Hi Daniel,
I will see if I can grow my balls big enough to upgrade :-)
Thanks for your help and time.
Regards,
Kristian.
On Wednesday 14 January 2015 14:58:22 Daniel-Constantin Mierla wrote:
Hello,
ok, so it is from the pid.
At this moment I don't see any reason in that part of code to
Hello,
the operation done by that code is practically a sql query, fetching the
result and packing it for nathelper usage. It is no race between it and
handling any sip packet, no shared memory accessed.
Do you get only one corefile?
From the backtrace it looks like row pointer is not ok, can
Hello,
the string_val field look ok. But p variable which is assigned that
value a line before the reported crash is not, as reported by bt full:
p = 0xcfcdffaf Address 0xcfcdffaf out of bounds
Do you have enabled core files per process/pid? I still wonder if it is
a overwritten core.
Hi Daniel,
# kamailio -V
version: kamailio 3.1.1 (x86_64/linux) c274ce
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,
DBG_QM_MALLOC, USE_FUTEX, FAST_LOCK-ADAPTIVE_WAIT, USE_DNS_CACHE,
Hello,
ok, so it is from the pid.
At this moment I don't see any reason in that part of code to crash,
unless hardware problems.
Maybe you can at least upgrade to latest version in 3.1.x series, to
rule out the side effects of the bugs found in 3.1 branch at that time.
It should be same config
Hi Daniel,
(gdb) p *row
$1 = {values = 0x9bd2b8, n = 13}
Regards,
Kristian.
On Wednesday 14 January 2015 10:01:41 Daniel-Constantin Mierla wrote:
Hello,
the operation done by that code is practically a sql query, fetching the
result and packing it for nathelper usage. It is no race
Hi Daniel,
Sorry. Forgot to answer your question.
I am quite sure, I only got one core file.
Regards,
Kristian Høgh
Uni-tel
On Wednesday 14 January 2015 10:01:41 Daniel-Constantin Mierla wrote:
Hello,
the operation done by that code is practically a sql query, fetching the
result and
Hello,
get also:
p *row-values
p i
Cheers,
Daniel
On 14/01/15 10:50, Kristian F. Høgh wrote:
Hi Daniel,
(gdb) p *row
$1 = {values = 0x9bd2b8, n = 13}
Regards,
Kristian.
On Wednesday 14 January 2015 10:01:41 Daniel-Constantin Mierla wrote:
Hello,
the operation done by that code is
Hi Daniel,
(gdb) p *row-values
$1 = {type = DB1_STRING, nul = 0, free = 0, val = {int_val = 58055112, ll_val =
58055112, double_val = 2.8683036404665902e-316, time_val = 58055112, string_val
= 0x375d9c8
sip:44221xxx@192.168.1.157:37612;rinstance=8c613319a30caff6;transport=TCP,
str_val = {
On 14/01/15 14:21, Kristian F. Høgh wrote:
Hi Daniel,
# kamailio -V
version: kamailio 3.1.1 (x86_64/linux) c274ce
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,
DBG_QM_MALLOC, USE_FUTEX,
10 matches
Mail list logo
