Github user mridulm commented on the pull request:
https://github.com/apache/spark/pull/8760#issuecomment-165304077
@squito Unless an executor or node is dying/has resource issues, in most
cases, the failure is not repeatable across tasksets due to differing
execution/resource characterstics. So blanket blacklist can be extremely
suboptimal.
For example: taskset requiring disk usage (shuffle) vs taskset for
executing collect() - the latter is not impacted by disk space being full/being
reclaimed in background while former will keep failing.
An additional state might most probably handle the requirement - will be
able to comment more when I actually see the design ofcourse :-)
---
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]
