[SR-Users] kamailio 4.2.3 Segmentation fault when using db_cluster module

[Olle E. Johansson]

On 09 Mar 2015, at 17:16, Jan Hazenberg <jenus at cyberchaos.nl> wrote:

> Hello,
> 
> I have a issue on kamailio 4.2.3 when using the db_cluster module. I have the following config:
> 
> #!define DBURL "cluster://cls1"
> 
> # ----- db_cluster params -----
> modparam("db_cluster", "connection",
>             "con1=>mysql://kamailio:kamailiorw@10.121.0.120/kamailio")
> modparam("db_cluster", "connection",
>             "con2=>mysql://kamailio:kamailiorw@10.121.0.121/kamailio")
> modparam("db_cluster", "connection",
>             "con3=>mysql://kamailio:kamailiorw@10.121.0.122/kamailio")
> modparam("db_cluster", "cluster", "cls1=>con1=9r8r;con2=9r8r;con3=9r8r")
> modparam("db_cluster", "inactive_interval", 180)
> modparam("db_cluster", "max_query_length", 5)
> 
> # ----- sqlops params -----
> modparam("sqlops","sqlcon","ca=>cluster://cls1")
> 
> 
> When i start kamailio it fails with:
> 
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: rr [../outbound/api.h:54]: ob_load_api(): Failed to import bind_ob
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: rr [rr_mod.c:160]: mod_init(): outbound module not available
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: usrloc [hslot.c:53]: ul_init_locks(): locks array size 1024
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: permissions [parse_config.c:251]: parse_config_file(): file not found: /usr/local/etc/kamailio/permissions.allow
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: permissions [permissions.c:608]: mod_init(): default allow file (/usr/local/etc/kamailio/permissions.allow) not found => empty rule set
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: permissions [parse_config.c:251]: parse_config_file(): file not found: /usr/local/etc/kamailio/permissions.deny
> Mar  9 17:11:49 localhost kamailio[31063]: INFO: permissions [permissions.c:617]: mod_init(): default deny file (/usr/local/etc/kamailio/permissions.deny) not found => empty rule set
> Mar  9 17:11:49 localhost kernel: kamailio[31063]: segfault at 80 ip 00007f7f4f9ffddc sp 00007fffde2f88a0 error 4 in db_cluster.so[7f7f4f9ec000+28000]
> Mar  9 17:11:50 localhost kamailio: ERROR: <core> [daemonize.c:315]: daemonize(): Main process exited before writing to pipe
> 
> 
> Could this be a bug or is this a configuration issue? I tested the dbnodes and they seem to respond fine to any query's i send.

Configurations should never create segmentation faults, so it's clearly a bug. Can you find the core dump file and produce a backtrace?

/O