[PATCH] Re: MO Concatenation

[Paul Bagyenda]

Oops! Yes there was a small bug once duplicate copies of the same message part were received (trying to debug-dump the sender address after free-ing the message structure). I've corrected this -- patch attached.

Please test and put through some votes!

P.

I've also included a patch (also posted earlier) to the AT2 module that is still missing from CVS and is crucial: The way UDH is parsed/ stored by the AT2 module is different from how other modules (at least SMPP) do it, which means that you see a different UDH byte sequence internally if you use AT2, from what you see with say SMPP. Without this patch too, the concat module would not work with AT2.

concat.diff
Description: Binary data

On Jan 06, 2007, at 23:00, Δημήτρης Ευμορφόπουλος wrote:

Here is the first crash ... It takes a lot of effort to find the logs ... there are about 3GB of logging in gz format ... Unfortunately logs are in not in debug.

Dimitris Evmorfopoulos
-----Original Message-----
From: Paul Bagyenda [mailto:[EMAIL PROTECTED]
Sent: Saturday, January 06, 2007 2:46 PM
To: Δημήτρης Ευμορφόπουλος
Subject: Re: MO Concatenation

Please provide the logs, would be interested to see why it crashed.
On Jan 06, 2007, at 15:05, Δημήτρης Ευμορφόπουλος
wrote:

2006-12-03 18:41:51 [32092] [12] PANIC: gwlib/octstr.c:2451: seems_valid_real: Assertion `ostr->len >= 0' failed. (Called from gw/bb_boxc.c:1679:check_concatenation.) 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart(gw_panic+0xe6) [0x80c28a6] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart [0x80c30f7] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart(octstr_get_cstr_real+0x26) [0x80c3206] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart(route_incoming_to_boxc+0x225) [0x80543e5] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart(bb_smscconn_receive+0x339) [0x805a279] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart [0x8082e42] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart [0x8084c09] 2006-12-03 18:41:51 [32092] [12] PANIC: /usr/sbin/ bearerbox.multipart [0x80b93d1] 2006-12-03 18:41:51 [32092] [12] PANIC: /lib/libpthread.so.0 [0xdcd3b6] 2006-12-03 18:41:51 [32092] [12] PANIC: /lib/libc.so.6(__clone +0x5e) [0x1c933e]

Actually the patch crashed on me twice but never got into trying to
debug or report since this was on a production system and we
decided to do the concatenation part on our own.

Yet the patch crashes on heavy traffic.

If anyone needs logs of the crashes I'll be happy to provide them.
I cannot recall if the logs where on debug at the time.

Dimitris Evmorfopoulos

-----Original Message-----
From: Alex Kinch [mailto:[EMAIL PROTECTED]
Sent: Friday, January 05, 2007 7:00 PM
To: Rogelio Serrano; Paul Bagyenda; Kannel Devel
Subject: Re: MO Concatenation

+1 for me too, on a practical and functional level. Can't comment on
the way it works as my c isn't up to much! Alex

On 1/5/07, Rogelio Serrano <[EMAIL PROTECTED]> wrote:

On 1/5/07, Paul Bagyenda <[EMAIL PROTECTED]> wrote:

I think Alex is talking about something different. You are talking
about MT concatenation.

To answer Alex's question: Yes I submitted the patch a while back.

might change that...

Paul.

AFAIK it is not yet on CVS. More votes in favour from you and others

I vote in favour of that too. I have been looking for that patch too.

On Jan 05, 2007, at 10:11, Enver ALTIN wrote:

Hi Alex,

On 1/3/07, Alex Kinch <[EMAIL PROTECTED]> wrote:

I'm sure I saw a patch floating around a while ago to re-assemble

still in CVS?

concatenated MO within Kannel. Did it make 1.4.1 stable - or is it

SMSC configuration and it should work.
--
Enver

Kannel supported that for ages, just enable concatenation=1 in your

--
the thing i like with my linux pc is that i can sum up my
complaints in 5
items