Displaying report 1-1 of 1.
Reports until 10:41, Friday 26 May 2017
H1 TCS (TCS)
aidan.brooks@LIGO.ORG - posted 10:41, Friday 26 May 2017 - last comment - 10:51, Friday 26 May 2017(36442)
Removed duplicate HWS channels from EPICS due to H1HWSMSR environment variables

In order to fix an EPICS channel conflict between H1HWSMSR and H1HWSMSR1, I killed the HWSX EPICS softIOC on H1HWSMSR and restarted the HWS code.

The EPICS channel names are specified uniquely on each HWS machine by modifications to the environment variables of that machine in the BASHRC file.

The HWS code currently generates two sets of EPICS channels per instantiation. This is an artifact of the corner station code running two HWS on a single machine. Currently, we're running each Hartmann sensor on a separate machine. In order to avoid creating duplicate channel names on different machines, we added fake prefixes into the names in BASHRC (ITMY0 on the HWSX machine and vice versa).

However, we missed re-running the BASHRC file on H1HWSMSR (the HWSX machine). So two machines were producing channels with the prefix H1:TCS-ITMY_HWS.

So, between killing the EPICS softIOC and restarting the HWSX code, I logged into the TMUX session on H1HWSMSR and reran source ~/.bashrc to fix the problem. The results can be seen here:

This should permanently fix the problem in TMUX. To be super-sure, I'll get Dave Barker to reboot the machine.

 

Images attached to this report
Comments related to this report
david.barker@LIGO.ORG - 10:49, Friday 26 May 2017 (36446)

h1hwsmsr was rebooted at 10:43 PDT

aidan.brooks@LIGO.ORG - 10:51, Friday 26 May 2017 (36447)

Dave has rebooted the H1HWSMSR computer and I have restarted the HWSX code in a new TMUX session.

Displaying report 1-1 of 1.