EPICS Home

Experimental Physics and Industrial Control System


 
1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  <20132014  2015  2016  2017  2018  2019  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  <20132014  2015  2016  2017  2018  2019 
<== Date ==> <== Thread ==>

Subject: Re: asyn+streamDevice hangs with synchronousLock:Yes
From: "Brian McAllister" <mcallister@mit.edu>
To: Mark Rivers <rivers@cars.uchicago.edu>
Cc: "tech-talk@aps.anl.gov" <tech-talk@aps.anl.gov>
Date: Thu, 04 Jul 2013 10:18:00 -0400
>>> On 7/4/2013 at 6:38:42 -0000, Mark Rivers wrote:

  > Hi Brian,

  > I didn't see a response to my previous 2 questions:

I was working with other parts of the system.  Then my desktop died, then
we had a power failure...

I left it running last night.  It took longer than most previous runs, but
eventually did (~10 hours).  One difference at the moment is that I have an
asynRecord instance watching the same input - initially with I/O Event, but
then Passive for the duration.  I tried that before and (small sample size)
it seems to delay the problem.

  >> When it is hung please do
  >> epicsMutexShowAll 1

istCooling> epicsMutexShowAll 1
ellCount(&mutexList) 117 ellCount(&freeList) 12
epicsMutexId 0x894b988 source ../../asyn/asynDriver/asynManager.c line 1868
epicsMutexId 0x89a2710 source ../../../src/libCom/osi/epicsMutex.cpp line 222

The IOC is still in that state, if there's a way to get more info from it.

  >> If you simplfy your problem to a single record does the problem go
  >> away?

When I can, I intend to try all variations.  As it takes hours (at least)
for this to fail, concentrating on it makes it difficult to progress with
the other devices (reboots required).

Is there any possiblity that making only one of the records I/O Event and
FLNKing to the others could change the behavior ?

- brian

References:
RE: asyn+streamDevice hangs with synchronousLock:Yes Mark Rivers

Navigate by Date:
Prev: Re: pvGet timeout in sequencer 2.1.12 Benjamin Franksen
Next: Re: pvGet timeout in sequencer 2.1.12 Benjamin Franksen
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  <20132014  2015  2016  2017  2018  2019 
Navigate by Thread:
Prev: RE: asyn+streamDevice hangs with synchronousLock:Yes Mark Rivers
Next: Re: asyn+streamDevice hangs with synchronousLock:Yes Brian McAllister
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  <20132014  2015  2016  2017  2018  2019