Github user andrewor14 commented on the pull request:
https://github.com/apache/spark/pull/2432#issuecomment-57555822
@sarutak I left a range of comments. The biggest things are: (1) I don't
see why we need to pass in the base event logging directory to the task
scheduler and scheduler backend. Reverting this will simplify a lot of the
changes. (2) It would be good to add return types and java docs for many of the
new methods you introduce in this patch. It may seem like a hassle, but in my
experience it really helps newcomers to understand the code. (3) I don't
understand why `MetricsSystem` needs to read in an identifier that is either
`spark.app.name`, or `spark.app.id`, or `spark.executor.id` from the conf.
Can't it accept a unique identifier from the person who instantiates it?
Accepting an identifier that could be either of the three configs may become a
source of pain point in the future.
---
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]
