On Mon, Dec 27, 2004 at 12:54:01AM -0500, Mitchell Laks wrote: > On Monday 27 December 2004 12:32 am, Roberto Sanchez wrote: > > I think the issue is that packages are not directly uploaded to testing. > > So it is possible to have version X of package A installed in testing. > >... > > If a serious or grave bug is filed, the package simply will not make it > > into testing. Likewise, if the package fails to build for *any* of > > the supported Debian architectures, it will not go into testing (unless > > it as architecture specific package, like a kernel). You could > > potentially be running insecure software for an indefinite period of > > time. > >... > > team. Thus, updates will be made as quickly as feasible. You simply do > > not have this guarantee with unstable or testing (except when testing > > gets security team support in preparation for release). > > Thank you so much Roberto for you clarification of the process. I wonder, > as > we get closer and closer to a sarge release, will using Sarge become less > risky? I see for instance that Stanford University has begun using some > debian sarge servers (1). Will there be major structual revisions before > sarge release, or are we getting more and more settled structurally? > Mitchell > (1) http://www.eyrie.org/~eagle/writing/debian-server.html
Until sarge gets security support, it's utterly unfit for any "server" use. Really it shouldn't be used on any system that goes anywhere near the internet. -- For every sprinkle I find, I shall kill you! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]