Experimental Physics and
Industrial Control System

"Allison, Stephanie" <[email protected]> · Fri, 18 Jan 2002 10:02:26 -0800

Hi Geoff,

> One suggestion to handle 
> these situations was to only signal an error afer N number of bad 
> readings in a row.  Has anyone attempted to handle this 
> situation with EPICS?  Any suggestions?

I know you don't use cmlog, but we use a special version of cmlog 
throttling for situations like this where you just want to prevent
the error log from being flooded until the problem is resolved.
The special throttling allows on-the-fly add/remove message throttling
based on a substring of the text being logged.  Right now, one has
to login to the ioc and type a command at the prompt.  We plan to
provide an EPICS interface (subroutine records) for throttling
from an EPICS display.

Anyway, Jie Chen is considering adding this feature into the next
version of cmlog.  Mike Laznovshy and Ron MacKenzie did the work
and have documentation with more detail:

Programmer's Guide:
http://www.slac.stanford.edu/grp/cd/soft/cmlog/slac_cmlog_throttle.html

Requirements:
http://www.slac.stanford.edu/grp/cd/soft/cmlog/cmlog_throttling_needs.html

Stephanie Allison

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System