[3462] in SIPB_Linux_Development
Re: outstanding bugs
daemon@ATHENA.MIT.EDU (Angie Kelic)
Sat Apr 14 15:06:42 2001
Message-Id: <200104141906.PAA01390@tomcat.mit.edu>
To: Derek Atkins <warlord@MIT.EDU>
cc: linux-dev@MIT.EDU
In-reply-to: Your message of "14 Apr 2001 11:39:03 EDT."
<sjmbspzjxt4.fsf@rcn.ihtfp.org>
Date: Sat, 14 Apr 2001 15:06:28 -0400
From: Angie Kelic <sly@MIT.EDU>
Yes, smp machines typically have both installed. Currently
the update will update just kernel-* and not kernel-smp-* which
means when it reruns lilo you end up with a machine that boots
the new single processor kernel. It works, it's just nonideal
and requires you to go off and install the new smp kernel.
--Angie
>Ok, then we can probably generate a package list via 'rpm -qa ... |
>grep -v ^kernel-smp'
>
>-derek
>
>Greg Hudson <ghudson@MIT.EDU> writes:
>
>> > I'm worried about the case where packagelist says nothing, and
>> > kernel-smp is installed. When rpmupdate runs, it should see
>> > 'kernel' in the new package-list, and notice that it isn't in the
>> > installed package-list. So, wouldn't rpmupdate try to install the
>> > "new" kernel package (even if it does nothing with the kernel-smp
>> > package)?
>>
>> Ah. I have been advised in the past that most SMP machines have both
>> kernel and kernel-smp installed, and that this is not harmful
>> (although rpmupdate tends to break your lilo.conf when it upgrades the
>> kernel package--a known bug).
>
>--
> Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
> Member, MIT Student Information Processing Board (SIPB)
> URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH
> warlord@MIT.EDU PGP key available
