Make New Entry,
Make Followup Entry
User name R. Michaels
Log entry time 14:54:18 on March28,2014
Entry number 387886
keyword=ET insertions stopping
Something is causing the ET-inserted events (EPICS and Scaler event type 140) to
stop after some time. E.g on run 1285 they stopped after about 1.2 hours.
The scripts that loops are evidently dying. Not sure why yet. For the scalers I
modifed the script to put stderr to /dev/null so it probably won't drop out of the
loop. It might be our infamous "run out of socket" problem again. This can
be alleviated by asking for data less often, so now I ask at 0.05 Hz instead of 0.5 Hz.
We'll see if any of this helps.
One issue with the EPICS script is that both L-HRS and R-HRS runs on one account
and they are stomping on each other by writing the same temporary file.
In general, these scripts are a mess because we made in several cases 2 copies,
one with "L" and one with "R" in the name, but there should be only ONE script
and it can differentiate spectrometers from the $SESSION -- the same way the
DAQs do. It is easier to maintain one script than two. I'll need to rewrite this
after this run.