segfaults in libthr

2009-07-23 Thread Andrew Berry

Hi,

I just upgraded a server from 7.2-RELEASE to 7.2-RELEASE-p2. I'm  
having issues with any port which links to libthr. This includes apr,  
subversion, and sqlite, so it's pretty common for httpd to segfault. I  
rebuilt libthr with debugging symbols, and I get the following  
backtrace. This is the result of an 'svn ls' operation.


Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x28e60700 (LWP 100176)]
0x28287e24 in mutex_lock_common (curthread=0x28e60700, m=0x297b3100,  
abstime=0x0) at /usr/src/lib/libthr/thread/thr_mutex.c:422

422 ENQUEUE_MUTEX(curthread, m);
(gdb) bt
#0  0x28287e24 in mutex_lock_common (curthread=0x28e60700,  
m=0x297b3100, abstime=0x0) at /usr/src/lib/libthr/thread/thr_mutex.c:422
#1  0x28287d2f in __pthread_mutex_lock (mutex=0x28d56c00) at /usr/src/ 
lib/libthr/thread/thr_mutex.c:451
#2  0x28d16862 in pthreadMutexEnter () from /usr/local/lib/ 
libsqlite3.so.8
#3  0x28d1674b in sqlite3MutexInit () from /usr/local/lib/ 
libsqlite3.so.8
#4  0x28d14875 in sqlite3_initialize () from /usr/local/lib/ 
libsqlite3.so.8

#5  0x28c993fc in init_sqlite () from /usr/local/lib/libsvn_subr-1.so.0
#6  0x28c7b259 in svn_atomic__init_once () from /usr/local/lib/ 
libsvn_subr-1.so.0

#7  0x0001 in ?? ()
#8  0x in ?? ()
#9  0x28cb22f0 in __JCR_LIST__ () from /usr/local/lib/libsvn_subr-1.so.0
#10 0x297bf020 in ?? ()
#11 0x297bf06c in ?? ()
#12 0xbfbfe538 in ?? ()
#13 0x28c9a179 in svn_sqlite__open () from /usr/local/lib/ 
libsvn_subr-1.so.0

Previous frame inner to this frame (corrupt stack?)

Anyone have any suggestions? I've all ready rebuilt affected ports  
without any change. Can I possibly rebuild the -p1 or -RELEASE version  
of libthr without horribly breaking the system?


Thanks,
--Andrew
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"


segfaults in libthr

2009-07-23 Thread Andrew Berry

(sorry for any duplicates, I'm having email issues)

Hi,

I just upgraded a server from 7.2-RELEASE to 7.2-RELEASE-p2. I'm  
having issues with any port which links to libthr. This includes apr,  
subversion, and sqlite, so it's pretty common for httpd to segfault. I  
rebuilt libthr with debugging symbols, and I get the following  
backtrace. This is the result of running httpd under gdb with 'run -X'.


(gdb) run -X
Starting program: /usr/local/sbin/httpd -X
[New LWP 100110]
[New Thread 0x28401040 (LWP 100110)]
[Thu Jul 23 12:45:59 2009] [warn] (2)No such file or directory: Failed  
to enable the 'httpready' Accept Filter
[Thu Jul 23 12:45:59 2009] [warn] (2)No such file or directory: Failed  
to enable the 'httpready' Accept Filter


Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x28e61700 (LWP 100110)]
0x282a5e24 in mutex_lock_common (curthread=0x28e61700, m=0x28409d40,  
abstime=0x0) at /usr/src/lib/libthr/thread/thr_mutex.c:422

422 ENQUEUE_MUTEX(curthread, m);
(gdb) bt
#0  0x282a5e24 in mutex_lock_common (curthread=0x28e61700,  
m=0x28409d40, abstime=0x0) at /usr/src/lib/libthr/thread/thr_mutex.c:422
#1  0x282a5d2f in __pthread_mutex_lock (mutex=0x28e54c7c) at /usr/src/ 
lib/libthr/thread/thr_mutex.c:451
#2  0x2826c5d0 in apr_thread_mutex_lock (mutex=0x28e54c78) at locks/ 
unix/thread_mutex.c:92

#3  0x2866dde2 in ?? () from /usr/local/libexec/apache22/mod_ssl.so
#4  0x287d0e05 in CRYPTO_lock () from /lib/libcrypto.so.5
#5  0x286b0a82 in SSL_COMP_get_name () from /usr/lib/libssl.so.5
#6  0x286b0da7 in SSL_COMP_get_compression_methods () from /usr/lib/ 
libssl.so.5

#7  0x2868a730 in SSL_library_init () from /usr/lib/libssl.so.5
#8  0x28dd0503 in zm_startup_openssl () from /usr/local/lib/php/ 
20060613/openssl.so
#9  0x289f3d70 in zend_startup_module_ex () from /usr/local/libexec/ 
apache22/libphp5.so
#10 0x289f8d1c in zend_hash_apply () from /usr/local/libexec/apache22/ 
libphp5.so
#11 0x289f279c in zend_startup_modules () from /usr/local/libexec/ 
apache22/libphp5.so
#12 0x289ae80d in php_module_startup () from /usr/local/libexec/ 
apache22/libphp5.so
#13 0x28a6b743 in php_apache2_startup () from /usr/local/libexec/ 
apache22/libphp5.so
#14 0x28b05140 in apache2_sapi_module () from /usr/local/libexec/ 
apache22/libphp5.so

#15 0x0001 in ?? ()
#16 0x in ?? ()
#17 0x0001 in ?? ()
#18 0x2840ff10 in ?? ()
#19 0xbfbfebf8 in ?? ()
#20 0x28a6c094 in php_apache_server_startup () from /usr/local/libexec/ 
apache22/libphp5.so
#21 0x28af10a6 in labels.37937 () from /usr/local/libexec/apache22/ 
libphp5.so

#22 0x2840c018 in ?? ()
#23 0x2882017b in status_init (p=0x28a6b720, plog=0x289ad3e0,  
ptemp=0x0, s=0x0) at mod_status.c:851

Previous frame inner to this frame (corrupt stack?)

Anyone have any suggestions? I've all ready rebuilt affected ports  
without any change. Can I possibly rebuild the -p1 or -RELEASE version  
of libthr without horribly breaking the system?


Thanks,
--Andrew
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"