Jira for issues tracking:

If you have an issue during OpcUaLarLtdbServer (peripheral) operations please:

1. Create a sub task on ATLLARONL-435 ticket
2. Assigne Florent BERNON

Shell

/det/lar/project/LTDB All command are made for a bash shell

Troubleshooting

• User Troubleshooter
• Expert Troubleshooter

Change firmware on felix

please contact experts ( Etienne FORTIN or Daniele LAUGIER )

2020/09/08 for pc-lar-felix-ltdb-00 / 01 / 03

• firmware is FLX712 LTDB RM4.9 12-6-2020 at 10h39

• driver:4.5.0

FLX-712/production setup

WIP: felixstar

FelixStarP1LAr

More... Close

Tips to know which server has to be launch

to check which server need to be launch: command "status" ==> create this alias status='/det/lar/project/LTDB/tools/misc/status.sh'
!! this is just an information about running server and return pid / log file path
But no information if the communication between servers is ok ==> check log file

Launch felixcore

Logs are in /data/logs/felixcore If you need to relaunch felixcore after a crash just do :

# For pc-lar-felix-ltdb-05 for example
cd /det/lar/project/LTDB/Felix
./configurationFelixScript_productionServer.sh
./start.sh

When you start felixcore, restart also OpcUaSca and OpcUaLar to avoid having errors.

Troubleshooting

Known issue : Sticky LOL register =0x02 => TTC error, no BCR

Killing felixcore

If you have to kill felixcore for some reason, log into the pc you're interested in (e.g pc-lar-felix-ltdb-04), then do:

source /det/dcs/Development/ATLAS_DCS_LAR/bin/tools/misc/status.sh
sudo kill -9 <process id for felixcore>

</sticky>

Launch OpcUaScaServer OpcUaLArLtdb Server

Logs are in /data/logs/OpcUa(LarLtdb |Sca)Server.

If you need to start OpcUaScaServer :

#On pc-lar-felix-ltdb-0X
cd /det/lar/project/LTDB/OpcUaScaServer
./start.sh

Check the log file: the name of the log file is an output of the start.sh script , if ko, try again

If some sca are off, use the tools to recover it (mettre ici la recette DL)

If you need to start OpcUaLarLtdbServer :

#On pc-lar-felix-ltdb-0X
cd /det/lar/project/LTDB/OpcUaLarLtdbServer
./start.sh 

Check the log file: the name of the log file is an output of the start.sh script, if ko, try again

log ok = tail "logFile"

To stop them :

sudo kill -9 ID(OPC) # with ID(OPC) the pid of the server

Use WinCCOAProject

BEFORE RUN we have to check that all scas are online and if not restart opcuasca

Use WinCCOAProject With Remina:

open Remote Desktop on the cerntsatldcs(cerntsatldcs.cern.ch) and make the next 5 steps. (for example remmina as remote desktop)

config remote desktop:

gateway: CERN.CHcerntsatldcs.cern.ch login: your cern mail (and not your cern login !!) password: your cern password
choose FSM

With direct mode :

ssh -X lxplus

ssh -X atlasgw

and connect to pcatllar05

source /det/dcs/linuxScripts/set_env_316.sh

WCCOAui -proj ATLLARLTDB -p main.pnl or WCCOAui -proj ATLLARLTDB -m gedi

On the windows menu:
Jcobframework/access control /login
select main.pnl and execute...

startConsole if it is not=> for expert

How to see which FEC (Front End Crate) are ON:

https://atlasop.cern.ch/dcs/

https://atlasop.cern.ch/sys/lar.php

Which felix server is connected to which quadran :

https://atlas-larmon.cern.ch/larphase1topologydb/web/felix_setup.php
https://atlas-larmon.cern.ch/larphase1topologydb/web/wendy_setup.php

Quadran listes: from wendy setup

pc-lar-felix-ltdb-xx	EMB (ElectroMagnetic Barrel)	EMEC (ElectroMagnetic End Cap)
00	A Q1 => I01L to I05R	A Q1 => A01 to A04F0
01	A Q2 => I05L to I09R	A Q2 => A04F1 to A07
02	A Q3 => I09L to I13R	A Q3 => A08 to A10
03	A Q4 => I013L to I01R	A Q4 => A11 to A13
04	C Q1 => H01R to H05L	C Q1 => C01 to C04F1
05	C Q2 => H05R to H09L	C Q2 => C04F0 to C07
06	C Q3 => H09R to H13L	C Q3 => C08 to C10
07	C Q4 => H013R to H01L	C Q4 => C11 to C13

How to use the LAr GUI:

First LAr LTDB controle screen:

Zone 1:RunControl menu

Select your Partition: EMEC A, EMEC C, BARRELA, BARRELC. In this picture, Barrel A is selected and the zone is ligth on the detector pictogram
You can select all partitions in the same time

a)Transitions:
You can controle the LTDB with a Finite State Machine with the following states

Transitions that have been performed are in grey
Every thing is ok if no red messages / no red zone.
The state of the LTDB you are looking should be coherent with the global state: a reminder at the top of the panel.

b)Links status: Button to check links with Latomes: only when system is configured ==> read the latomes registers...

c)Copy states: Copy the full state into the clipboard to paste it into elogs or mails to experts

Zone 2: Servers status

a) Check that the OpcUaScaServers are green; At least for the partition you want (check in chapter Quadrants list, just before)...
b) Check LarLtdb field are 1 At least for the partition you want else it should be launched

Red or orange= there is an issue
Green = ok
Black= unknown status

Zone3: Views of the LTDB states on the wheel

Color convention on the panel

RED FLASHING: there is something wrong ===> put the mousse on the LTDB square, a message appears with the error

Status of two LTDB in one crate, Green is good, Red is Bad.
- You can click on it to have the crate details panels. (for troubleshooting)
- put the mousse on it, wait and a message appears giving the errors status

if There is juste the error "state = partition state" ==> click on the LTDB and performs the transitions you need to put it on the good state

GLOBAL STATE: Indicate the global state and the time of the last transition

Be careful: when transition is running, don't care the state of the panel....status could be red during this step....

WARNING:

1) You need to select felix servers you will use in "MODE EXPERT" ENABLE/ DISABLE => after choosing the felix servers: go back previous screen and Power OFF and power ON to start the set up with all felix servers in the same time

2) Scan LOCx2 if Crate powerCycle

zone 4: log windows

Zone Expert:

RESERVED FOR EXPERTS:
-adding felix servers (ENABLE/DISABLE)
-LOCx2 Scan

HOW TO FOCUS on a specific LTDB and see Monitoring:

Clicking on the LTDB zone, you can have a look to differents parametres that are monitored: for the moment sca temperatures

Attention: After doing "CONFIG", it's better to click on the "Links Status" to update it.

* LTDB control panel (Reminder: Only if the LDTB zone is not Blue, in that case, you can see only monitoring):

* monitoring LTDB panel:

Errors messages:

1) Follows instructions in the message: For example go back to the OFF state that means do transitions UNCONFIG and SHUTDOWN in this windows and after returns to the current Global state (here CONFIGURED) doing INIT and CONFIG in this windows

Attention: if Link status BAD, and no message in the log, and OpcUaSca is on ===> LATOMES are off!!

2) the status of the LTDB is RED= After CONFIGURE, "Links Status " is BAD => Expert can do a scan of the LOCx2 (Expert mode )

3) the log window: Liste of error code and recovery mesure: OpcUaLarLtdbServerTroubleshooter

How to perform a Clkdes scan

When you did a power cycle, you MUST do a CLKDES scan===> Expert Mode in the Control panel

LtdbHowToDoClkDesScan

RC_FORMATION_14_APRIL_21.pdf

OBSOLETE

How to perform a scan Locx2 (Experts only)

When you did a power cycle, you MUST do a scan of the Locx2===> Expert Mode in the Control panel
1- Start the scan with the "scan Locx2" button
- Wait until the array is full
- you are happy with green and red zones repartion Or you detect an error....
-If error, you need to fixe it befor continuing (Experts)

2- "=>" compute the best delay for each fiber and display them the thin array to the right of the arrow
3- "Select all actives fibres", 1 is displayed for all actives fibers in the thin array and data are ready to be sent to the server

4- Update the server with those value with the "Update Server" button

WinCCOA project use ( same for all setups )

Please see the EMF Twiki LtdbConfigurationEmfPeripheral

Try all LTDB

Troubleshooting

Some Ltdbs are not online

• Kill Felixcore
• Kill OpcUaScaServer
• start Felixcore
• start OpcUaScaServer

-- EtienneMarieFortin - 2019-06-17