Hi Daniel,

In failure route, I used t_reply(). Shall I try send_reply()? What’s the difference?

Cheers,

Allen

From: Daniel-Constantin Mierla [mailto:miconda@gmail.com]
Sent: Monday, 23 June 2014 8:31 p.m.
To: Allen Zhang; Kamailio (SER) - Users Mailing List
Subject: Re: [SR-Users] kamailio not sending 202 after 408 is sent

Hello,

what function are you using inside kamailio.cfg to send the 202? Iirc, m_store() doesn't send anything internally, it is done via config. Be sure you use send_reply() or t_reply() inside the failure route block.

Cheers,
Daniel

On 22/06/14 23:32, Allen Zhang wrote:

Hi Daniel,

R sent 408 because the recipient was  REGISTERed but the testing device had a power failure and hence couldn’t receive the MESSAGE.

After it timed out,  m_store() was called in the failure route.

So the sequence is:

MESSAGE delivery failed because device powered off

R timed out the request

R send 408

In R’s failure route

R calls m_store()

MESSAGE stored

R send 202

Did I do something wrong?

Cheers,

Allen

From: sr-users-bounces@lists.sip-router.org [mailto:sr-users-bounces@lists.sip-router.org] On Behalf Of Daniel-Constantin Mierla
Sent: Friday, 20 June 2014 5:37 p.m.
To: Kamailio (SER) - Users Mailing List
Subject: Re: [SR-Users] kamailio not sending 202 after 408 is sent

Hello,

why is R sending the 408? You should catch it there in a failure route, or where do you execute m_store() inside R config?

Cheers,
Daniel

On 20/06/14 07:20, Allen Zhang wrote:

Hi all,

I have an edge proxy (E) and a registrar (R) behind it.

In R, if a MESSAGE failed to deliver for any reason, R stores the MESSAGE in msilo.

If the MESSAGE timed out, R sends a 408 time out first and then send a 202 Accepted after the MESSAGE is stored.

The problem is, E happily forwards the 408 to the sender of the MESSAGE, but decides not to forward the 202.

The log:

DEBUG: t_reply_matching: hash 16662 label 0 branch 0

DEBUG: tm [t_lookup.c:1032]: DEBUG: t_reply_matching: reply matched (T=0x7f99eebb2ae0)!

DEBUG: tm [t_lookup.c:1164]: DEBUG: t_check_msg: msg id=3 global id=3 T end=0x7f99eebb2ae0

DEBUG: tm [t_reply.c:2085]: DEBUG: reply_received: org. status uas=408, uac[0]=408 local=0 is_invite=0)

DEBUG: tm [t_reply.c:1207]: ->>>>>>>>> T_code=408, new_code=202

DEBUG: tm [t_reply.c:1706]: DEBUG: relay_reply: branch=0, save=0, relay=-1

Can I change this behaviour to also forward the 202 withouting hacking in the source code?

Regards,

Allen

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