[Ace-users] Re: [tao-bugs] Notify_Service Ctrl-C hangs

[Hans Utz]

Hi Johnny,

filed as bug #3054.
Cheers,
Hans

Johnny Willemsen wrote:
> Hi Hans,
> 
> Can you store this problem in bugzilla (see
> http://deuce.doc.wustl.edu/bugzilla/index.cgi )
> 
> Regards,
> 
> Johnny Willemsen
> Remedy IT
> Postbus 101
> 2650 AC  Berkel en Rodenrijs
> The Netherlands
> www.theaceorb.nl / www.remedy.nl  
> 
> *** Integrated compile and test statistics see
> http://scoreboard.theaceorb.nl ***
> *** Commercial service and support for ACE/TAO/CIAO             ***
> *** See http://www.theaceorb.nl/en/support.html                 ***
> 
>>       TAO VERSION: 1.5.10
>>       ACE VERSION: 5.5.10
>>
>>       HOST MACHINE and OPERATING SYSTEM: Linux RH WS 4
>>       COMPILER NAME AND VERSION (AND PATCHLEVEL): gcc 3.4.6
>>
>>       THE $ACE_ROOT/ace/config.h FILE:
>>         config-linux.h
>>       THE $ACE_ROOT/include/makeinclude/platform_macros.GNU FILE:
>>         platform_linux.GNU
>>
>>       AREA/CLASS/EXAMPLE AFFECTED: Notification Service
>>
>>       DOES THE PROBLEM AFFECT:
>>            EXECUTION
>>
>>       SYNOPSIS:
>>
>> When running Notify_Service with asynchronous dispatching (as 
>> enabled in 
>> the svc.conf file in TAO/orbsvcs/Notify_Service) Ctrl-C will hang, if 
>> there is still a consumer registered at the service.
>>
>>       SAMPLE FIX/WORKAROUND:
>>
>> The root problem is, that the Notify_Service_Driver::run() blocks on 
>> worker_thr_mgr()->wait(). Notify_Service_Driver::shutdown() calls 
>> orb_->shutdown(), but does not cancel the dispatching threads, which 
>> seem to be managed under the same thread manager.
>>
>> Consumers can not disconnect after Ctrl-C anymore either, as the orb 
>> event loop is already down.
>>
>> Workaround: Initializing -ORBRunThreads 0 causes ::run() to end up in 
>> orb_->run() instead of ...->wait(). This at least works for 
>> the "single 
>> threaded orb" configuration.
>>
>> Cheers,
>> Hans
>>
>>
>> -- 
>> Hans Utz - USRA/RIACS at NASA Ames
>> I do not speak for NASA, and NASA does not speak for me.
>>
>> _______________________________________________
>> tao-bugs mailing list
>> tao-bugs at mail.cse.wustl.edu
>> http://mail.cse.wustl.edu/mailman/listinfo/tao-bugs
>>
> 


-- 
Hans Utz - USRA/RIACS at NASA Ames
I do not speak for NASA, and NASA does not speak for me.