Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/1387#issuecomment-54382137
Sorry to revive this again, but I am still confused as to what this PR
achieves after reading through the conversation. When an executor JVM dies
because of an OOM, the driver will attempt to trigger a GC and trigger cleaning
tasks immediately. However, the executor that OOM'ed already died, and will
probably not be able to respond to cleaning requests from the driver.
Is the purpose of this to prevent other executors that haven't OOM'ed yet
from dying as well? For instance, say I have 5 executors, and 1 of them runs
OOM first. Is the purpose of this to prevent the other 4 from dying of the same
cause preemptively?
---
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]
