2009/5/29 Uwe Kastens :
> Hello,
>
> I have two units unable to communicate. Unit A sends an Invite without
> SDP, Unit B drops the call, because there is no sdp in the Invite.
An INVITE with no SDP offer is 100% valid. Then the SDP offer should
take place in the 200 Ok from the callee and the SDP
Hello,
I have two units unable to communicate. Unit A sends an Invite without
SDP, Unit B drops the call, because there is no sdp in the Invite.
Somehow I was wondering if its possible to fake a kind of initial SDP?
If not I have to but an asterisk in the middle.
BR
Uwe
--
kiste lat: 54.3226
Hi Bogdan,
I have upgrade libldap to 2.3.
However, there are some other errors pop up. Hope you can help.
make[1]: Entering directory `/home/lli/opensips_1_5/modules/ldap'
make[1]: Leaving directory `/home/lli/opensips_1_5/modules/ldap'
make[1]: Entering directory `/home/lli/opensips_1_5/module
On Thursday 28 May 2009, Thomas Gelf wrote:
> Hi list,
>
> I really like engage_media_proxy() as it really makes life easier.
> However, currently it is dialog-based and cannot be called in a
> branch_route. This has some side-effect:
>
> a) if a call is forked to two users, one behind NAT and on
Hello,
I have package OpenSIPS-CP 2.0 on Ubuntu 9.04.
I got below error when I press "Apply changes to server" and Mi Command.
sorry -- cannot connect to xmlrpc server
111 - Connection refused
What I missed configuration ?
Sincerely,
Tseveen.
___
Us
On Thursday 28 May 2009, Jeff Pyle wrote:
> Dan,
>
> With your patch Mediaproxy 2.3.2 built packages without any problems. On an
> x86 CentOS install I run 2.3.2 with the older version of python-application.
> But, installing the package, I get:
>
> Starting MediaProxy relay: media-relayCannot s
On Thursday 28 May 2009, Jeff Pyle wrote:
> Hmm, it appears I spoke too soon. It wants python-application version
> 1.1.1, but only 1.0.9-4 is available for this architecture.
My guess is that you mean distribution. python-application is architecture
independent so it is available with the same
On Thursday 28 May 2009, Jeff Pyle wrote:
> Hmm, it appears I spoke too soon. It wants python-application version
> 1.1.1, but only 1.0.9-4 is available for this architecture. Is this
> crucial?
Yes. mediaproxy-2.3.4 uses some of the new stuff in 1.1.1
--
Dan
_
Sorry, of course. I meant just conntrack.
Thanks.
On 28/05/2009, Dan Pascu wrote:
> On Wednesday 27 May 2009, Gavin Henry wrote:
>> Thanks Dan. I have done, read all the lists and figured this part out
>> myself
>> ;-)
>>
>> Does it use conntrack_sip?
>
> No. Why would it? It relays media not si
Dan,
With your patch Mediaproxy 2.3.2 built packages without any problems. On an
x86 CentOS install I run 2.3.2 with the older version of python-application.
But, installing the package, I get:
Starting MediaProxy relay: media-relayCannot start MediaProxy Relay. A Linux
host is required for oper
Hmm, it appears I spoke too soon. It wants python-application version
1.1.1, but only 1.0.9-4 is available for this architecture. Is this
crucial?
- Jeff
On 5/28/09 3:51 PM, "Jeff Pyle" wrote:
> That worked.
>
>
> On 5/28/09 9:19 AM, "Dan Pascu" wrote:
>
>> On Thursday 28 May 2009, Jef
That worked.
On 5/28/09 9:19 AM, "Dan Pascu" wrote:
> On Thursday 28 May 2009, Jeff Pyle wrote:
>> Hi Dan,
>>
>> Output was:
>> linux2-sparc Linux
>>
>> Iptables and conntrack are supported; the libnetfilter-conntrack-dev package
>> installed properly as part of the dependency packages list
Thanks for your input, Brett. I tried to initialize the counter from the
main route block, but it didn't work - somehow I ended up with rURI in
$avp(i:500). Then I rewrote the failure_route exactly like in the
example I have provided earlier - and it worked like a magic. Previously
I've had the
Hi,
I've got a performance-related question:
Is there any good way to get a statistic of how busy opensips
processes are on average? While the processes are blocked to process a
message, is there a way to check "what the udp-sip processes are doing
-> idle / working" without going into stuff like m
There isn't any magic with the web interfaces.. just get the web interface
working with the db, then get opensips working with the db. Approach them at
first as separate projects. Then you may need some small modifications to
both to mash the versions up.
-Brett
On Wed, May 27, 2009 at 9:50 PM, g
i just want a web link to start with, so that i count for the required packages
on CentOs 5.1 , the procedure i have is based on installing apache tomcat and
Debian linux and openser 1.2,i work on CentOs 5 .1 , opensips 1.5 ,
- Original Message -
From: Brett Nemeroff
To: gm
There are many of us out here that can do that.. perhaps you can ask a
specific question?-Brett
On Wed, May 27, 2009 at 9:38 PM, gmail wrote:
> Does anybody know how to install SerMyadmin for opensips 1.5
>
> ___
> Users mailing list
> Users@lists.op
Does anybody know how to install SerMyadmin for opensips 1.5___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Hi all,
Just wondering . .
We've got set of relays doing on average of 17 to 19 cps. When I get
traffic spikes my cps can hit just shy of 30 cps. When the cps gets to
about 24 - 26 cps we start to get loads of choppy audio and in some
cases one-way audio.
No errors in any of the logs on th
Hi Richard,
thank you for your reply! I stopped using other routes to force
OpenSIPS into doing "forbidden" things a while ago. Mostly this
caused other issues - and usually there IS a reason why certain
methods are only allowed in certain route types ;-)
In this special case moving engage_media_
For scenario b I suspect you could just call a route from within the
failure route and engage the media proxy there. I know you will be
able to make the function call without error, just never tested it to
see if it has any undesired side effects. Like not actually working.
Richard Revels
Hi list,
I really like engage_media_proxy() as it really makes life easier.
However, currently it is dialog-based and cannot be called in a
branch_route. This has some side-effect:
a) if a call is forked to two users, one behind NAT and one with
non-symmetric router and STUN or on public IP, m
Oh, heaven knows I don't use the perl approach anymore. I was just
remembering.
Jeff Pyle wrote:
> It’s during business hours. We sipp-tested it at 40 cps without an
> issue. I probably should have mentioned that. In our case we have to
> evaluate the costs returned by some stored procedur
It¹s during business hours. We sipp-tested it at 40 cps without an issue.
I probably should have mentioned that. In our case we have to evaluate the
costs returned by some stored procedures, mix in some spices, fluff the egg
whites, flash fry and return a route list.
Brett¹s right. In your appl
Right, but we're talking about the Perl module.
Brett Nemeroff wrote:
> gflags + avp_db_load should be able to do much much more than this
> without a problem.. even gflags + the perl module would be better
>
>
> On Thu, May 28, 2009 at 9:18 AM, Alex Balashov
> mailto:abalas...@evaristesys.co
gflags + avp_db_load should be able to do much much more than this without a
problem.. even gflags + the perl module would be better
On Thu, May 28, 2009 at 9:18 AM, Alex Balashov wrote:
> 60,000 times a day is about 41 times a minute (or, a little less than
> one operation per second), assuming
60,000 times a day is about 41 times a minute (or, a little less than
one operation per second), assuming uniform distribution.
That probably won't stress it too badly. My issues were with bigger loads.
Jeff Pyle wrote:
> For what it's worth, I use the perl module to execute some custom databa
oh fine... thanks... it is the picture :)
can you sent me the picture a little bit larger?
Regards
Michael
Original-Nachricht
> Datum: Thu, 28 May 2009 11:55:51 +0200
> Von: Adrian Georgescu
> An: "Michael Ciupka"
> CC: users@lists.opensips.org
> Betreff: Re: [OpenSIPS-Users
On Thursday 28 May 2009, Jeff Pyle wrote:
> Hi Dan,
>
> Output was:
> linux2-sparc Linux
>
> Iptables and conntrack are supported; the libnetfilter-conntrack-dev package
> installed properly as part of the dependency packages listed in the install
> file.
>
> The build file is attached. Looki
For what it's worth, I use the perl module to execute some custom database
operations for custom route decision making. It runs about 60k times per
day in a Xen VM with no memory or performance issues. I've been quite
pleased.
- Jeff
On 5/28/09 8:46 AM, "Alex Balashov" wrote:
> Admittedly,
I kinda recall the perl module maintainer stating on the list that he no
longer has time to maintain the module. Does anyone else remember that?
That's too bad. I love perl. :)
-Brett
On Thu, May 28, 2009 at 7:46 AM, Alex Balashov wrote:
> Admittedly, OpenSER 1.3.x was the last time I tried the
I don't entirely remember the way the arming of the flags works and how they
persist across failure_routes. Maybe someone can comment on that?
Instead of using flags, call:
$avp(i:500) = 1; # Initialize counter of failed attempts
from your main route block, before failure route is c
Hi Uwe.
I faced a similar problem; what I tried was applying a transformation on the
AVP and assign it to another AVP:
$avp(i:10)=$(avp(s:billing){s.int});
and then, use $avp(i:10) for dp_translate.
Regards.
Sergio.
On Thu, May 28, 2009 at 6:26 AM, Uwe Kastens wrote:
> Hi,
>
> If I call dp_
Admittedly, OpenSER 1.3.x was the last time I tried the Perl module, but
my performance results with the Perl module have been very bad, and
there are memory leaks in it as well.
I think much of it has to do with the fact that the maintainer has been
effectively unavailable for several years.
Brett Nemeroff wrote:
> You should be able to track this with avps. What version are you running? I
> know that in at least the newer versions of opensips, this
> is definitely supported:http://www.opensips.org/Resources/DocsCoreVar14
> "The AVPS will be visible in all routes where any message (rep
You should be able to track this with avps. What version are you running? I
know that in at least the newer versions of opensips, this
is definitely supported:http://www.opensips.org/Resources/DocsCoreVar14
"The AVPS will be visible in all routes where any message (reply or request)
of the transact
Hello,
How do I limit the number of gateways tried in lcr module? Say, try only
2 (or 3, 4) gateways even if there are 5 matching gateways in the database?
I can't do that by using the script variable because failure route is
reenterable and variable doesn't keep its value between entering. I a
Hi,
The query for the AVP gives back an VARCHAR. dp_translate seems to
expect an Integer. Could that be?
If so, it would be good to change the documentation for that module
http://www.opensips.org/html/docs/modules/1.5.x/dialplan.html
1.7.1
from
Will try to translate the src string into dest st
Hi,
If I call dp_translate with an integer, everything is working as expected:
if (!dp_translate("12345", "$fU/$avp(s:service)")) {
If I try the same with an AVP, I will get the following error:
xlog (" billingid $avp(s:billing)");
if (!dp_translate("$avp(s:billing)", "$fU/$avp(s:service)")) {
Hi Dan,
Output was:
linux2-sparc Linux
Iptables and conntrack are supported; the libnetfilter-conntrack-dev package
installed properly as part of the dependency packages listed in the install
file.
The build file is attached. Looking through it I notice a few things that
cause me to raise an
On Wednesday 27 May 2009, Jeff Pyle wrote:
> Hi Dan,
>
> It's Debian 5.0.1 on sparc architecture, a fresh installation. Debhelper is
> the at the newest version, according to apt-get.
>
> apt-get build-dep mediaproxy-relay seems okay. It reported "0 upgraded, 0
> newly installed, 0 to remove an
On Wednesday 27 May 2009, Gavin Henry wrote:
> Thanks Dan. I have done, read all the lists and figured this part out myself
> ;-)
>
> Does it use conntrack_sip?
No. Why would it? It relays media not signaling. It uses generic conntrack
rules with both SNAT and DNAT elements.
--
Dan
_
On Wednesday 27 May 2009, Jeff Pyle wrote:
> Hi Dan,
>
> It's Debian 5.0.1 on sparc architecture, a fresh installation. Debhelper is
> the at the newest version, according to apt-get.
>
> apt-get build-dep mediaproxy-relay seems okay. It reported "0 upgraded, 0
> newly installed, 0 to remove an
nobody have the old picture from the openxcap homepage?
or somebody know where i can download the "old" png file?
Original-Nachricht
Datum: Wed, 27 May 2009 22:39:43 +0200
Von: "Michael Ciupka"
An: users@lists.opensips.org
Betreff: [OpenSIPS-Users] OpenXCAP
Hello,
i need abso
Hi,
So you do :
fix_nater_register();
save("location");
and you do not get any "received" value for the registered contacts? do
you get any errors?
Regards,
Bogdan
oso che bol wrote:
> Hi Bogdan,
>
> I do not have field "Received" after registered user.
>
> Thanks,
> LN
>
> On Mon, Ma
Hi Sharath,
if you use a a combination of avp_db_query() and memcache, you can load
from DB when you first need the value and cache it for later usage. A
tutorial about memcache is available under
http://www.opensips.org/Resources/DocsTutMemcache
Soon, a "on_start" route will be available for
Hi Inaki,
yes, it is under devel. See:
http://lists.opensips.org/pipermail/users/2009-April/004946.html
Regards,
Bogdan
Iñaki Baz Castillo wrote:
> El Jueves, 28 de Mayo de 2009, Brett Nemeroff escribió:
>
>> I think I saw something about this for future releases.. as in an init
>> route or
It's been a while since I've needed this but I think gflags + avpops might
be slightly more elegant. YMMV.
I'm not sure what the overheard to calling a perl function is (however i
hear with the perl mod, it's very fast). I'd love to hear feedback on
this
-Brett
On Thu, May 28, 2009 at 1:42 A
48 matches
Mail list logo
