[
https://issues.apache.org/jira/browse/MAPREDUCE-849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12757024#action_12757024
]
Sharad Agarwal commented on MAPREDUCE-849:
------------------------------------------
Glanced over the patch. Mostly looking fine. Some minor nits:
To be consistent let all properties start with TT_ in TTConfig. Similarly for
JTConfig, all should start from JT_
JTConfig.JT could have a better name
mapreduce.jobtracker.taskalloc.capacitypad is scheduler property. Should we
rename it to mapreduce.jobtracker.scheduler.taskalloc.capacitypad to be more
clear
JTConfig.HOSTS and JTConfig.HOSTS_EXCLUDE takes file name. It will be better if
we have that in the name. Something like
mapreduce.jobtracker.hosts.exclude.filename
In MRConfig, MAX_MAPMEMORY_MB and MAX_REDUCEMEMORY_MB are used only in
jobtracker. so these should be moved to JTConfig
I don't see any references of MRConfig.MAX_TRACKER_BLACKLISTS.
Also currently mapred-default.xml doesn't document all the properties. We
should file a documentation jira to add all missing properties in
mapred-default.xml
> Renaming of configuration property names in mapreduce
> -----------------------------------------------------
>
> Key: MAPREDUCE-849
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-849
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: Amareshwari Sriramadasu
> Assignee: Amareshwari Sriramadasu
> Fix For: 0.21.0
>
> Attachments: Config changes.xls, Config changes.xls, patch-849-1.txt,
> patch-849-2.txt, patch-849.txt
>
>
> In-line with HDFS-531, property names in configuration files should be
> standardized in MAPREDUCE.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
