This Document contains procedure on how to enable control and readout of the AMCs in the event the server loose connection to the uTCA crates or the crates are connected for the first time. Documentation uses the TDE test crate IP in the examples.

1.

First, ensure the host machine for reading out the AMCs is correctly setup (readout server for the test crate is np02-srv-002).

dpdk-devbind.py -s

Should show some NICs using the DPDK compatible drivers. If there are none, then run

sudo /opt/setup-100g-vfio.sh

2.

Next, make sure the uTCA crate is pingable from the host machine for controlling the AMCs (np04-srv-011). if you get no response doing

arping 10.73.32.129 # or 1073.32.X.129 for another crate

then the 10G interface needs to be configured through nmcli (contact sysadmins). (was there any other steps to do here?)

3.

If the above is correctly setup, perform a first time boot of the AMCs. This requires sending ARP commands through the same network interface used to readout the the AMCs. To do so, start a run with the only the readout application enabled for the. For the test crate this involves disabling the tde crate application in the segment file (segments/tde-testcrate.data.xml):

<obj class="Segment" id="tde-testcrate-segment">
 <rel name="applications">
  <ref class="ReadoutApplication" id="runp02srv002eth1"/>
  <ref class="TDECrateApplication" id="tde-testcrate-application"/>  <!-- comment this out -->
 </rel>
 <rel name="controller" class="RCApplication" id="tde-testcrate-controller"/>
</obj>

and run drunc

drunc-unified-shell ssh-CERN-kafka ehn1-daqconfigs/sessions/tde-testcrate.data.xml tde-testcrate-session $USER-tde-test
 
# in the drunc shell
boot
conf
start --run-type test

Now commands can be sent to the AMCs to initialize them (how is this done?)

The run can now be stopped

# in the drunc shell
drain-dataflow
stop
scrap
terminate
exit

It should now be possible to send commands to the AMCs using the modules provided in tdemodules.

Undo the changes to segments/tde-testcrate.data.xml!

4.

You should be able to send commands and get the status of the AMCs using test_amc_status.py from the host machine for controlling the AMCs (np04-srv-011)

test_amc_satus.py -c 10.73.32.128 # to check all possible IPs from a crate

test_amc_satus.py -c 10.73.32.2 # too check a single AMC

and this should not produce any error messages in the printout.

5.

Ensure the timing is initialized for the frontend electronics

6.

The system should be fully setup and a run can be taken. For the TDE test crate:

drunc-unified-shell ssh-CERN-kafka ehn1-daqconfigs/sessions/tde-testcrate.data.xml tde-testcrate-session $USER-tde-test
 
# in the drunc shell
boot
conf
start --run-type test
enable-triggers