[sr-dev] Segfaults in mem subsystem

[Daniel-Constantin Mierla]

Hello,

you are running a development version from the latest stable series -- 
meanwhile there were fixes and stable releases out of that code. Upgrade 
to either latest stable in branch 4.0 or master branch to be sure is not 
something that was fixed already.

Cheers,
Daniel

On 6/22/13 9:10 PM, Николай wrote:
> Hi *,
>
> we are using kamailio built from source (with minor changes not in mem
> subsystem):
>
> # /usr/local/sbin/kamailio -V
> version: kamailio 3.4.0-dev6 (x86_64/linux) cd2b26-dirty
> flags: STATS: Off, USE_IPV6, USE_TCP, USE_TLS, TLS_HOOKS, USE_RAW_SOCKS,
> DISABLE_NAGLE, USE_MCAST, DNS_IP_HACK, SHM_MEM, SHM_MMAP, PKG_MALLOC,
> USE_FUTEX, FAST_LOCK-ADAPTIVE_WAIT, USE_DNS_CACHE, USE_DNS_FAILOVER,
> USE_NAPTR, USE_DST_BLACKLIST, HAVE_RESOLV_RES
> ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16,
> MAX_URI_SIZE 1024, BUF_SIZE 65535, DEFAULT PKG_SIZE 4MB
> poll method support: poll, epoll_lt, epoll_et, sigio_rt, select.
> id: cd2b26 -dirty
> compiled on 17:41:52 Jan  6 2013 with gcc 4.6.3
>
>
> and facing segfaults in memory control subsystem in different places.
> Some examples follow:
>
> 1)
> (gdb) back
> #0  del_nonshm_lump (lump_list=0x7f750e650600) at data_lump.c:668
> #1  0x00007f750e413d9c in ?? ()
> #2  0x00000000008db480 in mem_pool ()
> #3  0x00007f750e4137e1 in ?? ()
> #4  0x0000000000000004 in ?? ()
> #5  0x0000000001f200a7 in ?? ()
> #6  0x00007f740b5e40f0 in ?? ()
> #7  0x00007f740dc6d2a0 in ?? ()
> #8  0x00007f740d9d0af0 in ?? ()
> #9  0x00007f740c605900 in ?? ()
> #10 0x0000000000000001 in ?? ()
> #11 0x0000000000000000 in ?? ()
> (gdb) info local
> r = 0x83b7b789de298df2
> crt = 0x7f740c43ba60
> prev = 0x7f750e650600
> prev_r = 0x7f740c43ba60
>
> 2)
> (gdb) back
> #0  free_to_params (tb=0x1) at parser/parse_to.c:824
> #1  free_to (tb=0x1) at parser/parse_to.c:836
> #2  0x00007f4249ba2e92 in ?? ()
> #3  0x00000000008db480 in mem_pool ()
> #4  0x00007f4249ba27e1 in ?? ()
> #5  0x0000000000000004 in ?? ()
> #6  0x00000000027a5a37 in ?? ()
> #7  0x00007f4146d831e0 in ?? ()
> #8  0x00007f4148b7da50 in ?? ()
> #9  0x00007f4147eec600 in ?? ()
> #10 0x00007f4147f19350 in ?? ()
> #11 0x0000000000000001 in ?? ()
> #12 0x0000000000000000 in ?? ()
> (gdb) info local
> tp = <value optimized out>
>
> 3)
>
> (gdb) back
> #0  free_lump_list (lump_list=<value optimized out>) at data_lump.c:504
> #1  del_nonshm_lump (lump_list=<value optimized out>) at data_lump.c:661
> #2  0x00007f0631a69d9c in ?? ()
> #3  0x00000000008db480 in mem_pool ()
> #4  0x00007f0631a697e1 in ?? ()
> #5  0x0000000000000004 in ?? ()
> #6  0x0000000000f4eb77 in ?? ()
> #7  0x00007f052ec5a5b0 in ?? ()
> #8  0x00007f05303071d0 in ?? ()
> #9  0x00007f05309374c0 in ?? ()
> #10 0x00007f0530e0ccd0 in ?? ()
> #11 0x0000000000000001 in ?? ()
> #12 0x0000000000000000 in ?? ()
> (gdb) info local
> r = 0x7fa2312af83ca962
>
>
> Are the issues already known/fixed? Should I post a bugreport?
> The memory subsystem looks pretty buggy... Does kamailio implement/use
> its own mem allocating system?
>
> PS I'm new to the list - sorry for possibly bad email
>
> Thanks
> ----
> Nikolay
>
>
> _______________________________________________
> sr-dev mailing list
> sr-dev at lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev

-- 
Daniel-Constantin Mierla - http://www.asipto.com
http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda