Differences

This shows you the differences between two versions of the page.

--- daq:daq [2012/10/17 11:26]
recchia [DAQ cold restart]
+++ daq:daq [2024/02/12 05:23]
weisshaa [FRIB online build window]
@@ Line 1: / Line 1: @@
-GRETINA DAQ system
+======= FRIB online build window ======
+Issue the following command prior to starting the ReadoutGUI: export COINCIDENCE_TICKS=300
+(e.g. in bin/startrdo)
-====== HowTo for experimenter in charge ========
+======= Turn on/off Mode3 Data collection ======
+  * To start/stop mode3 data collection on the EPICS GUI page click the button Cluster. In the newly opened window click the button **2 only** for mode2 data only or the **2 and 3** for both mode2 and mode3 data.
+  * Alternatively in a terminal on a2.gam run the command **"caput Data_CS_Save2And3 0/1"** to turn mode3 collection off/on.
+  * After doing this the decomp scripts should be modified as running the syncrofix command could overwrite this.
+  * For GRETINA standalone the script to modify is is **grdecompSA.sh** and for GRETINA-S800, **grdecompParticleGamma.sh**.
+  * For both scripts change the line **"caput Data_CS_Save2And3 0/1"** to turn mode3 collection off/on
+======= New in 2022 ======
+start run window is obsolete. Two scripts are available for changing the data directory. Daq needs to be stopped, directories need to exist:
+  * changeDataDir_maintainRunNum  <data dir>  : keeps the Run number the same, for when we need to move physics experiment data collection to another disk
+  * changeDataDir_resetRunNum <data dir>  : finds the lowest available Run number without any larger RunDir #s
-  * Online SpecTcl [[:onlinespectcl|onlinespectcl]]
-====== Changelog ========
-ANY change to the DAQ system out here: [[:daqchange|daqchange]]
-**NOTE**: No change pass July 13th 2012 can be done without explicit
+======= This is DAQ page from previous campaigns ======
-announcement and approval (by Dirk).
+The DAQ page from the former S800-GRETINA campaign in 2012 can be found at
+https://wikihost.nscl.msu.edu/gretina/doku.php/daq:daq_2012
+  * S800-Gretina control [[docsS800Grt|docsS800Grt]]
+  * {{ :daq:ulm-e16011_names.png?linkonly |S800 Trigger GUI}}
+======= Troubleshoot ======
+  * S800-Gretina not correlated: Check CC setfile ~s800/converged_daq/Scripts/CC0105Begin.tcl for extsynch and extclock
+  * S800 is losing events. Check the network speed of spdaq32 by /sbin/ethtool eth0 and /sbin/ethtool eth1 , both should show speed as 1000Mb/s, if one shows 100Mb/s then the network is too slow and S800 cannot send data to GEB fast enough which will cause event loss happen.
+  * Check the ringbuffer status, $DAQBIN/ringbuffer status - -host=spdaq32 - -all , if the free colum becomes zero, DAQ related people needs to be notified for diagnostics.
-====== issues ========
+======= Decomp Information ======
-ANY issue/breakdown put here: [[:daqtrouble|daqtrouble]]
+  * [[easy_decomp_calibrations|Easy Gretina Decomp File Creation]]
+  * [[decomp_calibrations|Gretina Decomp Calibrations Explanation]]
+======= Gretina Computing =======
+Gretina computer farm is maintained by LBNL computer group and local NSCL IT department. Nonetheless I put documents here, but we do NOT do anything to the GRETINA computers!
-ReadOutShell reports "S800 changed to inactive" [[:ReadOutShellS800Inactive|ReadOutShellS800Inactive]]
-====== start procedures ======
+  * May 2019, documentation provided from Jackie: [[https://docs.google.com/document/d/1Aq5NkuEKBc4_5fx-95d2RaSZM08VQwVleYWr42QTsFg/edit?invite=CNr_nogM&ts=5cd488d9|Gretina System Documentation]]
+  * Instruction to shut down switches for NSCL power shutdown 12/5/2015 {{:daq:gretinaswitchesshutdown.pdf|}}
-  * [[GRETINA:GRETINA]]
+======= NSCL GRETINA LaBr Array =======
-  *[[s800daq:s800daq]]
-  *[[source measurement:source measurement]]
-  * S800 monitor [[https://groups.nscl.msu.edu/opdevtech/wiki/index.php/S800_DAQ_status_tool|S800mon]]
-====== Documentation ======
+    * [[fastscint_daq| FastScint DAQ Quick Start]]
-  * NSCL DAQ general [[http://docs.nscl.msu.edu/daq]]
+    * [[fastscint_daq_long | FastScint DAQ Details]]
-  * NSCL DAQ GRETINA (ReadOutShell scheme) [[http://docs.nscl.msu.edu/daq/gretinauserdocs.pdf]]
-  * S800 DAQ [[https://groups.nscl.msu.edu/opdevtech/wiki/index.php/Main_Page#Electronics_and_DAQ]]
-====== Trigger connection ======
+======= Misc 2019 ======
+  * Dual homed NSCL machines are spdaq32 and spdaq07
-  * [[daq:triggerconnection|Trigger connection (Card 29)]]
+====== miscellaneous ======
+  * Validation/Trigger window width is set by Dig116_CS_X55WidthP, Dig116_CS_X55WidthN. N/P stands for positive/negative maximum difference relative to Trigger timestamp which will be read out.
+  * Bringing up just the EPICs screen: in gretina home **./startGRETINA_EPICSdisplay**
+  * Set trigger multiplicity: **caput Trig0_CS_FSCThr n** (n=0,1,2,3... for mult 1,2,3,4...)
+  * See trigger pulse width: **CheckLED.sh**
+  * Change trigger pulse width: **SETLEDWIN <# of 20ns ticks>**
+  * Check timestamps, report boards that are out-of-sync: **CheckTS**
+  * Force timestamps to reset, DO NOT USE DURING RUN!!!: **SINGLEIMPSYNC**
+  * Warning script: run on a2 as gretina in ~/scripts: **./warningMessage4.sh **
+  * Check decomp stats  ** recordDecomp.sh | head -n50 **
+  * [[Flash digitizer firmware]]
+  * Check digitizer firmware version: on a2 or a1 terminal, type: **DIGSNIDFW** ,all digitizers should have same version firmware
+  * [[Replace digitizer with different color ones]]
+  * [[Replace fuses of digitizer]]
+  * [[No valid decomposed events]]
+  * Verify connecedt digitizers, at IOC interface, type **dbior**, it will print all identified digitizers within one bank
+  * VME related scripts are located at /global/devel/boot/vmeArray
+  * Check connected PVs, at IOC interface, type "casr"
+  * [[ Move detectors to new Hole position ]]
+====== Reboot a2 ======
+  * If a2 is rebooted due to system failure, two scripts need to be brought up
+    * from ~gretina: nohup scripts/warningMessage4.sh &
+    * from ~gretina/scripts/watcherdir: nohup watcher &
+ The first is the alarm script that talks to users.  The second is responsible for two things: writing record{Counters,Decomp,Settings} in the current run directory and cleaning up cluster process issues (usually restarting broken processes) soon after starting or stopping runs.
-====== Status ======
+  * Switch to a1
-[[:status|status]]
+    * If a2 cannot be brought back. Gretina DAQ can still run on a1. In the file ~/.profile , environment variable **DAQSERVER** needs changed to be a1.gam . This parameter is used to restrict gretina daq critical scripts to run on specific machine.
-==== Files/trees on GRETINA cluster to go ====
-(Only to be done by Dirk)
-   /global/data1x/gretina:
-   drwxrwxr-x  64 gretina users  4096 Sep 19  2011 CR-1
-   drwxrwxr-x 528 gretina users 20480 Oct  4  2011 CR-2
-   drwxrwxr-x  12 gretina users  4096 Nov 16  2011 CR-5
-   /global/data1b/gretina
-.7T	CR-9
-==== File Structure on GRETINA cluster / files to be kept ====
-   Here offline decomp data gets stored:
-    /global/data1a/gretina/user1
-    /global/data1a/gretina/user2
-    /global/data1a/gretina/user2_1
-    /global/data1a/gretina/user4
-   Base + data to create them:
-   /global/data1a/
-   drwxrwxrwx  8 mario   bin   4096 Mar 27 12:52 q1a1
-   drwxr-xr-x  6 mario   bin   4096 Nov 24  2010 q1a3
-   drwxrwxrwx  9 mario   bin   4096 Mar 27 12:53 q1b1
-   drwxr-xr-x  7 mario   bin   4096 Mar 27 12:53 q1b2
-   drwxr-xr-x  9 mario   bin   4096 Mar 27 12:55 q2a5
-   drwxr-xr-x  9 mario   bin   4096 Mar 27 12:57 q2a6
-   drwxrwxrwx  7 mkp     bin   4096 Mar 27 13:01 q2b4
-   drwxr-xr-x  7 mkp     bin   4096 Mar 27 13:03 q2b6
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 13:05 q3a14
-   drwxr-xr-x  6 mario   bin   4096 Mar 27 13:06 q3a7
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 13:08 q4a4
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 13:09 q4a8
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 13:11 q4b7
-   drwxr-xr-x  6 mario   bin   4096 Mar 29 14:00 q4b8
-   /global/data1b/
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 14:26 q3b14
-   drwxr-xr-x  5 mario   bin   4096 Mar 27 14:27 q3b5
-   drwxr-xr-x  5 gretina bin   4096 Mar  6 18:07 q5a10
-   drwxr-xr-x  5 gretina bin   4096 Mar  6 18:17 q5a12
-   drwxr-xr-x  5 gretina bin   4096 Mar  7 13:14 q5b10
-   drwxr-xr-x  5 gretina bin   4096 Mar  6 18:27 q5b3
-   drwxr-xr-x  5 gretina bin   4096 Mar 27 14:29 q6a15
-   drwxr-xr-x  5 gretina bin   4096 Mar 27 14:29 q6a3
-   drwxr-xr-x  5 gretina bin   4096 Jun 10  2011 q6b11
-   drwxr-xr-x  5 gretina bin   4096 Jun 10  2011 q6b9
-   drwxr-xr-x  3 gretina bin   4096 Mar 27 14:31 q7a11
-   drwxr-xr-x  5 gretina bin   4096 Mar 27 14:32 q7a9
-   drwxr-xr-x  5 gretina bin   4096 Mar 27 14:33 q7b13
-   drwxr-xr-x  5 gretina bin   4096 Mar 27 14:33 q7b15
-G	DNLapril2012:   DNL data of digitizers
-====== Configuration files for decompostion ======
-The configuration for the decomposition is here: https://wikihost.nscl.msu.edu/gretina/doku.php/daq:decomp
-====== Positioning of the quads ======
-When a position of a detector is changed the following has to be updated:
-  * CS_DataId in https://wikihost.nscl.msu.edu/gretina/doku.php/daq:decomp
-  * CS_GblId in /global/devel/boot/vmeArray/vme*.cmd --- The Id is given by  hole_number * 16 + position_in_quad * 4 + board_number
-  * GTSort_shack/DetNum.dat --- self explanatory
-====== How to replace a broken node with a spare one ======
-To replace a node of the cluster (e.g. n000x) with a spare one (e.g. n000y) follow the following procedure:
-In the file /global/devel/gretTop/11-1/gretClust/bin/linux-x86/setup7modulesRaw.sh and in the file /global/devel/gretTop/11-1/gretClust/bin/linux-x86/setupEndtoEnd28.sh
-  * replace node000x_CS_Program? with node000y_CS_Program?
-  * from the "GRETINA CLUSTER node allocation" window identify the name of the Decomp process (middle click on the crystal number) and replace it in the files
-  * change the n000x to offline and the n000y to online using the variables Decomp???_CS_Enable
-====== DAQ cold restart ======
-  * Switch off all the vme crates.
-  * wait 30 s
-  * Switch on the vme crate of the trigger cards.
-  * wait till the red lights over the trigger board are on (~ 3 minutes)
-  * Switch on all the other vme crates
-  * wait till the green blinking lights appears on all the digitizers (~ 3 minutes)
-  * start the daq and check the synchronization (alarm window or TS window). If few of the digitizers are out of sync recover them using the "reboot IOCs button", otherwise power-cycle on their crate

Trace:

Differences

Search

Toolbox

QR Code