[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: isdnutils and 2.0.36 (was: Linux 2.0.36 in slink?)

Remco Blaakmeer <remco-blaakmeer@quicknet.nl> wrote:
>On Wed, 16 Dec 1998, Enrique Zanardi wrote:
>> On Tue, Dec 15, 1998 at 01:21:52PM -0800, Oscar Levi wrote:
>> > It is unwise to change the kernel just before we ship.  I support
>> > making it optional for the minority of people who use ISDN adaptors.
>> And what about the "minority of people" that care about security holes in
>> our default 2.0.35 kernel?
>Those people compile their own kernels anyway. :-)

The people who are concerned about security holes are usually vastly
outnumbered by the number of people who *should* be concerned about security

After all, as a distribution, we're supposed to be concerned about security
*for* people.

>How many people would be opposed to including a kernel-source package for
>the 2.0.36 kernel, leaving 2.0.35 the 'standard' kernel for slink? I don't
>think it would break anything, and compiling your own kernel really isn't
>that hard anymore.

I think it's no better than leaving 2.0.36 out. Please, let's ship 2.0.36
as the standard kernel, seeing as it *will* work for all but a few people
(the issue is, we don't know who yet), *has* been working for a lot of
people, *hasn't* had to have anything critical about it patched in the
2.0.37pre's, and generally has less problems. (I've been hearing rumors from
Johnie Ingram that the 2.0.35 NFS client on the bootdisks is broken.)

We can keep 2.0.35 in the distribution just in case.

As for the suggestions of patching functionality from potato isdnutils to
the slink isdnutils, you end up with an untested mix of code with very
little upstream support. If we need to put new code in slink (I really think
we do in this case), we should go with the more tested option. This would be
the existing potato isdnutils package.
Robert Woodcock - rcw@debian.org
"Unix and C are the ultimate computer viruses" -- Richard Gabriel

Reply to: