It appears that after a reboot of the h1guardian computer a week or so ago, the PEM_MAG_INJ node was requested to INJECTIONS_COMPLETE, meaning that at 14:20 UTC (07:20 PDT) this morning while H1 was locked at low noise, the injection suite ran for the following 20 minutes. This seems to have gone without issue, and the log lives in the normal injection directory, /ligo/www/www/exports/pem/WeeklyMagneticInjection/logs/

The magnetic injections do NOT appear to correspond with the lockloss at 14:45 UTC (the suite had finished 4-5 minutes prior). It's also worth noting that the in-lock SUS charge measurements did not run this morning, which would normally start at 07:45 local time, so that isn't the cause of the lockloss either.

I'll leave the magnetic injection suite set to run automatically on Tuesday mornings unless people have objections to it (it will not run unless the IFO is locked at low noise anyway).