kubeovn / kube-ovn / 24435222599 / 1
25%
master: 25%

DEFAULT BRANCH: master
Ran
Files 206
Run time 5s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 24.553%. Remained the same
24435222599.1

push

github

web-flow 
feat(chart): ship PrometheusRule and Grafana dashboards in kube-ovn-v2 (#6630)

* feat(chart): ship PrometheusRule and Grafana dashboards in kube-ovn-v2

- Add an opt-in PrometheusRule (default disabled) covering controller
  liveness, subnet IP exhaustion, pinger OVS/OVN/apiserver/DNS health
  and OVSDB slow requests. Both up==0 and absent() rules are bundled
  for the controller so scrape failures and full target loss are both
  caught without firing false positives the moment the feature is
  enabled.
- Add an opt-in Grafana dashboards feature that publishes one ConfigMap
  per file under dist/monitoring with the kube-prometheus-stack sidecar
  label. Dashboards are referenced via per-file symlinks so the source
  of truth in dist/monitoring stays single. The chart rewrites the
  legacy \${DS_PROMETHEUS} placeholder to a configurable datasource
  name at render time, otherwise sidecar-loaded panels would point at
  a non-existent datasource.
- Fix the long-standing ovs_vsersion typo in dist/monitoring/ovs-grafana.json
  so the OVS Version stat panel renders an actual version string.
- Regenerate charts/kube-ovn-v2/README.md via helm-docs.

Both feature gates default to false, so clusters without the
prometheus-operator CRDs or the Grafana sidecar are unaffected.

Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Mengxin Liu <liumengxinfly@gmail.com>

* feat(chart): address Copilot review feedback on monitoring bundle

- Clarify the prometheusRule prerequisite note: scrape-dependent alerts
  stay inactive without scrape, but absent()-style alerts (e.g.
  KubeOvnControllerAbsent) still fire by design. Point users at
  Alertmanager silences/inhibitions if they want to suppress those
  until scraping is wired up.
- Rewrite KubeOvnControllerAbsent's self-recovery guidance to recommend
  Alertmanager silences instead of prometheusRule.additionalGroups,
  which can only... (continued)

13831 of 56331 relevant lines covered (24.55%)

0.29 hits per line

Source Files on job 24435222599.1