Re: alwayson causes assertion

On Tue, 6 Jul 2010, Rich Mattes wrote:

> I tried the config file you attached and was able to recreate the problem.
> I'm not sure where it stems from, but there are significant differences
> between CameraCompress and Passthrough.  The largest of which is that they
> inherit separate driver classes: cameracompress uses the ThreadedDriver
> class but passthrough is using RemoteDriver, which looks like a wrapper
> around the Driver class.
>
> I'm able to get a good backtrace from the assertion error.  It probably
> should be fixed, but is there any reason you need the driver to be alwayson?
> If the underlying driver that passthrough is relaying is set to alwayson, it
> should mitigate any initialization time problems.
>
> Rich
>

Yup, it is unusual situation when passthrough needs to be alwayson and 
this might be the reason we didn't find the bug so far. I wanted to test 
if new code (patch is now on a tracker) affects communication capabilities 
of Player itself, so I needed to force data flow. Passthrough seemed to be 
natural choice for me and alwayson is good way to force a driver to do 
its job. Since cameracompress worked correctly, the test can be considered 
as passed, however we have new not-so-important bug in passthrough 
discovered.

Paul

> -----Original Message-----
> From: Paul Osmialowski [mailto:[email protected]]
> Sent: Monday, July 05, 2010 4:08 PM
> To: [email protected]
> Subject: [Playerstage-developers] alwayson causes assertion
>
> I don't know if anyone mentioned that so far, but looks like setting
> alwayson to 1 for passthrough driver ends always with assertion. I tried
> three different operating systems and it's always the same. Strangely
> enough, I had no such problem with cameracompress driver. How they differ
> inside?!
>
> example cfg file that always fails:
>
> driver
> (
>   name "passthrough"
>   requires ["vlab.pjwstk.edu.pl:6888:blobfinder:0"]
>   provides ["blobfinder:0"]
>   alwayson 1
> )
>
> result:
> Registering driver
> Player v.3.0.2
>
> * Part of the Player/Stage/Gazebo Project
> [http://playerstage.sourceforge.net].
> * Copyright (C) 2000 - 2009 Brian Gerkey, Richard Vaughan, Andrew Howard,
> * Nate Koenig, and contributors. Released under the GNU General Public
> License.
> * Player comes with ABSOLUTELY NO WARRANTY.  This is free software, and
> you
> * are welcome to redistribute it under certain conditions; see COPYING
> * for details.
>
> accepted TCP client 0 on port 6888, fd 3
> listening on 6665
> Listening on ports: 6665
> player: /root/psg3/src/player-3.0.2/libplayercore/message.cc:717:
> MessageQueue* QueuePointer::operator->(): Assertion `Queue' failed.
>
> Whenever I set alwayson to 0, it starts fine, I can connect playerv to it,
> subscribe and all blobfinder data are rendered correctly!
>
> example cfg file that works fine:
>
> driver
> (
>   name "cameracompress"
>   requires ["vlab.pjwstk.edu.pl:6888:camera:0"]
>   provides ["camera:0"]
>   alwayson 1
> )
>
> Paul
>
> ----------------------------------------------------------------------------
> --
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Playerstage-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/playerstage-developers
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Playerstage-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/playerstage-developers
>

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first