[SPARK-5417] Remove redundant executor-id set() call

[ryan-williams]

This happens inside SparkEnv initialization as of #4194

[SparkQA]

Test build #26132 has started for PR 4213 at commit b3e4f7b.

• This patch merges cleanly.

[SparkQA]

Test build #26132 has finished for PR 4213 at commit b3e4f7b.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • case class Rating(userId: Int, movieId: Int, rating: Float, timestamp: Long)
  • case class Movie(movieId: Int, title: String, genres: Seq[String])
  • case class Params(
  • class ALS extends Estimator[ALSModel] with ALSParams
  • case class RatingBlock(srcIds: Array[Int], dstIds: Array[Int], ratings: Array[Float])
  • protected case class Keyword(str: String)
  • class SqlLexical extends StdLexical

[AmplabJenkins]

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/26132/
Test PASSed.

[andrewor14]

LGTM I've been wanting to do the same merging into master.

[ryan-williams]

thanks @andrewor14, to be clear, afaik the removed line here only became redundant on Sunday with #4194; did you mean you've been wanting to do the same since then? Or has this been redundant for longer, unbeknownst to me?

[andrewor14]

No, only as of #4194. After I merged that I realized we still set it elsewhere redundantly but didn't really get to removing it myself.

[srowen]

Since the commit this depends on was in 1.2 (1f8b718) and this was tagged for back-port into 1.2, after 1.2.1 was released I presume (like #4224) I'm going to backport into 1.2.