CentOS - Apr 2005 - Problems installing Centos 3.4 & 4 using NFS + Kickstart

My hardware is Sun V65x servers, (6GB RAM/2xXeon3GHZ/36GB SCSI disk).
I have NFS repositories of Centos 3.1, Centos 3.4, Centos 4 & RedHat 9.0 
ISO distros set up. I use boot floppies for each o/s with the same 
kickstart file which is of course altered to point at the desired 
distro. With Centos 4 I have to use a boot CDROM, of course, rather than 
floppy disk.  I wish to set up servers with either Centos 3.4 or 4 
versions rather than RH9. Root disk partition is 24GB btw.

The kickstart boot works fine with RH9 & Centos 3.1. However it fails 
with Centos 3.4 & Centos 4 but in different ways:

Centos 3.4
=======The kickstart proceeds normally, formats the partitions & starts the
install. However at 52% of the way through,  it aborts, always on the 
same file: "glibc-kernheaders-nnnn.rpm"
The fatal error message is non-specific & suggests that media is 
defective or disk space inadequate. I have checked the md5 sums of the 
ISOs & they agree with the download site ones. I have also extracted the 
"glibc-kernheaders-nnnn.rpm" from the ISO image #1 & compared
its' md5
sum with the same module from the /os/ directory on the download site & 
they also agree. I have also downloaded the ISOs again & they are the 
same md5s as the first set :-\

Centos4
======This install gets to the point at where it would normally try & mount
up
the CENTOS4 NFS repository which is on the same server as all the other 
distros (this runs RH9). It then goes into a loop whereby it keeps 
prompting for the name of the NFS server & the CENTOS directory,  where 
it would normally just mount it & proceed with the installation.

Anyone out there with similar experience & a working fix for this issue 
please ?

Les Oswald

On Tue, Apr 19, 2005 at 03:37:16PM +0100, Dr R L Oswald, Cranfield University UK
wrote:
> My hardware is Sun V65x servers, (6GB RAM/2xXeon3GHZ/36GB SCSI disk).
...
> The kickstart boot works fine with RH9 & Centos 3.1. However it fails 
> with Centos 3.4 & Centos 4 but in different ways:
> 
> Centos 3.4
> =======> The kickstart proceeds normally, formats the partitions &
starts the
> install. However at 52% of the way through,  it aborts, always on the 
> same file: "glibc-kernheaders-nnnn.rpm"
large memory issue? can you try with mem=2048M

3.1 and 3.4 mostly differ on the kernel version, 
I have not seen any similar issue here, but not Sun and no 6GB RAM machines.

How does your ks.cfg look like? 

You did not tell your arch: i386 or x86_64? Not sure it really matters here.
> Centos4
> ======> This install gets to the point at where it would normally try
& mount up
> the CENTOS4 NFS repository which is on the same server as all the other 
> distros (this runs RH9). It then goes into a loop whereby it keeps 
> prompting for the name of the NFS server & the CENTOS directory,  where
> it would normally just mount it & proceed with the installation.
Do you have 2 network cards? The 2.6 kernel from CentOS-3 can swap the eth0/1
cards
compared to the 2.4 kernel from CentOS-3.

No problem here either.

Cheers,

Tru
-- 
Tru Huynh (CentOS-3 i386/x86_64 Package Maintenance)
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBEFA581B
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL:
<http://lists.centos.org/pipermail/centos/attachments/20050419/40f8321d/attachment-0001.sig>