directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Karasulu (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DIRSERVER-1166) a lot of Persistent searches may exhaust the available threads
Date Fri, 25 Apr 2008 01:05:57 GMT

    [ https://issues.apache.org/jira/browse/DIRSERVER-1166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12592235#action_12592235
] 

Alex Karasulu commented on DIRSERVER-1166:
------------------------------------------

It should not be the case but there could be some flaw in the implementation.  Really we just
react to change events and start writing to the corresponding session.  It's the thread causing
the change that pushes out change notifications which in turn drive the send back to clients
in a PS call.

If we have a problem and are tying up a thread they you're fear is very valid and need to
correct it asap. 

> a lot of Persistent searches may exhaust the available threads
> --------------------------------------------------------------
>
>                 Key: DIRSERVER-1166
>                 URL: https://issues.apache.org/jira/browse/DIRSERVER-1166
>             Project: Directory ApacheDS
>          Issue Type: Bug
>    Affects Versions: 1.5.2
>            Reporter: Emmanuel Lecharny
>            Priority: Blocker
>             Fix For: 1.5.3
>
>
> While just trying to understand how the network layer works, I was wondering if PersistentSearch
might be a problem.
> It seems that each persistent search is consuming a thread, in order to keep the communication
with the client open. As we are using a limited size pool of thread, I'm afraid that the server
may simply stop responding if we have as many PS as nuber of threads in the pool.
> It may be a false alarm...

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message