ws-daemon: ignore error when workspace persistent state location not exist

[jenting]

Description

The initial workspace failed at the beginning, so the workspace's persistent state location never exist.

Related Issue(s)

Fixes #11715

How to test

Not sure, but we did see some nodes path /var/gitpod/workspaces with the folders

• <instance-id>
• <instance-id>-daemon

but without the corresponding <instance-id>.workspace.json.

Release Notes

None

Documentation

None

Werft options:

• /werft with-preview

[werft-gitpod-dev-com]

started the job as gitpod-build-jenting-11715.1 because the annotations in the pull request description changed
(with .werft/ from main)

[jenting]

I'm not sure, should we keep running the following code? or directly return here?

gitpod/components/ws-daemon/pkg/internal/session/workspace.go

Lines 210 to 232 in 5fb85e4

	s.stateLock.Lock()
	s.state = WorkspaceDisposed
	s.operatingCondition.Broadcast()
	s.stateLock.Unlock()
	err = s.store.runLifecycleHooks(ctx, s)
	if err != nil {
	return err
	}
	if s.PersistentVolumeClaim {
	// nothing to dispose as files are on persistent volume claim
	return nil
	}
	if !s.FullWorkspaceBackup {
	err = os.RemoveAll(s.Location)
	}
	if err != nil {
	return xerrors.Errorf("cannot remove workspace all: %w", err)
	}
	return nil

[sagor999]

I think this is a good fix. LGTM.