Possible improvement for Watcher.Dispose()

[mikhail-barg]

Hi!

I think that current implementation of Watcher.Dispose() is a bit lacking:

csharp/src/KubernetesClient/Watcher.cs

Lines 217 to 229 in 541abb0

	protected virtual void Dispose(bool disposing)
	{
	if (!disposedValue)
	{
	if (disposing)
	{
	_cts?.Cancel();
	_cts?.Dispose();
	}
	disposedValue = true;
	}
	}

specifically between _cts?.Cancel() and _cts?.Dispose() calls, we should await for the _watcherLoop task to finish. Otherwise it might attempt to access the Token of already disposed _cts, which is generally undesirable.

Also possible exception would be unobserved, which in turn may be a problem for applications with strict policies on unobserved exceptions.

What do you think? Will you consider a PR like this:

protected virtual void Dispose(bool disposing)
{
    if (!disposedValue)
    {
        if (disposing)
        {
            _cts.Cancel();
            try 
            {
                _watcherLoop.GetAwaiter().GetResult();
            }
            catch 
            {
                //consume possible exception
            }
            _cts.Dispose();
        }

        disposedValue = true;
    }
}

[tg123]

i don't think waiting inside dispose is a good idea

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale