Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/2524#issuecomment-57074480
It's probably easiest to move the accumulator update to TaskSetManager or
to the part of DAGScheduler that reports the result to the user. It's right
below the current update in the code:
```
if (!job.finished(rt.outputId)) {
job.finished(rt.outputId) = true
...
```
That happens only once per task, so it's a good place to do the update for
ResultTask. For ShuffleMapTask you can do it in the corresponding match
statement as well.
---
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]
