Open shaoqiang1987 opened 5 years ago
nmirabal
commented
5 years ago Thanks for reporting this. There are some problems in gtobssim and the developers are trying to track down the issue. Until this is fixed, please note that there is a new Galactic diffuse background gll_iem_v07.fits that you can try with gtobssim
https://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html
Also, please make sure to install the latest version of the Fermitools.
shaoqiang1987
commented
5 years ago I installed the latest version of the Fermitools and adopted the new Galactic diffuse background (gll_iem_v07.fits). I simulated the data over a shorter time ( simtime=1e5), but find the error "FITSIO status = 202: keyword not found in header
Caught St13runtime_error at the top level: FitsImage::read_fits_image: CFITSIO error 202" . I rewrite the
gll_iem_v07.fits by changing the header file ( add the CTYPE3='Energy' and CRPIX3=1 to header), then the simulation run well.
I used a longer time(simtime=1e7) ,however, we still got the error " Caught St12out_of_range at the top level: vector::_M_range_check".
jaulbric
commented
5 years ago Are there any updates on this bug?
shaoqiang1987
commented
5 years ago No updates
在 2019年4月25日，下午10:39，Jaryd Ulbricht notifications@github.com 写道：
Are there any updates on this bug?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.
donhorner
commented
5 years ago Sorry about this. There is no one at the Fermi Science Support Center that is familiar with this tool anymore. We reached out to the Fermi collaboration for help, but no one has been able or had the time to provide assistance yet.
jaulbric
commented
5 years ago It looks like the bug is being caused by MapSource::samplePixel in celestialSources/genericSources/src/MapSource.cxx. There are two vectors that contain the latitude and longitude information from the fits image and this function is trying to grab elements from these vectors that are going out of range. I've been able to run simulations by setting the pixels near the boundaries of the diffuse model image to 1.e-37. My guess is that the simulator only tries to generate events coming from the edge of the diffuse model during long simulations because the flux from those areas is rather low. If the few degrees on the boundaries of the mapcube are set to really really small numbers this prevents the problem because the simulator will only generate photons originating from the interior of the fits image.
donhorner
commented
5 years ago @jaulbric Thanks for looking into it. I'll see if I can get someone to follow up on this.
Haiming19
commented
5 years ago @nmirabal Are there any one help to update on this bug now?
donhorner
commented
5 years ago I'm sorry to report that we have really done anything about it. The people who created tools like gtobssim have left the Fermi project, so we no longer have any expertise in that area. No one works full time on the Fermtiools either, and like @jaulbric said the linked issue, it's hard to make it a priority when it's not a commonly used tool, and no one in the collaboration has been able to step up to help.
We're happy to accept help (and pull requests) if someone in the community wants to look into it and submit a patch.
imereu
commented
4 years ago I have the same problem....
If we simulated the data of Galactic Background (gll_iem_v06.fits) over a shorter time (for instance, simtime=1e5) as follow: "gtobssim emin=6000. emax=500000. edisp=no infile=model.xml srclist=GalacticDiffuse_srcIds.txt scfile=spacecraft.fits evroot=GalacticDiffuse simtime=100000.0 tstart=239557418.0 maxtime=558895302.0 use_ac=yes ra=95.0 dec=13.0 radius=10.0 irfs=P8R3_SOURCE_V2 seed=635514 chatter=2", the program run well.
However, If we simulate a longer time (simtime=1e7) data of gll_iem_v06.fits, we will obtain the error "Caught St12out_of_range at the top level: vector".