JeffersonLab / halld_recon

Reconstruction for the GlueX Detector
7 stars 9 forks source link

DCTOFPoint_factory cited in hd_root crash from halld_recon-4.51.0 #853

Open nsjarvis opened 2 hours ago

nsjarvis commented 2 hours ago

I found that these two evio files crash hd_root with similar errors: hd_rawdata_100740_000.evio hd_rawdata_100789_000.evio

The hint lines from one job are below, but I will attach the entire log files.
For one I was running danarest and for the other I was running CDC_dedx because danarest already crashed with that file.

I was using version_5.21.0.xml on alma9 at CMU.

`#5 0x00000000011b88ac in DCTOFPoint_factory::evnt(jana::JEventLoop*, unsigned long) ()

6 0x0000000000ae6680 in jana::JFactory::Get(std::vector<DCTOFPoint const, std::allocator<DCTOFPoint const> >&) ()

7 0x0000000000ae93c5 in jana::JFactory jana::JEventLoop::GetFromFactory(std::vector<DCTOFPoint const, std::allocator<DCTOFPoint const> >&, char const, jana::JEventLoop::data_source_t&, bool) ()

8 0x0000000000ae96c9 in jana::JFactory jana::JEventLoop::Get(std::vector<DCTOFPoint const, std::allocator<DCTOFPoint const> >&, char const, bool) ()

9 0x00000000014b4790 in DEventWriterREST::Write_RESTEvent(jana::JEventLoop*, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) const ()

10 0x00007f6ce3cf901c in JEventProcessor_danarest::evnt(jana::JEventLoop*, unsigned long) () from /home/gluex/gluex_top9/halld_recon/halld_recon-4.51.0/Linux_Alma9-x86_64-gcc11.4.1/plugins/danarest.so

11 0x00000000014edb82 in jana::JEventLoop::OneEvent (this=0x7f6ccc000b60) at src/JANA/JEventLoop.cc:693

12 0x00000000014ee1a4 in jana::JEventLoop::Loop (this=this

entry=0x7f6ccc000b60) at src/JANA/JEventLoop.cc:496

13 0x00000000014c30d5 in LaunchThread (arg=0x7fff7b6b1f30) at src/JANA/JApplication.cc:1382

14 0x00007f6cf3489c02 in start_thread () from /lib64/libc.so.6

15 0x00007f6cf350ec40 in clone3 () from /lib64/libc.so.6`

log_std_100740_000.txt

log_std_dedx100789_000.txt

nsjarvis commented 2 hours ago

This also crashes at JLab, with the same JLab default version set. I did not try any other files.

hd_root /cache/halld/RunPeriod-2022-05/rawdata/Run100789/hd_rawdata_100789_000.evio -PPLUGINS=danarest -PNTHREADS=1 -PEVIO:NTHREADS=1

nsjarvis commented 2 hours ago

Same problem with hd_rawdata_100373_000.evio and also hd_rawdata_101622_000.evio.

I think a bug must have crawled in. With no plugins, or tracking-free plugins, it runs ok.