Github user rekhajoshm commented on the pull request:
https://github.com/apache/spark/pull/7602#issuecomment-126524343
Thanks @srowen for your comment.The recursive calls for cluster child and
nodes within makeDotSubgraph internally appends the result of each call and has
normal recursion stack flow.makeDotSubgraph has to return value as main method
makeDotFile needs it to form the final dot file.There is no additional
StringBuilder created by the patch which is not already getting created in
original.indent is set to " " in original as well.
What I was also thinking is there is possibility of stack overflow when
either there is possible lurking infinite recursion and/or if processing goes
beyond recursion depth workable on cluster. But as stack overflow can mask
itself as OOM, I was also thinking of increasing âXss in pom.xml. Or atleast
catch OOM error in SparkException and have message making this suggestion to
user to configure -Xss. Hence updated patch.
This patch at the very least, makes sure objects are less,and âXss
suggestion which can help in any case. Given that, please review, approve
@srowen @JoshRosen @andrewor14 thanks!
---
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]
