When prometheus is not installed (e.g. on clear crc cluster), namespaces page crashes with namespace.jsx:133 Error: Service Unavailable at validateStatus (co-fetch.js:44) at co-fetch.js:112 this is mainly painful when running e2e-tests. Expected results: k8s/cluster/namespaces does not crash without prometheus in dev
Hi Filip, I tried to verify this bug with disabling Prometheus after OCP cluster is up. 1. Disable Prometheus pods # cat <<EOF >version-patch-first-override.yaml > - op: add > path: /spec/overrides > value: > - kind: Deployment > group: apps/v1 > name: cluster-monitoring-operator > namespace: openshift-monitoring > unmanaged: true > EOF # oc patch clusterversion version --type json -p "$(cat version-patch-first-override.yaml)" # oc scale deployment cluster-monitoring-operator --replicas=0 -n openshift-monitoring deployment.apps/cluster-monitoring-operator scaled # oc scale deployment grafana --replicas=0 -n openshift-monitoring deployment.apps/grafana scaled # oc scale deployment kube-state-metrics --replicas=0 -n openshift-monitoring deployment.apps/kube-state-metrics scaled # oc scale deployment openshift-state-metrics --replicas=0 -n openshift-monitoring deployment.apps/openshift-state-metrics scaled # oc scale deployment prometheus-adapter --replicas=0 -n openshift-monitoring deployment.apps/prometheus-adapter scaled # oc scale deployment prometheus-operator --replicas=0 -n openshift-monitoring deployment.apps/prometheus-operator scaled # oc scale deployment telemeter-client --replicas=0 -n openshift-monitoring deployment.apps/telemeter-client scaled # oc scale deployment thanos-querier --replicas=0 -n openshift-monitoring deployment.apps/thanos-querier scaled # oc scale statefulset alertmanager-main --replicas=0 -n openshift-monitoring statefulset.apps/alertmanager-main scaled # oc scale statefulset prometheus-k8s --replicas=0 -n openshift-monitoring statefulset.apps/prometheus-k8s scaled 2. Visit k8s/cluster/namespaces URL, I can see js errors in browser console, but the page did not crash Unable to fetch pod metrics Error: Bad Gateway u main-chunk-6b551ed9b809b1b3d9c7.min.js:1 s main-chunk-6b551ed9b809b1b3d9c7.min.js:1 main-chunk-6b551ed9b809b1b3d9c7.min.js:1 GEThttps://console-openshift-console.xxxx.openshift.com/api/prometheus/api/v1/rules [HTTP/1.1 502 Bad Gateway 30332ms] XHRGEThttps://console-openshift-console.xxxx.openshift.com/api/prometheus/api/v1/query?&query=sum%20by(namespace)%20(container_memory_working_set_bytes{container=%22%22,pod!=%22%22}) [HTTP/1.1 504 Gateway Time-out 30332ms] XHRGEThttps://console-openshift-console.xxxx.openshift.com/api/prometheus/api/v1/query?&query=namespace:container_cpu_usage:sum [HTTP/1.1 502 Bad Gateway 30333ms]
Do you think I should also verify in `dev` mode since I see you said `k8s/cluster/namespaces does not crash without prometheus in dev`
It was crashing only in development mode when using webpack dev server (https://github.com/openshift/console#frontend-development) - which has a more strict of handling errors. Anyway the errors should still be visible in the browser's console.
1. Follow the steps in comment 2 to disable cluster monitoring 2. Then follow the steps in https://github.com/openshift/console#frontend-development to run in a dev mode 3. Visit http://localhost:9000/k8s/cluster/namespaces && http://localhost:9000/k8s/cluster/namespaces/<one_namespace> page, console doesn’t crash and errors are shown in browser’s console Verified on 4.7.0-0.nightly-2020-10-22-175439
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.7.0 security, bug fix, and enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2020:5633