>>> Which Felix regression are you referring to? This one:
>>> https://issues.apache.org/jira/browse/FELIX-2646
>> The b26 performance numbers are worse than b25. Tom mentioned there
>> is a problem in Felix (the OSGi cache is being being rebuilt twice so
>> the second start of the domain is really slow) and Richard is looking
>> at the issue.
> Yes, that's caused by the Felix bug I referenced earlier. Note, it
> only affects server startup time the second time around, not in third,
> fourth, or fifth attempt and it does not come into picture after the
> server is started. So, that can't be cause. There could very well be
> some other problem in Felix or in some other component.
Could be. Tom might know more..
> When you said worse, qualify.
The numbers can be found the in the performance wiki page (I will send
you the link offline).
- Hong
> OSGi Cache rebuilding takes 2-3 seconds - here we are talking about
> minutes.
>
>>>
>>> I don't think that can affect.
>>>
>>> You already have a test case (deploy hudson.war).
>> But that seems a different issue as Harald was seeing the same
>> problem for b24 where Steve's app was deployed ok there (please see
>> previous posts on this thread).
> May be.
>
> Sahoo
>>
>> - Hong
>>>
>>>
>>> On Thursday 04 November 2010 08:13 PM, Hong Zhang wrote:
>>>> There was a felix regression went in b26/b27 timeframe and
>>>> currently being looked at. But I don't think it would cause this
>>>> much difference. Is it possible to share your application or a test
>>>> case which shows the problem so we can take a look from our side?
>>>> Or you could continue working with b24 and once the felix
>>>> regression is fixed, you can try it again to see if resolves the
>>>> problem.
>>>>
>>>> Thanks,
>>>>
>>>> - Hong
>>>>
>>>> On 11/4/2010 10:26 AM, noreply_at_java.net wrote:
>>>>> Hi,
>>>>>
>>>>> I get the same effect no matter whether the application is
>>>>> deployed from the
>>>>> admin console, or from the command line (always 100% CPU). It's
>>>>> doing it
>>>>> right now - the deployment, which on b24 is about 15 seconds,
>>>>> takes about 10
>>>>> minutes on b26 or b27.
>>>>>
>>>>> Regards,
>>>>>
>>>>> Steve.
>>>>>
>>>>>
>>>>> --
>>>>>
>>>>> [Message sent by forum member 'suttridge_farm']
>>>>>
>>>>> View Post: http://forums.java.net/node/711853
>>>>>
>>>>>
>>>>>
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
>>>>> For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
>>>> For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>>>>
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
>>> For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe_at_glassfish.dev.java.net
>> For additional commands, e-mail: users-help_at_glassfish.dev.java.net
>>
>