The reflector watcher stopped silently

[bzhangyan11]

Describe the bug

My application cannot receive any events of my cr from the fabric8 informer occasionally.
I analyzed logs and found there was no more logs about my cr after reconnecting failed.

Then i analyzed the instances of reflector class(io.fabric8.kubernetes.client.informers.impl.cache) in memory. I found the watching field of the reflector is false but stopFuture is not completed.

It seems to there is a bug in reconnect scene.



If an exception occurs in processList, there is no logs will be recorded and the watcher will stop silently

Fabric8 Kubernetes Client version

6.13.4

Steps to reproduce

i cannot reproduce

Expected behavior

it should retry after reconnecting failed.

Runtime

Kubernetes (vanilla)

Kubernetes API Server version

other (please specify in additional context)

Environment

Linux

Fabric8 Kubernetes Client Logs

No response

Additional context

No response

[shawkins]

It's not clear from what you have posted that there is a bug. The watching boolean can be false during the normal operation of the relfector. As long as you see an active reconnect future along with that, then the reflector is still trying to re-establish the watch.

[bzhangyan11]

It's not clear from what you have posted that there is a bug. The watching boolean can be false during the normal operation of the relfector. As long as you see an active reconnect future along with that, then the reflector is still trying to re-establish the watch.

Ok, but my application indeed fails to receive any events before it's restarted, i will keep watching.
thx