Main INDEX
Monthly INDEX
PREV
NEXT
User name R. Michaels
Log entry time 12:53:14 on August 6,2001
Entry number 69594
keyword=Recovery from adaql2 crash
Recovering CODA after adaql2 was problematic. There were these problems:
1. The start-run TCL scripts would not pop up. This happened yesterday
also (same for end-run TCL). Solution is below.
2. coda_er would not report status. I had forgotten to start rcServer by hand.
3. When I restarted Parity DAQ I saw a "stale NFS handle" for /adaql2/data2.
I guess we should not rely on adaql2 for Parity DAQ. I rebooted adaqcp and
this was ok.
4. The parity DAQ run number had been reset somehow to 1765 and I unfortunately
over-wrote this file !. So Parity run 1765 is now junk. Reset to 1770.
Runs 2638 & 9 and 21648 & 9 are junk. Runs 2640 and 21640 are production.
-------- Recovery procedure ----------
1. Before ssh into adaql2, type "xhost +" from adaqs2, then
"setenv DISPLAY adaqs2:0.0". I know this is not supposed to be
necessary, but after this I solved the TCL window popup problem (twice
it was the "solution").
2. Remember to run "rcServer" interactively on Linux based DAQs.
(On SunOS it does not matter if you do or not.)
3. After configuring DAQ, I like to press "reset" button (top left corner).
My unscientific impression is this helps.