Mention POLLING_INTERVAL variable in documentation

[fdaligand]

Summary

I learned how to quickly start 1Password Connect and the Operator from your video link. However, restarting a pod due to a secret change in 1Password takes 10 minutes (600 seconds) by default.

Use cases

When following the steps in the video, you'll notice that pods, which embed secrets generated by the Kubernetes Operator, are restarted when their associated secrets change in 1Password. Nevertheless, this process is nearly instant in the video but takes 10 minutes with the default configuration. As the POLLING_INTERVAL parameter is not mentioned in either the documentation or the video, it can lead to prolonged debugging times, searching for why it is not working.

Proposed solution

Suggest adding information in the documentation at Configuring Automatic Rolling Restarts of Deployments that restarts will occur based on the POLLING_INTERVAL.

Is there a workaround to accomplish this today?

I personally use the Helm option -set operator.pollingInterval=60, and it works fine. However, it's worth noting that, due to this setting, the local experience doesn't align with what is shown in your video.

[yalattas]

Agree, I spent like 10 minutes trying to figure out why my pods doesn't restart and secrets was never updated till I killed the operator pod and restarted.

Then I found this pollingInternal and put it 10 seconds for testing. Almost instantly secret was update. Eventually, I set it to 30 seconds