geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rick McGuire (JIRA)" <>
Subject [jira] Resolved: (GERONIMO-4129) genorimo MimeMessage extensibility internal behaviour vs sun implementation OOM issue.
Date Thu, 19 Jun 2008 14:03:44 GMT


Rick McGuire resolved GERONIMO-4129.

       Resolution: Fixed
    Fix Version/s: 2.2

Committed revision 669490.

This uncovered 2 issues:

1)  The SharedFileInputStream was having problems with properly handling eof situations, which
was putting the parse logic into a loop, causing the OOM error. 

2)  The MineMessage.parse() method was not calling the protected createInternetHeaders() method
for creating the headers, which cause some extensibility problems for the MimeMessage subclass.

> genorimo MimeMessage extensibility internal behaviour vs sun implementation OOM issue.
> --------------------------------------------------------------------------------------
>                 Key: GERONIMO-4129
>                 URL:
>             Project: Geronimo
>          Issue Type: Bug
>      Security Level: public(Regular issues) 
>          Components: mail
>    Affects Versions: 2.1.1
>            Reporter: Stefano Bagnara
>            Assignee: Rick McGuire
>            Priority: Minor
>             Fix For: 2.2
> In Apache JAMES we use a MimeMessageWrapper class that is an extension for MimeMessage
providing some sort of lazy loading for streams coming from files and db records. To do that
we work with protected and public fields/methods from MimeMessage.
> Even the most simple test is current running OOM during the parse().
> In order to run the tests you will need the following files:
> This code also uses Disposable and ContainerUtils from avalon: they are simply a marker
interface for the dispose() method and ContainerUtils.dispose() is a method to call dispose
only if the object is Disposable.
> if you don't care of unclosed resources you can remove Disposable and the ContainerUtils.
> RFC2822Headers from mailets is instead a constants collection for rfc headers: you can
safely replace that with literals.

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

View raw message