Github user vanzin commented on the issue:
https://github.com/apache/spark/pull/14931
Hmm... as far as I understand, `SlaveLost` does not mean "remote machine
has died", it means "remote executor has died", which does not mean that the
shuffle files were lost. Unless you actually try to connect to the shuffle
service and verify whether it's alive, you might be throwing away work.
It's possible that "machine has died" might be the most common cause of
"SlaveLost", but I'm a little wary of treating both as the same.
@squito has been looking at node blacklisting and might have better ideas
here than I do.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
