dodjdnh opened a new pull request, #1573: URL: https://github.com/apache/dubbo-admin/pull/1573
## Summary Complete the provisioning connections among the existing Kubernetes Observability resources: Grafana automatically discovers the six Dubbo dashboards, a Jaeger datasource is provisioned, and Kubernetes Admin is configured with the six corresponding dashboard URLs. This change depends on the Grafana Service DNS fix in [D002 PR #1571](https://github.com/apache/dubbo-admin/pull/1571). That PR has now been merged; D003 has been rebased onto its merged `develop` commit, so this PR contains only the three D003 files below. ## Changes - `release/kubernetes/monitoring/grafana-deployment.yaml`: reuse the existing dashboard mount pattern, adding a volume and volume mount for each of the six Dubbo ConfigMaps under `/grafana-dashboard-definitions/0/<dashboard name>`. Dashboard JSON, UIDs, variables, and queries are unchanged. - `release/kubernetes/monitoring/grafana-dashboardDatasources.yaml`: retain the existing Prometheus configuration and add `name: jaeger`, `uid: fe670vj54q48wc`, `type: jaeger`, `access: proxy`, and `url: http://jaeger.monitoring.svc:16686`. The UID is the datasource value currently saved in the three trace dashboards. Reusing it is a compatibility choice, not restoration of a previously existing formal datasource configuration. - `release/kubernetes/dubbo-system/dubbo-admin.yaml`: add the six `console.metricDashboards` / `console.traceDashboards` URLs using `http://grafana.monitoring.svc:3000/d/<corresponding dashboard UID>?kiosk=1&theme=light`. Preserve the DNS corrected by D002 and the existing ConfigMap → volume → mount → `-c` loading path. ## Validation The same external static Provisioning Oracle, with unchanged decision logic, reported: ```text Unfixed baseline: A1–A6 FAIL; B1 PASS, B2–B4 FAIL; C1–C6 FAIL; OVERALL FAIL D003 changes: A1–A6 PASS; B1–B4 PASS; C1–C6 PASS; OVERALL PASS ``` YAML parsing of the three changed files, `kubectl create --dry-run=client --validate=false`, and `git diff --check` passed. A delegated Open Code Review of the three changed files found no clear blocking issue. **Runtime provisioning has not been verified and is not claimed as PASS.** No confirmed safe disposable Kubernetes environment was available for that validation. In such an environment, it remains necessary to verify that Grafana loads the six dashboards and the Jaeger datasource automatically, the trace dashboards resolve the datasource variable on first load, and the running Admin reads all six URLs. A static Oracle PASS does not replace these checks. ## Dependency D002 PR #1571 has been squash-merged into `apache/dubbo-admin:develop`. D003 was rebased onto that merged commit. The PR diff contains only the three D003 files listed above and does not repeat the D002 changes. ## Related issue Fixes #1572 ## Non-goals This PR does not include the D002 baseline deployment fixes themselves, Service or Instance Trace query semantics, the OTel Collector, `k8sattributes`, dashboard JSON rewrites, Log, Trace → Log / Log → Trace / Trace → Metric, or a claim that loading a dashboard proves business-level E2E queries pass. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
