HI Sivakumar,
I don't know whether my issue (
https://glassfish.dev.java.net/issues/show_bug.cgi?id=13928) is related. A redeployment actually solves frees the memory in my case.
The memory leak occurs just by enabling CDI (with <beans></beans>) inside a JSF 2 application and generating the load. It doesn't occur outside JSF 2. So you can inject CDI managed beans into @Stateless without any problems,
thanks!,
adam
On 12.10.2010, at 17:38, Sivakumar Thyagarajan wrote:
> Hi Adam
>
> On Tuesday 12 October 2010 08:15 PM, Alexis Moussine-Pouchkine wrote:
>> Please make sure you also file this against Weld (and share the issue I'd). Let's hope the version we're about to integrate has a fix...
>
> Thanks for reporting this.
>
> This has already been filed in WELD issue tracker as https://jira.jboss.org/browse/WELD-570 and https://jira.jboss.org/browse/WELD-476. I heard today from the Weld lead that this has been fixed in their trunk and WELD 1.1 BETA2 promotion would happen this week or the next. So, when we integrate it in GlassFish 3.1, this should be fixed.
>
> This is also tracked in GlassFish as issue 12368.
>
> Thanks
> --Siva.
>>
>> Adam Bien<abien_at_adam-bien.com> wrote:
>>
>>> Hi All,
>>>
>>> I even filed an issue. A simple binding between JSF and CDI causes memory leaks - everything (NetBeans Project, Profiler snapshot, JMeter load script) is attached: https://glassfish.dev.java.net/issues/show_bug.cgi?id=13928
>>> It is an weld issue - JBoss6m5 behaves the same way.
>>>
>>> Any thoughts?
>>>
>>> thanks in advance!,
>>>
>>> adam
>>>
>
Consultant, Author, Speaker, Java Champion
Weblog: blog.adam-bien.com
press: press.adam-bien.com
eMail: abien_at_adam-bien.com
twitter: twitter.com/adambien
Mobile: 0049(0)170 280 3144
Author: 7 (Java SE/EE, SOA) Books, about 100 articles