🐛 Event watches picking up outdated events when installing twice into the same cluster

[TylerGillson]

Description

All instances of watcher, err := toolsWatch.NewRetryWatcher("1", &cache.ListWatch{WatchFunc: watchFunc}) should begin watching from the current resource version of the relevant k8s object, not 1, otherwise stale events are processed.

Reproduction steps

1. Deploy a kind cluster
2. Run validator install
3. Delete plugin deployments
4. Attempt reinstallation

Environment

validator and/or plugin versions:
kubectl version:
kind version:
etc...

Screenshots

Logs

No response

[dosubot]

Hi, @TylerGillson. I'm helping the validatorctl team manage their backlog and am marking this issue as stale.

You reported a bug regarding event watchers in a Kubernetes cluster that incorrectly process outdated events when a plugin is installed multiple times. The issue arises from the watchers being initialized with a resource version of 1 instead of the current version, leading to stale event handling. You provided steps to reproduce the issue and requested additional environment details for further investigation.

Could you please let us know if this issue is still relevant to the latest version of the validatorctl repository? If it is, feel free to comment here to keep the issue open. Otherwise, you can close it yourself, or it will be automatically closed in 7 days. Thank you!