Experimental Physics and
Industrial Control System

We have seen a similar problem recently, but only for one IOC.
We found that the date wasn't updating on that particular IOC
- probably to do with our Tornado 2.2 build and/or the ppc
architecture (it's a bit of grasping at straws but it's a
possibility).
Have you checked that the date is updating? Craig found that
running one particular function (I don't remember what it was)
causes the date to stop updating in our case.

Maren

James F Ross wrote:

Hello all,
Another archiver problem discovered, unrelated to my previous messages.
I noticed today that I'm getting "back-in-time stamp" and "Invalid/null
time stamp 0" errors from all my IOCs, starting from April 27th.  The
channels are all connected and the IOCs are running and have been all
this time, but no variables have been archived since this time.  I
wouldn't have noticed if I hadn't been digging into the log files for
another reason.

I don't know much about epics time stamps, nor how the archiver could
fail to read all of them at once for multiple engines.  Any help would
be appreciated.
Thanks!
James