[Spce-user] Call does not stop after hang-up

[Marcos Pytel]

Hi all!

I tried again to actívate TOPOS. Apparently it fixed the problem with record router headers empy, but... After a few minutes of actívate It, the LB Service went down! I restarted it but the error appears again, it fails.

The log of LB shows:

Jan 22 17:05:28 sipwise lb[29645]: CRITICAL: <core> [core/pass_fd.c:277]: receive_fd(): EOF on 20
Jan 22 17:05:28 sipwise lb[29552]: ALERT: <core> [main.c:745]: handle_sigs(): child process 29596 exited by a signal 6
Jan 22 17:05:28 sipwise lb[29552]: ALERT: <core> [main.c:748]: handle_sigs(): core was generated

Restart LB: 

Jan 22 17:06:32 sipwise lb[843]: WARNING: tls [tls_init.c:778]: init_tls_h(): openssl bug #1491 (crash/mem leaks on low memory) workaround enabled (on low memory tls operations will fail preemptively) with free memory thresholds 26738688 and 13369344 bytes
Jan 22 17:07:01 sipwise lb[893]: CRITICAL: <core> [core/pass_fd.c:277]: receive_fd(): EOF on 23
Jan 22 17:07:01 sipwise lb[843]: ALERT: <core> [main.c:745]: handle_sigs(): child process 847 exited by a signal 6
Jan 22 17:07:01 sipwise lb[843]: ALERT: <core> [main.c:748]: handle_sigs(): core was generated


I did a rollback and now is working normally.

What can be happening?

Thank you in advance!
Marcos.

-----Mensaje original-----
De: Andrew Pogrebennyk <apogrebennyk at sipwise.com> 
Enviado el: miércoles 9 de enero del 2019 07:02
Para: Marcos Pytel <marcos.pytel at cotesma.com.ar>; 'Alex Lutay' <alutay at sipwise.com>; spce-user at lists.sipwise.com
Asunto: Re: [Spce-user] Call does not stop after hang-up

On 01/08/2019 08:18 PM, Marcos Pytel wrote:
> Jan  8 16:11:35 sipwise lb[4196]: ERROR: rr [loose.c:188]: 
> find_next_route(): failed to parse Route body Jan  8 16:11:35 sipwise 
> lb[4196]: ERROR: rr [loose.c:855]: after_loose(): failed to find next 
> route Jan  8 16:13:14 sipwise lb[4211]: ERROR: <core> 
> [core/udp_server.c:607]: udp_send(): 
> sendto(sock,0x7f64290db8f0,1089,0,10.7.255.250:5060,16): Invalid 
> argument(22) Jan  8 16:13:14 sipwise lb[4211]: CRITICAL: <core> 
> [core/udp_server.c:612]: udp_send(): invalid sendtoparameters#012one 
> possible reason is the server is bound to localhost and#012attempts to 
> send to the net Jan  8 16:13:38 sipwise lb[4213]: ERROR: <core> 
> [core/forward.h:219]: msg_send_buffer(): udp_send failed

Hi Marcos,
yes, this is the problem and most probably it's caused by wrong route or contact headers from the end device.
Unfortunately, I still can't open the new pcap file you've provided ("The capture file appears to be damaged or corrupt.
(pcap: File has 823214080-byte packet, bigger than maximum of 262144)") How did you capture it? Did you press Ctrl+C to stop capture before copying the file from your machine?
Anyway, I would suggest to check the headers, especially in the 200 OK received from callee. There must be something wrong.

Also, as a generic hint sometimes it helps activating topology stripping feature (set kamailio.lb.security.topos.enable: 'yes' in config.yml if you are on mr6.5.x release) because then we are stripping the rec-route headers. But it will not help with bad INVITE or 200 from the UA.

Hope this helps,
Andrew