URGENT BUG in DMTGen affecting replay
DMTGen seems to be giving different values for a channel based on whether FrameLength is set to 4 or 1:
On lowlatency@cit, run
DMTGen -conf DMTGen-Virgo_Data.cfg
with the contents of DMTGen-Virgo_Data.cfg set to:
Parameter StartGPS 1187388000
Parameter EndGPS 1187388004
Parameter Compression "zero-suppress-or-gzip"
Parameter OutputDirectory .
Parameter FrameLength 4
Parameter GenerStride 4
Parameter VerboseLog ./var/log/llpipeline/DMTGen-Virgo-verbose.log
Source source0 "FrameData(Channel=V1:DQ_ANALYSIS_STATE_VECTOR,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:DQ_ANALYSIS_STATE_VECTOR source0
Source source1 "FrameData(Channel=V1:Hrec_STATE_VECTOR,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:Hrec_STATE_VECTOR source1
Source source2 "FrameData(Channel=V1:Hrec_hoft_16384Hz,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:Hrec_hoft_16384Hz_O2Replay source2
Then check the values in the channel
$ FrameDataDump -I$(ls *.gwf | tail -1) -CV1:DQ_ANALYSIS_STATE_VECTOR -a -d
4095
...
Now, go back and change to "FrameLength 1" and "GenerStride 1", i.e.
Parameter StartGPS 1187388000
Parameter EndGPS 1187388004
Parameter Compression "zero-suppress-or-gzip"
Parameter OutputDirectory .
Parameter FrameLength 1
Parameter GenerStride 1
Parameter VerboseLog ./var/log/llpipeline/DMTGen-Virgo-verbose.log
Source source0 "FrameData(Channel=V1:DQ_ANALYSIS_STATE_VECTOR,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:DQ_ANALYSIS_STATE_VECTOR source0
Source source1 "FrameData(Channel=V1:Hrec_STATE_VECTOR,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:Hrec_STATE_VECTOR source1
Source source2 "FrameData(Channel=V1:Hrec_hoft_16384Hz,Files=/data/O2/fill/V1/*.gwf)"
Channel V1:Hrec_hoft_16384Hz_O2Replay source2
and again look at the values in the channel
$ FrameDataDump -I$(ls *.gwf | tail -1) -CV1:DQ_ANALYSIS_STATE_VECTOR -a -d
3928819656
Looking at the original file,
/data/O2/fill/V1/V-gaps_filled-1187388000-2000.gwf
the actual value would seem to probably be 4095.