-
Notifications
You must be signed in to change notification settings - Fork 28.6k
[SPARK-2815]: Compilation failed upon the hadoop version 2.0.0-cdh4.5.0 #1754
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
QA tests have started for PR 1754. This patch merges cleanly. |
This piece of code in the build is legacy, as the note in the following line says. I don't think it needs to be changed. The OP is building with the wrong command; that's the problem. |
This patch is to be consistent with the 1.0 version. |
QA results for PR 1754: |
I think this is an intermediate YARN version that is different from both the yarn-alpha and yarn-stable API's. @witgo what if you apply the patch here - does it work? https://github.com/apache/spark/pull/151/files In terms of merging this - I'm not sure we want to support 3 different YARN API's in the build and I think CDH itself said that YARN was not stable/supported here, so I'm not sure if we want to merge that patch. I am curious though whether it fixes it for you. |
This PR makes sbt' s behavior is consistent with building-with-maven.md description.
|
@@ -71,7 +71,7 @@ object SparkBuild extends PomBuild { | |||
} | |||
Properties.envOrNone("SPARK_HADOOP_VERSION") match { | |||
case Some(v) => | |||
if (v.matches("0.23.*")) isAlphaYarn = true | |||
if ("^2\\.[2-9]+".r.findFirstIn(v) == None) isAlphaYarn = true |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If there is ever a Hadoop 2.10, this pattern would consider it a YARN alpha version. Better to match against 2\\.[01]\\.[0-9]+
or something. This won't address the actual issue in the OP. It looks like a bit of cleanup, but in a deprecated code path. It would let Hadoop 2.0 / 2.1 be "supported" but they aren't actually.
@pwendell #151 compilation fails.
|
We need to explicitly pointed out that spark does not support the version |
I don't mind putting this one in (it's simple enough and might lower the bar for anyone trying to go this route). But the regex needs to be fixed, otherwise it will match 2.10.X. |
@pwendell I have updated the regular expression. |
QA tests have started for PR 1754 at commit
|
QA tests have finished for PR 1754 at commit
|
…pache#1754) * rdar://109188323 [Boson] Enable Boson scan in Spark 3.4 by default * fix
No description provided.