should update pediacluster status when conditions or SyncResources changed rather than timed execution

[duanmengkk]

These logs make me confused.At first I thought the cluster status was constantly changing(sync resources Or cluster healthy status),and I research the code and find the status of pediacluster is updated every 5s.See code https://github.com/clusterpedia-io/clusterpedia/blob/b06cc6d0f6f06af50269a9533202211552c63252/pkg/synchromanager/clustersynchro/cluster_monitor.go#L36

Do you need to optimize it? @Iceber

[Iceber]

What is the version of clusterpedia? Although it is checked once every 5 seconds, the status update log is not printed if the status has not changed.

Can you give me a more complete log, the picture has cut off part of the log.

[duanmengkk]

The latest code in main branch.

[duanmengkk]

The lastTransitionTime between latest status and updated status is diffrent

[Iceber]

I'm running clustersynchro manager with the main branch

I0905 18:01:22.612425    8227 clustersynchro_manager.go:498] "Update Cluster Status" cluster="cluster-example" conditions=[{Type:Validated Status:True ObservedGeneration:0 LastTransitionTime:2022-08-02 12:46:34 +0800 CST Reason:Validated Message:} {Type:SynchroRunning Status:True ObservedGeneration:0 LastTransitionTime:2022-09-05 18:01:14 +0800 CST Reason:Running Message:cluster synchro is running} {Type:ClusterHealthy Status:True ObservedGeneration:0 LastTransitionTime:2022-09-05 18:01:15 +0800 CST Reason:Healthy Message:cluster health responded with ok} {Type:Ready Status:True ObservedGeneration:0 LastTransitionTime:2022-09-05 18:01:15 +0800 CST Reason:Ready Message:}]



Now: 09-05 18:03:18

I did not find frequent status updates.

Can you print the version information?

# replace the pod name of clusterpedia clustersynchro-manager
$ kubectl -n clusterpedia-system exec {clustersynchro-manager-pod-name} -- clustersynchro-manager --version=raw

In a previous commit, it did trigger frequent updates, but this has been fixed.

[duanmengkk]

I run it in my local env. I guess the status in ResourceSynchro and status in pediaclusterListener may have some time delay. The informer doesn't handle event in time,so the status get from pediaclusterListener is always delayed by few seconds.

It is just my guess,not sure.

[Iceber]

The infromer delay may result in one or two additional status updates, but not a continuous status update

Do other environments have this problem?