[
https://issues.apache.org/jira/browse/MAPREDUCE-7047?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16395559#comment-16395559
]
Robert Kanter commented on MAPREDUCE-7047:
------------------------------------------
Thanks [~xgong]. A few minor things:
# This code can be simplified, and should return a non-zero exit code:
{code:java}
boolean createdWorkingDirsSuccess = true;
for (Path workingDir : workingDirs) {
if (!prepareWorkingDir(fs, workingDir)) {
createdWorkingDirsSuccess = false;
LOG.error("Failed to create the workingDir:"
+ workingDir.toString());
break;
}
}
if (!createdWorkingDirsSuccess) {
return 0;
}
{code}
to
{code:java}
for (Path workingDir : workingDirs) {
if (!prepareWorkingDir(fs, workingDir)) {
createdWorkingDirsSuccess = false;
LOG.error("Failed to create the workingDir:"
+ workingDir.toString());
return 1
}
}
{code}
# When cleaning up the working directories, is it necessary to check that the
directories exist before trying to delete them? That's going to result in
additional calls. IIRC, delete will do a no-op if the directory doesn't exist.
# Looks like we accidentally lost a blank line above the
{{runDistributedShell}} method
# {{equals}} should also consider the other fields added to {{AppInfo}}. I
don't think we'll have any cases where this would cause a problem, but it's
safer if we take those into account.
# It would be better if the test used different values for the applications
(e.g. different working dirs, etc)
> Make HAR tool support IndexedLogAggregtionController
> ----------------------------------------------------
>
> Key: MAPREDUCE-7047
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-7047
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: Xuan Gong
> Assignee: Xuan Gong
> Priority: Major
> Attachments: MAPREDUCE-7047.trunk.1.patch
>
>
> In https://issues.apache.org/jira/browse/MAPREDUCE-6415, we have created a
> tool to combine aggregated logs into HAR files which currently only work for
> TFileLogAggregationFileController. We should make it support
> IndexedLogAggregtionController as well.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
