Re: Purple Module load causes opensips core dump

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Purple Module load causes opensips core dump

mani sivaraman
I compiled and installed opensips 1.5.1 with mysql db and works fine. I compiled and installed the new purple module from outside and installed the same. if I try to load the purple.so module and add modprobe for purple params, opensips core dumps saying

Jun 12 12:06:38 [22121] CRITICAL:core:receive_fd: EOF on 15
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: child process 22110 exited by a signal 11
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: core was generated
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: terminating due to SIGCHLD
Jun 12 12:06:38 [22104] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22105] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22106] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22107] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22108] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22114] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22109] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22115] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22116] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22117] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22119] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22121] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22112] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22103] NOTICE:presence:destroy: destroy module ...


The gdb back trace is not very useful either.

(no debugging symbols found)
Core was generated by `opensips'.
Program terminated with signal 11, Segmentation fault.
[New process 22083]
#0  0x085a9c6d in ?? ()
(gdb) backtrace full
#0  0x085a9c6d in ?? ()
No symbol table info available.
(gdb)

Help Any one , Please.


_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Reply | Threaded
Open this post in threaded view
|

fix_nated_sdp

Anil M Pannikode (hotmail)

Does fix_nated_sdp("2") function only works on SIP INVITE ?

 

I am trying to rewrite the media IP and looks like it is working only for SIP INVITE. (Not for any subsequent messages like 200 OK)

 

In my case I get an invite from an external client and my server re-invites for T.38. The client responds with 200 OK and I want to rewrite the media IP address. (Since the client is currently passing an internal ip for media IP)

 

 

Anil

 

 


_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Reply | Threaded
Open this post in threaded view
|

Re: fix_nated_sdp

Alex Balashov
fix_nated_sdp() operates on any request or reply containing an SDP payload.

If you wish to apply it to a 200 OK with SDP payload, you need to call
it from a reply route that catches 200 OKs;  200 OKs are replies, not
requests.  For example:

route {
    ...

    # Relay INVITE.

    t_on_reply("1");

    if(!t_relay())
       sl_reply_error();

    exit;
}

onreply_route[1] {
    if(t_check_status("200")) {
       if(nat_uac_test("8") && search("Content-type: application/sdp")) {
          fix_nated_sdp("2");
       }
    }
}

Anil M Pannikode (hotmail) wrote:

> Does fix_nated_sdp("2") function only works on SIP INVITE ?
>
>  
>
> I am trying to rewrite the media IP and looks like it is working only
> for SIP INVITE. (Not for any subsequent messages like 200 OK)
>
>  
>
> In my case I get an invite from an external client and my server
> re-invites for T.38. The client responds with 200 OK and I want to
> rewrite the media IP address. (Since the client is currently passing an
> internal ip for media IP)
>
>  
>
>  
>
> Anil
>
>  
>
>  
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> [hidden email]
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users


--
Alex Balashov
Evariste Systems
Web    : http://www.evaristesys.com/
Tel    : (+1) (678) 954-0670
Direct : (+1) (678) 954-0671
Mobile : (+1) (678) 237-1775

_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Reply | Threaded
Open this post in threaded view
|

Re: fix_nated_sdp

Anil M Pannikode (hotmail)
Thanks that worked.

Anil


-----Original Message-----
From: Alex Balashov [mailto:[hidden email]]
Sent: Friday, June 12, 2009 5:13 PM
To: Anil M Pannikode (hotmail)
Cc: [hidden email]
Subject: Re: [OpenSIPS-Users] fix_nated_sdp

fix_nated_sdp() operates on any request or reply containing an SDP payload.

If you wish to apply it to a 200 OK with SDP payload, you need to call
it from a reply route that catches 200 OKs;  200 OKs are replies, not
requests.  For example:

route {
    ...

    # Relay INVITE.

    t_on_reply("1");

    if(!t_relay())
       sl_reply_error();

    exit;
}

onreply_route[1] {
    if(t_check_status("200")) {
       if(nat_uac_test("8") && search("Content-type: application/sdp")) {
          fix_nated_sdp("2");
       }
    }
}

Anil M Pannikode (hotmail) wrote:

> Does fix_nated_sdp("2") function only works on SIP INVITE ?
>
>  
>
> I am trying to rewrite the media IP and looks like it is working only
> for SIP INVITE. (Not for any subsequent messages like 200 OK)
>
>  
>
> In my case I get an invite from an external client and my server
> re-invites for T.38. The client responds with 200 OK and I want to
> rewrite the media IP address. (Since the client is currently passing an
> internal ip for media IP)
>
>  
>
>  
>
> Anil
>
>  
>
>  
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> [hidden email]
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users


--
Alex Balashov
Evariste Systems
Web    : http://www.evaristesys.com/
Tel    : (+1) (678) 954-0670
Direct : (+1) (678) 954-0671
Mobile : (+1) (678) 237-1775


_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Reply | Threaded
Open this post in threaded view
|

Re: Purple Module load causes opensips core dump

Eric PTAK
In reply to this post by mani sivaraman
Hi mani,
 
purple module hasn't been tested with opensips, and I can't say if it should work.
but just a question : how did you compile the module ? eg, within which project ?
 
regards,
eric.

2009/6/12 mani sivaraman <[hidden email]>
I compiled and installed opensips 1.5.1 with mysql db and works fine. I compiled and installed the new purple module from outside and installed the same. if I try to load the purple.so module and add modprobe for purple params, opensips core dumps saying

Jun 12 12:06:38 [22121] CRITICAL:core:receive_fd: EOF on 15
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: child process 22110 exited by a signal 11
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: core was generated
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: terminating due to SIGCHLD
Jun 12 12:06:38 [22104] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22105] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22106] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22107] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22108] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22114] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22109] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22115] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22116] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22117] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22119] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22121] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22112] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22103] NOTICE:presence:destroy: destroy module ...


The gdb back trace is not very useful either.

(no debugging symbols found)
Core was generated by `opensips'.
Program terminated with signal 11, Segmentation fault.
[New process 22083]
#0  0x085a9c6d in ?? ()
(gdb) backtrace full
#0  0x085a9c6d in ?? ()
No symbol table info available.
(gdb)

Help Any one , Please.


_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users



_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Reply | Threaded
Open this post in threaded view
|

Re: Purple Module load causes opensips core dump

Eric PTAK
Did you added LD_PRELOAD=/path/to/your/libpurple.so env var in your ctl script ?
 
Eric.

2009/6/15 mani sivaraman <[hidden email]>
Thanks for getting back Eric. After opensips I tested the latest Kamailio 1.5.1 code with purple module. I get the same SIGCHLD core dump on Kamailio as well. I saw Kamailio comes with purple source code. If I load purple.so and then add the modprobe for purple, I get the core dump. Is there any thing I can check or look for in the debug to make sure it works. If I unload purple and remove the respective modprobes every thing seems to work fine.

Thank you


On Mon, Jun 15, 2009 at 3:15 AM, Eric PTAK <eric.ptak.fr@gmail.com> wrote:
Hi mani,
 
purple module hasn't been tested with opensips, and I can't say if it should work.
but just a question : how did you compile the module ? eg, within which project ?
 
regards,
eric.

2009/6/12 mani sivaraman <[hidden email]>
I compiled and installed opensips 1.5.1 with mysql db and works fine. I compiled and installed the new purple module from outside and installed the same. if I try to load the purple.so module and add modprobe for purple params, opensips core dumps saying

Jun 12 12:06:38 [22121] CRITICAL:core:receive_fd: EOF on 15
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: child process 22110 exited by a signal 11
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: core was generated
Jun 12 12:06:38 [22103] INFO:core:handle_sigs: terminating due to SIGCHLD
Jun 12 12:06:38 [22104] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22105] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22106] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22107] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22108] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22114] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22109] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22115] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22116] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22117] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22119] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22121] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22112] INFO:core:sig_usr: signal 15 received
Jun 12 12:06:38 [22103] NOTICE:presence:destroy: destroy module ...


The gdb back trace is not very useful either.

(no debugging symbols found)
Core was generated by `opensips'.
Program terminated with signal 11, Segmentation fault.
[New process 22083]
#0  0x085a9c6d in ?? ()
(gdb) backtrace full
#0  0x085a9c6d in ?? ()
No symbol table info available.
(gdb)

Help Any one , Please.


_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users





_______________________________________________
Users mailing list
[hidden email]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users