2009 December

December 2009


My friend and co-teacher Daniel-Constatin Mierla has written a series of articles on the soon-to-be released Kamailio 3.0 that highlights new features. The main thing is that this new version of the old OpenSER SIP server is merged with the original SER, as part of the SIP-router project. This gives Kamailio a large set of new features, most notably the rewritten kernel that has more stability, scalability and performance than the old OpenSER. Please visit Miconda’s blogg and learn about all the new features!

Best of New in Kamailio 3.0.0 - #1: include fileI’m starting a series of posts, to highlight the best new features in Kamailio (OpenSER) 3.0.0, of course, from my point of view, hoping to cover most of them before full 3.0.0 is out (RC3 was done yesterday).

Yesterday, Russell Bryant finally made up his mind and confirmed on the asterisk-dev mailing list that the next release of Asterisk will be 1.8, which will also be a Long Term Support (LTS) release. This also means that the 1.4 is now officially classed as a LTS release too.

What is a long term release, LTS?

Long Term Releases are releases that are going to be supported for four years. Standard releases, like 1.6.x, are going to be supported for one year, with one additional year of security fixes. This means that the support for 1.6.0 will cease in october 2010. There’s a new release schedule on the Asterisk.org web site that explains the details.

Open Source Projects have to be easy to understand and use

I feel that this is a very good solution for the whole Asterisk community and that we all will benefit from it. I have personally not been happy with the 1.6.x release schedule, which has been very misunderstood and has confused a large group of users. Hopefully, we can now continue with a release schedule that the world understands and that makes sense for everyone. While I understand the need for releasing quicker than we’ve done in the past, the detail about the naming, the actual release numbers (1.6.0, 1.6.1 etc) was very hard to explain to people. With years of experience of doing Asterisk and VoIP training, I have a lot of respect of the need of being able to easily explain things, from configuration details to release schedule…

Time to focus on defining Asterisk 1.8

Now we, the Asterisk community, need to focus quickly on the new release and plan what’s going in there. If you have code for new features lying around (as I have tons of in various branches of my svn repository), now is the proper time to step forward, contribute it to the bug tracker and get it evaluated, discussed and maybe finally included in Asterisk. Whatever goes into 1.8 at release time, will be what we will have for production use for  a long time.

Please dedicate time for testing during Q1 and Q2 2010!

We also ask you to dedicate time during next year to help the Asterisk project with testing. You don’t have to be a developer to test – and we need tests of everything from documentation to configuration and technichal issues. We don’t have all of the equipment you have, we don’t have your dialplans, we don’t have all the applications you integrate Asterisk with. If Asterisk is important to your organization, please make sure that you dedicate time during the first half of 2010 to do regular testing of the new release betas and release candidates. We do need your help to make Asterisk 1.8 a good release, worthy to replace the 1.4 as a new LTS release.If you’re a member of a Linux or Asterisk group, please help in organizing Asterisk 1.8 test-partys. If you need help with ideas, please contact our community liason, John Todd. Meeting other Asterisk users, testing stuff together is one of the best ways to expand your knowledge of Asterisk. Sharing ideas and how-to’s in real time while setting up test labs and scenarious is really, really fun.

Asterisk 1.8 will make a difference

Asterisk has added a lot of new features and internal scalability and stability since 1.4. The 1.6.x releases are to me test releases to show and run practical tests with all of these changes. The core has changed, the API’s has changed and the internal PBX is practically new. We’ve proven scalability to over 10.000 calls on one server. We’ve proven interoperability with many, many products out there. We’ve changed the way we do development and we’ve moved Asterisk into the world of non-PSTN wideband audio. Of course, there’s a lot of more things we can do, but if we consider all of the changes since 1.4, Asterisk 1.8 LTS will be a really cool telephony toolkit.

At the first Astricon I was very happy to see Marc Blanchet as one of the attendees. I knew he was one of the IPv6 gurus and wanted someone to show some interest in Asterisk and IPv6.Well, he did not only get interested in it, but started coding on it. The results have been available for quite some time at http://www.asteriskv6.org/ and Marc has tested it at several SIPits for interoperability.

This patch is very large and affects large areas of Asterisk. In order to support IPv6, we need to update the way we interact with sockets, with DNS, with URI’s. The SIP channel needs to handle multiple UDP as well as TCP sockets in both protocols. The ACL’s we use for all VoIP protocols and manager needs support for IPv6. And much more.

Marc hasn’t been able to spend time to keep it up to date with the everchanging trunk. I feel we need to move this forward and try to divide the large patch into smaller pieces that can be reviewed separately by the developer team and  be merged gradually. First, Marcs branch needs a serious overhaul to get up to date with trunk.

In order to work on this, Marc and I needs funding .I have a few interested parties, but need more interested parties that can commit to funding during the first half of 2010 for this project.  It’s not a small task, the current estimate is at least one month’s work for each of us for updating, cutting it up, merging, going through the review process, testing and finalizing with new tests at SIPit or a similar event.

If your organization is interested, please let me know off list and we’ll discuss from there. My e-mail is as always oej@edvina.net. Please don’t hesitate to mail me with any questions you might have about this project.

/Olle