Bug #901
Error with Fermi LAT caldb in ctools
| Status: | New | Start date: | 06/12/2013 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assigned To: |  Knödlseder Jürgen | % Done: | 0% | |
| Category: | - | |||
| Target version: | - | |||
| Duration: |
Description
If you have an old version of ctools and update to a new one you might get the following error:
RuntimeError: *** ERROR in GFits::open(std::string&): Unable to open FITS file "/Users/schulz/software/ctools/share/caldb/data/glast/lat/bcf/ea/aeff_P7SOURCE_V6_front.fits" (status=104)
the problem is that the CALDB for fermi lat is no longer in the ctools but in gammalib. so you have to adapt the $CALDB path and include the gammalib caldb path:
export CALDB=$GAMMALIB/share/caldb/:$CALDB
Recurrence
No recurrence.
History
#1
Updated by Schulz Anneli over 11 years ago
- Description updated (diff)
#2
Updated by Schulz Anneli over 11 years ago
This also leads to the fact that the example scripts ($CTOOLS/share/examples/python/make_unbinned_analysis.py and ...binned... ) in python don’t work anymore with the default parameters since the paths are wrongly specified.
First you have to adjust the IRF name to a filename which is in $GAMMALIB/share/caldb/cta/
e.g. from “cta_dummy_irf” to “kb_A_50h_v3”
(Alternatively copy cta_dummy_irf from $CTOOLS... to $GAMMALIB...)
Second you need to adjust the model path: from model_name="${GAMMALIB}/share/models/crab.xml” to "${CTOOLS}/share/models/crab.xml”
and in model.xml: from "$GAMMALIB/share/models/bkg_dummy.txt” to "$CTOOLS/share/models/bkg_dummy.txt”
( Alternatively you can just decide to have the models folder also in $GAMMALIB/share/)
#3
Updated by Knödlseder Jürgen over 11 years ago
- Assigned To set to Knödlseder Jürgen
Can you try using
GAMMALIB_CALDB=/usr/local/gamma/share/caldb
In fact,
GAMMALIB_CALDB takes now precedence over CALDB to avoid conflicts with the Fermi Science Tools.
But I guess you problem is more related to the fact that you installed gammalib and ctools in different paths, which is likely not tested in detail. I’ll look into this when I find some time.