v0.14.0rc2 crash on shutdown #9881

issue dooglus opened this issue on February 27, 2017

dooglus commented at 8:22 PM on February 27, 2017: contributor

I had rc2 running overnight. It was fully synced. My computer was being slow, which is sometimes caused by bitcoin-qt doing a lot of disk access. I clicked the X in the top right of the bitcoin-qt gui window to shut it down to see if that would help. Nothing happened - the gui window didn't go away. I figured I had misclicked (using a tap-to-click touchpad it happens) so I clicked the X again. bitcoin-qt crashed, like this:

(gdb) run
Starting program: /home/chris/Programs/bitcoin.git/src/qt/bitcoin-qt 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffe9d6b700 (LWP 7353)]
[New Thread 0x7fffdd8ff700 (LWP 7356)]
[New Thread 0x7fffd5ffe700 (LWP 7357)]
[New Thread 0x7fffd57ad700 (LWP 7358)]
[New Thread 0x7fffd4f5c700 (LWP 7359)]
[New Thread 0x7fffcffff700 (LWP 7360)]
[New Thread 0x7fffcf7ae700 (LWP 7367)]
[New Thread 0x7fffcef5d700 (LWP 7368)]
[New Thread 0x7fffce70c700 (LWP 7369)]
[New Thread 0x7fffcdebb700 (LWP 7370)]
[New Thread 0x7fffcd66a700 (LWP 7371)]
[New Thread 0x7fffb972f700 (LWP 7375)]
[New Thread 0x7fff9d18f700 (LWP 7399)]
[New Thread 0x7fff9c93e700 (LWP 7400)]
[New Thread 0x7fff97fff700 (LWP 7401)]
[New Thread 0x7fff977ae700 (LWP 7402)]
[New Thread 0x7fff96f5d700 (LWP 7403)]
[New Thread 0x7fff9670c700 (LWP 7404)]
[New Thread 0x7fff95ebb700 (LWP 7405)]
[New Thread 0x7fff95033700 (LWP 7407)]
[New Thread 0x7fff77fff700 (LWP 7408)]
[Thread 0x7fff97fff700 (LWP 7401) exited]
[Thread 0x7fff9d18f700 (LWP 7399) exited]
[Thread 0x7fff95033700 (LWP 7407) exited]
[Thread 0x7fffce70c700 (LWP 7369) exited]
[Thread 0x7fffcd66a700 (LWP 7371) exited]
[Thread 0x7fffcef5d700 (LWP 7368) exited]
[Thread 0x7fffcdebb700 (LWP 7370) exited]
[Thread 0x7fffcf7ae700 (LWP 7367) exited]
[Thread 0x7fff96f5d700 (LWP 7403) exited]
[Thread 0x7fff977ae700 (LWP 7402) exited]
[Thread 0x7fff9670c700 (LWP 7404) exited]
[Thread 0x7fff9c93e700 (LWP 7400) exited]
[Thread 0x7fffd5ffe700 (LWP 7357) exited]
[Thread 0x7fff95ebb700 (LWP 7405) exited]
[Thread 0x7fffd4f5c700 (LWP 7359) exited]
[Thread 0x7fffcffff700 (LWP 7360) exited]
[Thread 0x7fffd57ad700 (LWP 7358) exited]
[Thread 0x7fffdd8ff700 (LWP 7356) exited]
[Thread 0x7fff77fff700 (LWP 7408) exited]
[Thread 0x7fffe9d6b700 (LWP 7353) exited]
bitcoin-qt: /usr/include/boost/thread/pthread/recursive_mutex.hpp:101: boost::recursive_mutex::~recursive_mutex(): Assertion `!pthread_mutex_destroy(&m)' failed.

Program received signal SIGABRT, Aborted.
0x00007ffff3292067 in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
56	../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) thread apply all where

Thread 13 (Thread 0x7fffb972f700 (LWP 7375)):
[#0](/bitcoin-bitcoin/0/)  pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
[#1](/bitcoin-bitcoin/1/)  0x00005555559bbb7b in BGThread (this=0x7fffd84d1990) at leveldb/util/env_posix.cc:574
[#2](/bitcoin-bitcoin/2/)  leveldb::(anonymous namespace)::PosixEnv::BGThreadWrapper (arg=0x7fffd84d1990) at leveldb/util/env_posix.cc:521
[#3](/bitcoin-bitcoin/3/)  0x00007ffff4cc30a4 in start_thread (arg=0x7fffb972f700) at pthread_create.c:309
[#4](/bitcoin-bitcoin/4/)  0x00007ffff334562d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Thread 1 (Thread 0x7ffff7fb8780 (LWP 7340)):
[#0](/bitcoin-bitcoin/0/)  0x00007ffff3292067 in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
[#1](/bitcoin-bitcoin/1/)  0x00007ffff3293448 in __GI_abort () at abort.c:89
[#2](/bitcoin-bitcoin/2/)  0x00007ffff328b266 in __assert_fail_base (fmt=0x7ffff33c3f18 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", 
    assertion=assertion@entry=0x555555d9b940 "!pthread_mutex_destroy(&m)", file=file@entry=0x5555559ebc80 "/usr/include/boost/thread/pthread/recursive_mutex.hpp", 
    line=line@entry=101, 
    function=function@entry=0x555555ec1e20 <_ZZN5boost15recursive_mutexD4EvE19__PRETTY_FUNCTION__> "boost::recursive_mutex::~recursive_mutex()") at assert.c:92
[#3](/bitcoin-bitcoin/3/)  0x00007ffff328b312 in __GI___assert_fail (assertion=0x555555d9b940 "!pthread_mutex_destroy(&m)", 
    file=0x5555559ebc80 "/usr/include/boost/thread/pthread/recursive_mutex.hpp", line=101, 
    function=0x555555ec1e20 <_ZZN5boost15recursive_mutexD4EvE19__PRETTY_FUNCTION__> "boost::recursive_mutex::~recursive_mutex()") at assert.c:101
[#4](/bitcoin-bitcoin/4/)  0x00005555555c5601 in boost::recursive_mutex::~recursive_mutex (this=0x555556206220, __in_chrg=<optimized out>)
    at /usr/include/boost/thread/pthread/recursive_mutex.hpp:101
[#5](/bitcoin-bitcoin/5/)  0x000055555594fc78 in ~recursive_mutex (this=0x555556206220, __in_chrg=<optimized out>) at util.cpp:165
[#6](/bitcoin-bitcoin/6/)  ~AnnotatedMixin (this=0x555556206220, __in_chrg=<optimized out>) at sync.h:55
[#7](/bitcoin-bitcoin/7/)  ~CCriticalSection (this=0x555556206220, __in_chrg=<optimized out>) at sync.h:95
[#8](/bitcoin-bitcoin/8/)  CInit::~CInit (this=<optimized out>, __in_chrg=<optimized out>) at util.cpp:164
[#9](/bitcoin-bitcoin/9/)  0x00007ffff3294b29 in __run_exit_handlers (status=0, listp=0x7ffff36025a8 <__exit_funcs>, run_list_atexit=run_list_atexit@entry=true) at exit.c:82
[#10](/bitcoin-bitcoin/10/) 0x00007ffff3294b75 in __GI_exit (status=<optimized out>) at exit.c:104
[#11](/bitcoin-bitcoin/11/) 0x00007ffff327eb4c in __libc_start_main (main=0x5555555c65c0 <main(int, char**)>, argc=1, argv=0x7fffffffe5b8, init=<optimized out>, fini=<optimized out>, 
    rtld_fini=<optimized out>, stack_end=0x7fffffffe5a8) at libc-start.c:321
[#12](/bitcoin-bitcoin/12/) 0x00005555555d7b08 in _start ()
(gdb)

I don't have a lot of trust in my computer. It sometimes fails to run g++ correctly for example. See #9814 (comment) for other discussion of this. But this issue looks real to me.

laanwj commented at 9:28 PM on February 27, 2017: member
These are line 164 and 165 of util.cpp in 0.14.0rc2:
```
        for (int i = 0; i < CRYPTO_num_locks(); i++)
            delete ppmutexOpenSSL[i];
```
Don't see how this could fail, really :/
laanwj added the label Bug on Feb 27, 2017
dooglus commented at 10:23 PM on February 27, 2017: contributor

I've installed memtest86+ ("thorough real-mode memory tester") and will run it next time I reboot the machine. Again, this could be another "ghost" bug caused by a hardware error but I thought it best to report it anyway.

Edit: memtest86+ found no memory errors, though it does hang after 2.5 minutes if I let it use all 4 cores in parallel while testing. No idea what that means.
fanquake commented at 2:17 AM on August 13, 2017: member

Going to close this for the same reason as #9814. If we ever get another report it can be reopened.
fanquake closed this on Aug 13, 2017
DrahtBot locked this on Sep 8, 2021

Contributors

Labels

Linked (view graph)

#1 JSON-RPC support for mobile devices ("ultra-lightweight" clients)#2 Long-term, safe, store-of-value #3 Encrypt wallet #4 Export/Import wallet in a human readable, future-proof format #5 Make the version number the protocol version and not the client version #6 Treat wallet as a generic keystore #7 Block-header-only, faster startup client #8 RPC command to sign text with wallet private key #9 Fix for GUI on Macs and latest wxWidgets #10 Add address to listtransactions output #11 Nolisten patch #12 Monitor transactions and/or blocks #9814 0.14rc1 coredump in QScreen::handle()