[
https://issues.apache.org/jira/browse/HADOOP-12358?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14723767#comment-14723767
]
Xiaoyu Yao commented on HADOOP-12358:
-------------------------------------
Thanks all for the review and feedbacks. Update to patch v6 with the following
summary of changes based on the feedback:
1. add "-safely" option to -rm command as [~aw] suggested.
2. Reduce the configuration complexity as [~arpitagarwal] suggested.
Only one key "hadoop.shell.delete.limit.num.files" is used.
3. Document 'hadoop.shell.delete.limit.num.files' in core-default.xml.
4. Make this feature optional and off by default to avoid breaking any existing
automations. It is enabled only if all the three criteria are met:
* Trash is not enabled or unable to protect the directory to be deleted
* and -safely is used in the rm command
* and hadoop.shell.delete.limit.num.files > 0
This way, the admin can choose if they think the feature is useful for certain
use cases. Especially with HADOOP-11353, admin can alias 'hadoop -rm' to
'hadoop -rm -safely' in .hadooprc like 'rm' to 'rm -i' in Linux deployments
when necessary.
[~arpitagarwal]: Given HDFS-4995 and HDFS-8046 have improved the NN locking
issue of getContentSummary, is it OK to investigate for performance
improvement in a separate JIRA?
> FSShell should prompt before deleting directories bigger than a configured
> size
> -------------------------------------------------------------------------------
>
> Key: HADOOP-12358
> URL: https://issues.apache.org/jira/browse/HADOOP-12358
> Project: Hadoop Common
> Issue Type: Bug
> Components: fs
> Reporter: Xiaoyu Yao
> Assignee: Xiaoyu Yao
> Attachments: HADOOP-12358.00.patch, HADOOP-12358.01.patch,
> HADOOP-12358.02.patch, HADOOP-12358.03.patch, HADOOP-12358.04.patch,
> HADOOP-12358.05.patch, HADOOP-12358.06.patch
>
>
> We have seen many cases with customers deleting data inadvertently with
> -skipTrash. The FSShell should prompt user if the size of the data or the
> number of files being deleted is bigger than a threshold even though
> -skipTrash is being used.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)