You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2020/11/26 17:39:40 UTC

[GitHub] [spark] warrenzhu25 commented on pull request #29113: [SPARK-32314][SHS] Add config to control whether log old format of stacktrace

warrenzhu25 commented on pull request #29113:
URL: https://github.com/apache/spark/pull/29113#issuecomment-734420927


   > Briefly looking at spark code, this is currently still in use - looks like that comment in `TaskEndReason` might not be valid anymore.
   > See [here](https://github.com/apache/spark/blob/594c7c613a8ef80ab6b3725832579f12d40b02c8/core/src/main/scala/org/apache/spark/executor/Executor.scala#L694) for example.
   > 
   > There are other uses as well, given that I am not in favor of this change. If it had been a truely dead code, I would have been more inclined to cleanup (with a caveat that this is part of public api, and so we need to be careful with breaking behavior).
   
   Thanks for comment. For this example, it's about how to create ExceptionFailure. This pr is only about when writing into event log, also affecting TaskEndReason after replay event. And there is no api directly use `stackTrace: Array[StackTraceElement]`, this only used in below:
   
   ```
     override def toErrorString: String =
       if (fullStackTrace == null) {
         // fullStackTrace is added in 1.2.0
         // If fullStackTrace is null, use the old error string for backward compatibility
         exceptionString(className, description, stackTrace)
       } else {
         fullStackTrace
       }
   ```


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org