[Serdev] Fwd: [Users] NEW FEATURES: t_replay() enhancements

Klaus Darilion klaus.mailinglists at pernau.at
Thu Dec 7 16:34:34 UTC 2006


Andrei Pelinescu-Onciul wrote:
> On Dec 06, 2006 at 20:34, Weiter Leiter <bp4mls at googlemail.com> wrote:
>> Would this make sense as feature request? Me, for one, I find it useful.
> 
> Yes, in fact some us of discussed the automatic-error-reply problem,
> although in private. It was not fixed because of the iminent code freeze
> (at the time) and lack of time, but it will be fixed in the next release.
> 
> The draft fix is to delay sending the reply to the end of script. When
> the script ends, if no reply was sent for the current transaction, a
>  reply will be sent automatically. If a reply was already sent, nothing
>  will happen. No script change will be necessary.
> 
> The stop-sending-100 fix is a separate feature, which I don't find very
> usefull (I don't think there are a lot of scenarios where you send first 
>  a stateless reply and then go statefull), but if people need it we can
>  easilly add it too.

I think this is very essential. Stop the client sending retransmissions 
which may overload the server, especially if the server is already slow. 
This happens if you have time consuming tasks (DB lookups, radius 
authentication ...) before t_relay, but do not want to go stateful.

regards
klaus


> 
> Andrei
> 
> 
>> ---------- Forwarded message ----------
>> From: Bogdan-Andrei Iancu <bogdan at voice-system.ro>
>> Date: Dec 6, 2006 6:15 PM
>> Subject: [Users] NEW FEATURES: t_replay() enhancements
>> To: "users openser.org" <users at openser.org>, devel <devel at openser.org>
>>
>> Hi everybody,
>>
>> following several discussion threads from some time ago, I added a
>> support for controlling the behaviour of t_relay() functions. This
>> addresses:
>>
>> 1) to ability to stop t_relay() from automatically sending the 100
>> trying provisional reply before relaying the request. This is useful in
>> scenarios where you already send a stateless 100 trying from the script.
>>
>> 2) ability to do serial forking in failover scenarios - as original
>> t_relay() automatically sends a negative reply in case of a failure
>> (unresolvable URI, TCP connect refused, etc), it was impossible to
>> continue with the next destination. Now, the auto-reply can be disabled.
>>
>> backward compatibility is kept.
>>
>> For more details (how to do it, options, etc), please see the online docs:
>>    http://www.openser.org/docs/modules/1.2.x/tm.html#AEN340
>>
>>
>>
>> regards,
>> bogdan
>>
>> _______________________________________________
>> Users mailing list
>> Users at openser.org
>> http://openser.org/cgi-bin/mailman/listinfo/users
> 
>> _______________________________________________
>> Serdev mailing list
>> Serdev at lists.iptel.org
>> http://lists.iptel.org/mailman/listinfo/serdev
> 
> _______________________________________________
> Serdev mailing list
> Serdev at lists.iptel.org
> http://lists.iptel.org/mailman/listinfo/serdev


-- 
Klaus Darilion
nic.at



More information about the Serdev mailing list