mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Edouard De Oliveira (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DIRMINA-716) Get rid of "finalize" method in AbstractIoFilterChain.java
Date Tue, 02 Jun 2009 09:03:07 GMT

    [ https://issues.apache.org/jira/browse/DIRMINA-716?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12715432#action_12715432

Edouard De Oliveira commented on DIRMINA-716:

The finalize() catch is a good one i tried to see it for myself and i confirm the poor memory
management it causes (you can look at http://tedorgwp.free.fr/?p=450 which contains graphs
of the generated heap load).

I do also fully agree with the copy on change method as i think it only happens in specific
implementations. Other implementations will gain some additionnal free memory from this tweak.

> Get rid of "finalize" method in AbstractIoFilterChain.java
> ----------------------------------------------------------
>                 Key: DIRMINA-716
>                 URL: https://issues.apache.org/jira/browse/DIRMINA-716
>             Project: MINA
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 2.0.0-M5
>            Reporter: Yongxing Wang
>            Assignee: Emmanuel Lecharny
> The finalize method in AbstractIoFilterChain.java can possibly cause OOM error when the
system is under very heavy load. The negative impact of "finalize" method can be found at:
> http://www.fasterj.com/articles/finalizer1.shtml
> AbstractIoFilterChain holds a reference to Session object which can hold a list of unwritten
data. The deterministic run of Finalizer can very well cause memory not fred up on time under
heavy load. And it can certainly cause very bad memory usage pattern and contribute to the
overall heavy system load.

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

View raw message