Re: GWRewriteE164 IN

"Jan Riedinger" <riedinger@xxxxxxxxxxxxxx> · Mon, 7 Feb 2005 01:08:57 +0100

Hello Ian,

I just want to mention that I had similar problems, because the 
documentation is still not updated and I was glad to find this email for an 
explanation.

Furthermore, I think it should be discussed if it makes sense to use the 
defined alias of a GW instead of its H323 id to decide which in rule should 
be applied. I have two arguments for it.

1.

You should be able to do rewritting of incoming numbers independently of the 
configuration of the originating GW. If I define a routing config for a 
carrier I dont want that it gets unusable, when the H323 id is changed at 
the originating GW. Often the admin of this GW won't know, what difficulties 
are caused if he changes the id of his GW, because it is seldom relevant for 
terminating GWs.

2.

The defined aliases are relevant for the decision, which out rule is 
applied. It is not clear, why in and out rules are not treated in the same 
manner.

Regards
  Jan

----- Original Message ----- 
From: <Ian.Campbell@xxxxxxxxxx>

To: <openh323gk-users@xxxxxxxxxxxxxxxxxxxxx>

Sent: Thursday, September 16, 2004 3:44 PM

Subject: RE:  GWRewriteE164 IN

Igor,
Glad it worked! I'll have a look at updating the documentation to
make this it a little clearer ;-)

Ian

-----Original Message-----
From: Kompnet [mailto:kompnet@xxxxxxxxx]
Sent: 16 September 2004 13:12
To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Subject: Re:  GWRewriteE164 IN

Ian, thank you very much!
It works even without 'komp-net' alias in PermanentEndpoints section.
From ini-file:
[RasSrv::PermanentEndpoints]
# For gateway
192.168.110.224=KOMPNET;0022#78632,4321#

[RasSrv::GWRewriteE164]
komp-net=in=1234#=5678#
KOMPNET=out=4321#=0022#

[RasSrv::RewriteE164]
5678#=4321#

And all rules applies in described sequence.
From logfile:
2004/09/16 16:06:26.218 4       ProxyChannel.cxx(1151)          GWRewrite
source: setup H323 ID or E164
2004/09/16 16:06:26.218 2            Toolkit.cxx(347)
GWRewriteTool::RewritePString: 1234#78632347611 to 5678#78632347611
2004/09/16 16:06:26.218 2            Toolkit.cxx(289)
RewritePString: 5678#78632347611 to 4321#78632347611
2004/09/16 16:06:26.219 2            Toolkit.cxx(347)
GWRewriteTool::RewritePString: 1234#78632347611 to 5678#78632347611
2004/09/16 16:06:26.219 2            Toolkit.cxx(289)
RewritePString: 5678#78632347611 to 4321#78632347611
2004/09/16 16:06:26.219 3             gkauth.cxx(1224)  GKAUTH  default
Setup
check ok
2004/09/16 16:06:26.219 5              Routing.h(144)   ROUTING Checking
policy Explicit for request Setup CRV=1000
2004/09/16 16:06:26.220 5              Routing.h(144)   ROUTING Checking
policy Internal for request Setup CRV=1000
2004/09/16 16:06:26.220 2             RasTbl.cxx(534)   GK      Gateway
7209_endp match 4321#
2004/09/16 16:06:26.220 4             RasTbl.cxx(891)   Alias match for GW
192.168.110.224:1720
2004/09/16 16:06:26.220 5              Routing.h(150)   ROUTING Policy
Internal applied to the request Setup CRV=1000

It would be absolutely not bad, if in the documentation it has been
reflected,
that for 'input' rewriting it is necessary to configure not an alias of an
endpoint, but the real h323id.

Regards,
Igor

On Thursday 16 September 2004 14:52, Ian.Campbell@xxxxxxxxxx wrote:
Igor,
The current problem with getting the inbound rule working is that
the H323 Id of the gateway that it is sending and the name configured in
the config file are different - the code uses the ID in the H225 section
of
the Setup for the name used for the inbound rewrite:

Currently you are getting:

h225pdu = {
    h323_uu_pdu = {
      h323_message_body = setup {
        protocolIdentifier = 0.0.8.2250.0.4
        sourceAddress = 1 entries {
          [0]=h323_ID  8 characters {
            006b 006f 006d 0070 002d 006e 0065 0074   komp-net
          }
        }

You have configured:

KOMPNET=in=1234#=5678#

so there is no match. If you change it to:

[RasSrv::PermanentEndpoints]
# For gateway
192.168.110.224=komp-net;0022#78632,0000#,1234#

[RasSrv::GWRewriteE164]
komp-net=in=1234#=5678#

Should work. For the outbound rewrite the code uses the name specified in

the config file to select the rules to use. Like I said in the previous

email the outbound rewriting is performed after the route selection, 
which

in your case is the prefixes in the [RasSrv::PermanentEndpoints] section.

I hope this helps.

Regards,

Ian Campbell
Band-X Ltd.

-----Original Message-----
From: Kompnet [mailto:kompnet@xxxxxxxxx]
Sent: 16 September 2004 11:11
To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Subject: Re:  GWRewriteE164 IN

Ian, look for my comments.

On Thursday 16 September 2004 12:20, Ian.Campbell@xxxxxxxxxx wrote:
> Hello Igor,
> As I understand you are presently trying to take a call in from
> KOMPNET and send it back to KOMPNET - if so, never tried this! Not sure
> exactly what will happen - but we can debug the situation further by
> turning on the logging and watching exactly what is in the Setup
messages

-

> if you could do this with -ttttt that would be good and then post it
> here.

Yes, log of two calls (1-st for IN rule, 2-nd for OUT rule) and ini-file
are
in attachment.

> Another useful tool would be Ethereal to see exactly what the contents
of
> the Q931 and H225 sections of the Setup actually have in them when they
> arrive.

Can't understand how can it help you in this simle situation. All 
required

information is in logfile.

> As far as the output rule stuff goes that is working as designed, the
> sequence of processing is as follows:
>
> 1. Process Inbound Rules for KOMPNET (1234#78632 -> 1234#78632)

But why??? The rule is (for incoming call) 1234#78632 ->5678#78632.

> 2. Process Global Rules (1234#78632 -> 1234#78632)

> 3. Select Call Route (1234#78632 select permanent endpoint KOMPNET)

> 4. Process Outbound Rules for KOMPNET (1234#78632 -> 5678#78632 )

>

> So that is ok, I think what you are missing is that route selection is

> performed at 3.

>

> For the config file stuff I'm not aware of a limit on how long the 
> lines

> are, perhaps someone who knows Pwlib could answer that?

>

> Regards,

>

> Ian Campbell

> Band-X Ltd.

>

> -----Original Message-----

> From: Kompnet [mailto:kompnet@xxxxxxxxx]

> Sent: 16 September 2004 07:35

> To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx

> Subject: Re:  GWRewriteE164 IN

>

>

> Ian, rules for 'in' gw-rewriting everyone do not work.

> And it is not clear, in what sequence rules for rewriting from 
> different

> sectons are applied.

> I've tried this config:

>

> [RasSrv::PermanentEndpoints]

> 80.80.200.200=PARTNER;1,2,3,4,5,6,7,8,9

> 192.168.110.224=KOMPNET;0022#78632,0000#,1234#

> [RasSrv::GWRewriteE164]

> KOMPNET=in=1234#=5678#

> [RasSrv::RewriteE164]

> 5678#=0022#

>

> The call comes from "KOMPNET" with a prefix 1234#78632.

> In the documentation it is told, that the first apply input per-GW
rules,
> then
> global rules, and at last output per-GW rules.
> I want to exchange prefix first from 1234#78632 to 5678#, then from
> 5678#78632
> to 0022#78632 (global rule), and at last I want to send Setup with
prefix
> 0022#78632 to KOMPNET.

> But in reality gnugk matches call with the alias KOMPNET and places 
> call

to

> KOMPNET with prefix 1234#78632.
>
> On the other hand, if the output rule is configured and there isn't
input
> rules at all

> [RasSrv::GWRewriteE164]

> KOMPNET=out=1234#=5678#

>

> then gnugk rewrites prefix from 1234#78632 to 5678#78632 and sends 
> Setup

to

> KOMPNET with prefix 5678#78632. But why to KOMPNET? This prefix

corresponds

> with another alis with prefix 5.
>
> I can't understand anything.
>
> On Wednesday 15 September 2004 16:27, Ian.Campbell@xxxxxxxxxx wrote:
> > Igor,
> > Sorry for the confusion on the syntax - originally this worked the
> > way you had it - but the Win32 version of Pwlib parses the multiple

lines

> > incorrectly. So I switched to the semicolon seperated version.
> >
> > Regards,
> >
> > Ian Campbell
> > Band-X Ltd.
> >
> > -----Original Message-----
> > From: Kompnet [mailto:kompnet@xxxxxxxxx]
> > Sent: 15 September 2004 13:10
> > To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
> > Subject: Re:  GWRewriteE164 IN
> >
> >
> > Thank you, Ian
> > The problem was in misunderstanding of syntax.
> > It was
> > [RasSrv::GWRewriteE164]
> > test=out=00#=0022#
> > TTT1=out=1234#7863=1757#8863
> > TTT1=out=1234#380=1757#380
> >
> > When I've changed config to
> > TTT1=out=1234#7863=1757#8863;out=1234#380=1757#380
> > all has risen on the places.
> >
> > On Wednesday 15 September 2004 14:57, Ian.Campbell@xxxxxxxxxx wrote:
> > > Igor,
> > > As Rafat has said this should be trivial to configure, here
is a
> > > brief example for a permanent endpoint:
> > >
> > > [RasSrv::PermanentEndpoints]
> > > x.x.x.x:1720=gw-a;33 <---- configure
>
> x.x.x.x as
>
> > > gw-a send prefix 33 to it
> > >
> > > [RasSrv::GWRewriteE164]
> > > gw-a=out=33=44;in==33 <---- define rules for gw-a
> > > rewriting for outbound calls map 33 to 44 and stick 33 on the front
> > > of the number on inbound calls.
> > >
> > > I just saw your most recent email - what exactly is the problem you

are

> > > experiencing at the moment? If you can let me have the 
> > > configuration

>

> file

>

> > > for it I can provide more assistance.

> > >

> > > Regards,

> > >

> > > Ian Campbell

> > > Band-X Ltd.

> > >

> > > -----Original Message-----

> > > From: Kompnet [mailto:kompnet@xxxxxxxxx]

> > > Sent: 15 September 2004 09:01

> > > To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx

> > > Subject:  GWRewriteE164 IN

> > >

> > >

> > > Whether probably to apply IN per-GW E164-rewriting in case of use 
> > > of

> > > permanent

> > > endpoints and simple Q931 Setup?

> > > If so, which configuration must be used?

> > > How to bind  in [RasSrv::PermanentEndpoints] or in anothrer section

the

> > > incoming prefix to the GW from which there comes a call?
> > > I can only apply OUT per-GW rewriting.

--
Best regards,
Igor Prokhorov

-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

_______________________________________________________

List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Archive: http://sourceforge.net/mailarchive/forum.php?forum_id…49
Homepage: http://www.gnugk.org/

-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

_______________________________________________________

List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549
Homepage: http://www.gnugk.org/

-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl

_______________________________________________________

List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
Archive: http://sourceforge.net/mailarchive/forum.php?forum_id…49
Homepage: http://www.gnugk.org/