Experimental Physics and
Industrial Control System

Tim Mooney <[email protected]> · Wed, 26 Sep 2007 08:17:04 -0500

Heinrich du Toit wrote:

In auto save documentation it says "Autosave is not completely
bulletproof" and that at APS you have problems at times.
Can you maybe explain to me what the effect of these problems are?
Does the IOC stop working? does it load "corrupted" data or something.
Or does it simply not completely restore?
Thanx
-Heinrich

Not restore.
Most of the data I have is for the version of autosave in synApps 4.6.
For that version, problems include failure to save parameter values
because the save task crashed, the task had a stale file handle, one
save operation took long enough to overlap with the next save operation,
or unknown causes.  I don't have any record of a released version of
autosave restoring corrupted data, or causing some other problem in an
ioc.  It's all about failure to restore parameter values, and this is
nearly always caused by an earlier failure to save them.
For recent versions of autosave, I have so far two instances of failure
to save parameter values, and complaints about unwarranted error messages.
There is more version-specific information on the autosave web page, and
in release notes:
www.aps.anl.gov/aod/bcda/synApps/autosave/autosave.html

--
Tim Mooney ([email protected]) (630)252-5417
Beamline Controls & Data Acquisition Group
Advanced Photon Source, Argonne National Lab.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System