Frankly, as a member of the user community (and not really the 
developer community in any meaningful way), I think it's more 
significant that point releases be available as soon as important bugs 
are fixed. I know that Slony-I is a relatively young product, so I tend 
to give it a 24-hour window after release before I start testing, 
anyway.

For use in production environments, core functionality is just as 
important as security, and security advisories tend to come out in many 
products as soon as a patch is applied.

I'm just happy to have a replication solution for postgres.

Thanks!

-tfo

--
Thomas F. O'Connell
Co-Founder, Information Architect
Sitening, LLC
http://www.sitening.com/
110 30th Avenue North, Suite 6
Nashville, TN 37203-6320
615-260-0005

On Oct 22, 2004, at 11:15 AM, Darcy Buskermolen wrote:

> As everyone here knows, Slony-I is still a relatively new project and 
> as a
> project we are still working out kinks in the change approval process,
> release managemnet, announcements, and the rest of the mangemnet 
> tasks, we
> are always open to feed back from our users and supported on how to 
> help
> streamline this process when ever possible, and all we ask in return 
> is you
> provide quality feedback when it comes to identifying bugs, and testing
> solutions to these bugs.
>
> Perhaps it's time to mention a few things regarding the quick 1.0.3, 
> 1.0.4,
> 1.0.5.  Jann sent out a pre-announcement notice to -general stating the
> intent to package 1.0.3 and that people were encouraged to test 
> -STABLE to
> see if there were any outstanding issues. After a few days with no 
> feedback
> 1.0.3 was tagged and prepred for release, bundled and packaged. Only 
> after
> making 1.0.3 available did Vivek Khera find out that 1.0.3 didn't 
> compile
> cleanly on FreeBSD 4.10, so the change was made, and 1.0.4 was created.
> Testing began on a few additional platforms that we developers have 
> access
> to, so that an official announcement could be made with statements of 
> 1.0.4
> has been tested successfully on X.y.z running a.b.c .. During this 
> time I
> mannaged to trip bug 896 (where merge set was issued before the 
> subscription
> was complete) resulting in broken replication in a production database.
> After speaking with Jann on IRC about this, He came to the following
> conclusion: "If someone with the experience of Darcy steps onto that
> landmine, a way to guard against it is worth a release by itself."  
> And this
> in it's self gave birth to 1.0.5.
>
> None of us on the development team are feeling especially happy with 
> this
> release schedule mish mosh, but as always our primary focus is 
> providing a
> stable, working full featured product, and I belive our attention to 
> this
> detail is evedent in this bang-bang-bang of releases.
>
> Slony-I development team.
>
> -- 
> Darcy Buskermolen
> Wavefire Technologies Corp.
> ph: 250.717.0200
> fx:  250.763.1759
> http://www.wavefire.com
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at gborg.postgresql.org
> http://gborg.postgresql.org/mailman/listinfo/slony1-general