[biday prometheus] feat: remove Prometheus dependency, collect all metrics via nodemon.#376

Open

Parthiba-Hazra wants to merge 14 commits into

mainfrom

ph/cut-the-prom-dep

Commits on May 5, 2026

feat: remove Prometheus dependency, collect all metrics via nodemon DaemonSet
Parthiba-Hazra
committed
ci: remove Prometheus and metrics-server from CI workflows
Parthiba-Hazra
committed
ci: add nodemon build and deploy to k8s-compatibility-test
Parthiba-Hazra
committed

Commits on May 6, 2026

fix: populate GPU metrics from nodemon unified cache
Parthiba-Hazra
committed
fix: use node-level CPU/memory from kubelet stats/summary instead of container aggregation
Parthiba-Hazra
committed
fix: use strings.SplitN instead of fmt.Sscanf for slash-delimited cache key parsing
Parthiba-Hazra
committed
fix: pass parent context to historical percentile cache to prevent goroutine leak
Parthiba-Hazra
committed
fix: exclude Secret from installer_updater.yaml to prevent token overwrite on update
Parthiba-Hazra
committed
fix: template namespace in backend-install and installer_updater for DAKR backend
Parthiba-Hazra
committed
feat: add one-time Prometheus cleanup migration job to installer
Parthiba-Hazra
committed
feat(helm): add Prometheus cleanup as post-install/post-upgrade hook
Parthiba-Hazra
committed
feat: add standalone nodemon cleanup to migration job
Parthiba-Hazra
committed
fix: remove standalone nodemon cleanup from kubectl migration job
Parthiba-Hazra
committed

Commits on May 7, 2026

fix: read GPU request/limit counts from pod spec for dashboard display
Parthiba-Hazra
committed