Seaside hangs - what can I do?

> I also do a WAServerAdaptor allSubInstancesDo: [ :each | each kill]. in
> addition to WARegistry clearAllHandlers. Then I do a Smalltalk
> garbageCollectMost, and that usually clears up any issues that I may have
> caused, usually during debugger sessions.
>
> On Wed, Jan 13, 2010 at 10:27, Mariano Martinez Peck <[hidden email]>
> wrote:
>>
>>
>> On Wed, Jan 13, 2010 at 4:18 PM, Matthias Berth
>> <[hidden email]> wrote:
>>>
>>> Hi,
>>>
>>>
>>> I have encountered some error in a seaside image that lead to a
>>> hanging request. Then I opened a process browser and terminated the
>>> process that was associated to the request. Everything seemed to go
>>> fine after that. However, now (after some more requests) the image is
>>> not responding to any HTTP requests anymore. Everything else (browsing
>>> code, inspecting) works OK. I have also tried other applications, e.g.
>>> /seaside/config - not working.
>>>
>>> I am using Pier, with some persistency that is subclassed from
>>> PRImagePersistency. The process I terminated was in some semaphore
>>> critical section (guess: when doing the persistency). I am working
>>> from Pharo core 1.0rc1.
>>>
>>> My guess is that there is some semaphore involved, and maybe some
>>> process that is needed for handling http requests is no longer there.
>>>
>>> Is there a way to "reset" everything so I get HTTP request handling
>>> back? I have tried NetNameResolver initializeNetwork without success.
>>>
>>> Is there anything that I can do to prevent situations like this in the
>>> future? Anyone have a list of "do#s and don'ts " when using the
>>> process browser?
>>>
>>
>> I have no idea, but sometimes WARegistry clearAllHandlers.   help me.
>>
>>
>>
>>>
>>> Thanks a lot
>>>
>>> Matthias
>>> _______________________________________________
>>> seaside mailing list
>>> [hidden email]
>>> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
>>
>>
>> _______________________________________________
>> seaside mailing list
>> [hidden email]
>> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
>>
>
>
>
> --
> -JT
>
>
>
> _______________________________________________
> seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
>
>

>> Is there anything that I can do to prevent situations like this in the
>> future? Anyone have a list of "do#s and don'ts " when using the
>> process browser?
>
> Go back to the process browser and check at which semaphore the
> incoming requests hang. Signal this semaphore (there is a menu on that
> process in the process browser).
>
> Lukas
>
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
> _______________________________________________
> seaside mailing list
> [hidden email]
> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
>