[
https://issues.apache.org/jira/browse/ZOOKEEPER-2956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17000138#comment-17000138
]
Andor Molnar commented on ZOOKEEPER-2956:
-----------------------------------------
We had the same issue in a production environment recently.
I wonder if there was any plan on implementing recursive delete on server side.
> Cannot delete znode that owns too many child znodes by `rmr` command
> --------------------------------------------------------------------
>
> Key: ZOOKEEPER-2956
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2956
> Project: ZooKeeper
> Issue Type: Bug
> Components: server
> Reporter: Benedict Jin
> Assignee: Benedict Jin
> Priority: Major
> Attachments: 172MB_child_znodes_list.png, rmr_failed.png
>
>
> We cannot delete znode that owns too many child znodes by `rmr` command, due
> to the list of child znodes could be 172 MB, which is too huge for the
> default value of `jute.maxbuffer` (1MB). In fact, we shouldn't be effected by
> the number of child znodes when we want to delete some special znodes
> recursively.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
