g+
g+ Communities
Argonne National Laboratory

Experimental Physics and
Industrial Control System

2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014  Index 2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
<== Date ==> <== Thread ==>

Subject: IOC Redundancy in R3.14.10
From: "Schoeneburg, Bernd" <bernd.schoeneburg@desy.de>
To: core-talk-bounces@aps.anl.gov
Cc: Bob Dalesio <bdalesio1@comcast.net>
Date: Thu, 03 Jul 2008 16:01:13 +0200
Dear Andrew, dear all,

I want to remind you that we are highly interested in base modifications for IOC redundancy. For sites which are not interested in redundant ioc operations there would be no additional code overhead if the additional code is in "#ifdef"'s. Nine files of the epics base are effected by these changes (mainly small ones).

The proposed model gives maximal freedom of usage.

-- If the define (REDUNDANCY) is not done, Episc runs as usual and no additional object code is loaded.

-- If REDUNDANCY is defined (base extensions included) but not any additional redundancy software (RMT) loaded, Epics is running as usual. If you want to use some of the new features on some but not all your ioc's, then this is important.

-- If you add the redundancy software RMT (= redundancy monitor task), even on non-redundant ioc's, you are able to get information from all tasks which have registered to the RMT registry. If you use the sequencer extended to support RMT, you can debug your SNL programs. A CSS-plugin is available for this. Debugging allows you to:
	- view programs, state sets, variables in a tree
	- view and change (force) unconnected and connected variables
	- view and change current state
	- start, stop and single-step each state-set
	- use CSS object contribution with variables

-- If you have a second IOC as a redundant hardware, the RMT on the ioc will manage the redundancy state of the ioc. It will control updating the non-selected one and starts or stops activities during switch-over. During normal operation records and state programs are synchronized between the ioc's so that a switch-over is relativly smooth.

Some more details have been described on the last Epics Meetings.

We think that our changes are ready(*) so far. We have continuously tested the redundant ioc pair for 40 days on a Intel-based PCI plattform. In November we had already a discussion with Andrew and others. We have fixed some weak points meanwhile.

We are convinced that we now have reached the state in which redundancy should become part of the next Epics release. Any feedback is welcome.


Bernd


(*) The definition of the hook pointer still has to be placed into the proper file of the Epics base (task for Andrew).
Replies:
Re: IOC Redundancy in R3.14.10 Andrew Johnson

Navigate by Date:
Prev: Re: Plans and Dates for 3.14.10 Release Andrew Johnson
Next: Re: IOC Redundancy in R3.14.10 Andrew Johnson
Index: 2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
Navigate by Thread:
Prev: 64-bit patch committed Andrew Johnson
Next: Re: IOC Redundancy in R3.14.10 Andrew Johnson
Index: 2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
ANJ, 02 Feb 2012 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· EPICSv4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·