Github user davies commented on the pull request:
https://github.com/apache/spark/pull/10045#issuecomment-160777345
Good point, we may also want to fail fast in local cluster (having only one
executor). Ideally, the tasks should be retried on other failure domains (other
executors) first, then back to previous one. The difficult part is that the
list of executors is dynamic. To simplify this, we can assume that the list of
executors is static when schedule a task (or retry). So your suggestion is
better (the timeout behavior depends on the number of available executors).
---
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]
