Re: Compton tripped [BLAST_SHIFTS] 06/12/03 A-shift summary

From: Bill Franklin (wafrankl@mit.edu)
Date: Thu Jun 12 2003 - 12:19:19 EDT


When I went to the counting bay a little while ago, I found that the
Compton DAQ was hung. This happens when it is trying to broadcast events
but can't establish a connection with the ca_server. We are still
learning about the performance of the event transfer mode, but the surest
recovery sequence in this case is to:

        1. Kill the processes for the Compton DAQ and
         analyzer(pol-control and caRepeater on dblast09).

        2. Reboot the Compton IOC.

        3. Restart CODA and ET.

        4. Restart pol-control for the Compton.

        5. Start runs for CODA and the Compton. Make sure that Compton
        events are being received.

Improved documentation for the Compton is underway and a set of updated
instructions for operating the polarimeter will be in the counting bay by
the end of the day.

Bill

On Thu, 12 Jun 2003, Karen Dow wrote:

>
> warnevent will tell you which type of event (physics, scaler, EPICS or
> Compton) is missing, for each 10 second cycle.
>
> Bill Franklin tells me Compton ships data to ET about every 10
> seconds. Since warnevent works on a 10 second interval, it's possible
> that Compton information could be missing for a cycle, producing the
> clapping. But Compton shouldn't be missing for EVERY cycle. And
> Compton writes a message about an et failure on the blast03 screen when
> there is truly a problem.
>
> Any time you restart ET you have to restart visual_scal, ca_server and
> pol-control. Presumably this was done after run 834 when CODA was
> restarted due to ER problems.
>
> I exited Compton and restarted pol-control after run 842 this morning.
> Compton attached to ET successfully. Still do sometimes get clapping
> from warnevent, but not every cycle. Maybe warnevent should be changed
> to a 15 second cycle, to guarantee at least 1 Compton per cycle?
>
> Karen
>
> zhangchi wrote:
> >
> > We just realized that at certain point during the shift problem did really
> > occured to compton. Because the clapping warning was on since the very
> > beginning of the shift, while compton was OK, we failed to notice it when
> > compton really failed.
> >
> > Chi, Adam
> >
> > On Thu, 12 Jun 2003, zhangchi wrote:
> >
> > >
> > > Continued data taking with unpolarized H2, half wave plate in and out by
> > > fills. First two runs 833, 834 were bad due to problems with
> > > EventRecorder. restarted Coda and problem seemed go away. took data and
> > > crunched into /net/data/4/Analysis/pro2003/before_ff as the last shift
> > > did.
> > >
> > > Although Compton seems to being running, the warning system continued to
> > > complain about no compton data. Had to turn down the volume and ignore the
> > > warning.
> > >
> > > Adam, Chi
> > >
>

-- 
**********************************************************************
Wilbur A. Franklin			*  Phone: 617-253-9518
Bates Linear Accelerator Center		*  FAX:	  617-253-9599
P.O. Box 846				*  Email: wafrankl@mit.edu
Middleton, MA 01949			*
**********************************************************************



This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST