Followup: Re: [CentOS] Hosed by 4.2
kerplop at sbcglobal.net
Wed Oct 19 20:57:44 UTC 2005
> Karanbir Singh wrote:
>> Hi Robert,
>> Robert wrote:
>>>> why did you manually need to install the kernel ?
>>> Because a "yum update kernel" offered to install the -SMP kernel.
>>> This is, no doubt, an artifact of anaconda & associates deciding at
>>> the time CentOS4 was first installed that an SMP kernel was
>>> appropriate for an Athlon XP in an ASUS A7NX8 ver.2 deluxe m/b,
>>> compounded by my packrat reluctance to throw it away at the outset.
>> if you remove the kernel-smp ( which, based on your statement - you
>> dont seem to be using) yum should not update it :) technically, only
>> packages already installed are updated ( or pkgs that satisfy depends
>> for other pkgs ).
>> Anyway, if anaconda left behind a smp kenel on UP machine, sounds
>> like a bugreport to me.....
> Not only leaves it behind but configures GRUB to make SMP the default.
> Drives NTP nuts! I should be shot for not removing it when I first
> installed CentOS4. It'll be gone when the smoke clears from my
> upcoming exercise.
>>> I'm reasonably sure everything is gonna be O.K. Yum is one of the
>>> packages that gets reported twice:
>> sounds like you are going to have a fun filled Monday morning. I
>> forsee rpm and coffee in your immediate future. Be a good idea to
>> backup the rpmdb somewhere. Just in case.
>> remove everything apart from the Packages file from /var/lib/rpm -
>> then rebuild the db ( rpm -v --rebuilddb ). then try work with the -V
>> option to verify what you have what rpm thinks you have.
>> - K
> Thanks for the advice. My Monday and Tuesday have already been
> scheduled for other joyous tasks, so I'll attack this fiasco
> Wednesday. (Now that I'm retired, I wonder almost daily how the hell
> I ever had time to work!)
> At any rate, I've already burned /var/lib/rpm to a CD and I'll have a
> brand new full backup by about 3:00 AM Wednesday.
I rebuilt the rpm database but when I verified the packages, the number
of output lines approached the vaalue of 1/0 as a limit. But that's
neither here nor there. I'll get over the fresh install of CentOS4.2.
What I was really interested in commenting on is anaconda's penchant for
installing the SMP kernel *and making it the default* when the UP is
clearly (to humans, anyhow) indicated. Here is my
[root at mavis ~]# cat /boot/grub/grub.conf
# grub.conf generated by anaconda
# Note that you do not have to rerun grub after making changes to this file
# NOTICE: You have a /boot partition. This means that
# all kernel and initrd paths are relative to /boot/, eg.
# root (hd0,0)
# kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00
# initrd /initrd-version.img
title CentOS-4 i386 (2.6.9-22.ELsmp)
kernel /vmlinuz-2.6.9-22.ELsmp ro root=/dev/VolGroup00/LogVol00
title CentOS-4 i386-up (2.6.9-22.EL)
kernel /vmlinuz-2.6.9-22.EL ro root=/dev/VolGroup00/LogVol00
[root at mavis ~]#
[root at mavis ~]# grep ' kernel' install.log
[root at mavis ~]#
This excerpt from /var/log/dmesg shows that anaconda correctly
identifies both the m/b
and, a few lines from the end of this snippet, the CPU.
HighMem zone: 0 pages, LIFO batch:1
DMI 2.2 present.
Asus A7N8X v2 detected: BIOS IRQ0 pin2 override will be ignored
ACPI: RSDP (v000 Nvidia ) @ 0x000f75e0
ACPI: RSDT (v001 Nvidia AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x1fff3000
ACPI: FADT (v001 Nvidia AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x1fff3040
ACPI: MADT (v001 Nvidia AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x1fff74c0
ACPI: DSDT (v001 NVIDIA AWRDACPI 0x00001000 MSFT 0x0100000e) @ 0x00000000
ACPI: PM-Timer IO Port: 0x4008
Built 1 zonelists
Kernel command line: ro root=/dev/VolGroup00/LogVol00 rhgb quiet
CPU 0 irqstacks, hard=c03e5000 soft=c03e4000
PID hash table entries: 2048 (order: 11, 32768 bytes)
Detected 1913.094 MHz processor.
Using tsc for high-res timesource
Console: colour VGA+ 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 514240k/524224k available (2111k kernel code, 9412k reserved,
144k init, 0k highmem)
Calibrating delay loop... 3776.51 BogoMIPS (lpj=1888256)
Security Scaffold v1.0.0 initialized
SELinux: Starting in permissive mode
There is already a security framework initialized, register_security failed.
selinux_register_security: Registering secondary module capability
Capability LSM initialized as secondary
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
CPU: After generic identify, caps: 0383fbff c1c3fbff 00000000 00000000
CPU: After vendor identify, caps: 0383fbff c1c3fbff 00000000 00000000
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 512K (64 bytes/line)
CPU: After all inits, caps: 0383f3ff c1c3fbff 00000000 00000020
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: AMD Athlon(tm) XP 2600+ stepping 00
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
ACPI: IRQ9 SCI: Edge set to Level Trigger.
checking if image is initramfs... it is
I'm NOT trying to blame anaconda or anyone else for the exercise I've
been through today, merely pointing out the reason that I manually
installed the kernel before doing the upgrade.
Best regards to all who have made CentOS happen!
More information about the CentOS