KAGRA Logbook

About missing data on mx_stream a.k.a. IPC glitches: "IPC" on RCG means Shmem, Dolphin, and RFM communications e.g. the 5th bit of STATE_WORD on GDS_TP screens and they are distinguished from interprocess communication between RCG and DAQ. And also, we already know "IPC" on RCG is not a direct and only cause of this issue. So note that "IPC glitches" is slightly confusing name about this issue (though it's not a wrong name as a general meaning of IPC).

I noticed that which model to use to investigate this glitch issue depends on used signals, so I created glitch segments for each model. For example, online h(t) is affected by glitches on K1CALCS. On the other hand offline h(t) may be affected by ones also on K1LSC, K1CALEX and so on. For cleaned h(t), glitch information also about witness channels is necessary.

Now it's manually provided but it will be automated and provided on Kashiwa cluster. Total number glitches in science mode of O4c (~50days) on each model are shown in the attachment. In general, this issue occurred multiple models at the same time, so total number (the bottom line in the attachment) is not same as sum of numbers on each model. Glitch rate is drastically improved from O4a but now many glitches occurr on K1VISETMYLSC and K1SENDBEACONEY.