[
https://issues.apache.org/jira/browse/HADOOP-9747?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16290306#comment-16290306
]
Bharat Viswanadham commented on HADOOP-9747:
--------------------------------------------
[~daryn]
One more thing I found is
we need to remove testCheckTGTAfterLoginFromSubject from branch-2 patch also,
this has been addressed in trunk patch.
Because in latest patch we have only simple and kerberos configurations.
where as previously we used to have simple, user-kerberos(when login from
subject) and keytab-kerberos. user-kerberos app config used to have
os_specific_login and krb5login configurations, so if no kerberosprincipal is
found it used to use os_specific_login user and loginUserFromSubject() used to
pass, but now this will not work, as it will say User cannot be null, as the
subject will be empty (as there is no user of type KerberosPrincipal).
I am providing multiple comments, as more i am understanding the code, getting
more questions, hope you don't mind.
Please let me know if my analysis looks correct or am i missing something here.
> Reduce unnecessary UGI synchronization
> --------------------------------------
>
> Key: HADOOP-9747
> URL: https://issues.apache.org/jira/browse/HADOOP-9747
> Project: Hadoop Common
> Issue Type: Bug
> Components: security
> Affects Versions: 0.23.0, 2.0.0-alpha, 3.0.0-alpha1
> Reporter: Daryn Sharp
> Assignee: Daryn Sharp
> Priority: Critical
> Attachments: HADOOP-9747-trunk.01.patch,
> HADOOP-9747.2.branch-2.patch, HADOOP-9747.2.trunk.patch,
> HADOOP-9747.branch-2.patch, HADOOP-9747.trunk.patch
>
>
> Jstacks of heavily loaded NNs show up to dozens of threads blocking in the
> UGI.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
