[SPARK-18643][SPARKR] SparkR hangs at session start when installed as a package without Spark #16077
+9
−4
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Member
Author
|
Test build #69386 has finished for PR 16077 at commit
|
Contributor
|
Is this means that it will install spark package each time when starting a new session in interactive mode? Thanks. |
Member
Author
|
install.spark() checks if SPARK_HOME is set or the Spark Jar can be found in the cache location - it only downloads if Spark isn't found.
|
Contributor
|
LGTM |
Contributor
|
Just to clarify this limits the auto-install feature to SparkR running on an interactive shell -- Is that right ? I think thats mostly fine as its the use-case we were targeting but it might good to check our documentation and update it appropriately. |
Member
Author
|
Right, I was just reviewing possible code paths for this in the last few days and I'm pretty confident that this change will not run install.spark in cluster modes (which would have Spark/JVM already running). Also you are right, I just found that we didn't really talk about how install.spark would be called in sparkR.session() - I'll add that. |
|
Test build #69654 has finished for PR 16077 at commit
|
Contributor
|
LGTM. Merging this to master, branch-2.1 |
asfgit
pushed a commit
that referenced
this pull request
Dec 5, 2016
… a package without Spark ## What changes were proposed in this pull request? If SparkR is running as a package and it has previously downloaded Spark Jar it should be able to run as before without having to set SPARK_HOME. Basically with this bug the auto install Spark will only work in the first session. This seems to be a regression on the earlier behavior. Fix is to always try to install or check for the cached Spark if running in an interactive session. As discussed before, we should probably only install Spark iff running in an interactive session (R shell, RStudio etc) ## How was this patch tested? Manually Author: Felix Cheung <[email protected]> Closes #16077 from felixcheung/rsessioninteractive. (cherry picked from commit b019b3a) Signed-off-by: Shivaram Venkataraman <[email protected]>
robert3005
pushed a commit
to palantir/spark
that referenced
this pull request
Dec 15, 2016
… a package without Spark ## What changes were proposed in this pull request? If SparkR is running as a package and it has previously downloaded Spark Jar it should be able to run as before without having to set SPARK_HOME. Basically with this bug the auto install Spark will only work in the first session. This seems to be a regression on the earlier behavior. Fix is to always try to install or check for the cached Spark if running in an interactive session. As discussed before, we should probably only install Spark iff running in an interactive session (R shell, RStudio etc) ## How was this patch tested? Manually Author: Felix Cheung <[email protected]> Closes apache#16077 from felixcheung/rsessioninteractive.
uzadude
pushed a commit
to uzadude/spark
that referenced
this pull request
Jan 27, 2017
… a package without Spark ## What changes were proposed in this pull request? If SparkR is running as a package and it has previously downloaded Spark Jar it should be able to run as before without having to set SPARK_HOME. Basically with this bug the auto install Spark will only work in the first session. This seems to be a regression on the earlier behavior. Fix is to always try to install or check for the cached Spark if running in an interactive session. As discussed before, we should probably only install Spark iff running in an interactive session (R shell, RStudio etc) ## How was this patch tested? Manually Author: Felix Cheung <[email protected]> Closes apache#16077 from felixcheung/rsessioninteractive.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
If SparkR is running as a package and it has previously downloaded Spark Jar it should be able to run as before without having to set SPARK_HOME. Basically with this bug the auto install Spark will only work in the first session.
This seems to be a regression on the earlier behavior.
Fix is to always try to install or check for the cached Spark if running in an interactive session.
As discussed before, we should probably only install Spark iff running in an interactive session (R shell, RStudio etc)
How was this patch tested?
Manually