Closed DominicOram closed 3 years ago
The problem turned out not to be related to the collimator, Galil controllers nor the MOXA NPort they are connected to. The additional NPort that is in the screened room had had its network cable unplugged at the main switch, and because VIs were trying to talk to devices via it (including the PSU in #6496), slowed down the rest of LabVIEW system. It is not currently understood why, nor why the front panel of said NPort failed to show any errors - it even still had a green ready LED and displayed a valid ISIS IP address.
Performance was restored to usual levels once the cable had been reconnected, and remained so when all Galil controllers had been re-enabled.
I suspect that someone had set the MOXA to a fixed IP address? - whether as part of some attempt at diagnosis or just in the belief that it would be more reliable - I wouldn't expect the MOXA to complain about being off a network with a red light if it considered itself to be the network :-).
I suspect that someone had set the MOXA to a fixed IP address? - whether as part of some attempt at diagnosis or just in the belief that it would be more reliable - I wouldn't expect the MOXA to complain about being off a network with a red light if it considered itself to be the network :-).
No, definitely still set to DHCP. After reconnecting the network cable and power-cycling the MOXA, it came back with a recognisable ISIS IP address.
OK, that's good. I now think I remember that after testing this explicitly in my office a few years back, the MOXAs do keep their DHCP address if they have no access to the DHCP server (and only lose it if they are restarted). This does work if the DHCP server is lost for some reason - but I guess keeping the address is not so much use if the MOXA is no longer connected at all!
Have added a note to the SECI Trouble-Shooting page, although this may also apply to IBEX instruments.
Great, thanks @davidkeymer
As an instrument scientist on CHIPIR I would like my collimator to start moving as soon as I tell it to. Currently there is a long delay between me telling it to move and getting a response. Prior to the work done in https://github.com/ISISComputingGroup/IBEX/issues/6496 there was no delay.
Notes from troubleshooting over the weekend:
Acceptance Criteria