When I restarted the calibration pipeline last Thursday, June 8 (see https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=36727 ), I noticed that gstlal-calibration-1.1.7 had been installed on h1dmt0 and h1dmt2. This apparently happened automatically during Tuesday maintenance and was not intentional. However, this change did not affect the process or the output until the restart on June 8. Due to the amount of testing already done with this version (the last batch of C01 was made with it; also see https://bugs.ligo.org/redmine/issues/5531) and the difficulty of reverting to version 1.1.4, it was decided to continue using 1.1.7 at LHO. We have had 5 days of online running since this restart, and no problems have occurred. LLO plans to continue using 1.1.4 for one more week before upgrading to 1.1.7 on June 20.

New features/bug fixes included in 1.1.7 as compared to 1.1.4:
* Algorithm to compute SRC detuning parameters; 4 more 16 Hz channels:
        - {IFO}:GDS-CALIB_F_S
        - {IFO}:GDS-CALIB_F_S_NOGATE
        - {IFO}:GDS-CALIB_SRC_Q_INVERSE
        - {IFO}:GDS-CALIB_SRC_Q_INVERSE_NOGATE
* Command line option to remove calibration lines (not yet being used, but would add a 16 kHz channel, {IFO}:GDS-CALIB_STRAIN_CLEAN )
* Latency reduced by 5 seconds (was 10 - 14 seconds, now 5 - 9 seconds)
* Bug fix to mark 3 seconds before and after raw data dropouts bad in the CALIB_STATE_VECTOR. This 3 seconds is due to the settling of the FIR filters. The risk of running without this big fix is that there can be glitches around raw data dropouts that need to be manually vetoed (see https://alog.ligo-la.caltech.edu/aLOG/index.php?callRep=32486 )