Github user uce commented on the pull request:
https://github.com/apache/flink/pull/1153#issuecomment-146297139
I've addressed your inline comments (see commits). Not rebased yet. I think
it's easier to review this way.
The major points:
- Followed all renaming suggestions.
- Removed the checkpoint ID counter and replaced with .
- Left the submission to fail in case of concurrent modifications to have
clear error messages. The behaviour in such cases is rather undefined at the
moment.
- Your description about the terminally failing behaviour is correct. The
behaviour was on purpose for the initial version. The root zNode essentially
identifies the cluster at the moment. We can fix this is to use a per cluster
random root zNode, e.g. `/flink/random/...` instead of `/flink/...` What do you
think? In any case, I would do this as a quick follow up.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
