Hi, Matt.
I think the run numbers in the recon data are wrong after the first file. This is because a pre start event was used to set the run number in EvioToLcio, but this is not present in subsequent files. So the run number gets a default value of zero for most of the files. (I think this was the problem Luca was having where the conditions system suddenly started seeing run number 0 in his job.)
When running recon for a pass, you can set this yourself using an override.
There is the “-R” switch to EvioToLcio….
EvioToLcio -R 1234 […]
Or you could use ConditionsDriver with a variable in the steering file.
<driver name=“ConditionsDriver” type=“…”>
<runNumber>${runNumber}</runNumber>
<detectorName>SomeFancyDetector</detectorName>
<freeze>true</freeze>
</driver>
And then
EvioToLcio -DrunNumber=1234 […]
Either way you’re going to need to extract the run number from the EVIO file name (should be easy I think using a grep or shell command?) and then set it manually using one of the above methods in the next recon passes.
I have also added extraction of run numbers from the head bank that appears every ~40 events in the data. But not all of the runs have these. It will be less of a problem in the future as this should happen more automatically. For now, we should assume not all of the runs have the head banks.
—Jeremy
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
|