I'm not looking back to see the old conversation so forgive me if I'm
answering the wrong question here. So, I think you want to send the call
to a customer, say at their desk. If they don't answer then send the call
to, for example, their cellphone. In that case you would want serial
forking r
If I set a nonce password on a opensips 3.x proxy and the same one on
opensips 2.x proxy it is expected behaviour that it still wont match if
call starts on opensips 2, is challenged, then INVITE is sent to opensips 3
proxy?
[image: BandwidthMaroon.png]
Richard Revels • System Architect II
t; OpenSIPS Summit 27-30 Sept 2022, Athens
> https://www.opensips.org/events/Summit-2022Athens/
>
> On 9/20/22 12:39 AM, Richard Revels wrote:
>
> It appears to me that if I set the request uri in a route block and then
> use t_relay(,"somedestination proxy") to send th
imply send
straight to the domain in the request uri but that has since reverted so my
INVITE comes back to my proxy on loopback.
I can adjust my config but want to be sure i understand what is happening
first.
Richard Revels
___
Users mailing
directory are indeed created
4) work with old files in different directory
[image: BandwidthMaroon.png]
Richard Revels • System Architect II
900 Main Campus Drive, Suite 100, Raleigh, NC 27606
m: 919-578-3421 • o: 919-727-4614
e: rrev...@bandwidth.com
On Fri, Oct 2, 2020 at 12:53 AM
.com
>
> On 6/11/20 4:15 PM, Richard Revels wrote:
>
> I feel like this function would be a lot more useful if the last argument
> would accept a variable rather than a static int.
>
> https://opensips.org/html/docs/modules/2.1.x/avpops.html#idp5679392
>
> Richard Revels
>
I feel like this function would be a lot more useful if the last argument
would accept a variable rather than a static int.
https://opensips.org/html/docs/modules/2.1.x/avpops.html#idp5679392
Richard Revels
___
Users mailing list
Users
been
looking at for the last few hours.
[image: BandwidthMaroon.png]
Richard Revels • System Architect II
900 Main Campus Drive, Suite 100, Raleigh, NC 27606
m: 919-578-3421 • o: 919-727-4614
e: rrev...@bandwidth.com
On Thu, Jul 4, 2019 at 4:21 AM johan de clercq wrote:
>
Try enclosing the 1 in quotes
dp_translate("1","$ruri.user/$var(rU)")
[image: BandwidthMaroon.png]
Richard Revels • System Architect II
900 Main Campus Drive, Suite 100, Raleigh, NC 27606
m: 919-578-3421 • o: 919-727-4614
e: rrev...@bandwidth.com
On Mon, Ju
scenarios I expect to be
doing some of that over the next few days. Just let me know what still
needs to be looked at and I'll try to get it in.
[image: BandwidthMaroon.png]
Richard Revels • System Architect II
900 Main Campus Drive, Suite 100, Raleigh, NC 27606
m: 919-578-3421 •
> OpenSIPS Developerhttp://www.opensips-solutions.com
>
> On 03.08.2018 03:46, Richard Revels wrote:
>
> I'm using opensips version 2.3 from rpm package on fedora. I've noticed
> that if I use the wrong auth info for the mysql db and the connection fails
> then opensips does a
I'm using opensips version 2.3 from rpm package on fedora. I've noticed
that if I use the wrong auth info for the mysql db and the connection fails
then opensips does a core dump in db_table_version in both the acc and
permissions modules. Probably more, I didn't bother going further down the
lis
I'm using a webserver response rather than cache_fetch but this may help.
#we got a valid response from the web server.
convert from string to json object
$json(resp-obj) := $avp(router-resp);
if( $avp(ok_log_this) == "true" )
I am unsure of the expected behaviour in the config switch / case block
when the break is not defined between a defined case and the default case
but what I'm seeing right now is not what I expect. Wanted to get some
input before opening a bug tracker on it.
--code--
route[testswitch]
{
s
http://www.opensips.org/html/docs/modules/2.2.x/registrar.html#id294862
There are several functions in the registrar module to help you determine
if a Contact is already registered using a given AOR. After that you can
use flags to the save command to indicate if you want to set a max number
of c
Are you using subnets in your address table? How often do you reload the
address table?
There was a package memory leak that was fixed in the development tree that
might explain what you are seeing.
On Wed, Nov 11, 2015 at 4:26 AM, dpa wrote:
> Hello!
>
>
>
> Is there any assumption about prob
Cassandra came to my attention a couple days ago. I've never used it so
thought I would give it a try. To get it to compile on my centos servers I
had to modify the CXXFLAGS line in modules/cachedb_cassandra/Makefile to:
CXXFLAGS=$(CFLAGS:-Wno-deprecated option=) -DHAVE_NETINET_IN_H
Hope this s
Here is another example or two.
Set module parameter to grab an extra column (tag) in a dbaliases table:
modparam("avpops", "db_scheme",
"alias_scheme:table=dbaliases;username_col=username;domain_col=domain;value_col=tag;value_type=string")
In the config use the alias scheme (send in the uri of t
d it - not sure what I was thinking". :
>
Richard
On Jun 24, 2011, at 9:22 AM, Bogdan-Andrei Iancu wrote:
> Hi Richard,
>
> On 06/22/2011 04:36 PM, Richard Revels wrote:
>>
>> Bogdan and all,
>>
>> Greetings. I have been using revision 7602 in
Bogdan and all,
Greetings. I have been using revision 7602 in Production and have found it
quite stable w/ most of the functionality I need. Everything else aside
though, the support for an Event socket is going to be enough to cause me to
start the update process.
One thing I would really l
Err, I guess there are varying levels of "sure" as well. It overwrites least
used data rather than oldest data.
http://code.google.com/p/memcached/wiki/FAQ#Item_Expiration
Richard
On Apr 20, 2011, at 10:34 AM, Richard Revels wrote:
> I am sure it will overwrite old data if it
expire an object on january 1st of next year, this is how you do that.
And the wiki link.
http://code.google.com/p/memcached/wiki/NewProgramming#Expiration
Richard
On Apr 20, 2011, at 10:14 AM, Brett Nemeroff wrote:
> On Wed, Apr 20, 2011 at 9:09 AM, Richard Revels wrote:
> Forever is a l
Forever is a long time. If memcached fills up it will start booting out old
entries to allow new ones to be stored. I seem to remember there being a month
time limit on entries as well but that might be only if you are setting a
timeout value. Too bad I can't used memcache to replace my feebl
Hope this saves someone time at some point. The type of value stored in the
groupID avp (if used) when calling do_routing() must be an int. There is no
attempt to convert in the function. I store everything in usr_preferences as a
string and then use transforms where needed in the config scri
Simply means it's already there. Should be good to go.
Richard
On Mar 10, 2011, at 5:12 AM, Toyima Dias wrote:
> Just one more thing friends,
> When i try to install MDB2, i got following errors
>
> /#pear install MDB2
> Skipping package "pear/MDB2", already installed as version 2.4.1
> No val
k that revision was between the 1.6.3 release tag and the 1.6.4 release
by the way.
Richard
On Dec 30, 2010, at 3:05 AM, Saúl Ibarra Corretgé wrote:
> On 28/12/10 7:49 PM, Richard Revels wrote:
>> Are there any known issues with media dispatcher 2.4.2 communicating with
>> opensips r
Are there any known issues with media dispatcher 2.4.2 communicating with
opensips revision 7377? I am getting errors back to the call to
engage_media_proxy on two attempts from within my script. This doesn't seem to
occur immediately on start-up most times but does get to the point of 100%
f
t; Hi Richard,
>
> On 12/04/2010 08:35 PM, Richard Revels wrote:
>> It's possible that the b2bua doesn't yet support what I'm trying to do with
>> it but I find I have some basic questions about the XML layout as a result
>> of trying. If I want to handle
It's possible that the b2bua doesn't yet support what I'm trying to do with
it but I find I have some basic questions about the XML layout as a result
of trying. If I want to handle two different types of request within a
scenario does it require two blocks or a single block
containing the two m
The Opensips website has a section that outlines the process for generating
registers (and provides the scenario files) via sipp too.
http://www.opensips.org/Resources/PerformanceTests
On Dec 2, 2010, at 7:59 PM, Jeff Pyle wrote:
> I'm embarrassed to say how long I've wished for this capability
probably be tomorrow before I get that full trace to you.
Richard
On Dec 2, 2010, at 1:42 PM, Richard Revels wrote:
> Yeah, I used the function call on both the message and response. This
> platform is production and does a lot of traffic each day. I know this
> doesn't h
each other.
I do have a complete trace for the call. Let me obfuscate all the IP addresses
and I'll mail it.
Richard
On Dec 2, 2010, at 1:16 PM, Saúl Ibarra Corretgé wrote:
> Hi,
>
> On 12/02/2010 07:04 PM, Richard Revels wrote:
>> Here is the m line from an INVITE/200 af
Here is the m line from an INVITE/200 after the messages were modified by
use_media_proxy in each direction. The call happened to be mine and although
it did not stay up long enough for me to be 100% sure, I think the person on
the other end of the line was someone other than the person who cal
I'm not sure that process persistence is what Brett was looking for but rather
Dialog persistence. I have found that local memcache support is very fast and
takes care of this type of need quite well.
Using a unique key, made up perhaps of the SIP call-id and type of value like
"B2Bwhatever_
http://www.opensips.org/index.php?n=Resources.DocsTutLoadbalancing
On Sep 25, 2010, at 6:22 AM, Stefano Sasso wrote:
> Hello folks,
> my company has experience in setting up single asterisk setup, but
> recently one of our customers asked us to set up an asterisk cluster,
> that must be Hi
Media relay uses linux connection tracking. The rules are bi-directional. If
packets are still coming from the called party side back to the (no longer
there) calling party side then the packet counters are still incrementing.
I've been looking at the same type of issue with users on sketchy n
I take that back. ping_from is different. ping_sock is either in a newer
version than I have or doesn't exist.
Richard
On Sep 5, 2010, at 10:25 AM, Richard Revels wrote:
> Another little harmless thing is the 1.6 documentation calls out ds_ping_sock
> but the param is ds_ping_fr
Another little harmless thing is the 1.6 documentation calls out ds_ping_sock
but the param is ds_ping_from. Didn't see this mentioned on the list so I
don't think this has changed in more recent revisions.
Richard
On Apr 16, 2010, at 6:48 AM, Bogdan-Andrei Iancu wrote:
> Hi Jock,
>
> ok, wh
I'm kind of thinking that if the BYE is not accepted (error response) the
failed_transaction_flag flag might need to be set at the start of the BYE
transaction as well in order to pick it up.
On Sep 1, 2010, at 10:41 AM, Alex Massover wrote:
> Probably you don't trigger accounting for BYE re
I think this also happens (route header on initial invite) when you set an
outbound proxy on some clients. My soft phone does this. I just yank route
headers with the remove header field command on initial invites as I don't want
clients deciding how to route. As TR mentioned, that can be bad
h this ALG device early
this week and send the traces.
Richard
On Jul 30, 2010, at 4:35 AM, Saúl Ibarra Corretgé wrote:
> Hi Richard,
>
> On 28/07/10 23:04, Richard Revels wrote:
>> No need to apologize. I'm intruding on your time. Just wanted to make sure
>> I didn&
orretgé wrote:
> On 28/07/10 14:01, Richard Revels wrote:
>> Saul,
>>
>> Could you verify you got the trace file I sent? I mailed it directly to you
>> so I wouldn't have to worry about obfuscation on the IP addresses and want
>> to make sure I didn&#x
BG_QM_MALLOC above DDBG_F_MALLOC in the file
and THEN uncomment DDBG_QM_MALLOC.
Hopefully this will help someone down the road. Should be obvious but got past
me for quite a while.
DOAH!
Richard
On May 27, 2010, at 10:35 AM, Bogdan-Andrei Iancu wrote:
> Hi Richard,
>
>
Saul,
Could you verify you got the trace file I sent? I mailed it directly to you so
I wouldn't have to worry about obfuscation on the IP addresses and want to make
sure I didn't get caught in a spam filter or anything.
Richard
On Jul 27, 2010, at 7:48 AM, Richard Revels wrote:
&g
d
On Jul 27, 2010, at 4:21 AM, Saúl Ibarra Corretgé wrote:
> Hi Richard,
>
> On 24/07/10 23:05, Richard Revels wrote:
>> Saul (and everyone) good afternoon.
>>
>> I think I've come across a call flow that Mediaproxy would be expected to
>> handle but is
Saul (and everyone) good afternoon.
I think I've come across a call flow that Mediaproxy would be expected to
handle but is not. Umm, it's like sacrilege I know; but I think Mediaproxy may
have a bug.
1) The originator sends INVITE with a couple of codec choices.
2) engage_media proxy is calle
over the header Expire.
>
> Regards,
> Bogdan
>
> Richard Revels wrote:
>>#this next part logs an error if the message has no expires
>> info (scanner messages sent directly to proxy for instance)
>>if(is_present_hf("Expires"))
>>
#this next part logs an error if the message has no expires info
(scanner messages sent directly to proxy for instance)
if(is_present_hf("Expires"))
$avp(i:10) := $hdr(Expires);
else
$avp(i:10) := $ct.fields(expires);
if( $avp(i:10)
Possibly this?
AVPs are persistent per SIP transaction, being available in "route",
"branch_route" and "failure_route". To make them available in "onreply_route"
armed via TM module, set "onreply_avp_mode" parameter of TM module (note that
in the default "onreply_route", the AVPs of the transac
Sas wrote:
> Try 'make realclean' before a 'make install'. This should clean up
> completely the repo.
>
> Regards,
> Ovidiu Sas
>
> On Thu, May 27, 2010 at 11:45 AM, Richard Revels
> wrote:
>> Yep. Did a make clean and make all before make
One other thing you might want to do is set a couple symbolic links with
different names to opensips binary and call them for each instance. It makes
it easier to kill one instance without disturbing the others.
And then copy opensipsctl to individual names so you can set opensipsctlrc to
the
Yep. Did a make clean and make all before make install. I think its a little
deeper than that.
On May 27, 2010, at 11:30 AM, Bogdan-Andrei Iancu wrote:
> Richard Revels wrote:
>> I got an error when starting up with only qm_malloc enabled. Was planning
>> to mention that
AGLE, USE_MCAST, SHM_MEM,
SHM_MMAP, PKG_MALLOC, F_MALLOC, DBG_QM_MALLOC, DBG_F_MALLOC,
FAST_LOCK-ADAPTIVE_WAIT
Richard
On May 27, 2010, at 10:35 AM, Bogdan-Andrei Iancu wrote:
> Hi Richard,
>
> Richard Revels wrote:
>> In Makefile.defs uncomment
>>
>> -
In Makefile.defs uncomment
-DDBG_QM_MALLOC \
-DDBG_F_MALLOC \
In script set
debug=6
memlog=6
Restart and let run for a while. Then
cat /var/log/opensips-msg | egrep 'freeing|DBG:core:fm_malloc.*called' | sed -e
's/.*free.*\: \(.*\)/\1-mfree/' -e 's/.*malloc.*\: \(.*\)/\1
If all you have is user one, user two, and opensips, you will never have to
worry about this because the call will not set up unless the two endpoints can
settle on a codec they both support. No transcoding needed.
Richard
On May 20, 2010, at 6:06 AM, samoh wrote:
>
> Hi everyone,
>
> How c
Sorry. Should have wondered what the avp is set to when the dialog is created
rather than when it is destroyed.
On May 10, 2010, at 12:33 PM, Richard Revels wrote:
> I wonder what the timeout_avp is set to when the dialog goes away? Not to
> get off topic, but if anyone knows, can th
I wonder what the timeout_avp is set to when the dialog goes away? Not to get
off topic, but if anyone knows, can that avp be modified via opensipsctl? Been
meaning to investigate this question and keep forgetting.
Richard
On May 10, 2010, at 9:17 AM, Neo Anderson wrote:
> Hello Bogdan,
>
There are several parts of the config where this type of problem can be checked
and caught. An easy one, provided you aren't running an asterisk on the same
IP that talks to opensips or something, is this source ip and request domain
check
if( $si == $rd )
{
Pretty sure I must be missing something here but when you do the
alias_db_lookup, the destination host and port will already be set to whatever
you set the domain column to in the dbaliases table. Just call t_relay() and
it goes.
As Adrian mentioned, there are a ba-zillion ways to do this. Yo
Being able to make changes on a branch and have those changes disappear when
the branch does is very handy.
So far, the issue of not being able to change a header in script because it had
already been changed in a function call hasn't been a major issue for me
either. I wonder if the example g
This comment from the opensips mailing list archive may help with the
redundancy setup.
> 2) Set ip_nonlocal_bind via sysctl, which will allow you to bind to an
> interface even if it's not available on that host. You can have a hot
> standby host this way (on the virtual address), because it wil
though, I'm using $fn to check the CLID, however, $fn has
> double-quotes encasing it. Is there a way that I can strip the double quotes
> from $fn (I'll copy it to an AVP first obviously) before passing it to
> dp_translate, otherwise, it doesn't match my regexes.
>
dialplan module
On Apr 16, 2010, at 3:46 PM, Douglas Lane wrote:
> Hi Guys,
>
> Firstly, big thanks for assisting me with the username and raelm issue I was
> having - seems to have worked itself out nicely.
>
> I'm looking for some guidance on how to do the following the "right" way:
>
> We
I think if you don't populate the ha1b column in the subscriber table it would
also solve this problem by not passing the auth check. Depending on how you
populate the table this may be done outside your control in which case you
could use the password_column_2 modparam and set it to a column l
Apr 5, 2010, at 3:28 PM, Marco Nesler wrote:
> http://www.smartvox.co.uk/serfaq_install_mediaproxy2.htm
>
> 2010/4/5 Richard Revels
> Here is the process I used with an up to date CentOS 5.4 server
>
> wget http://download.ag-projects.com/MediaProxy/mediaproxy-2.4.2.tar.gz
>
Here is the process I used with an up to date CentOS 5.4 server
wget http://download.ag-projects.com/MediaProxy/mediaproxy-2.4.2.tar.gz
tar -xzf mediaproxy-2.4.2.tar.gz
cd mediaproxy-2.4.2
yum install python-tools.x86_64 python-twisted-core.x86_64
python-zope-interface.x86_64 python.x86_64 pytho
or alias_db is designed exactly to handle looking up a username. if found you
have side benefit of having the request uri set for next hop.
On Apr 2, 2010, at 5:18 AM, Bogdan-Andrei Iancu wrote:
> or for any kind of mysql query, you may use the "avp_db_query" function
> from avpops module:
>
27;. If there are clients that reject this kind of tags, I could take
> this character out.
>
> Regards,
>
> --
> Anca Vamanu
> www.voice-system.ro
>
>
> Richard Revels wrote:
>> I started looking at B2BUA top hiding again last night. The from tag on the
&
I started looking at B2BUA top hiding again last night. The from tag on the
outbound side is being built from something (don't remember what) on the
original Invite that contains an @ symbol and domain. Can this be changed? I
am finding quite a few UAs that don't like having that symbol in a
he
> procs will keep syncronizing over the lock for the this vars. The stat
> vars are using atomic ops (internally) and they do not require locking.
>
> just my 2 cents on this,
> Bogdan
>
> Richard Revels wrote:
>> Most likely this is where Bogdan is headed but, i
Or you could do this too. : >
On Mar 4, 2010, at 5:48 AM, Bogdan-Andrei Iancu wrote:
> Paweł,
>
> this can be simpler done with an external script that runs on cron and
> periodically fetch the stats you need (via opensipsctl fifo
> get_statistics), dump the values somewhere and resets them
Most likely this is where Bogdan is headed but, if you only want to increment
and reset some variables in the script there are global vars available to do
that with. cfgutils module.
#first a var that is shared across all processes
modparam("cfgutils", "shvset", "didtracker=i:0")
#now one that
I'm having a problem that I think is the same as this discussion. When a call
from a natted user comes in to my opensips proxy, my config does auth and then
immediately fires up the nat ping and media proxy ( engage-mediaproxy() ) to
provide far end nat traversal. Now I'm trying to add top hid
t;
> On 8 Jan 2010, at 22:17, Richard Revels wrote:
>
>> I've been meaning to come back to this. I'll try to get something
>> coherent together this weekend for CentOS users. The process
>> outlined at the link below is somewhat overkill as most of the
&g
I've been meaning to come back to this. I'll try to get something coherent
together this weekend for CentOS users. The process outlined at the link below
is somewhat overkill as most of the python stuff can be done via easy_install
with just a couple tweaks to the readme. The gnutls must be c
Interesting. I think you may have mentioned this before Dan but I didn't catch
it for some reason. That makes installing media proxy on CentOS / RHEL 5.x
easier. I've found that creating an RPM to install python 2.5 (along with the
2.4 rather than upgrading it) and then using the virtualenv p
Thank you for looking into this. Didn't mean to cause anyone work on
the weekend. I'll update from SVN tonight or tomorrow.
Richard Revels
On Sep 26, 2009, at 1:56 PM, Bogdan-Andrei Iancu wrote:
> Hi Richard,
>
> I found some bug related to flushing (of output MI tree)
ed to write -line too
long!!!
Sep 25 19:36:21 wtrunking /usr/local/opensips/sbin/opensips[10316]:
ERROR:mi_datagram:mi_datagram_server: failed to build the response
Richard Revels
___
Users mailing list
Users@lists.opensips.org
al close to the top of my "pretty cool
people" list. Thank you.
Richard Revels
On Jul 15, 2009, at 2:01 PM, Brett Nemeroff wrote:
I think it's worth re-iterating that memcache is NOT meant to be a
reliable data store and you should essentially build your
applications
Andrei,
I'll read the documentation shortly but I wonder if you could give me
a quick booster here. Does this module allow for two or more opensips
proxies to access the same memory cached data on the distributed cache?
Richard Revels
On Jul 15, 2009, at 9:57 AM, andrei dragus
What's your budget?
http://www.ditechnetworks.com/products/packet-voice-processor.html
Richard Revels
On Jul 3, 2009, at 11:33 AM, Brett Nemeroff wrote:
I too am curious about this.. High volume transcoding.. something
that ISNT asterisk. :)
-Brett
On Fri, Jul 3, 2009 at 10:
ing in the site-packages directory I
really wasn't expecting the dispatcher to load without further work.
But it did so I'm happy. Do I need to get rid of the 0.10.6 version
and if so what is the method of going about that?
Richard Revels
On Jul 3, 2009, at 10:00 AM, Ruud Klaver wrote:
t-0.10.6-py2.4.egg/sqlobject/mysql/
mysqlconnection.py", line 103, in _executeRetry
Jul 3 13:03:32 opensips-hou media-dispatcher[6005]: query =
unicode(query, self.dbEncoding)
Jul 3 13:03:32 opensips-hou media-dispatcher[6005]:
exceptions.TypeError: unicode() argument 2 m
I expect you will find that calls per second and cpu have a lot less
to do with media relay performance than do concurrent sessions and
codec (in other words, packets per second). Check out this article
and then we can discuss tools for checking network limitations if
needed.
http://staff
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
rent
> configuration, at
> least to the best of my knowledge. I'll read up on the different
> types of
> routes to see if I can figure this one out. Do you have any quick
> suggestions on how to adapt a branch-less config into something
> where I can
> add and delet
Add the headers in branch routes. Headers added in primary routing
can't be removed in later processing.
On Mar 6, 2009, at 10:40 PM, Jeff Pyle wrote:
> Hello,
>
> I’m using serial forking to send requests to multiple PSTN
> carriers. Some
> of these carriers want P-Asserted-Identity/Priva
Man Pages are your friend. When you divert opensips messages to
another logging file or destination you also need to stop them from
going to the standard messages file. Here are a couple of pertinent
entries from a syslog-ng (gentoo) and rsyslog (CentOS) config. The
remote destination en
Khan,
Here is a link to a pretty nice document on SIP:
http://www.tm.uka.de/itm/uploads/folien/100/MMK-05-SIP-4up.pdf
Two very good tools for looking at SIP requests and replies are tshark
and ngrep. On the opensips proxy try running "ngrep -q -W byline port
5060" from the command line.
Ju
I didn't really have any plans to comment on the wish list being
created as everyone has their own needs. However, since you went
there first, here is what I believe about XML.
I don't think there is a valid reason to ever have a file on disk that
contains XML formatted data. XML was thoug
I'm guessing modparam is inside a route and shouldn't be or assignment
is outside a route and shouldn't be.
Sent from my iPhone
On Nov 16, 2008, at 6:26 AM, "chamo" <[EMAIL PROTECTED]> wrote:
>
> Nov 16 12:17:07 [49967] CRITICAL:core:yyerror: parse error in config
> file, line 166, column 7-16
91 matches
Mail list logo
