Hello,
version 1.2.0 of OpenSER has been released. It is major release, which means that you get along a lot of new features and improvements to existing ones in old versions. The configuration file and database structure suffered some changers, please read careful release notes to learn how to update.
Just to list few of the new features: - SIMPLE Presence support (http://openser.org/dokuwiki/doku.php/presence:presence-module) - SNMP (http://openser.org/docs/modules/devel/snmpstats.html) - PERL API (http://openser.org/docs/modules/devel/perl.html) - JAVA SIP Servlet Application Server (http://wesip.eu) - script variables and transformations (http://openser.org/dokuwiki/doku.php/transformations:devel) - DNS failover - IP blacklists - XMPP IM gateway - XMLRPC - SIP session timers - secure federation peering - retransmission timer accuracy - usrloc refurbished for better performances - automatic error handling - AVPs in reply routes
Read full release notes at: http://www.openser.org/mos/view/OpenSER-v1.2.x-Release-Notes/
Source tarball can be downloaded from: http://www.openser.org/pub/openser/latest/src/ Or SourceForge: http://sourceforge.net/project/showfiles.php?group_id=139143&package_id=...
This is the first release coming out o SVN. Old CVS repository was migrated to SVN, and with these, a few changes apply to get sources from repository. Please read instructions from download page: http://www.openser.org/mos/view/Download/
Migration instructions (although not complete, I'm sure) are available at: http://openser.org/dokuwiki/doku.php/install:1.1.x-to-1.2.x
The documentation for modules is posted at: http://openser.org/docs/modules/1.2.x/
We would like to thank to all developers and contributors for their work with coding and testing for this new release.
Cheers, Daniel
congratulations!
thanks for the hard work.
regards klaus
PS: someone should make a snapshot of http://www.openser.org/dokuwiki/doku.php/pseudovariables:devel and name it pseudovariables:1.2.x and update openser.org/docs
regards klaus
Daniel-Constantin Mierla wrote:
Hi,
The dialog PV are missing from the 1.2 and devel PV webpage. http://openser.org/docs/modules/1.2.x/dialog.html#AEN185
Regards, Ovidiu Sas
On 3/13/07, Daniel-Constantin Mierla daniel@voice-system.ro wrote:
Hi Klaus,
this was a collective effort - There was a huge volume of help for testing, troubleshooting and fixing problems, help coming from all over the community. Also we thanks to everybody contributing to openser , disregarding if it is about code, patches, reports, ideas, docs, etc...All this put together made this great result!
BTW, we did some tests with OpenSER 1.2.0 and today we will publish the results ;)
regards, bogdan
Klaus Darilion wrote:
While I tried to modify the code, I found it is a big challenge for the newbie to understand the make and build system.
Just wondering if anybody is considering use CMake to replace the autotool. KDE4 is now using this user-friendly and cross-system build system, so I assume it will become more popular.
Wanding. Everything is possible. Everytime I am thinking.
On 3/14/07, Bogdan-Andrei Iancu bogdan@voice-system.ro wrote:
Hi,
Just throwing wild ideas ... but if we moved to autotools (we need to look for a guru ... I agree, more than autotools is auto-chaos) we could easily cross-compile ... And that said, it just comes to mind the chance of cross-compiling with mingw and have a windows openser, just for the non-linux user and expand our borders. It may be a cut down version to start with, but wouldn't it be great?
Cesc
On 3/28/07, wanding wang wang.wanding@gmail.com wrote:
On Friday 30 March 2007 01:00, Cesc wrote:
Hi Cesc,
i don't think that it would be so easy to port openser to windows, there are many Unix/ Linux specific thinks specified deep in the code.
More easy cross-compiling could be nice. But then we should use cmake instead of autotools, like the KDE project, because this is usable even for non-gurus like me. :-)
Cheers,
Henning
On 3/30/07, Henning Westerholt henning.westerholt@1und1.de wrote:
Hi,
Well, never used cmake myself, so can't tell. Autotools is complicated if starting from scratch, but if using existing configs from some other project, the jumpstart is very big.
As for the port ... i would expect some difficulties in some parts, like locking, loging and memory mgmnt ... also some database access libraries ... but maybe a simple, very cut-down version, with basic functionality, could be realized ...
Just brainstorming ...
Cesc
On Friday 30 March 2007 01:00, Cesc wrote:
No. Posix based software was in general never intended to run on Windows. More often than not the additional effort isn't worth it and another thing to pull away from core development. This is not even to mention the packaging effort which goes with it as the MS crowd expects binaries. Finding a developer which will solely handle the entire task is difficult and once you open the box someone is stuck with it from there on. Feel free to ignore this as it's just my observation from what I have seen in quite a few projects which this was requested and attempted.
Cheers,
Christopher
On Friday 30 March 2007 12:51, C. Bergström wrote:
The effort could be quite high, you're right. And i don't think it is right thing to do to port every free software project on the planet to the MS Windows platform, if they want to use e.g. OpenSER, they should use linux or a other unix system. :-)
If anybody want to pay a developer for this, then it is another question. But it is also my opionion that this work should not be done from the core development team.
Best regards,
Henning
On 3/30/07, Henning Westerholt henning.westerholt@1und1.de wrote:
I agree that this could mean quite some effort, that it need not be directly supported by the core team, but there is a lot of satelite developers who may do that in their "free time", as a hobby. I don't agree in the monolithic version of "only linux" ... running on another platform, being portable, opens a whole new bunch of oportunities to any project ... For one, in my company, we run embedded linux AND windows embedded ... but we cannot have the same proxy on both platforms, so you have to support two software units ... that is work!
Cesc
On 3/30/07, Juha Heinanen jh@tutpro.com wrote:
:) yes, i know, there is always two faces to a story ...
Look, the "my-work" example was just to show that being flexible and open is not bad ... of course it takes more effort; this is like all in life, pros and cons.
Basically, going back to the origins of this thread: I think we should not bluntly discard new building ways just because one of the outcomes may be a windows-based openser. Now, if all the porting needed can be done with the current makefiles ... ok then.
Cesc
So why not try to see the beautiful scene if Openser can run on every kind of server platform?
Anyway, Windows Server system will not disappear in the seen future, and what's more, some organizations and companies still seriously rely on Windows, I mean if they found Openser can be integrated into their Enterprise application systems on hand, they will probably be very excited. Even as the common user, I believe our planet is looking forward to a power ,stable and free SIP server, just like Apache HTTP server!
In order to fully understand the whole picture of Openser, I try to find some architecture documents, but I failed, who can do me a favor?
So I try to figure out the relationship of some key modules as the attachment, any comments will be appreciated.
Wanding. Everythig is possible. Everytime I am thinking.
On 3/30/07, Cesc cesc.santa@gmail.com wrote:
wanding wang writes:
So why not try to see the beautiful scene if Openser can run on every kind of server platform?
because that will eat resources that would otherwise go into development of new features, enhancing performance, fixing bugs, responding to mailing list messages, etc.
-- juha
On Sunday 01 April 2007 03:25, wanding wang wrote:
Hi Wanding,
i read the "SIP Express Router v0.8.8 - Developer’s Guide" from iptel.org as introduction into the code/ design. But this is outdated in many areas, like the fifo mechanism. So just read the code. ;-)
If there are any newer documents about the design/ internal workings of openser, i would appreciate a pointer to it too.
Cheers,
Henning
Hi Daniel,
Core team support ... this work would need not be done by the core team, but if in-depth support in the form of inside knowledge was offered, that should be enough to start with ... Once the cross-compile environment is setup and running, to be used by everyone, this would be a job of everyone to maintain.
Cesc
On 3/30/07, Daniel-Constantin Mierla daniel@voice-system.ro wrote:
On Friday 30 March 2007 11:30, Cesc wrote:
As long as you don't need to change anything with autotools, and you can stay with your copied configuration, then autotools work fine. I follow the KDE project closely, and they made good experiences with cmake after a long period of autotools problems, http://dot.kde.org/1172083974/ . I look a little bit into this, and it seems ok.
Do you expect a huge interest in a windows port?
Henning
Hi Cesc,
that is totally true - the key question is actually if we need autotools - the current makefile is self-configuring (autodetection) and so far worked mainly with no problems. So - are there some major problems to push us to autotools?? otherwise, without a guru with free time, it will be something that will make our lives more complicated.
regards, Bogdan
Cesc wrote:
Just throwing wild ideas ... but if we moved to autotools (we need to look for a guru ... I agree, more than autotools is auto-chaos)