EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Updated Bug Report for EPICS 3.13.5, 3.13.6, 3.13.7, 3.14.beta.N, and 3.14.1
From: "Jeff Hill" <[email protected]>
To: <[email protected]>
Date: Mon, 3 Feb 2003 11:21:15 -0700
> NOTE 3.13.5 may also be subject to this Bug. Jeff can provide
> an answer next week.

A serious bug has been identified in iocCore (see Marty's bug
report from last Friday). I am sending a second report for the
same bug so that I can give some additional details following my
practice for previous patches. Sorry about the delay and
replication - there was some confusion about whether this would
be sent out before or after when a patch release was made
available at APS.

Scope
-----
The bug exists in EPICS 3.13.5, 3.13.6, 3.13.7, 3.14.beta.N, and
3.14.1

Symptoms
--------
An IOC core CA server thread is found to be using excessive CPU
possibly at elevated priority resulting from vxWorks priority
inheritance scheduling. All attached CA clients will permanently
disconnect from the IOC, and record processing may be disrupted
if the threads priority is elevated. Elevated priority was
observed in the one situation where the bug has been reproduced
and identified so far.

Circumstances Related to Relative Rates of Occurrence
-----------------------------------------------------
This bug will occur only in circumstances where the CA client
deletes a monitor subscription and the client's CA event queue in
iocCore is 100% full. Since there is always space for each
monitor subscription to place at least one update on the queue
then the queue can only be full if *all* subscriptions for a
particular client are sending subscription update events at a
rate that is faster than the client is reading them. Therefore,
specialized clients that have a small number of regularly updated
subscriptions and that do not regularly call ca_poll() or
ca_pend_event() are more likely to reproduce the problem.

Fix
---
A patch has been committed to CVS for both R3.13 and R3.14. Patch
releases will be announced soon.

Jeff

> -----Original Message-----
> From: Marty Kraimer [mailto:[email protected]]
> Sent: Friday, January 31, 2003 2:04 PM
> To: [email protected]
> Subject: Bug in 3.13.6, 3.13.7, and 3.14
> 
> A bug exists in these releases that can cause an infinite loop
> on iocs.
> The condition can happen if a Channel Access clients has many
> undelivered
> monitors when it cancels events.
> 
> The "Bug and Fix reports" for each release provides a link to a
> patch file which
> fixes the problem. This can be used to patch existing releases.
> We will also
> generate release R3.13.8 next week. A new 3.14 release will be
> generated when
> changes being made to the PCAS are completed.
> 
> 


References:
Bug in 3.13.6, 3.13.7, and 3.14 Marty Kraimer

Navigate by Date:
Prev: Building Epics base with Tornado 2.2 Ronald L. Sluiter
Next: EPICS base release 3.13.8 Marty Kraimer
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Bug in 3.13.6, 3.13.7, and 3.14 Marty Kraimer
Next: Building Epics base with Tornado 2.2 Ronald L. Sluiter
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·