You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by liancheng <gi...@git.apache.org> on 2014/08/26 21:55:31 UTC

[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

GitHub user liancheng opened a pull request:

    https://github.com/apache/spark/pull/2145

    [SPARK-2608][Core] Fixed command line option passing issue over Mesos

    This PR is based on #1986, authored by @scwf.
    
    The basic idea is to start Mesos executors directly with `java` (similar to what we do for standalone executors). This PR is different from #1986 in two major aspects:
    
    1. Environment variables in `sc.executorEnvs` are properly set for Mesos executors
    2. `PYTHONPATH` is properly set for fine grained Mesos executors by calling `sbin/mesos-pyenv.sh`
    
       `sbin/spark-executor` is renamed to `sbin/mesos-pyenv.sh` since it's only responsible for setting `PYTHONPATH` now, and the executor is started by `MesosSchedulerBackend` after the change.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/liancheng/spark fix-mesos-opts

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/spark/pull/2145.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #2145
    
----
commit 478ccda56e4a791a8b3783b51740fec52502ac4f
Author: Cheng Lian <li...@gmail.com>
Date:   2014-08-26T09:17:37Z

    Fixed command line option passing issue over Mesos
    
    Conflicts:
    	core/src/main/scala/org/apache/spark/scheduler/cluster/mesos/CoarseMesosSchedulerBackend.scala

----


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2145#discussion_r16741070
  
    --- Diff: sbin/mesos-pyenv.sh ---
    @@ -17,10 +17,10 @@
     # limitations under the License.
     #
     
    +#
    +# This script is only used for fine grained Mesos executors
    +
     FWDIR="$(cd `dirname $0`/..; pwd)"
     
     export PYTHONPATH=$FWDIR/python:$PYTHONPATH
     export PYTHONPATH=$FWDIR/python/lib/py4j-0.8.2.1-src.zip:$PYTHONPATH
    -
    -echo "Running spark-executor with framework dir = $FWDIR"
    -exec $FWDIR/bin/spark-class org.apache.spark.executor.MesosExecutorBackend
    --- End diff --
    
    It seems that we don't even need this file if we're doing this. We can just export the `PYTHONPATH` inside the mesos backend classes themselves.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53505309
  
    QA tests have started for PR 2145. This patch merges cleanly. <br>View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/19247/consoleFull


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53505551
  
    The last build failure was caused by Spark Streaming, should be irrelevant.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53521159
  
    @andrewor14 A summary of differences between all 4 PRs:
    
    1. #1513 by @scwf, the first PR for this issue
    1. #1986 by @scwf, rebased #1513 to the master
    1. #2103 by @tnachen, same as #1986 except that `sc.sparkHome` is used to locate Spark home directory on both driver side and executor side
    1. #2145 by @liancheng, described in the PR description


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53516030
  
    I think we need to consolidate all our fixes now :) So I think we all understand the problem and get the idea, the problem is really I think our fixes as we reviewed each other's stuff for now isn't completely fixing the problem.
    
    What used to happen as we know now is that we're passing in extra options like java settings to as one of the param to spark-class which is incorrect.
    
    Now your fix here although uses CommandUtils which is the consolidated util for generating command, it's still running compute-classpath and setting the classpath information directly in the CommandInfo's value. I tested my fix #2103 earlier with mesos master and slave on the same host, and although it ran it still not enough when running a mesos cluster with master and slaves in separate hosts as the classpath was wrong.
    
    I think we should not assume the classpath from the framework launching tasks should be the same as the slave. 
    
    I think we should either 1) still use spark-class and let spark-class resolve classpaths 2) Update command utils to not run compute-classpath on the spot,  but let the slave run compute-classpath to run the /usr/bin/java with.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53611716
  
    We should eventually make `-Dfoo="bar bar"` work, though the top priority now is just to fix the core functionality of the Mesos code. This is a nice-to-have addition, but the lack of it should not block the release.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53520529
  
    @pwendall let's not merge this yet as I'll try to run it on a mesos cluster as I don't think it will work. I'll try to have something tonight or tomorrow mornjng 


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by scwf <gi...@git.apache.org>.
Github user scwf commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53539011
  
    hi @liancheng , is there a situation we should cover ` -Dfoo="bar bar" ` ?


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by pwendell <gi...@git.apache.org>.
Github user pwendell commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53519972
  
    Hey @liancheng @tnachen @scwf - don't worry to much about consolidating the patches. I can just merge this one and give all three of you author credits on it... this is how we normally do it.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53510522
  
    QA results for PR 2145:<br>- This patch FAILED unit tests.<br>- This patch merges cleanly<br>- This patch adds no public classes<br><br>For more information see test ouptut:<br>https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/19247/consoleFull


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by pwendell <gi...@git.apache.org>.
Github user pwendell commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2145#discussion_r16741799
  
    --- Diff: sbin/mesos-pyenv.sh ---
    @@ -17,10 +17,10 @@
     # limitations under the License.
     #
     
    +#
    +# This script is only used for fine grained Mesos executors
    +
     FWDIR="$(cd `dirname $0`/..; pwd)"
     
     export PYTHONPATH=$FWDIR/python:$PYTHONPATH
     export PYTHONPATH=$FWDIR/python/lib/py4j-0.8.2.1-src.zip:$PYTHONPATH
    -
    -echo "Running spark-executor with framework dir = $FWDIR"
    -exec $FWDIR/bin/spark-class org.apache.spark.executor.MesosExecutorBackend
    --- End diff --
    
    We can use `sparkHome`... we need it anyways to find this script.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53521325
  
    @tnachen Ah, I see where I'm wrong, although I pass `"."` as Spark home to run `compute-classpath.sh`, it converts `"."` to absolute path internally, thus resulted classpaths are all bound to driver side environment.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53619649
  
    Sure.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53535667
  
    Ok I just tried it on a mesos cluster and it didn't work. 
    
    The classpath it put in the Mesos command is where spark lives in the spark shell in another host, but not in just pulled down spark-executor from the tar.
    
    sh -c 'cd spark-1*; "/usr/bin/java" "-cp" "::/home/jclouds/src/spark/conf:/home/jclouds/src/spark/assembly/target/scala-2.10/spark-assembly-1.1.0-SNAPSHOT-hadoop1.0.4.jar" "-XX:MaxPermSize=128m" "-Xms512M" "-Xmx512M" "org.apache.spark.executor.CoarseGrainedExecutorBackend" "akka.tcp://sparkDriver@highly-available-182-e54.c.modern-saga-648.internal:47860/user/CoarseGrainedScheduler" "20140818-071808-3483423754-5050-2070-4" "10.151.50.130" "2"'
    
    So we must still compute classpaths after it's pulled down, not wherever the spark-shell is being executed and assume it's going to run the tared spark.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2145#discussion_r16741443
  
    --- Diff: sbin/mesos-pyenv.sh ---
    @@ -17,10 +17,10 @@
     # limitations under the License.
     #
     
    +#
    +# This script is only used for fine grained Mesos executors
    +
     FWDIR="$(cd `dirname $0`/..; pwd)"
     
     export PYTHONPATH=$FWDIR/python:$PYTHONPATH
     export PYTHONPATH=$FWDIR/python/lib/py4j-0.8.2.1-src.zip:$PYTHONPATH
    -
    -echo "Running spark-executor with framework dir = $FWDIR"
    -exec $FWDIR/bin/spark-class org.apache.spark.executor.MesosExecutorBackend
    --- End diff --
    
    I tried to, but still left this file here because it seemed non-trivial to figure out `$FWDIR` from Scala code.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53480498
  
    /cc @andrewor14


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53536988
  
    @tnachen Thanks a lot! I'm currently working on another version that can figure out executor side classpath correctly. The basic idea is:
    
    1. we still start the executor with `spark-class`, and
    1. we pass `extraJavaOpts` and `extraLibraryPath` via `SPARK_EXECUTOR_OPTS`, which is recognized by `spark-class` and not used anywhere else.
    
    You may find the WIP version [here](https://github.com/liancheng/spark/compare/apache:branch-1.1...liancheng:mesos-fix-with-env-var?expand=1#diff-d425d35aa23c47a62fbb538554f2f2cfR123). Discussed with @pwendell about this solution tonight, and it seems workable. And it's also much simpler. For now, the only issue is that it cannot handle quoted string with spaces correctly (i.e. `-Dfoo="bar bar"`). It might be buggy in other ways though, still testing it.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by pwendell <gi...@git.apache.org>.
Github user pwendell commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2145#discussion_r16756703
  
    --- Diff: core/src/main/scala/org/apache/spark/scheduler/cluster/mesos/CoarseMesosSchedulerBackend.scala ---
    @@ -18,19 +18,21 @@
     package org.apache.spark.scheduler.cluster.mesos
     
     import java.io.File
    -import java.util.{List => JList}
    -import java.util.Collections
    +import java.util.{Collections, List => JList}
     
     import scala.collection.JavaConversions._
     import scala.collection.mutable.{HashMap, HashSet}
     
    -import org.apache.mesos.{Scheduler => MScheduler}
    -import org.apache.mesos._
     import org.apache.mesos.Protos.{TaskInfo => MesosTaskInfo, TaskState => MesosTaskState, _}
    +import org.apache.mesos.{Scheduler => MScheduler, _}
     
    -import org.apache.spark.{Logging, SparkContext, SparkEnv, SparkException}
    +import org.apache.spark.deploy.Command
    +import org.apache.spark.deploy.worker.CommandUtils
    --- End diff --
    
    This set of utilities is not intended to be used outside of the `deploy` code (i.e. Spark's standalone scheduler) that's why it's causing issues here.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2145#discussion_r16746707
  
    --- Diff: sbin/mesos-pyenv.sh ---
    @@ -17,10 +17,10 @@
     # limitations under the License.
     #
     
    +#
    +# This script is only used for fine grained Mesos executors
    +
     FWDIR="$(cd `dirname $0`/..; pwd)"
     
     export PYTHONPATH=$FWDIR/python:$PYTHONPATH
     export PYTHONPATH=$FWDIR/python/lib/py4j-0.8.2.1-src.zip:$PYTHONPATH
    -
    -echo "Running spark-executor with framework dir = $FWDIR"
    -exec $FWDIR/bin/spark-class org.apache.spark.executor.MesosExecutorBackend
    --- End diff --
    
    But `sparkHome` is driver side Spark home directory, which can't be used on driver side to assemble the Mesos executor side command line. On the other side, `sbin/mesos-pyenv.sh` is always executed on Mesos executor side, thus `$FWDIR` always points to the right position.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53538424
  
    Also I think I didn't mention it explicitly, I've been testing with having a spark tar ball available through a HTTP server and setting SPARK_EXECUTOR_URI to that, and slaves have no spark installed. I know folks are using both cases where the executor uri is either set or unset, which defaults to spark_home. 


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by pwendell <gi...@git.apache.org>.
Github user pwendell commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53537143
  
    Hey guys, yeah this is an issue with the approach of using the utilities from the standalone deploy mode for this - it makes assumptions that don't hold in mesos mode. I spoke a bit offline with @liancheng and I think there is a much simpler/surgical fix that will unblock the Spark 1.1 release. But we should have a nicer way of building up the command in Scala like is done here. It might mean we slightly re-factor things so that parts of the utility functions for standalone mode can be used here.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53498810
  
    QA results for PR 2145:<br>- This patch FAILED unit tests.<br>- This patch merges cleanly<br>- This patch adds no public classes<br><br>For more information see test ouptut:<br>https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/19238/consoleFull


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53537607
  
    I'm glad we're having these conversations :) Really helping folks that have bad experience using Mesos with Spark. I'm looking forward for the fix and once it's updated I can verify the fix with our mesos cluster. I'm chatting with Mesos committers about different issues people are hitting and I'll be addressing those in future patches.


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by tnachen <gi...@git.apache.org>.
Github user tnachen commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53516127
  
    Btw, have you guys tested with a actual mesos and spark cluster? I have one setup now to make sure things will run


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53617288
  
    @liancheng Given that there is now a newer PR that supersedes this one, would you mind closing this?


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53499077
  
    test this please


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53487152
  
    QA tests have started for PR 2145. This patch merges cleanly. <br>View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/19238/consoleFull


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng closed the pull request at:

    https://github.com/apache/spark/pull/2145


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by liancheng <gi...@git.apache.org>.
Github user liancheng commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53517950
  
    Hey @tnachen and @scwf, the only reason I opened this PR separately is that we're already late for the Spark 1.1 RC release, and we need a fix quickly. We would definitely list both of you two in the contributor list of this release :)
    
    In this PR I used `"."` as executor side Spark home when executor URI is provided (see [here](https://github.com/apache/spark/pull/2145/files#diff-d425d35aa23c47a62fbb538554f2f2cfR162) and [here](https://github.com/apache/spark/pull/2145/files#diff-e3a5e67b8de2069ce99801372e214b8eR136)). This should be valid because we first `cd` into the right Spark home directory (with the `basename` trick explained [here](https://github.com/apache/spark/pull/2145/files#diff-d425d35aa23c47a62fbb538554f2f2cfR151)).
    
    As for testing, @tnachen you really hit the point :) I only tested this PR with a local single node Mesos cluster, which cannot simulate the situation that Mesos slave nodes doesn't have Spark installed in the same path as the driver side. It would be greatly appreciated if you can help testing this PR with your real distributed Spark over Mesos cluster. Thanks in advance!


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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


[GitHub] spark pull request: [SPARK-2608][Core] Fixed command line option p...

Posted by andrewor14 <gi...@git.apache.org>.
Github user andrewor14 commented on the pull request:

    https://github.com/apache/spark/pull/2145#issuecomment-53516887
  
    Hi @liancheng @tnachen @scwf. It seems that there are many patches on fixing this that are based on each other (#2103, #1986). Can we flatten out the differences and consolidate all the changes within one PR?


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

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