01/13/2003: Added directions to reset Latchup in VME and silicon. Also, not required to veto triggers before starting a run, so updated the START_RUN section.

01/23/2003: Added GUI documentation.

Instructions for running DAQ using the GUI

General Information:

Since a picture is worth a thousand words, here's an annotated screendump showing different parts of the GUI. The big red labels describe the approximate functions of different sections of the GUI.

Most control operations are carried out from the top left 2 panels, for example:

Setting up a Run type
Starting / Stopping Runs
Using the Expert Menus to Reset Silicon Latchup or Restart various DAQ components

Launching the GUI

If not already logged in, Login to the DAQ X-Term computer in the center of console with USER='runctrl', PASSWD='10oclock'
You should see an icon titled 'PhatDAQRunCtrl' in the center of your desktop (desktop ONE).
Double-Click on this icon to launch the GUI. Hit 'OK' on the options dialog that appears, and you should see a 'Run' window start up.
The Run window should display the following messages:
In particular, the six logfiles for the various subsystems should be opened automatically, and the two command connections to PhatDAQ and EMM should be initiated. If the messages stop at any point in the above 'Run' window, this usually indicates that particular subsystem is busy/refusing connection; you should call the DAQ expert on call in this case, and let him/her know the contents of the above window.
If all the connections noted in the 'Run' window are made successfully, the PhatDAQRunControl GUI window should start up.
All the SubSystem windows should update with various messages (the details of these messages are not important for the end-user). All initializations are performed automatically over a period of about 30 sec, and the various messages indicate these initializations are in progress.
System readiness to take data is indicated by the following message in the 'Silicon ROC Messages' window:
At this point, you can proceed to the Start/Stop Run Section.

Starting/Stopping a RUN

Select Runtype from Combo box
Choose to Write/Not Write to Disk
Select Appropriate Trigger configuration for current Running.

Guide to selecting Run Types:

Run Type	Detector Setup	Special considerations (See details below)
ALLBEAM	Silicon + Plastic (Normal data taking)	Si BUSY should be removed for first running after Silicon is switched ON.
PADDLECOSM	Plastic Only, NO Silicon (Beam studies with Silicon turned OFF)	None.
SICOSM	PCAL Cosmic running, (Silicon OFF, Don't care about Plastics)	PCAL HV should set appropriately, Time Calibrator should be disabled from Trigger
FBTIME/FBPED	FastBus Time Calibrator and FastBus Pedestal runs	Start with Master VETO set, then disable VETO to get high rate.

Considerations for Each RunType:

ALLBEAM: In general, you should STOP RUN before performing any operation with EITHER Silicon Low Voltage or High Voltage. You can START RUN immediately after the operation, but it is important for DAQ stability that you don't have a RUN going while you are ramping either Low Voltage or High Voltage.

After Silicon Low Voltage is first switched ON, and you are ready to start running, you need to temporarily disable the Silicon Busy signal. This is done by going into the Electronics Room and unplugging a cable marked as 'Silicon BUSY' located in the bottom left trigger rack. You can then hit START RUN, ensure that you are getting events, and go back in to the electronics room and plug the Silicon BUSY back in again. Note that this procedure needs to be followed only ONCE, after you first switch on Silicon Low Voltage, and you must remember to plug back in the Silicon BUSY after starting the RUN.
PADDLECOSM: No special considerations.
SICOSM: To switch to PCAL cosmic running when there is no beam in the machine, you need to load different High Voltage settings into the PCAL (consult PCAL documentation for this). Don't forget to switch back to normal high voltage settings after you are done with the PCAL cosmic running.
FBTIME/FBPED: To run FastBus Time Calibrator and FastBus Pedestal runs, have the TOF expert on call set the range and threshold as appropriate. Before starting the run, set the Master Trigger VETO signal located in the Trigger Rack to block all triggers except heartbeats. After issuing START RUN, you can then reset this VETO so that you start getting triggers through.

After following the above considerations, you can hit the large friendly green button labelled 'START RUN' to start the run. A beep indicates run has started, and the DAQ Status window should start updating with the current RUN in green, the event rate etc. The Run number is 0 if you are not writing to disk.

Contrary to popular belief, this is not the time to open your laptop and check your email! Proceed to see the section on 'What to check for while running', to see what are the things to monitor while DAQ is running and writing data to disk.

Stopping a Run is as simple as hitting the large friendly red button labelled 'STOP RUN'. Wait ~ 5 seconds after stopping a Run to let all the sub-systems return back to steady state before attempting to start a new run. Final system readiness is indicated by the same message 'Waiting for CMD....' in the SiliconROC messages window.

What to check for while running

Check for Green status and increasing event counter in the GUI. The Event Rate indicator should be up at ~ 230 Hz (the maximum rate accepted by the DAQ in case of good beam conditions).
If running ALLBEAM type with Writing to DISK enabled, you should hear a beep everytime the sequence switches.
You should also monitor the bottom two scalers on the visual scaler display which monitor the input/output Level-0 triggers to the DAQ. Both should be up at ~230 Hz commensurate with what is reported in the DAQ status pane.
If silicon is ON, you also need to keep close, continuous watch on the radiation levels reported by RadMon (chipmunk display). Consult Silicon documentation for details on monitoring Silicon health.
You will see the occasional latchup from various Silicon FEC's, and you will need to reset them using the procedure outlined below.

Recovering from Silicon Latchup

When the central alarmer alarms that Latchup has occured, acknowledge the alarm and STOP RUN.
Start a new run without writing so that you keep getting updated data from the Silicon FEC's.
Click on the 'Si-Expert' -> 'Restart ROCDB' menu item to restart RocDB and get a fresh reading from ROCdB containing the latest Silicon slow monitoring values. RocDB takes about a minute to put in a new reading into the database.
Go to the 'Si Latchup VFP Summary' link on the PHOBOS portal to determine which FEC ports have latched up. If there are too many (greater than 8 ports), consult Silicon expert for further instructions - donot reset latchup if there are too many ports latched up as this could trip the Low Voltage power supplies. Consult Silicon Expert in the case of large number of latchups due to adverse beam conditions.
If less than 8 ports are latched up, Click on 'Si-Expert'->'Reset Si Latchup' to reset the latchup.
Click on the 'Si-Expert' -> 'Restart ROCDB' menu item to restart RocDB again and get the latest Silicon slow monitoring values.
Go back to the portal page to ensure that the latched up ports have recovered.
If they have recovered, you can START RUN again.

Known Errors and modes of failure:

‘Bad interrupt 0’ messages in VME ROC window

DAQExpert/Close VME ROC Connection
Push RST button on the leftmost module in VME crate F.
Wait 2’39” or until green light the rightmost module is on and red is off.
DAQExpert/Open VME ROC Connection

‘Cannot bind to port 7002’ messages in PhatDAQ

You have multiple instances of phatdaq running and need to reboot usp.phobos.bnl.gov:

How to reboot usp.phobos.bnl.gov

1. Locate USP window on PhAndrei workstation. (icon USP in left quarter of the taskbar at the bottom of the screen)

2. Login as root (password is in the drawer)

3. #reboot

4. wait until ‘The system is ready’

FASTBUS Dead

Indication:
1. No response to stop run command in ‘Fastbus ROC Messages’.
2. Red light L2 on Fastbus crate, module STR340/SFI

Workaround:
1. Stop run
2. Reboot Fastbus by pressing RST button on module STR340/SFI. It will take 3 minutes.
3. Wait for red light L4 on STR340/SFI.
4. DAQExpert/Close Fastbus
5. DAQExpert/Open Fastbus
6. Start run

GUI is not responding

This is probably due to network problems on usp.phobos.bnl.gov.

Indication:
Open terminal window and execute command:
>ping usp

Repeat pinging until it start responding. It usually takes 1-2 minutes. If more than 5 minutes – call Andrei.

Usually it should continue taking data by itself.

How to run DAQ in console command line mode

General Information:

ALL control operations are done on the DAQ X-Term console (left-center) on the PHOBOS main console. You may occasionally need to go over to the PHOBOSNT computer monitor (on Andrei's desk) to check on the status of a few subsystems.
ALL passwords for logging in to various consoles mentioned below are 10oclock
If you are not already logged in on the DAQ X-Term, login with username 'runctrl'

Setting up console windows on the DAQ xterm.:

You should see an icon on the DAQ xterm desktop called 'StartDAQ_CommandLine', double-click on it and say 'OK' to whatever dialog box appears. A window called 'Run' will appear, that you can close by double-clicking on its top left corner (we won't be using it)
You should now see three windows starting on the desktop, each requiring you to enter the same password. The three windows should have titles 'SiliconROC', 'RunControl' and 'PhatDAQ'. .
After login is complete, you need to change directory to a specific location in each window, by typing in the commands indicated in the following picture:

To Start DAQ:

You need to enter commands in each of the three windows to start scripts as follows. The DAQ programs will be launched, indicated by many verbose messages in each window.

In Window Named	Type in
SiliconROC	source startROC.csh
PhatDAQ	source start.csh
RunControl	source start.csh

In the RunControl window, you will see messages to the following effect, indicating that the control connection is ready:

*******************************************

* *

* W E L C O M E to R O O T *

* *

* Version 2.23/12 1 February 2000 *

* *

* You are welcome to visit our Web site *

* http://root.cern.ch *

* *

*******************************************

CINT/ROOT C/C++ Interpreter version 5.14.25, Nov 25 1999

Type ? for help. Commands must be C++ statements.

Enclose multiple statements between { }.

root [0]

Processing runcontrol.C...

Note: posix.dll is not found. Do 'sh setup' in $CINTSYSDIR/lib/posix directory if you use UNIX.

phatdaq control session

Cannot connect to eb port 7001. Start phatdaq first!

Connecting to 130.199.65.20 ...Done

started.

Use rc_command.C(command,argument) to control phatdaq.

root [1]

System readiness to take data is indicated by a message in the SiliconROC window saying

MasterCE Waiting for START_RUN.....
Master Waiting for command....

You should check with the on-call trigger expert at this point to ensure that the trigger settings are as desired, mainly to ensure that the data you are taking is not crap. During detector commissioning phase, the trigger settings are under study, and you will have to co-ordinate with the trigger expert who may direct you to take different data sets with different trigger settings.
If silicon system has been turned on (Low Voltage and High Voltage) you can now start a run by typing in at the 'RunControl' window: (in the ROOT prompt)

root [2] rc_start(RUN_ALLBEAM+WRITING)
For running without the Silicon detectors for trigger studies, you need to check with the trigger expert on call that the trigger settings are as desired, and type in at the RunControl window:

root [2] rc_start(RUN_PADDLECOSM+WRITING)
For PCAL cosmic running, you need to check with the PCAL expert on call that the PCAL HV settings are as desired, and type in at the RunControl window:

root [2] rc_start(RUN_SICOSM+WRITING)

Also, the time calibrator must be switched off in hardware while running PCAL cosmic.
This should cause triggers to start getting through to the DAQ, indicated for example by yellow and green lights flashing on top of the FastBus crate, located to the left of the trigger rack.
You can also check the run status periodically by typing 'rc_status()' command on the commandline. The online displays should also update continuously with new data, new run and new event numbers.

To Stop the Run When DAQ is Working

root [3] rc_stop_run()

To Start a Run When DAQ is Working

When Silicon is on:

root [2] rc_start(RUN_ALLBEAM+WRITING)

When Silicon is off:

root [2] rc_start(RUN_PADDLECOSM+WRITING)

To Exit from the DAQ:

You can stop the run by typing in at the 'RunControl' window: (in the ROOT prompt)

root [3] rc_stop_run()
root [5] .q
Next, you must hit Ctrl-C once in the SiliconROC window. This should cause the SiliconROC program to stop.
You now need to launch a small sub-program in the SiliconROC to clean up all stale DAQ threads. You do this by typing in the SiliconROC window.

source cleanup.csh

When this cleanup program starts up, the PhatDAQ process in the 'PhatDAQ' window should exit cleanly, saying 'Pollable event occured' and return to the command prompt.
Once the cleanup is finished, the SiliconROC window shows messages like 'Other party is no longer connected...'. At this point you should also kill the cleanup program by pressing Ctrl-C once in the SiliconROC window.
The system is now back in clean state, and you can go back and start a new Run as above.

Known Problems/Errors in Event Building

Warnings (that can be ignored)

PhatDAQ reports occassionally '***Break**** write on a pipe' type messages. These are harmless (online distributor reconnecting) and can be ignored.
A message is printed every 1 sec or so by both PhatDAQ and SiliconROC stating current rate/status etc

Errors (that should NOT be ignored)

Occasionally, the PhatDAQ program will exit with an error message like:
fpdp_event_wait: Error in fpdp_event_wait, Data Stamp DEADFACE != B0EF0805
fpdp_event_wait: Trying to adjust..

This indicates that PhatDAQ has crashed with an unrecoverable error. In this case, you only need to restart PhatDAQ itself, without restarting the SiliconROC. The procedure is:

1) Stop the current run: (in the RunControl window) type: root [2] rc_stop_run

2) Quit from RunControl: (in the RunControl window) type: root [2] .q

3) Restart PhatDAQ: (in the PhatDAQ window) type: source start.csh

4) Restart RunControl: (in the RunControl window) type: source start.csh

During data taking with collisions, the lights on top of the Fastbus crate in the electronics room (visible from the console) should remain continuously lit - they are actually flashing at a high rate indicating data taking is in progress. The online software should continuously update with increasing event numbers. If there are no collisions and you are taking cosmic data, you should see at least heartbeat triggers flashing at 1Hz or so. If this is not the case, check DAQ status by typing 'rc_status()' command in the 'RunControl' window. Consult DAQ expert on call to debug further.
Even after running the cleanup.csh script in the SiliconROC window, PhatDAQ does not exit back to the command prompt. In this case, you should try running the cleanup.csh script again. If this doesn't produce results, you need to contact the DAQ expert on call since this may require a hard reset of the subsystems.

How to Restart rocdb

If Si Hybrids reading on PHOBOS portal gets old for 10 min, but DAQ is taking data and silicon is ON it is probable that the rocdb program, that pulls monitoring information from DAQ and writes it to database has crashed.

To restart this program, do the following:

Locate ROCDB window on DAQ X-Term
(or open new x-terminal by clcking on the 'terminal' icon in the desktop controls.
ssh phobos@usp.phobos.bnl.gov
<passwd will be required>
cd $DAQ/dqout/bin)

Kill currently running rocdb using Ctrl C. If it does not respond then use following command in RunControl window:
.!kill_rocdb

And enter the command:
rocdb 1 0 30

Here the first argument enables connection to the database, second argument enables writing to the file rocdblogMMDD.csv
The program detects if phatdaq is running, in this case it requests the whole event from the phatdaq, extracts monitoring information from the FEC trailer blocks and sends it to the datadase.

To change reporting interval to 60 seconds:
rocdb 1 0 60

How to find out that VME ADC is working properly

2. telnet 130.199.65.18
Trying 130.199.65.18...
Connected to 130.199.65.18.
Escape character is '^]'.

3. vmeroc >ds_po=0x100.

You will see something like:

[0]readV792.2@fa060000: [60]Status1,2=0043,28,EvCnt=000a77
0000:4a002000 4800406c 4810408d 48014078 48114086 4802406d 4812405c 4803407c 0020:48134079 48044086 4814407f 4805407b 48154085 4806407e 4816407d 48074089 0040:4817409e 48084088 48184098 48094082 48194086 480a407f 481a4072 480b409b 0060:481b4061 480c4096 481c408b 480d409d 481d4084 480e4077 481e4078 480f4085 0080:481f4031 4e000a77

V792.2 – correspond to ADC2.
First word 4a002000– header
Last word 4e000a77 – footer
The rest are codes. For example 48 in each word means slot 9. The lower 13 bits are values.

In the above example we see normal pedestals.

4. When you finished, set the printout mode back to normal:
vmeroc >ds_po=0

How to Reset Latchup in a VME ADC

NOTE: This documentation is somewhat obsolete. Andrei will update it.

For now, ignore steps 1&2 and go right to 3. It is also occasionally necessary to cycle the VME power (red button buried under cables on the right side of the power supply which is right below the crate). Note item 5 below also applies if you recycle the power.

1. Locate the troubled ADC, it will have red OVC PWR.

2. Using ‘ADC LATCHUP RESET TOOL’ push a miniature push-button switch PWR (between two cable connectors). You should do it twice until OVC PWR turns green.

3. Reboot the crate by pressing RST button on the leftmost module. Observe that on the rightmost module (AVME 941X) turned red FAIL.

4. In DAQ GUI / DAQEsxpert select Close VME ROC Connection.

5. Wait at least 2 minutes 29 seconds after the pressing RST until the red light FAIL is off and green light PASS is on (both are on the rightmost module in the crate: AVME 941X)

6. In DAQ GUI / DAQEsxpert select Open VME ROC Connection.

7. Stop the run and watch that VME ROC Messages is responding.

If, after starting a run, you see in VME ROC messages: “Bad VME interrupt 0” that means that rebooting was not successful and you need to reboot the crate by pressing ‘RST’ button on the leftmost module. Go to 3.

How to Reset Silicon Latchup

Locate VMEROC window at DAQ X-Term computer and enter the command:
type in the command: (or use Esc+k to bring back recently typed command)

latchup_reset();

How to run without VME (only Silicon and Fastbus)

Use ‘phatdaq 3’ instead of ‘phatdaq’

Micro DAQ

Micro DAQ runs on the PHOBOS VME crate. It waits for the gate on the first ADC, reads all modules and sends event to the usp.phobos.bnl.gov using UDP protocol. The reader program udq running on usp receives the events and write them to the file usp.phobos.bnl.gov:/data/0/udq_yymmdd_hhmmss.udq.

ssh phobos@usp.phobos.bnl.gov
cd /data/0/udqdata
udq 0 60

The first argument controls event printing; if it is 1 then program will print every event. The second argument is the file-change period in minutes.
The program will respond with:

udq - Micro DAQ spooler. Printout=0, File change every 60 min

file udq_031120_165647.udq opened

data socket opened at port 7500 size 30720

StartedlEv:0,rEv:0,d:0,kB:0,ev/s:0

lEv:5,rEv:1650,d:0,kB:2,ev/s:0

lEv:16,rEv:1661,d:0,kB:7,ev/s:1

…

If it does not respond with incrementing events then probably the server program is not running on vmeroc.phobos.bnl.gov. The simplest way to start it – recycle the power on the vme crate. 5 minutes after recycling the udq should report incoming events.

Starting the Micro DAQ when system is set to run normal DAQ

If DAQ is running - close connection from the DAQ GUI to VME ROC using [DAQExpert][Close VME ROC Connection]
telnet 130.199.65.18
Trying 130.199.65.18...
Connected to 130.199.65.18.
Escape character is '^]'

If you get message: “Sorry, this system is engaged” then recycle the power on the VME crate wait 3 minutes and try again.
Disable reacting to DAQ triggers using:
mEMSD_enable(0)
Spoon Micro DAQ:
sp udq_roc_start
Apply gates to ADC0. It should start blinking

When you finished with uDAQ, recycle the power on the crate to have it back for normal DAQ.

Please send all comments and suggestions regarding this page to Andrei

level2 lfo23; tab-stops:list 1.0in'>3. #reboot

4. wait until ‘The system is ready’

FASTBUS Dead

Indication:
1. No response to stop run command in ‘Fastbus ROC Messages’.
2. Red light L2 on Fastbus crate, module STR340/SFI

GUI is not responding

This is probably due to network problems on usp.phobos.bnl.gov.

Indication:
Open terminal window and execute command:
>ping usp

Repeat pinging until it start responding. It usually takes 1-2 minutes. If more than 5 minutes – call Andrei.

Usually it should continue taking data by itself.

How to run DAQ in console command line mode

General Information:

ALL control operations are done on the DAQ X-Term console (left-center) on the PHOBOS main console. You may occasionally need to go over to the PHOBOSNT computer monitor (on Andrei's desk) to check on the status of a few subsystems.
ALL passwords for logging in to various consoles mentioned below are 10oclock
If you are not already logged in on the DAQ X-Term, login with username 'runctrl'

Setting up console windows on the DAQ xterm.:

You should see an icon on the DAQ xterm desktop called 'StartDAQ_CommandLine', double-click on it and say 'OK' to whatever dialog box appears. A window called 'Run' will appear, that you can close by double-clicking on its top left corner (we won't be using it)
You should now see three windows starting on the desktop, each requiring you to enter the same password. The three windows should have titles 'SiliconROC', 'RunControl' and 'PhatDAQ'. .
After login is complete, you need to change directory to a specific location in each window, by typing in the commands indicated in the following picture:

To Start DAQ:

You need to enter commands in each of the three windows to start scripts as follows. The DAQ programs will be launched, indicated by many verbose messages in each window.

In Window Named	Type in
SiliconROC	source startROC.csh
PhatDAQ	source start.csh
RunControl	source start.csh

In the RunControl window, you will see messages to the following effect, indicating that the control connection is ready:

*******************************************

* *

* W E L C O M E to R O O T *

* *

* Version 2.23/12 1 February 2000 *

* *

* You are welcome to visit our Web site *

* http://root.cern.ch *

* *

*******************************************

CINT/ROOT C/C++ Interpreter version 5.14.25, Nov 25 1999

Type ? for help. Commands must be C++ statements.

Enclose multiple statements between { }.

root [0]

Processing runcontrol.C...

Note: posix.dll is not found. Do 'sh setup' in $CINTSYSDIR/lib/posix directory if you use UNIX.

phatdaq control session

Cannot connect to eb port 7001. Start phatdaq first!

Connecting to 130.199.65.20 ...Done

started.

Use rc_command.C(command,argument) to control phatdaq.

root [1]

System readiness to take data is indicated by a message in the SiliconROC window saying

MasterCE Waiting for START_RUN.....
Master Waiting for command....

You should check with the on-call trigger expert at this point to ensure that the trigger settings are as desired, mainly to ensure that the data you are taking is not crap. During detector commissioning phase, the trigger settings are under study, and you will have to co-ordinate with the trigger expert who may direct you to take different data sets with different trigger settings.
If silicon system has been turned on (Low Voltage and High Voltage) you can now start a run by typing in at the 'RunControl' window: (in the ROOT prompt)

root [2] rc_start(RUN_ALLBEAM+WRITING)
For running without the Silicon detectors for trigger studies, you need to check with the trigger expert on call that the trigger settings are as desired, and type in at the RunControl window:

root [2] rc_start(RUN_PADDLECOSM+WRITING)
For PCAL cosmic running, you need to check with the PCAL expert on call that the PCAL HV settings are as desired, and type in at the RunControl window:

root [2] rc_start(RUN_SICOSM+WRITING)

Also, the time calibrator must be switched off in hardware while running PCAL cosmic.
This should cause triggers to start getting through to the DAQ, indicated for example by yellow and green lights flashing on top of the FastBus crate, located to the left of the trigger rack.
You can also check the run status periodically by typing 'rc_status()' command on the commandline. The online displays should also update continuously with new data, new run and new event numbers.

To Stop the Run When DAQ is Working

root [3] rc_stop_run()

To Start a Run When DAQ is Working

When Silicon is on:

root [2] rc_start(RUN_ALLBEAM+WRITING)

When Silicon is off:

root [2] rc_start(RUN_PADDLECOSM+WRITING)

To Exit from the DAQ:

You can stop the run by typing in at the 'RunControl' window: (in the ROOT prompt)

root [3] rc_stop_run()
root [5] .q
Next, you must hit Ctrl-C once in the SiliconROC window. This should cause the SiliconROC program to stop.
You now need to launch a small sub-program in the SiliconROC to clean up all stale DAQ threads. You do this by typing in the SiliconROC window.

source cleanup.csh

When this cleanup program starts up, the PhatDAQ process in the 'PhatDAQ' window should exit cleanly, saying 'Pollable event occured' and return to the command prompt.
Once the cleanup is finished, the SiliconROC window shows messages like 'Other party is no longer connected...'. At this point you should also kill the cleanup program by pressing Ctrl-C once in the SiliconROC window.
The system is now back in clean state, and you can go back and start a new Run as above.

Known Problems/Errors in Event Building

Warnings (that can be ignored)

PhatDAQ reports occassionally '***Break**** write on a pipe' type messages. These are harmless (online distributor reconnecting) and can be ignored.
A message is printed every 1 sec or so by both PhatDAQ and SiliconROC stating current rate/status etc

Errors (that should NOT be ignored)

Occasionally, the PhatDAQ program will exit with an error message like:
fpdp_event_wait: Error in fpdp_event_wait, Data Stamp DEADFACE != B0EF0805
fpdp_event_wait: Trying to adjust..

This indicates that PhatDAQ has crashed with an unrecoverable error. In this case, you only need to restart PhatDAQ itself, without restarting the SiliconROC. The procedure is:

1) Stop the current run: (in the RunControl window) type: root [2] rc_stop_run

2) Quit from RunControl: (in the RunControl window) type: root [2] .q

3) Restart PhatDAQ: (in the PhatDAQ window) type: source start.csh

4) Restart RunControl: (in the RunControl window) type: source start.csh

During data taking with collisions, the lights on top of the Fastbus crate in the electronics room (visible from the console) should remain continuously lit - they are actually flashing at a high rate indicating data taking is in progress. The online software should continuously update with increasing event numbers. If there are no collisions and you are taking cosmic data, you should see at least heartbeat triggers flashing at 1Hz or so. If this is not the case, check DAQ status by typing 'rc_status()' command in the 'RunControl' window. Consult DAQ expert on call to debug further.
Even after running the cleanup.csh script in the SiliconROC window, PhatDAQ does not exit back to the command prompt. In this case, you should try running the cleanup.csh script again. If this doesn't produce results, you need to contact the DAQ expert on call since this may require a hard reset of the subsystems.

How to Restart rocdb

Kill currently running rocdb using Ctrl C. If it does not respond then use following command in RunControl window:
.!kill_rocdb

To change reporting interval to 60 seconds:
rocdb 1 0 60

How to find out that VME ADC is working properly

2. telnet 130.199.65.18
Trying 130.199.65.18...
Connected to 130.199.65.18.
Escape character is '^]'.

3. vmeroc >ds_po=0x100.

You will see something like:

V792.2 – correspond to ADC2.
First word 4a002000– header
Last word 4e000a77 – footer
The rest are codes. For example 48 in each word means slot 9. The lower 13 bits are values.

In the above example we see normal pedestals.

4. When you finished, set the printout mode back to normal:
vmeroc >ds_po=0

How to Reset Latchup in a VME ADC

1. Locate the troubled ADC, it will have red OVC PWR.

2. Using ‘ADC LATCHUP RESET TOOL’ push a miniature push-button switch PWR (between two cable connectors). You should do it twice until OVC PWR turns green.

3. Reboot the crate by pressing RST button on the leftmost module. Observe that on the rightmost module (AVME 941X) turned red FAIL.

4. In DAQ GUI / DAQEsxpert select Close VME ROC Connection.

5. Wait at least 2 minutes 29 seconds after the pressing RST until the red light FAIL is off and green light PASS is on (both are on the rightmost module in the crate: AVME 941X)

6. In DAQ GUI / DAQEsxpert select Open VME ROC Connection.

7. Stop the run and watch that VME ROC Messages is responding.

How to Reset Silicon Latchup

Locate VMEROC window at DAQ X-Term computer and enter the command:
type in the command: (or use Esc+k to bring back recently typed command)

latchup_reset();

How to run without VME (only Silicon and Fastbus)

Use ‘phatdaq 3’ instead of ‘phatdaq’

Micro DAQ

ssh phobos@usp.phobos.bnl.gov
cd /data/0
udq 0 60

The first argument controls event printing; if it is 1 then program will print every event. The second argument is the file-change period in minutes.
The program will respond with:

udq - Micro DAQ spooler. Printout=0, File change every 60 min

file udq_031120_165647.udq opened

data socket opened at port 7500 size 30720

StartedlEv:0,rEv:0,d:0,kB:0,ev/s:0

lEv:5,rEv:1650,d:0,kB:2,ev/s:0

lEv:16,rEv:1661,d:0,kB:7,ev/s:1

…

Starting the Micro DAQ when system is set to run normal DAQ

Close connection from the DAQ GUI to VME ROC using [DAQExpert][Close VME ROC Connection]
telnet 130.199.65.18
Trying 130.199.65.18...
Connected to 130.199.65.18.
Escape character is '^]'
Disable reacting to DAQ triggers using:
mEMSD_enable(0)
Spoon Micro DAQ:
sp udq_roc_start

When you finished with uDAQ, recycle the power on the crate to have it back for normal DAQ.

DAQ, General Information/Help

Contents

Status of this Document

Instructions for running DAQ using the GUI

General Information:

Launching the GUI

Starting/Stopping a RUN

What to check for while running

Recovering from Silicon Latchup

Known Errors and modes of failure:

&lsquo;Bad interrupt 0&rsquo; messages in VME ROC window

&lsquo;Cannot bind to port 7002&rsquo; messages in PhatDAQ

How to reboot usp.phobos.bnl.gov

FASTBUS Dead

GUI is not responding

How to run DAQ in console command line mode

General Information:

Setting up console windows on the DAQ xterm.:

To Start DAQ:

To Stop the Run When DAQ is Working

To Start a Run When DAQ is Working

To Exit from the DAQ:

Known Problems/Errors in Event Building

How to Restart rocdb

How to find out that VME ADC is working properly

How to Reset Latchup in a VME ADC

How to Reset Silicon Latchup

How to run without VME (only Silicon and Fastbus)

Micro DAQ

StartedlEv:0,rEv:0,d:0,kB:0,ev/s:0

Starting the Micro DAQ when system is set to run normal DAQ

Please send all comments and suggestions regarding this page to Andrei

FASTBUS Dead

GUI is not responding

To Stop the Run When DAQ is Working

To Start a Run When DAQ is Working

StartedlEv:0,rEv:0,d:0,kB:0,ev/s:0

Please send all comments and suggestions regarding this page to Andrei

‘Bad interrupt 0’ messages in VME ROC window

‘Cannot bind to port 7002’ messages in PhatDAQ