[
https://issues.apache.org/jira/browse/HDFS-7701?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14387087#comment-14387087
]
Xiaoyu Yao commented on HDFS-7701:
----------------------------------
Patch v3 looks pretty good to me. Thanks Peter for addressing the comments and
unit test failures. I Just have a couple of Nits below.
*ContentSummary.java*
line 292: Can you rename getStorageHeader() to getStorageTypeHeader()?
Line 296: Can you remove the comment on RAM_DISK_QUOTA as it is the type that
does not support quota by storage type. Also the 14/18 format for
quota/remaining quota print might also need to adjust?
Line 353: Can you add JavaDoc for the new ContentSummary#toString() method?
Line 362: Can you consolidate the string literals as static const of the class
for "none" and "inf"?
Line 400: Can you remove the extra empty line?
*CommandFormat.java*
There is an extra empty line in line 244.
Can you reword the comment on line 247? "Used when a duplicated option is
supplied to a command.
*Count.java*
Can you remove the extra empty line in line: 120 and 145?
Can you break the line at line: 150 so that each line is within 80.
*TestCount.java*
Line:30 There is an extra empty line
Can you remove the unused import "Import org.apache.hadoop.util.StringUtils;"
> Support reporting per storage type quota and usage with hadoop/hdfs shell
> -------------------------------------------------------------------------
>
> Key: HDFS-7701
> URL: https://issues.apache.org/jira/browse/HDFS-7701
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Reporter: Xiaoyu Yao
> Assignee: Peter Shi
> Attachments: HDFS-7701.01.patch, HDFS-7701.02.patch,
> HDFS-7701.03.patch
>
>
> "hadoop fs -count -q" or "hdfs dfs -count -q" currently shows name space/disk
> space quota and remaining quota information. With HDFS-7584, we want to
> display per storage type quota and its remaining information as well.
> The current output format as shown below may not easily accomodate 6 more
> columns = 3 (existing storage types) * 2 (quota/remaining quota). With new
> storage types added in future, this will make the output even more crowded.
> There are also compatibility issues as we don't want to break any existing
> scripts monitoring hadoop fs -count -q output.
> $ hadoop fs -count -q -v /test
> QUOTA REM_QUOTA SPACE_QUOTA REM_SPACE_QUOTA DIR_COUNT
> FILE_COUNT CONTENT_SIZE PATHNAME
> none inf 524288000 524266569 1
> 15 21431 /test
> Propose to add -t parameter to display ONLY the storage type quota
> information of the directory in the separately. This way, existing scripts
> will work as-is without using -t parameter.
> 1) When -t is not followed by a specific storage type, quota and usage
> information for all storage types will be displayed.
> $ hadoop fs -count -q -t -h -v /test
> SSD_QUOTA REM_SSD_QUOTA DISK_QUOTA REM_DISK_QUOTA
> ARCHIVAL_QUOTA REM_ARCHIVAL_QUOTA PATHNAME
> 512MB 256MB none inf none
> inf /test
> 2) If -t is followed by a storage type, only the quota and remaining quota of
> the storage type is displayed.
> $ hadoop fs -count -q -t SSD -h -v /test
>
> SSD_QUOTA REM_SSD_QUOTA PATHNAME
> 512 MB 256 MB /test
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
