i just committed to kamailio trunk an enhanced version of lcr module.
changes from kamailio wiki page
http://www.kamailio.org/dokuwiki/doku.php/features:new-in-1.5.x
are as follows:
* New high-performance implementation that keeps lcr information in
in-memory hash table, whose size c
Klaus Darilion writes:
> > When a SUBSCRIBE message is received, OpenSER creates an entries in
> > the ActiveWatcher table. This is used to generate NOTIFY message to
> > watchers. The problem I have seen is that OpenSER keeps on creating
> > new entries to the ActiveWatchers if the SIP UA der
Graham Wooden writes:
> Yeah, I see that 407 Proxy Auth too, but I am not sure where to check for
> that.
>
> Here is my only challenge statement:
>
> if (is_method("REGISTER"))
> {
> if (!www_authorize("", "subscriber"))
> {
>
your cap looks weird. your proxy receives invite from (i think) gw,
forwards it, and then sends 407 to gw. try to check from your config
file, why your proxy challenges the invite after it has forwarded it.
perhaps your cap does not include all packets (all interfaces).
-- juha
___
Graham Wooden writes:
> Yup, I see the invite and the corresponding ACK back from them - but that's
> it. About a second later after getting the ACK back, the caller gets
> dropped.
i still don't get it. it should go like this:
pstn gw -> invite -> kamailio -> invite -> uas1
Graham Wooden writes:
> Is this invite supposed to be like any other call, originating from
> kamailio?
it is not a new call, get_redirects adds new branches to existing call
from pstn. when you call t_relay after get_redirects, you should see
with wireshark an invite (or invites in case of mu
Graham Wooden writes:
> That being said - does anyone know any tricks to get this working - in such
> a way that I can create a new call leg and have the call go back out,
> preferably within kamailio? Thanks,
perhaps i didn't understand what you try to do, but if call comes from pstn
to kaami
Alex Balashov writes:
> What about some sort of hashing algorithm that operates on regular
> expression sums or is otherwise tuned by regex elements?
the problem is: given a string, find from set of prefixes the longest
one that matches the string. if prefixes are known strings, they can be
o
by accident i sent this message to devel list when i intended to send it
to users list.
i'm thinking to speed up lcr load_gws() implementation by storing prefixes
into hash table. it would allow load_gws() to execute (if necessary) in
constant time independent of the number of prefixes. also, i'
sam writes:
> I want to be a SIP providers like broadvoice gimzo ,where can i get DID
> numbers ,do we need to have contact with telecommunication
> Department.
this came to the wrong lists. also, do not post the same question to
both lists. you need to contact the telecom authority of the c
Graham Wooden writes:
> if (t_check_status("302")) {
> xlog("Got a 302 - redirecting");
> get_redirects("*");
> t_relay();
> }
>
> Pretty straight forward. Maybe the issue is with them still?
> Thanks,
check with wireshark
toly writes:
> In domain module, command is_domain_local - using parameter_fixup where
> there is a call pkg_malloc,
> but nowhere function free_fixup is used.
> This will be memory leak, or I'm missing something.
for completeness, memory for the parameter should indeed be freed when
opense
Yoo Chan Jeon writes:
> I do not have any problems to run the server.
> Which database should I add [EMAIL PROTECTED] <[EMAIL PROTECTED]> to?
> How do I add the resource list?
> What kind of format do I need to add a list if I want to add [EMAIL
> PROTECTED],
> [EMAIL PROTECTED], [EMAIL PRO
Iñaki Baz Castillo writes:
> Well, since NOTIFY is in-dialog it's just needed that the proxy does
> RR in the initial SUBSCRIBE before forwarding it to presence server,
> am I right?
inaki,
you should be right, but i haven't tested it. there is 'record_route'
field in active_watchers table.
Daniel-Constantin Mierla writes:
> > I would think that using one gateway primarily, with a secondary as a
> > standby, is a very, very common use case. What is the best way to do
> > this with dispatcher?
i haven't followed this thread, but is there a reason why you can't use
lcr module fo
Elena-Ramona Modroiu writes:
> I just committed a new module: htable
> http://www.kamailio.org/docs/modules/devel/htable.html
elena,
thanks for the new module.
> As a roadmap for the new module, planned for addition:
..
> - ability to save/load in/from database
> - ability to init items
michel freiha writes:
> Hi All,I need please to configure kamailio in order to force the user to
> register to my it from a specific IP...It's like creation of an access
> list...If the customer tries to register from another IP frm the one
> specified in kamailio he'll be denied
using radius
Klaus Darilion writes:
> I've checked RFC 3265 but could not find a description. Probably you
> should ask at the sip-implementors list.
i looked at rfc 3265 and found this in section 3.3.6:
Note that the NOTIFY messages triggered by SUBSCRIBE messages with
"Expires" headers of 0 will c
a user complained that un-subscribing a pending subscription is also
pending, i.e., kamailio responds with 202 instead of 200. i verified
the behavior.
now the question is, what do the relevant RFCs say about it? it appears
logical to me that if a user gives up waiting to get his/her
subscript
in order to understand this dialog event package, i read introduction
section of rfc 4235. what i don't get is why it is not enough that user
who wants to know when another user becomes available, subscriber to
that user's presence event package. if the user is in a phone call,
his/her user prese
George Lee writes:
> But according to rfc3824, publish method is not part of the message
> flow for mwi. How do I generate a publish with the message-summary
> event?
your voicemail app needs to do that using pua_mi interface.
-- juha
___
Users mai
George Lee writes:
> I know Kamailio SIP server can handle SUBSCRIBE for message-waiting
> indication. Is there a way to ask the SIP server to generate a NOTIFY
> with message summary event and its related headers like
> Message-waiting, Message-Account, Voice-Message, etc.?
kamailio will gen
Daniel-Constantin Mierla writes:
> there are already several important fixes since 1.4.0, so we should look
> to build the first maintenance release for branch 1.4. I am thinking of
> 2 weeks time frame from now. Make sure you point outstanding issues you
> are aware of as soon as possible
Darren Sessions writes:
> Juha has a point about Debian, although Ubuntu i386 is a breeze as
> well (easier than even Debian I thought and Ubuntu seems to have newer
> stable packages all the way around).
ubuntu works too. regarding newer packages, i have to disagree. i
recently tried t
Peter Lemenkov writes:
> I've got different advise. After quick googling I found that these
> constants are related to netfilter kernel module, so you probably need
> to install kernel-headers and pass properly additional includedirs (if
> required) to mediaproxy buildsystem.
sure, but if you
Ricky Gutierrez writes:
> any suggestion is appreciated?
my suggestion is that you use debian lenny.
-- juha
___
Users mailing list
Users@lists.kamailio.org
http://lists.kamailio.org/cgi-bin/mailman/listinfo/users
Alex Balashov writes:
> Do you have any perspective on how this performs compared to various
> high-end commercial SBCs?
if they don't have any custom hw support to do the forwarding, then i
don't see how they could perform much better on pc hardware.
-- juha
Alex Balashov writes:
> How does it use the kernel? Some sort of kernel-side module that has
> plumbing into the IP routing/forwarding subsystem? Or some sort of ALG
> module with hooks?
it uses linux conntrack modules. don't ask me about details.
-- juha
___
Alex Balashov writes:
> If one is written in Python and one in C, it seems to me that the
> performance and scalability makes the choice of rtpproxy a
> no-brainer.
alex,
mediaproxy 2.0 is using linux kernel (not python code) to relay media
packets and the claim is close to wire speed perfor
Nuno Marques writes:
>- What's the difference between the dispatcher and the relay?
dispatcher talks with proxy and selects a relay that actually relays rtp
packets. i'm pretty sure that this is documented in mediaproxy package
files.
-- juha
__
henning,
i agree that the things you proposed should be done, but they all deal
with internal matters inside the code and are not really visible as new
features to the end user.
there should be a balance on how much effort goes into internal
improvements/restructuring and how much to developing n
Ali Jawad writes:
> True..However my problem occurs when the mobile phone closes the call,
> so in this case PSTN GW send's it in UDP to Openser, and Openser will
> send it in TLS to the softphone, right ?
yes, it request uri from pstn gw specifies tls transport.
-- juha
Ali Jawad writes:
> Sorry my fault..I think got brain busted..
> The picture should be
>
> Softphone --TLS-->Openser ---UDP-->PSTNGW
that does not change my response. softphone sends bye to openser using
tls and openser sends bye to pstngw using whatever transport protocol
request uri spec
Ali Jawad writes:
> The returning party in this case would be the PSTN GW and it talks UDP.
>
> I.e.
>
> Softphone --TLS-->Openser ---UDP--->Softphone
>
> Should't Openser be responsible for the translation of these
> parameters?
i don't see a PSTN gateway anywhere in the figure. open
Ali Jawad writes:
> However I did notice the following in the logs
>
> Aug 28 13:41:02 [8564] DBG:core:parse_msg: uri:
> 8F980980E97E42F8EC;nat=yes>
the party sending bye should use as request uri contact uri of
the other party that it learned during dialog setup (invite, 200 ok).
check th
Klaus Darilion writes:
> IMO this would be interesting:
> - query time statistics
> - DNS/ENUM response codes:
> - server failure
> - NXDOMAIN
> - no NAPTR records
> - NAPTR records found, but service not matched
> - NAPTR records found, service matched but reg
ram writes:
> then what version of Freeradius is best to use
it is best to use the latest version, i.e., now 2.0.5 unless newer has
already appeared.
-- juha
___
Users mailing list
Users@lists.kamailio.org
http://lists.kamailio.org/cgi-bin/mailman/li
Nuno Marques writes:
>Flavio (the author) says we need to apply a patch to the freeradius to
> "fix a problem with accounting type 15".
>My question is - is it still necessary to apply that patch or the new
> version of freeradius allready covers it?
last week i upgraded my freeradius
Henning Westerholt writes:
> > I am thinking to keep in core
> > - avps
> > - headers
> > - var
> > - R-URI, From URI and To URI and destination URI related variables
> > - auth tokens (username, realm)
as an example of difficulty of making the selection, From URI is not
more important than
Henning Westerholt writes:
> > But the enum module does not export any statistics now. As an example,
> > for usrloc module:
> > kamctl fifo get_statistics usrloc:
what kind of statistics you would need? like delay statistics of
queries or what?
-- juha
_
Nuno Marques writes:
> +++---+--+--+-+
> | id | src_ip | proto | from_pattern | tag | version |
> +++---+--+--+-+
> | 1 | 192.168.137.20 | Any | ^sip:.*$ | NULL | 3 |
> ++
Daniel-Constantin Mierla writes:
> I have seen on mailing lists discussions about needs that will be solved
> by this new feature. Any comments?
daniel,
so far i have not seen a need to access branches in config script. lcr
module functions have been enough for me. of course i don't have
an
Juha Heinanen writes:
> i'm using
>
> t_on_branch("0");
>
> to disable calling of branch route. this has been documented or
> discussed somewhere and is ok to me.
i want to add, that i don't want any change in the current way of
resetting branch r
Iñaki Baz Castillo writes:
> Hi, for now, the "normal" way to detect if an in-dialog request/response
> beyonds to a natted dialog is by adding a uri parameter ("nat=yes") in
> the "Contact" header of the first request and response and macthing that
> parameter in the RURI of subsequent in-
Daniel-Constantin Mierla writes:
> > I really don't know if this is intuitive or not. The only way to
> > "dissable" branch_route[1] in the failure route is by adding:
i'm using
t_on_branch("0");
to disable calling of branch route. this has been documented or
discussed somewhere and is ok to
Daniel-Constantin Mierla writes:
> So, I am asking:
> - how do you see a proper management structure for the future
i thought that people in an open source project are trustworthy and that
there is no need for a legal entity. unfortunately this turned out to
be false and a legal entity is need
Ruchir writes:
> I'm trying to configure openser & freeradius with AAA. I'm using custom
> mysql tables for authentication & accounting. I've modified queries in
> sql.conf in raddb folder. Authentication works fine. I've added additional
> fields in radius_extra parameter of acc module. I've
Alex Balashov writes:
> And you do not feel that a revamping of the ground rules in the
> incumbent project could have been achieved? In other words, couldn't
> you have said to everyone: "Look, what we've got right now isn't
> working [for the following reasons] and it must change, or I
Bogdan-Andrei Iancu writes:
> Form my perspective,
> looking back over the last year, I do not see any evolution with OpenSER
> - just take a look at release 1.2 and release 1.4 (future) and tell me
> it is the same - going in the same direction is dead-end.
bogdan,
again, i don't see any
Bogdan-Andrei Iancu writes:
> From business point of view, what do you prefer:
>1) stick to kamilio and face angry and unsatisfied customers because
> of the software/project condition (starting with 1.3 we had a lot of
> complains/bad experience with quality)
bogdan,
were these compl
Bogdan-Andrei Iancu writes:
> This is way I moved form SER to OpenSER and further to OpenSIPS - just
> to keep delivering the project (whatever name it has) envisioned by
> openser 3 years ago.
bogdan,
it would be interesting to know what prevented you from delivering to
OpenSER project. t
Andreas Granig writes:
> Please elaborate on the issues you have with the trade mark
> (if you're allowed to do so).
anders,
in case no one has answered to this yet, OpenSER project got a few
months ago a notice from a company who has registered trademarks on
various SER related names and tha
Bogdan-Andrei Iancu writes:
> Also, I'm ready to overlook this "not-nice" action, so granting back my
> admin rights on the OpenSER project will be an act of respect.
bogdan,
i don't know about the rights removal thing, but it would be interesting
for the community to know what your plans reg
Alex Balashov writes:
> I don't, of course, have any sort of inside perspective as I am not
> involved in development, so I can't presume to judge the merits or
> veracity of the various justifications given for this adventure. Apart
> from Bogdan-Andrei's brief treatment of the subject, e
hi bogdan,
i'm trying to understand this in order figure out what to do in the
future.
> - the need for a new reliable release - delayed or low-quality releases
> are affecting the credibility of the project
are all the open dialog and tm related issues on openser tracker fixed
in opensips?
Alex Balashov writes:
> > canceling of b branch should happen already after step 4, but perhaps 4 and
> > 5 take place almost simultaneously and there is some race condition
> > related bug in tm module.
>
> I think it's just the order of events. According to my packet capture:
>
> - Pac
Alex Balashov writes:
> 3. Proxy bifurcates the call into two branches 'branch A' (to A) and
> 'branch B' (to B). Rewrites RURI, relays INVITE.
>
> 4. A answers with 200 OK.
>
> 5. B answers with 200 OK.
>
> 6. Proxy passes back 200 OK to SBC for A. Then for B.
>
> 7. SBC issues i
301 - 357 of 357 matches
Mail list logo
