[
https://issues.apache.org/jira/browse/HDFS-8855?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14660012#comment-14660012
]
Bob Hansen commented on HDFS-8855:
----------------------------------
I tried running this script against an HDFS-2.6.0 patched with [~daryn]'s
original HDFS-7597 patch (just to ensure that my refactoring wasn't a
confounding factor), and the issue seems to persist. It is entirely possible
that I messed up my configuration and was accidentally running unpatched code,
so take that with a grain of salt.
Looking at the source, my inclination would be to have an LRU (much like the
HDFS-7597 patch) attached to the WebHDFS session that would map ugi->HDFS
client. This would keep the client's information and connection state around
for re-use, but shut it down once the HTTP session ended. What do you think,
[~xiaobingo]?
> Webhdfs client leaks active NameNode connections
> ------------------------------------------------
>
> Key: HDFS-8855
> URL: https://issues.apache.org/jira/browse/HDFS-8855
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: webhdfs
> Environment: HDP 2.2
> Reporter: Bob Hansen
> Assignee: Xiaobing Zhou
>
> The attached script simulates a process opening ~50 files via webhdfs and
> performing random reads. Note that there are at most 50 concurrent reads,
> and all webhdfs sessions are kept open. Each read is ~64k at a random
> position.
> The script periodically (once per second) shells into the NameNode and
> produces a summary of the socket states. For my test cluster with 5 nodes,
> it took ~30 seconds for the NameNode to have ~25000 active connections and
> fails.
> It appears that each request to the webhdfs client is opening a new
> connection to the NameNode and keeping it open after the request is complete.
> If the process continues to run, eventually (~30-60 seconds), all of the
> open connections are closed and the NameNode recovers.
> This smells like SoftReference reaping. Are we using SoftReferences in the
> webhdfs client to cache NameNode connections but never re-using them?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
