[SR-Users] Issue while initiating database of my new module in kamailio

[Daniel-Constantin Mierla]

Hello,

On 12/6/11 5:28 AM, Jagdeep S wrote:
> Hi Daniel,
>
> Thanks for the reply. I was just seeing differences in various module
> initialization over the weekend and saw that the exports structure was
> slightly different from what i had taken from the template in the
> kamailio website.
ok, probably that doc is for older version. Was it a wiki page? If you 
can give the link, we can update it and make a note about versions and 
validity of the content.

Cheers,
Daniel

>   I think the extra parameters/methods added in the
> latest versions were missing in my exports structure due to which my
> initialization used to fail. And during initialization i pass "pseudo
> variable" as argument to my method but i was not  having the pseudo
> variable member in the structure. Since it was not initialized the db
> initialization failed.
>
> I just changed the following structure from:
>
> struct module_exports exports= {
>    "mymodule",	/* module's name */
>    DEFAULT_DLFLAGS, /*!<  dlopen flags */
>    cmds,          /* exported functions */
>    mod_params,    /* param exports */
>    mod_stats,     /* exported statistics */
>    mod_init,      /* module initialization function */
>    0,             /* reply processing function FIXME Not sure when this
> is used */
>    mod_destroy,   /* Destroy function */
>    child_init     /* per-child init function */
> };
>
> to the following structure:
> struct module_exports exports= {
>         "mymodule",   /* module's name */
>          DEFAULT_DLFLAGS, /* dlopen flags */
>          cmds,       /* Exported functions */
>          mod_params, /* Exported parameters */
>          mod_stats,  /* exported statistics */
>          0,          /* exported MI functions */
>          0,          /* exported pseudo-variables */
>          0,          /* extra processes */
>          mod_init,   /* module initialization function */
>          0,          /* response function */
>          mod_destroy,/* destroy function */
>          child_init  /* child initialization function */
> };
>
> Now i am able to start the kamailio server with db initialized :)
> Anyways thanks for the response.
>
> Regards,
> Jagdeep S
>
>
>
> On 12/6/11, Daniel-Constantin Mierla<miconda at gmail.com>  wrote:
>> Hello,
>>
>> can you paste here the mod_init() and child_init() functions and say
>> which one is line 239 (inside child init)? I expect some db init step
>> for your module was not done.
>>
>> Also, the output of 'bt full' might be useful.
>>
>> Cheers,
>> Daniel
>>
>> On 12/3/11 6:35 AM, Jagdeep S wrote:
>>> Hi All,
>>>
>>> I am trying to create a new module under the kamailio source base, as
>>> part of understanding the framework. But when i start the Kamailio, i
>>> am seeing a core dump occuring when my new module is getting initiated
>>> and when the database is getting initialised. PFA the files attached
>>> containing the config file and source files used for creating module
>>> files.
>>>
>>> I am using the "DEFAULT_DB_URL" as database uri. The core dump
>>> backtrace shows that the module init fails at the following point:
>>> (gdb) bt
>>> #0  0x00000000 in ?? ()
>>> #1  0x00ef390d in child_init (rank=9172352) at my_mod.c:239
>>> #2  0x081226be in init_mod (m=0xb7515a04) at sr_module.c:922
>>> #3  0x08122630 in init_mod (m=0xb7515c78) at sr_module.c:919
>>>
>>> The line 239 in my module file points to the child-init function which
>>> initializes the database connectivity ==>   "mymod_dbh =
>>> mymod_dbf.init(&db_url);"
>>>
>>> I have written the module using the template provided under the
>>> kamailio website. All the parameters are defined properly. Have no
>>> idea as to what is going wrong.
>>>
>>> I tried debugging the core dump using gdb but of no much help. I am
>>> not able to run the gdb during runti sr-users at lists.sip-router.orgme
>>> due to which i am not able to add break points and check. Not sure
>>> what the problem is. That's why i generated core dump file and
>>> analysed the backtrace, but i can't add break points and run it again
>>> as the executable is not involved.
>>>
>>> Can anyone please let me know what is the issue is or any pointers if
>>> i have missed some part of the code due to which the connectivity to
>>> db is not happening. Due to this i am not able to proceed further with
>>> the testing of my new module. Also any tips on how to do get the gdb
>>> debugging during runtime will also be very helpful.
>>>
>>> Any help is appreciated :).
>>>
>>>
>>>
>>> _______________________________________________
>>> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
>>> sr-users at lists.sip-router.org
>>> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>> --
>> Daniel-Constantin Mierla -- http://www.asipto.com
>> http://linkedin.com/in/miconda -- http://twitter.com/miconda
>>
>>
>

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