Get a Segfault on 5000 SRR files

[fataltes]

Hi,

Thanks for your interesting tool.

I could successfully run your tool on the original brain, breast, blood dataset from SBT.

I tried to run SeqOthello on 5000 raw read experiments and it failed with a segfault after 9 hours.

I made sure that it’s not because of a broken input file by running SeqOthello on 5 batches of 1000 splits of the original 5000 experiments and all of them passed.
It’s also not because of out of memory because /usr/bin/time reported only 22Gs of memory usage after the segfault while the system we use has way more memory than this.
It’s not because of out of disk space because the taken space on disk after failure was 131Gs while we have ~4TB available.

That would be great if you guys can help us out here to figure out what might have gone wrong.

Here is the gdb backtrace report in case it might help you understand the issue better:

#0  0x0000000000446574 in Othello<unsigned long>::testHash(unsigned int) ()                                                                                                          #1  0x0000000000449487 in Othello<unsigned long>::Othello(unsigned char, unsigned long*, unsigned int, bool, void*, unsigned long, int) ()
#2  0x000000000044d6d3 in L1Node::constructothello(unsigned int, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) ()
#3  0x000000000044df5a in std::thread::_Impl<std::_Bind_simple<std::_Mem_fn<void (L1Node::*)(unsigned int, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)> (L1Node*, unsigned int, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)> >::_M_run() ()
#4  0x00007ffff76987a0 in ?? () from /home/linuxbrew/.linuxbrew/lib/libstdc++.so.6
#5  0x00007ffff79a66ca in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0
#6  0x00007ffff6e03caf in clone () from /lib/x86_64-linux-gnu/libc.so.6

Thank you,
Fatemeh