Open ldigesu opened 9 months ago
Ciao Laura, unfortunately I have no real experience on the Mac, and I can't offer any informed suggestion.
The only real place where you can get a segmentation fault in ixpeobssim, I think, is when loading the XSPEC models. The first thing that I would do would be to navigate into ixpeobssim/xspec
and run a
cd ixpeobssim/xspec
./compile --cleanup
then try to run any ixpeobssim application. If you get a warning that the XSPEC models could not be loaded but everything else runs fine, then we're up to something. At that point you can try and recompile the models against the new release
cd ixpeobssim/xspec
./compile
The only other thing that comes to mind is if your update left a mixture of old and new compiled libraries and you are pointing to the wrong ones?
Caro Luca, Buon anno!
I manage to come back to this issue i.e. getting a segmentation fault form iixpeobbssim application since the update of heasoft, 6.32.1.
I tried what you suggested. The compile.py --cleanup gives this main error (full log output attached):
make[2]: No rule to make target `hd-std-clean-subdirs'. Stop. make[1]: [clean-subdirs] Error 2 make: *** [clean] Error 2
The segmentation fault error persists after the cleanup. Let me know If you could think to any other workaround.
Thanks!
Laura
Da: Luca Baldini @.> Inviato: venerdì 1 dicembre 2023 14:45 A: lucabaldini/ixpeobssim @.> Cc: Di Gesu Laura @.>; Author @.> Oggetto: Re: [lucabaldini/ixpeobssim] Conflict between ixpeobssim and heasarc (version 6.32.1) on Mac OS (12.6.6. Monterey) (Issue #720)
Non si ricevono spesso messaggi di posta elettronica da @.*** Informazioni sul perché è importantehttps://aka.ms/LearnAboutSenderIdentification
Ciao Laura, unfortunately I have no real experience on the Mac, and I can't offer any informed suggestion.
The only real place where you can get a segmentation fault in ixpeobssim, I think, is when loading the XSPEC models. The first thing that I would do would be to navigate into ixpeobssim/xspec and run a
cd ixpeobssim/xspec ./compile --cleanup
then try to run any ixpeobssim application. If you get a warning that the XSPEC models could not be loaded but everything else runs fine, then we're up to something. At that point you can try and recompile the models against the new release
cd ixpeobssim/xspec ./compile
The only other thing that comes to mind is if your update left a mixture of old and new compiled libraries and you are pointing to the wrong ones?
— Reply to this email directly, view it on GitHubhttps://github.com/lucabaldini/ixpeobssim/issues/720#issuecomment-1836144685, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BD6Y23RWR7DAIG5XOFJBF73YHHNQLAVCNFSM6AAAAABABKP5T6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMZWGE2DINRYGU. You are receiving this because you authored the thread.Message ID: @.***>
Hello, I recently updated the heasoft package (to version 6.32.1) on my Mac (12.6.6. Monterey). I am now noticing that when the heasoft is initialized in the bash, ixpeobssim fails.
Specifically, if in my .bashrc, I do the usual heasoft initialization:
export HEADAS=/Applications/heasoft-6.32.1/x86_64-apple-darwin21.6.0 source $HEADAS/headas-init.sh,
then I get "segmentation fault" when running simply e,g. xppimms.py.
When I comment out the heasoft initialization, ixpeobssim works normally.
Any idea of why is this?
Laura