Github user tgravescs commented on the issue:
https://github.com/apache/spark/pull/18213
I think this is a very hard thing for us to know, to many different failure
types. I agree that setting to 1 is better then us getting it wrong, although
I question a bit still if that is right either. You are now making it less
robust by default. Yes if it fails and was a user error it retries a couple of
times and wastes some time and resources. If user is watching they should see
that and kill it. That to me is not as bad as not retrying when you should.
personally I would rather see this stay, if you want it changed for your
distribution or cluster config then set it in the confs for your cluster.
If you have specific cases you are seeing this to much perhaps we can look
at those in more detail to handle specially.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
