2 Oct
2012
2 Oct
'12
11:15 p.m.
You can make a simple config, relaying to local host port 9 and in
failure route send a t_reply() with your code.
Btw, I just added to dialog module a new parameter that allows disabling
usage of dialog module to store the values for From/To headers. As you
exposed, it might be a chance that not-tracked dialogs with changes to
From/To will not be handled properly - I will write another message to
be reviewed.
Anyhow, is good to have an option not to use dialog module. I used it
many times in memory-only mode, a restart can cause troubles for sure.
To be easy to backport without a risk for conflicts, I split the changes
in two commits, one for source code and the other for docs. If you want
to try, just cherry-pick this commit in 3.3:
http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=0d299fe…
Once the issue is sorted out, I will backport the patch if it proves to
fix some wrong behavior.
Cheers,
Daniel
On 10/2/12 10:05 PM, Alex Balashov wrote:
I will check. The problem is that this installation is
exotic and
deviant from my usual order of operations, which creates further
confusion.
-- Alex
--
Sent from my Samsung mobile, and thus lacking in the refinement one
might expect from a proper keyboard.
Alex Balashov - Principal
Evariste Systems LLC
235 E Ponce de Leon Ave
Suite 106
Decatur, GA 30030
Tel: +1-678-954-0670
Web: http://www.evaristesys.com/
Daniel-Constantin Mierla <miconda(a)gmail.com> wrote:
On 10/2/12 9:45 PM, Alex Balashov wrote:
--
Daniel-Constantin Mierla - http://www.asipto.com
http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Kamailio Advanced Training, Berlin, Nov 5-8, 2012 - http://asipto.com/u/kat
Kamailio Advanced Training, Miami, USA, Nov 12-14, 2012 - http://asipto.com/u/katu
Perhaps the source of my misunderstanding is
something to do with
branching.
The actual case I am troubleshooting is a 503 generated via t_reply
out of a failure route. The complaint is that this 503 contains a
modified From header, exposed to the originator of the call.
If you created the
transaction after calling msg_apply_changes(), then
I am quite sure it is how it worked all the time. Try with an older
version and see.
Cheers,
Daniel
-- Alex
--
Sent from my Samsung mobile, and thus lacking in the refinement one
might expect from a proper keyboard.
Alex Balashov - Principal
Evariste Systems LLC
235 E Ponce de Leon Ave
Suite 106
Decatur, GA 30030
Tel: +1-678-954-0670
Web: http://www.evaristesys.com/
Daniel-Constantin Mierla <miconda(a)gmail.com> wrote:
Those originated locally were never using the original value if you
called msg apply changes -- you can try with an older version, if it
does that, then it is a bug somewhere :-).
Btw, if you called msg apply changes after creating the transaction,
then SL is out of scope and tm local reply is using the initial
values, but that means the changes are only for that branch, a
failure route will get the initial request content.
Cheers,
Daniel
On 10/2/12 9:30 PM, Alex Balashov wrote:
--
Daniel-Constantin Mierla -http://www.asipto.com
http://twitter.com/#!/miconda -http://www.linkedin.com/in/miconda
Kamailio Advanced Training, Berlin, Nov 5-8, 2012 -http://asipto.com/u/kat
Kamailio Advanced Training, Miami, USA, Nov 12-14, 2012 -http://asipto.com/u/katu The expected behaviour is that all replies, both
relayed and
endogenously originated (via SL or TM), as well as sequential
requests directed back to the caller, will preserve the original
headers.
-- Alex
--
Sent from my Samsung mobile, and thus lacking in the refinement one
might expect from a proper keyboard.
Alex Balashov - Principal
Evariste Systems LLC
235 E Ponce de Leon Ave
Suite 106
Decatur, GA 30030
Tel: +1-678-954-0670
Web: http://www.evaristesys.com/
Daniel-Constantin Mierla <miconda(a)gmail.com> wrote:
Not sure what is 'work as you expect', provided the details I gave
in the previous mail related to local replies.
Otherwise, if you don't track all dialogs, could be an issue if some
of the replies are ok and some are not.
I will check the uac module and add a parameter to not use dialog
module even if loaded, if such option is not already there.
Cheers,
Daniel
On 10/2/12 9:11 PM, Alex Balashov wrote:
--
Daniel-Constantin Mierla -http://www.asipto.com
http://twitter.com/#!/miconda -http://www.linkedin.com/in/miconda
Kamailio Advanced Training, Berlin, Nov 5-8, 2012 -http://asipto.com/u/kat
Kamailio Advanced Training, Miami, USA, Nov 12-14, 2012 -http://asipto.com/u/katu So, you're saying that if all the dialogs
were tracked, everything
would work as I expect it to?
-- Alex
--
Sent from my Samsung mobile, and thus lacking in the refinement one
might expect from a proper keyboard.
Alex Balashov - Principal
Evariste Systems LLC
235 E Ponce de Leon Ave
Suite 106
Decatur, GA 30030
Tel: +1-678-954-0670
Web: http://www.evaristesys.com/
Daniel-Constantin Mierla <miconda(a)gmail.com> wrote:
I think there was an addition to use dialog module to store initial
and new values. I don't know if it can be turned off or is
automatically when the dialog module is loaded - just to see if
that is the cause.
However, if you do msg_apply_changes(), local replies will use the
new headers. I'm pretty sure it happened the same before -- e.g.,
it was no way to revert back so that sl can use older values of the
headers. Perhaps you can revert from config, changing the header to
initial value that you store in an avp, apply changes and send the
reply.
Received and relayed replies should be converted.
Cheers,
Daniel
On 10/2/12 8:57 PM, Alex Balashov wrote:
> I should be clearer: when I say "internally generated", I don't
> mean it in the sense of 408 or something else opaquely generated
> by the proxy. I mean internally generated as in originating from
> the proxy, e.g. something sent with sl_send_reply() or t_reply().
>
>
> -- Alex
>
> --
> Sent from my Samsung mobile, and thus lacking in the refinement
> one might expect from a proper keyboard.
>
> Alex Balashov - Principal
> Evariste Systems LLC
> 235 E Ponce de Leon Ave
> Suite 106
> Decatur, GA 30030
> Tel: +1-678-954-0670
> Web: http://www.evaristesys.com/
>
> Alex Balashov <abalashov(a)evaristesys.com> wrote:
> Hello,
>
> I am using the latest cloned 3.3 branch and performing a
> uac_replace_from() operation. After the operation, I do
> msg_apply_changes().
>
> My 'uac' modparams are:
>
> modparam("uac", "rr_from_store_param", "vsfc")
> modparam("uac", "restore_mode", "auto")
> modparam("uac", "restore_passwd", "xxxxxxxxx")
>
> However, the modified From URI is getting back to the originator,
> both
> in internally generated errors, and in relayed replies.
>
> This didn't used to happen before 3.3, but has been outstanding for
> quite some time (I didn't look at it much until now). Is there some
> possibility that there is a >= 3.3 bug? Or, have I simply
> misconfigured
> something?
>
> Thanks!
>
> -- Alex
>
> --
> Alex Balashov - Principal
> Evariste Systems LLC
> 235 E Ponce de Leon Ave
> Suite 106
> Decatur, GA 30030
> Tel: +1-678-954-0670
> Fax: +1-404-961-1892
> Web: http://www.evaristesys.com/, http://www.alexbalashov.com/
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing
> list
> sr-users(a)lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
> sr-users(a)lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
--
Daniel-Constantin Mierla -http://www.asipto.com
http://twitter.com/#!/miconda -http://www.linkedin.com/in/miconda
Kamailio Advanced Training, Berlin, Nov 5-8, 2012 -http://asipto.com/u/kat
Kamailio Advanced Training, Miami, USA, Nov 12-14, 2012 -http://asipto.com/u/katu
--
Daniel-Constantin Mierla -http://www.asipto.com
http://twitter.com/#!/miconda -http://www.linkedin.com/in/miconda
Kamailio Advanced Training, Berlin, Nov 5-8, 2012 -http://asipto.com/u/kat
Kamailio Advanced Training, Miami, USA, Nov 12-14, 2012 -http://asipto.com/u/katu