Add tags `ckf` and `seldon` to dashboard. Ref canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/notebook-operators
that referenced
this issue
Apr 9, 2024
* Remove `uid` from the `datasource` fields. * fix typo * Add tag `ckf` to dashboard. * Increase retry_for_attempts to 10 attempts * unpin prometheus Part of canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834
orfeas-k
added a commit
to canonical/katib-operators
that referenced
this issue
Apr 9, 2024
…from #172 (#173) * Add `ckf` and `katib` tag to katib-controller's grafana dashboard. * Fix `datasource` field * Fix current experiments & trials panel by: * adding a `legendFormat` field * converting it to a time series & mapping no value to 0 in order to account for the fact that katib-controller doesn't output any `current` metrics when there is no experiment or trial running. Ref canonical/bundle-kubeflow#856 Ref canonical/bundle-kubeflow#834
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Context
Among charms that have a grafana dashboard (see #834), fix the dashboards that are not working at the moment.
What needs to get done
Fix grafana dashboards not working
Definition of Done
All dashboards provided are working
The text was updated successfully, but these errors were encountered: