G2p DAQ How-To

From Hall A Wiki
Jump to: navigation, search

Normal DAQ Operations

DAQ Account Info

This assumes that runcontrol is running. If not, see Cold Start in section III.

  • For g2p and gep we are running two separate DAQs, one for each HRS arm.
    • LHRS DAQ is run on the adaql2 computer on the a-onl account.
    • RHRS DAQ is run on the adaql1 computer on the a-onl account.

Start and Stop Runs

  • Normally, runcontrol will be already running. If not, see section on Cold Start below (for the impatient: type “startcoda”).
    • To start and stop runs, push the buttons “Start” (6th button from left, looks like >>) and “End” (5th button, looks like a box).
      • Start is the same as the sequence prestart, go

CODA Configurations

  • We are using the LHRS configuration for the LHRS and RHRS configuration for the RHRS.
    • These are located in the “LHRS” and “RHRS” sessions respectively.
    • The configurations should be loaded by default.
      • To change configurations, use the “Configurations” menu and select COOL.
      • If you’ve been running, you will first have to push the “Reset” (<<) button. Choose the configuration from the dialog box, then press “Download” (looks like a disk drive).
      • A description of all CODA Configurations is available here

  • Re: the buttons in rcgui: If you let the mouse hover over the button for a few seconds you can discover what the buttons mean because a little dialog box comes up and tells you, e.g. which button is “Start”, etc.

Deadtime Monitor

  • Deadtime is displayed in the ”datamon” window – make sure you are using the one on the same computer as runcontrol.
  • Datamon is probably already running near the runcontrol window.
  • If it is not running, type “datamon”
    • You will need to do this for both LHRS/RHRS on adaql2/adaql1
  • If the deadtime is > 20% for an extended period of time you probably need to adjust the prescale factors

Editing Prescale Factors

  • Type “prescaleL” or “prescaleR" to open the prescale GUI. Make changes to prescale values and click "save" before starting the next run.

Checking for Gates

  • Go to the raw data file directory on the a-onl or adev account
  • Type xcefdmp
    • Click "View File"
    • Click "Data Source" >>"Open"
    • Select a data file to look at
    • Click "View Next" until a diagram with each ROC comes up
    • Select a ROC, clicking "View Next" allows you to scan through events
    • If the hex word 0xdc0000ff appears, this ROC is not getting gates!
      • Call an expert to fix the problem.

Trigger and Scaler Overview

How to start xscaler (HRS) / view the scalers

start xscaler :

  • Go to the hapc5 terminal in the middle of rack CH01A09. If it is not logged in, log in as user "adaq"
  • Type ssh adaq@adaql4 and enter the password when prompted.
  • Type goxscaler
  • Type ./xscaler Left(Right) depending on which HRS you want

to view the scalers :

  • The simplest : login to adaq l1 as adaq
  • Type xscaler
  • Follow what it says

Definition of g2p/gep Triggers

  • T1 = R-HRS main S1.and.S2m trigger
  • T2 = R-HRS efficiency (2/3 among S1,S2m,Cer)
  • T3 = L-HRS main S1.and.S2 trigger
  • T4 = L-HRS efficiency (2/3 among S1,S2m,Cer)
  • T5 = Nothing
  • T6 = Nothing
  • T7 = 103kHz pulser
  • T8 = 1024 Hz pulser
  • T9 = Nothing

Downloading/Checking the Trigger

Throughout the course of the experiment the trigger should not change, but if the CAMAC crates lose power then you will have to download the trigger again.
To download the trigger:

  1. Logon to an adaq Linux box like "adaql1" or "adaql2" as user "atrig"
  2. Navigate to the "trigger" directory
  3. trigsetup trigger_left(right).map
  4. In the window that just opened click "download all" and verify that there are no errors


Rebooting and Resetting

Quick Resets

Problems with CODA 2.6.1 can usually be solved with a simple Reset (<< in rcgui) or with a Cold Start. Call an expert if you are stuck. Do NOT waste an hour stuck on resets.

If you reboot the ROC, or if something on the workstation is hung up, try pushing the ”Reset” button in runcontrol (<<” in rcgui). Then Configure, Download, and Start a run as usual. If a “quick” reset doesn’t work, try a Cold Start (see below)

Rebooting a ROC

If a ROC seems to be hung up, you can reboot it. To do this follow these steps:

  1. Type ”reboot” at the vxWorks prompt (− > reboot) in the ROC's xterm window.
    1. Another possibility is to type “ctrl-X” at the vxWorks prompt; this sometimes works when typing reboot does not work
  2. Wait 2 minutes and telnet back in to verify its alive.

The name of the ROC computer is normally written in the name of the xterm (hallasfi1, hallasfi2, etc). Finally, if the above does not work one can try the remote crate resets below [[1]](or gain access to the cpu and press the reset button or power cycle the crates ).

Special Note on ROC3

If the LHRS crate is power cycled, the boot parameters for ROC3 will need to be re-entered. The boot parameters are located in doc/vmeram g2p.doc and also on the g2p Wiki [Boot_Parameters_for_g2p]. To change the boot parameters do the following:

  1. Go to the vxWorks prompt for ROC3 and type ”c”.
  2. This will allow you to enter the parameters, one line at a time.
  3. When finished, type ”p” to print the boot parameters (and check that they are correct)
  4. Type ”@” to reboot the ROC.

Cold Restarts (Starting CODA from scratch)

  1. To perform a cold start of the processes on the Linux computer (adaql1 or adaql2), simply type “startcoda” on the a-onl account. This performs a “kcoda” for you, as well as starts all the components you need.
  2. Eventually runcontrol (rcgui) will pop up.
    1. In runcontrol press the Platform button and then ”connect”.
    2. After “connect”, verify you have the correct session (LHRS/RHRS) and configuration (LHRS/RHRS). Usually you will, but if not you’ll need to select the session and config.
    3. You may now do the “configure” by pressing the button with crossed screwdriver and wrench in the upper left corner. Be patient, configure takes awhile.
    4. Then you may download, which is the next button to the right, it looks like a tape drive.
    5. In general, the button names can be discovered by putting the mouse on it and letting it sit there for a few seconds; a small box appears with the name of the button.

    6. To start a run, press the >> button.
    7. To end a run press the square box that’s sorta in the middle.
    8. To reset, press the << button.
  3. Start datamon by typing “datamon”


If you ever logout of the DAQ computer adaql1 or adaql2 or have rebooted it, here is how to restart the preferred setup. Login to the adaql1(adaql2) console as ”adaq” and then 

ssh -l a-onl adaql1(adaql2)
Reason is that if you login directly as a-onl, some GUI windows like start-run don’t pop up. Just do what I said. Start CODA by typing “startcoda” as explained above. Emacs the prescale file in the background: 
emacs /prescale/prescaleL(or)R.dat

Remote Crate Resets

The following is how to remotely reset the crates when the crates get hung up, and cannot be rebooted.

ROC1/ROC2/ROC6
  • Go to HALL-A tools screen on hacsbc2 (left-hand side of shift leader's computer)
  • Under "MISCELLANEOUS" section click on "Crate Resets" and under "Resets from the Electron Detector Stack (right) Crate" look for "SFI (Fastbus ROC1 and ROC2) #1" and click on "click to reset" button. After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.
  • If doing above doesn't help, then under "Resets from the Electron Detector Stack (right) Crate" look for "Lower Fastbus AC" and click on "click to reset" button.For ROC1,the boot parameters will need to be re-entered in the portserver terminal window, check the "special note for ROC1/ROC3" in this document above [[2]].
TS0
  • Go to HALL-A tools screen on hacsbc2 (left-hand side of shift leader's computer)
  • Under "MISCELLANEOUS" section click on "Crate Resets" and under "Resets from the Electron Detector Stack (right) Crate" look for "Trigger Supervisor" and click on "click to reset" button.
  • After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.
ROC 26
  • Direct a webbrowser to http://hareboot32, the password is located on the yellow paper attached under the screen of the third person`s computer.( If the URL is blocked,you can click on the "Edit" menu on firefox and select "preference", click on the "Advanced" category, click on "Network" tab, click on "Settings" under Connection, choose "No proxy", click on "OK", then go to the website http://hareboot32 .）
  • Under "Device manager" click "Control Action" and select "Immediate Reboot" for Outlet 8 "happex7 RHRS(ROC 26)", then click "apply".
  • After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.
ROC3/ROC4/ROC5
  • Go to HALL-A tools screen on hacsbc2 (left-hand side of shift leader's computer)
  • Under "MISCELLANEOUS" section click on "Crate Resets" and under "Resets from the Hadron Detector Stack (left) Crate" look for "SFI (Fastbus ROC3 & ROC4) #1" and click on "click to reset" button. After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.
  • If doing all above doesn't help then under "MISCELLANEOUS" section click on "Crate Resets" and under "Resets from the Hadron Detector Stack (left) Crate" look for "Lower Fastbus AC" and click on "click to reset" button. For ROC3,the boot parameters will need to be re-entered in the portserver terminal window, check the "special note for ROC1/ROC3" in this document above [[3]].
TS11
  • Go to HALL-A tools screen on hacsbc2 (left-hand side of shift leader's computer)
  • Under "MISCELLANEOUS" section click on "Crate Resets" and under "Resets from the Hadron Detector Stack (left) Crate" look for "Trigger Supervisor" and click on "click to reset" button.
  • After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.
ROC 25
  • Direct a webbrowser to http://hareboot28, the password is located on the yellow paper attached under the screen of the third person`s computer.( If the URL is blocked, you can click on the "Edit" menu on firefox and select "preference", click on the "Advanced" category, click on "Network" tab, click on "Settings" under Connection, choose "No proxy", click on "OK", then go to the website http://hareboot28 .)
  • Click "Control Action" and select "Immediate Reboot" for Outlet 1 "happex8 LHRS(ROC 25)", then click "apply".
  • After rebooting, in the runcontrol GUI, click Reset, click Download, and then start a run.

Trouble Shooting

How to solve 100% dead time problem

If there is a 100% deadtime for more than a min, do the following

1. Stop the run, kill coda (kcoda)

2. Restart the coda by logging into a-onl account(even if you are in a-onl account).

3. In most cases this will solve the problem and if this doesn't help then reboot TS crates (type "reboot" in the ROC terminal) and start coda again.

4. If this doesn't help then reboot all the crates and start coda again.

Run number synch problem

  • With CODA 2.6.1 there seems to be a small probability that the run number reported by rcgui is different from the one in the data file. Firstly, the one that is correct is the one in the datafile name, which also appears as a record in the datafile, and which derives from MSQL. Secondly, fixing this goofy problem is sometimes tricky; you can try Options ... Set Run Number, but the GUI needs to be in a proper state. Or just report it to the expert and let him/her fix it.


High Deadtime on RHRS

If the deadtime is >20% for rates of 4kHz or lower, follow these steps:

1. Post screenshot of RHRS scalers and Datamon

2. Post screenshot of component windows for all ROCs and TS, platform, EB and ER

3. Run "top" on a-onl@adaql1 (look for a terminal, it should already be running) and post in HALOG. Look for processes using a lot of CPU time.

4. Have an expert run blaster tests. Located on the adaq account in the Blast/ directory (use blaster.o.vw55).

Retrieved from "https://hallaweb.jlab.org/wiki/index.php?title=G2p_DAQ_How-To&oldid=15488"