Neuvector updater pod sidecar never exits

[corang]

Steps to reproduce

1. Have neuvector running for any extended period of time

Expected result

Updater pod sidecars exit along with the main job

Actual Result

The main container exits and the sidecar runs forever

Visual Proof (screenshots, videos, text, etc)

Just manually cleared them I can grab a screenshot when it happens again

Severity/Priority

Low/med I think its mostly visual but I wonder if it affects neuvector's functionality

[corang]

As a workaround I've been portforwarding port 15000 from the sidecar and running curl -X POST http://localhost:15000/quitquitquit but this is obviously manual and shouldn't be necessary

[mjnagel]

This is likely related to the pepr behavior you've seen with watches - pepr should be doing the sidecar termination (and has successfully in some long running environments from what I've seen).

Good to track this issue, just want to call out that its likely coupled with watch problems - I would check if cycling the pepr watcher picks up those extra pods and terminates them.

[mjnagel]

Linking defenseunicorns/pepr#745

[mjnagel]

Should be resolved by #419 - leaving open until that is released and we have feedback though.

[mjnagel]

Tentatively closing - please reopen or create a new issue if you encounter this problem in latest core versions.