Allow sharded cluster with custom directory/executable provider to be restarted

[tobi-laa]

Problem description

If a sharded cluster is started in a custom directory, stopped and then restarted again, it fails. The underlying reason appears to be that the files specified via the cluster-config-file directive cannot be reused.

Proposed solution

When a RedisInstance is stopped, its redis.conf is parsed and all cluster-config-file directives looked up. All of these cluster config files are resolved against the folder the Redis executable resides in and, if present, are deleted.

[codemonstur]

I like what it does but I don't like the way it does it. Not that there is anything immediately wrong, it is just not my style. I've been wanting to work on this but I just can't seem to find the time. I hope you are not waiting for this to get done soon, I'm swamped.

[tobi-laa]

Thanks for the feedback :) I am not waiting for it to be done soon. We just stumbled over it at work but have a workaround.

If you have any specific ideas as to how you would tackle this bug I could offer to implement another solution to it - once you have the time for it, of course!

Have a nice Sunday