aLIGO LHO Logbook

H1 CDS

patrick.thomas@LIGO.ORG - posted 18:23, Friday 20 April 2018 - last comment - 18:47, Friday 20 April 2018(41573)

restarted medm web screenshots, something is wrong

Gerardo, Patrick

Wanted to only restart the screenshots for the vacuum system but ran into trouble and ended up restarting all of them. Kept getting 'Xvfb failed to start' when attempting to run h0_start, but not when running the others. Deleted /tmp/.X99-lock and /tmp/.X6-lock after searching the web for answers, but this did not help. Upon Gerardo's suggestion we compared the scripts. h1_cds_psl_start, h1_sei_start, and h1_sus_start had 'exec xvfb-run -a', but h0start only had 'exec xvfb-run'. I copied h0start to h0start.original_20_apr_2018 and added '-a' to h0start. This allowed it to run, but now all of the screenshots have black rectangles over them.

Comments related to this report

patrick.thomas@LIGO.ORG - 18:35, Friday 20 April 2018 (41574)

Link

Ran 'ps -elf | grep Xvfb'. Found two processes:

controls@script0:screenshooter 0$ ps -elf | grep Xvfb
0 S controls  1172     1  0  80   0 - 185119 poll_s 2016 ?        18:55:17 Xvfb :6 -screen 0 1600x1200x24
0 S controls  2716     1 20  80   0 - 15987 poll_s 17:04 ?        00:17:22 Xvfb :99 -screen 0 1920x1400x24 -nolisten tcp
0 S controls 14854 31575  0  80   0 -  1918 pipe_w 18:29 pts/0    00:00:00 grep --color=auto Xvfb

Killed them both. Got mail in /var/mail/controls. Last line reads:

ligo/apps/simlink_webview/update_webview.cron: line 39:   618 Killed                  matlab -display :6 -logfile $USERAPPS_DIR/cds/common/scripts/webview.log -r 'cd $USERAPPS_DIR/cds/common/scripts/; webview_simlink_update'

Looks like I stopped something that I shouldn't have....

But now the original script to run the vacuum screenshots works.

patrick.thomas@LIGO.ORG - 18:47, Friday 20 April 2018 (41575)

Link

So it appears to be working now, but I may have killed the simulink web screenshots.