[asterisk-users] Stable-Stable Asterisk

Steve Totaro stotaro at first-notification.com
Fri Aug 24 15:33:52 CDT 2007


Russell Bryant wrote:
> James FitzGibbon wrote:
>   
>> Let me ask a question myself: what kind of regression test does * undergo
>> before release, and what level of traffic gets put through stuff like
>> app_queue?  I assume it's not real-world scale, else these hard to pin down
>> concurrency issues we're seeing would have been caught in test.
>>     
>
> Let *me* ask a question.  :)  What level of heavy regression testing would you
> *expect* of an open source development team?
>
> We really do try very hard to test all of our changes.  We have community
> members that work very hard to help test out the more invasive changes.
> Furthermore, we have a lot of people run the code from the release branches
> directly so that regressions are caught quicker, and hopefully before they make
> it in to a release.
>
> At Digium, we have a department dedicated to doing testing of our products,
> including Asterisk.  Every bug that is found as a part of this testing gets
> fixed in the open source branches as well.
>
>   
I don't think anyone is arguing that you guys are not trying your hardest. 

The point is that 1.4.x is not stable enough for production.  Start 
thinking in terms of the traditional telephony world and not in terms of 
the software world. 

Traditional PBXs and switches have years and years of testing and bugs 
are very minimal (there are many I have found in various older Avayas 
and Toshibas but still usually something like you cannot delete 
something after you create it, nothing that takes out a system).

Software is pushed out as fast as possible with known bugs.  I just 
don't really see that in the traditional telephony world. 

Almost every company relies on their phone system more than even email.  
Email could be down for a few hours, a day, even a week and business 
would get done.  Not so with a phone system.

Thanks,
Steve Totaro



More information about the asterisk-users mailing list