Restarting the TPOL DAQ

This page contains information how to restart the TPOL after a crash.

Restarting the online DAQ

NOTE: some of the tasks listed below have to be done from tpol02, some have to be done from lytpol04.
Also note that tpol01 is an alias for lytpol04, so You can use any of the two names to telnet to the VME processor.

Verify that the online statue-database process is still intact:
- Start the TpolMonitor
- If it connects to the TPOL DAQ, the database process is still fine
- If it does not connect, You have to restart the online status-database process before going on.

Try to kill the remaining parts to the DAQ: pol2000@tpol02:~/daq > linux/super password: (give the password for expert access) 08.03.2002 20:04:36 Sync@lytpol04.desy.de:0 MsgIdent SUCCESS connected to Super@tpol02.desy.de:2241 08.03.2002 20:04:36 Super@tpol02.desy.de:2241 MsgIdent SUCCESS connected to Sync@lytpol04.desy.de:4444 Exit 08.03.2002 20:05:14 local main LOCAL connection to Sync@lytpol04.desy.de:4444 down Note that Exit has to be typed with a capital E.

If the synchronize process is down or in a bad state, the program will terminate directly after typing in the password:

pol2000@tpol02:~/daq > linux/super
password: (give the password for expert access)
08.03.2002 20:08:41 local main LOCAL connection to Unknown@*:0 down

If You mis-type the password, the program will also terminate directly after typing the password, but it knows that it was connected to the Sync process: pol2000@tpol02:~/daq > linux/super password: (give the wrong password here) 11.03.2002 19:57:09 local MsgIdent SUCCESS connected to Sync@lytpol04.desy.de:4444 11.03.2002 19:57:09 local main LOCAL connection to Sync@lytpol04.desy.de:4444 down

verify that no DAQ process is hanging around on lytpol04: lytpol04:/home/pol2000/daq > ps x pid ppid pgrp pri text stk data time dev user S name 37 108 37 17 60 12 36 0.04 ttyp0 pol2000 C /bin/ps 108 119 108 17 512 12 136 0.80 ttyp0 pol2000 W /bin/bash 24216K/129988K free physical/virtual, 768K used (in this display) If some DAQ processes are running, it looks like: lytpol04:/home/pol2000/daq > ps x pid ppid pgrp pri text stk data time dev user S name 3 108 3 17 60 12 36 0.02 ttyp0 pol2000 C /bin/ps 6 1 23 17 1256 40 284 0.27 pol2000 W /home/pol2000/daq/lynxos/connec 8 1 23 17 1128 16 700 0.29 pol2000 W /home/pol2000/daq/lynxos/synchr 10 1 23 17 1100 16 168 0.18 pol2000 W /home/pol2000/daq/lynxos/vmerea 66 1 23 17 1000 8 128 0.04 pol2000 W /home/pol2000/daq/lynxos/scaler 108 119 108 17 512 12 136 0.81 ttyp0 pol2000 W /bin/bash 134 1 23 17 1288 16 824 0.12 pol2000 W /home/pol2000/daq/lynxos/slowde 143 1 23 17 2356 24 316 0.30 pol2000 W /home/pol2000/daq/lynxos/pilot 147 1 23 17 1176 32 184 0.12 pol2000 W /home/pol2000/daq/lynxos/analys 7812K/117928K free physical/virtual, 12828K used (in this display) In this case it is best to kill the synchronize process (process number 8 in the above example), then check again and kill the other processes if necessary: lytpol04:/home/pol2000/daq > kill -KILL 8 lytpol04:/home/pol2000/daq > ps x pid ppid pgrp pri text stk data time dev user S name 86 108 86 17 60 12 36 0.03 ttyp0 pol2000 C /bin/ps 108 119 108 17 512 12 136 0.90 ttyp0 pol2000 W /bin/bash 24216K/129988K free physical/virtual, 768K used (in this display)

restart the daq from the PC pol2000@tpol02:~/daq > linux/startdaq lytpol04: prio -5 lytpol04: cd /home/pol2000/daq lytpol04: prio +5 /home/pol2000/daq/lynxos/synchronise synchronise.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/scaler scaler.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/slowdev slowdev.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/connectmfcc connectmfcc.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/analysis analysis.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/vmeread vmeread.log 2>&1 & lytpol04: prio +5 /home/pol2000/daq/lynxos/pilot pilot.log 2>&1 & /home/pol2000/daq/linux/connecthera connecthera.log 2>&1 & /home/pol2000/daq/linux/cleanup cleanup.log 2>&1 &

switch on the HV and enable the auto-pilot

Restarting the online status-database

Try to shut down the old database process pol2000@tpol02:~/daq > linux/dbmon 08.03.2002 21:06:47 local MsgIdent SUCCESS connected to Statusdb@tpol02.desy.de:4443 Exit 08.03.2002 21:07:01 local main LOCAL connection to Statusdb@tpol02.desy.de:4443 down Note that Exit has to be typed with a capital E.

If the statusdb program is down or in a bad state, dbmon will terminate directly after starting it:

pol2000@tpol02:~/daq > linux/dbmon
08.03.2002 21:08:49 local main LOCAL connection to Unknown@*:0 down

Verify that no statusdb process is running on the PC pol2000@tpol02:~/daq > ps ax | grep statusdb 18961 pts/4 S 0:00 grep statusdb If a statusdb process is running, kill it, eg.: pol2000@tpol02:~/daq > ps ax | grep statusdb 18965 pts/4 S 0:00 linux/statusdb 18969 pts/4 S 0:00 grep statusdb pol2000@tpol02:~/daq > kill -KILL 18965 pol2000@tpol02:~/daq > ps ax | grep statusdb 18971 pts/4 S 0:00 grep statusdb

Restart the statusdb program pol2000@tpol02:~/daq > linux/startdb linux/statusdb statusdb.log 2>&1 &

After restarting the online-database, it is best to restart the DAQ as well.

David South

Last modified: Mon Nov 7 14:57:05 CET 2005