Eiger server can crash during prepareAcq
Seen at ID11, sending it here in case someone else can take it on (@papillon did not see what we are doing wrong yet):
Scan 149 Mon Jan 18 14:48:33 2021 /data/id11/nanoscope/blc12678/id11/dt/dt_testforpymca/dt_testforpymca.h5 nscope user = opid11 fscan rot 0 0.025 14400 0.001 0.0010001 Preparing ... ERROR 2021-01-18 14:49:29,789 bliss.scans: Exception caught in eiger.prepare (DevFailed[ DevError[ desc = COMM_FAILURE CORBA system exception: COMM_FAILURE_WaitingForReply origin = Connection::connect reason = API_CorbaException severity = ERR]
In detectors/eiger.yml the prepare_timeout is set to 300, but somehow it timed out with less than a minute rather than the 300/60 = 5 minutes expected.
EDIT (by Matias): Thanks to all tests done on this issue, what happens is that the Eiger server crashes during
prepare phase, and gets restarted by supervisor => this produces
COMM_FAILURE_WaitingForReply, which is
another kind of error (nothing to do with timeout)