ws-manager: Fix consistency of disposed ws between ws-daemon and ws-manager

[utam0k]

Description

Due to a retry timing issue, there were cases where the workspace had already been successfully backed up on the ws-daemon side, but ws-manager was trying to back it up again with an error.

More detail

Related Issue(s)

Fixes #11710

How to test

Is there any good way to do this?

Release Notes

Fixed a rare case in which prebuild was terminating successfully but failing

Documentation

Werft options:

• /werft with-preview

[jenting]

Can we move these lines upper?

gitpod/components/ws-manager/pkg/manager/monitor.go

Lines 1115 to 1126 in b8669f3

	ctx, cancelReq := context.WithTimeout(ctx, time.Duration(m.manager.Config.Timeouts.ContentFinalization))
	m.finalizerMap.Store(workspaceID, cancelReq)
	defer func() {
	// we're done disposing - remove from the finalizerMap
	val, ok := m.finalizerMap.LoadAndDelete(workspaceID)
	if !ok {
	return
	}
	cancelReq := val.(context.CancelFunc)
	cancelReq()
	}()

Therefore, we can reuse the defer function.

[utam0k]

@jenting Thanks for your feedback. fixed

[utam0k]

/werft with-preview

👎 unknown command: with-preview
Use /werft help to list the available commands

[werft-gitpod-dev-com]

started the job as gitpod-build-to-retry-prebuild.3 because the annotations in the pull request description changed
(with .werft/ from main)