madeline.wade@LIGO.ORG - posted 16:55, Saturday 15 December 2018 - last comment - 08:18, Thursday 20 December 2018(45972)
GDS-CALIB_STATE_VECTOR configuration change
M. Wade
There was an issue at LHO where the OBSERVATION_READY bit of the ODC-MASTER channel was not reporting properly. Since this is the channel used by the GDS calibration code to determine h(t)-OK in the GDS-CALIB_STATE_VECTOR, in consultation with Keita I made a configuration change so that h(t)-OK is now determined based on OBSERVATION_INTENT instead of OBSERVATION_READY. This change was made and the pipeline was restarted on the DMT machines at GPS time 1228956497.
Thank you very much Maddie.
I also masked out all INJ ODC bits because it was receiving nothing (all zero, meaning all bad, or maybe something else, see top row of the attached) and low latency pipelines were seeing bogus injection flags.
Operators: When H1 unlocks, DOWN state in ISC_LOCK should set H1:ODC-OPERATOR_OBSERVATION_READY back to zero. But pay attention to see if this really happened, and if it doesn't, manually press "commissioning" button in guardian overview.
Also see this LLO alog about their workaround.
https://alog.ligo-la.caltech.edu/aLOG/index.php?callRep=42394
Just to facilitate further discussion, I attach screenshots of the guardian overview screen, and the super-manager guardian node called simply "IFO." This node looks at the "OK" bit of all subordinate nodes (there is an "exclude" list for nodes that we know are bad / non-functional / unimportant / underconstruction), and generates its own bit, called H1:GRD-IFO_OK the subordinate nodes include - a check that all of the automated interferometer control system nodes are in their pre-programmed nominal state (save those that have been excluded), - a check by DIAG_SDF against *every* switchable EPICs setting (save those intentionally not monitored because they're switched by an above mentioned automated node), - a check by DIAG_EXC whether there are any unwanted excitations on, - a check by DIAG_CRIT which checks critical guardian functionality, and - a check by DIAG_MAIN several other automated checks of observatory conditions. The other (and only) human-determined bit is the above mention H1:ODC-OPERATOR_OBSERVATION_READY, which is to what we refer to as the "intent" bit, which is set by cognizent operators, engineers, and scientists when they ackowledge that no one is intending to modify or measure the instruement, and thus they believe it is ready for astrophysical consumption to the best of their knowledge.FYI, before injection bits were masked out, ODC MASTER screen looked identical to this (except the intent bit, which is masked out in the screen shot for some test, but it wasn't masked out in ER13).