OKD 4.22.0-okd-scos.3 Web GUI Diagram odd behaviour

[ensar-bht]

Describe the bug

After I did an update of my OKD cluster from 4.21.0-okd-scos.11 to 4.22.0-okd-scos.3, I noticed main statistics diagram in "Cluster utilization" section of main "Overview" tab is moving from left to right and no live stream updated statistics shows up. In some time, it disappeares totaly without any statistics slowly moving from left to right. If I do web page refresh, the statistics show up again (refreshes) and still disappeares in time moving from left to right without any update.

All pods in monitoring stack are running without any visible issue:

oc get pods -n openshift-monitoring

NAME READY STATUS RESTARTS AGE
alertmanager-main-0 6/6 Running 0 4d5h
alertmanager-main-1 6/6 Running 0 4d5h
cluster-monitoring-operator-8685c5476f-pfgp6 1/1 Running 0 4d4h
kube-state-metrics-5c7f94db6f-b8fpr 3/3 Running 0 4d5h
metrics-server-868d88b488-gpclw 1/1 Running 0 30h
metrics-server-868d88b488-nzljp 1/1 Running 0 30h
monitoring-plugin-5994fcd69f-jgtr8 1/1 Running 0 4d5h
monitoring-plugin-5994fcd69f-xlmtf 1/1 Running 0 4d5h
node-exporter-9df8r 2/2 Running 2 4d5h
node-exporter-cx87g 2/2 Running 2 4d5h
node-exporter-gmq6h 2/2 Running 2 4d5h
node-exporter-j5p2f 2/2 Running 2 4d5h
node-exporter-k7xkn 2/2 Running 2 4d5h
node-exporter-zszjb 2/2 Running 2 4d5h
openshift-state-metrics-57cc8cc95-l5vgt 3/3 Running 0 4d5h
prometheus-k8s-0 6/6 Running 0 29h
prometheus-k8s-1 6/6 Running 0 29h
prometheus-operator-56c65fc8bf-pd8z9 2/2 Running 0 4d4h
prometheus-operator-admission-webhook-bb7ff6fd9-rnm2s 1/1 Running 0 4d5h
prometheus-operator-admission-webhook-bb7ff6fd9-vzxqx 1/1 Running 0 4d5h
thanos-querier-6d7d4d8c4f-cbdj6 6/6 Running 0 29h
thanos-querier-6d7d4d8c4f-p2sn9 6/6 Running 0 29h

All cluster operators post update are healthy.

Occurs on:

• Home → Overview dashboard

Symptoms:

• graphs not rendering correctly / broken layout

Version

4.22.0-okd-scos.3

Reproducibility

100%

Log Bundle

No response

[ensar-bht]

I am doing the inspection of this behaviour in web browsers, I opened up the consoles in Google Chrome and Edge and the problem is the same. Opening "Inspection" and looking at the console and network I noticed this:

Unsatisfied version 18.3.1 from openshift-console of shared singleton module react (required ~0.14.2)
Unsatisfied version 5.0.1 from openshift-console of shared singleton module redux (required ^4.0.4)
[DEPRECATION WARNING] networking-console-plugin@0.0.1 was built for an older version of Console and may not work correctly in this version.
Unsatisfied version 0.0.0-fixed from openshift-console of shared singleton module @openshift-console/dynamic-plugin-sdk (required ^4.22.0-prerelease.2)

and also this:

Unsatisfied version 16.5.8 ... react-i18next
Unsatisfied version 0.0.0-fixed ... dynamic-plugin-sdk

oc get consoleplugins -A
NAME AGE
monitoring-plugin 588d
networking-console-plugin 270d

I also tried removing the monitoring-plugin but the problem persists.

I suspect that this is the main culprit:

react-i18next mismatch
dynamic-plugin-sdk mismatch

Could it be that frontend bundle (JS) and backend aren't fully in sync?

Additional info:

• Issue persists even when monitoring-plugin Console plugin is disabled
• Console operator reports healthy state (AVAILABLE=True, PROGRESSING=False, DEGRADED=False)
• No failing backend API requests observed – issue appears isolated to frontend module resolution

Conclusion:

This appears to be a frontend bundle inconsistency (Module Federation), where shared dependencies (react, redux, react-i18next, dynamic-plugin-sdk) are not aligned, causing rendering issues in the Overview dashboard.

[ensar-bht]

Hello everyone,

another update, I just did the clulster update from 4.22.0-okd-scos.3 to 4.22.0-okd-scos.5 and I can confirm that the problem is still present.

consumes:70 Unsatisfied version 16.5.8 from openshift-console of shared singleton module react-i18next (required ^11.12.0)
(anonymous) @ consumes:70
consumes:70 Unsatisfied version 0.0.0-fixed from openshift-console of shared singleton module @openshift-console/dynamic-plugin-sdk (required =4.20.0)
Unsatisfied version 5.0.1 from openshift-console of shared singleton module redux (required ^4.0.4)
(anonymous) @ consumes:70
plugin-init.ts:84 [DEPRECATION WARNING] networking-console-plugin@0.0.1 was built for an older version of Console and may not work correctly in this version.
(anonymous) @ plugin-init.ts:84
plugin-entry.dbcbc96819fb6ab00316.min.js?cacheBuster=2df8a890-93a7-4694-abf6-d4fc59243a41:1 Unsatisfied version 0.0.0-fixed from openshift-console of shared singleton module @openshift-console/dynamic-plugin-sdk (required ^4.22.0-prerelease.2)

[ensar-bht]

Problem still present even after the cluster update to 4.22.0-okd-scos.7