Feb 23 2004, v2.0
This text was adapted for Moeller Polarimeter DAQ from page
Hints on Hall A Data Taking with CODA 2.2 from Bob Michaels
This file: www.jlab.org/~moller/guide1_linux.html
I. How to Run CODA 2.2 Moeller Polarimeter DAQ on Linux box
Preface:
The Moeller Polarimeter DAQ uses computer 'adaql1'.
The MSQL database server still running on computer 'adaql1'.
The name of database is 'mollerpol_l' and session name is 'moller'.
The MSQL database listening port(MSQL_TCP_PORT) set to 8301.
The database is stored in directory /home/moller/MSQL (since Dec 2002)
If there is no MSQL server running on adaql1 one should login
to adaql1 using 'moller' account and type to start it:
'run_daemon_db msqld &'
We are using different database for running DAQ on Sun's and Linux mashines.
To change database one should modify boot file for Moller Detector
VME crate (hallavme5) : ~moller/vxworks/hallavme5_c2.boot.
Uncomment(delete symbol '#' at start of line) next lines:
......
#putenv "EXPID=mollerpol_l" - set name of MSQL database
......
#putenv "CMLOG_PORT=8229"
#putenv "CMLOG_HOST=adaql1"
......
#bigendian_out=0; - set bytes order to litle endian(PC)
Reboot VME crate 'hallavme5'.
Then need to synchronize run number for this database.
To do this one should start 'dbedit' , select database
with name "mollerpol_l" select table "sessions"
and modify run number to the last run number in database "mollerpol".
To change bytes order there is another possibility to do this remotely:
Type on any of 'adaq**' computer 'dpwish'.
Then in dpwish terminal type next command:
DP_ask ROC11 configure -bigendian_out 0
To check value of 'bigendian_out' type next command:
DP_ask ROC11 cget -bigendian_out
ROC11 - is name of 'hallavme5' in MSQL database.
Value 1 is corresponded ROC connections to big endian mashine(Sun)
Value 0 is corresponded ROC connections to litle endian mashine(PC).
1. Use the "moller" account on "adaql1" or "adaql2" computer.
2. Start Moller Polarimeter Slow Control program:
Before staring "mpc" program one should start "cdevGateway" server
on adaqs2 using moller account.
To start cdevGateway server login on 'adaqs2' and type "cdevGatewayStart".
The cdevGateway server port for connection is set now to 9573.
On "adaql1" change directory to '~/Java/msetting' and type "mpc".
The settings parameters for detector electronics loaded from file:
~moller/daq/coda2/DETSET_new
If computer that is used to start "mpc" program is different from adaql2,
one need to set DISPLAY enviroment to propper value.
There is also 'mollerSetting' program for detector electronics setting
on Linux platform.
If 'mollerSetting' is used instead of the program
for detector electronics setting in 'mpc' the last will not working and
one need to modify script:
'~moller/daq/coda2/scripts/getruninfo'
The next line should be uncomment(remove symbol '#' at begin of line):
.......
# if [ 0 -gt 1 ]
......
And the next line should be comment(place symbol '#' at begin of line):
.............
if $jv -classpath /home/moller/Java/msetting MPCclient $h $p \
"get detectorinfo" > $CODAPATH/RunInfo/detector.set
.............
It is necessary to get information about detector settings from
right source.
3. To start CODA on the workstation, type "runcontrol".
If ET system not running on adaql1, type in another
terminal 'et_start moller &'.
When CODA window pop-up, press the Connect button.
After "connect", wait 20 seconds, then to select or change
configurations, use the "Configuration" button.
Choose the configuration from the 'Run Type' dialog box,
then press "Download".
The relevant configurations are:
beam_pol -- to make beam polarization measurements,
(the sign of current in Helmholtz coils change to
opposite for every run)
moller_data -- the same as configuration 'beam_pol', but
current in Helmholtz coils setup to zero.
moller_puls -- to do a pulser run, testing of DAQ, and get pedestals.
moller_LED -- to do a run with flashing LED, and detector testing.
If you've been running, you will first have to push "Abort" button.
4. To start and stop runs, push the buttons "Start Run" and "End Run".
Start Run is the same as the sequence prestart, go.
After starting run you getting data with events rate about 300-500 Hz,
and data rate 30-50 kByte/sec.
* Where is the data ?
The all data save to the directory /adaql1/data1/moller/raw/
The names of data files for every type of run are:
Run type file name
"beam_pol" moller_data_NN.dat
"moller_data" moller_data_NN.dat
"molelr_puls" moller_data_puls_NN.dat
"molelr_LED" moller_data_led_NN.dat
Where NN is number of run.
* Disk Handling ?
Storage to MSS and purging of disks is done automatically.
Things to *NOT* do:
Do *NOT* erase or move data files (even if you think its junk).
Do *NOT* try to change run numbers.
Do *NOT* change names of run files.
Do *NOT* try to put data by hand into MSS.
It takes a couple hours for files to appear in MSS, and the
files remain on adaq disk for typically 1 or 2 days.
II. Resets of CODA and Cold Starts
Problems with CODA 2.2 can usually be solved with a simple
reset or with a Cold Start. If not call Bob Michaels or
Dave Abbott.
RESETS:
The DAQ uses hallavme5(ROC11). The halladaq14 is used only
for target control (to set current in Helmholtz coils, to do
the target magnetization measurement and the temperature of target
measurement).
If the ROC seems really frozen, press the green button on the
panel of buttons in the center room of the counting house.
The buttons are labelled "moller detector VME #2","moller target VME#1".
Hold the button about 10 seconds.
If you reboot the ROC, or if something on the workstation is
hung up, try pushing the "Reset" button in runcontrol (NOTE:
If "cancel" button is available, you should first push "cancel"
before trying "reset", else reset may do nothing). Then
configure by pressing "Configure" and choosing the configuration
you want (presumably the "beam_pol"), download, and
Start Run. That should work, but if not try a cold start (see
below).
COLD START
First kill all process on the workstation (adaql2)
by typing from anywhere on the adev account "kcoda". This stops
runcontrol, the event builder, event recorder, the runcontrol
server, and cleans up the ET system. Now you can start everything
again by typing "runcontrol" but it is first a good idea to
make sure the VME crates (i.e. the ROCs) are running.
One finds that when resetting, one can frequently avoid rebooting
the frontend crates, and just restart runcontrol, but if you must:
Press the green buttons for ROC11
as explained in the Reset section above.
Then press the Connect button.
After "connect", wait 20 seconds, then press "Configuration" and then
"Run Type"; a dialog
box pops up and you must choose the configuration you want, which
is presumably "beam_pol" for make beam polarization measurement.
Then press download and wait about 30 seconds. Now you can "Start Run"
to start a run.
On rare occassions you may need to "Reset" after configuring the
"Run Type" (if msql database gets confused about state of things).
THINGS THAT GO WRONG WITH DAQ or COMPUTERS:
1. Lots of deadtime -- As of July 98, we found a bad cable in
the SCSI chain which tricked the operating system into lowering
the rate on SCSI. With the solved, we hopefully won't see
deadtime fluctuations at high rate when we write to MSS.
Still, there are a few other causes: a) Rapid disk I/O to
the same disk partition as where we write data. We avoid
this now. b) Huge event sizes. Normal is ~1 kbyte. If
much larger, then maybe you have no pedestal suppression, or
oscillations in preamps, pileup in TDC, etc.
2. Computer problems: First try calling Bob Michaels.
You may also call the computer-center expert on-call
(ask the guard shacalling Bob Michaels.
You may also call the computer-center expert on-call
(ask the guard shack, extension 4444, to do it).
If Bob Michaels
isn't around, try calling Mark Jones or Dave Abbott
for CODA problems.
III. How to Program the RAM of the ROCs.
This is a skill you RARELY need but I add it here for completeness.
You may skip reading the rest of this.
Since we rotated VME computers in July 98 (the newest fastest are
deployed, the oldest slowest are spares), and since the 100 Mb VME
computers are on their own subnet, the old CODA 1.4 setup is dead.
It could be revived, if needed. Here's the 2.2 setup
To connect to VME crates the serial port server is used.
Type 'telnet hatsv5 2004' for connection to 'hallavme5'.
Type 'telnet hatsv5 2003' for connection to 'halladaq14'.
Enter username: adaq
And password: ******
For hallavme5
This is for Moller detector DAQ
VxWorks (for Motorola MVME2300 - MPC 604r) version 5.3.1.
Kernel: WIND version 2.5.
boot device dc
host adaqs3
file adaqs3:/home/moller/vxworks/vx2306_1
(e) 129.57.192.31:ffffff00
(h) 129.57.192.45
(u) moller
(f) 0x20
(tn) hallavme5
(s) /home/moller/vxworks/hallavme5_c2.boot
For halladaq14
This is for Moller target control
VxWorks (for Motorola MVME162) version 5.2.
Kernel: WIND version 2.4.
Made on Thu Aug 1 12:46:31 EDT 1996.
Boot line:
boot device ei(0,0)
host adaqs3
file adaqs3:/home/moller/vxworks/vx162-8MB
(e) 129.57.164.39
(h) 129.57.192.45
(u) moller
(f) 0x20
(tn) halladaq14
(s) ~moller/vxworks/halladaq14.boot
Reminder of vxWorks operations.
You are either in the mode with a prompt like
[VxWorks Boot]:
or a prompt
->
In the former case, you change the NVRAM by typing "c"
(without my quotes), and typing the new fields to the
right of the old fields. To reboot, you type "@".
In the latter case you type at the -> prompt, "bootChange"
to change NVRAM, and type "reboot" to reboot.
This page maintained by moller@jlab.org