Prior the recent #1263 , RUN::config.unixTime was being cached across events unless nonzero in EVIO, which was not nice for multi-threading the decoder. But now we likely have many events with unixTime=0, as that was apparently an old DAQ "fix"!
We may want to revert to the previous non-reproducible situation, or reassign unix time in post-processing.
Prior the recent #1263 ,
RUN::config.unixTimewas being cached across events unless nonzero in EVIO, which was not nice for multi-threading the decoder. But now we likely have many events withunixTime=0, as that was apparently an old DAQ "fix"!We may want to revert to the previous non-reproducible situation, or reassign unix time in post-processing.