Etienne Champetier

Also happening on Alma 9.3 / cockpit-300.3-1.el9_3.x86_64

I don't think the CI failure is related to this change

Rebased on master

If we wait for goroutines to finish we really need to ensure that they exit fast, else k8s will just kill the pod and we don't gain anything. Waiting for...

Hello @johananl @rata Do we really want this for 0.10.0 ? There is a good backlog of PR that need merging that provide way more value than cleaning on shutdown....

@Matoch Making MetalLB easier to troubleshoot, even if it's not directly an issue in MetalLB is always reasonable, please send a PR

If you think this is risky, you should run it more often, not less, but on your schedule, so you find out what fails and fix it. Given the lack...

NRI doesn't allow to edit `logPath`, to copy container logs to journald, is there a simpler way than having a runc-wrapper that edit the json, and building a modified version...

The problem here is that you/the current code assumes lost connection is a normal shutdown. For clean shutdown, containerd (nri adaptation) should do an rpc call to ask the plugin...

The big limitation of using syslog is that panic text is not logged, or at least need some more work (in 1.23 we will have SetCrashOutput)