CentOS

Hello and thanks for 5.3, my upgrade on existing boxes worked fine, just having some trouble installing new images from PXE and also network boot CD.

Each time I boot either from PXE or CentOS-5.3-i386-netinstall.iso the installer starts then after disk partitioning errors with traceback dump.

I get same issue if I boot from CentOS-5.3-i386-netinstall.iso and use a public mirror - same traceback call appears. It doesnt do this if I boot from the DVD directly.

Lastly, when I got the traceback error and hit "remote", I got errors and it didnt copy. My understanding of the "remote" button is that it uses scp to copy anacdump.txt to a remote host. From the network install image I could not find any trace of scp or ssh framework, so dont know if thats why it errored?

In /var/log/http/access_log:-
192.168.2.21 - - [05/Apr/2009:13:33:06 +0100] "GET /kickstart_files/centos53.ks HTTP/1.0" 200 1519 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:07 +0100] "GET /centos/5.3/os/i386/images/updates.img HTTP/1.0" 404 314 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:07 +0100] "GET /centos/5.3/os/i386/disc1/images/updates.img HTTP/1.0" 404 320 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:07 +0100] "GET /centos/5.3/os/i386/images/product.img HTTP/1.0" 404 314 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:07 +0100] "GET /centos/5.3/os/i386/disc1/images/product.img HTTP/1.0" 404 320 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:07 +0100] "GET /centos/5.3/os/i386/images/minstg2.img HTTP/1.0" 200 39669760 "-" "anaconda/11.1.2.168"
192.168.2.202 - - [05/Apr/2009:13:33:38 +0100] "GET /centos/5.3/os/i1/.discinfo HTTP/1.1" 404 303 "-" "urlgrabber/3.1.0"
192.168.2.202 - - [05/Apr/2009:13:33:38 +0100] "GET /centos/5.3/os/i386//disc1/.discinfo HTTP/1.1" 404 312 "-" "urlgrabber/3.1.0"
192.168.2.202 - - [05/Apr/2009:13:33:40 +0100] "GET /centos/5.3/os/i386/repodata/repomd.xml HTTP/1.1" 200 1140 "-" "urlgrabber/3.1.0 yum/3.2.19"
192.168.2.202 - - [05/Apr/2009:13:33:40 +0100] "GET /centos/5.3/os/i386/repodata/primary.xml.gz HTTP/1.1" 200 898857 "-" "urlgrabber/3.1.0 yum/3.2.19"
192.168.2.202 - - [05/Apr/2009:13:33:41 +0100] "GET /centos/5.3/os/i386/repodata/comps.xml HTTP/1.1" 200 935865 "-" "urlgrabber/3.1.0 yum/3.2.19"

In /var/log/http/error_log:-
[Sun Apr 05 13:33:07 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i386/images/updates.img
[Sun Apr 05 13:33:07 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i386/disc1
[Sun Apr 05 13:33:07 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i386/images/product.img
[Sun Apr 05 13:33:07 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i386/disc1
[Sun Apr 05 13:33:38 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i1
[Sun Apr 05 13:33:38 2009] [error] [client 192.168.2.202] File does not exist: /home/htdocs/centos/5.3/os/i386/disc1

anacdump.txt attached.

Grateful for any pointers.

R


------

For what its worth here's my config, but this works fine with 5.2. When I switch the relevant parts to centos 5.3 I get the above traceback.

- copied the boot kernel from images/pxeboot to /var/lib/tftpboot/centos53
-rw-r--r-- 1 root root 6302756 2009-04-05 12:18 initrd.img
-rw-r--r-- 1 root root 1826484 2009-04-05 12:18 vmlinuz

- lines in pxelinux.cfg look like this:-
label centos53
kernel /centos53/vmlinuz
append initrd=/centos53/initrd.img ramdisk_size=9216 ks=http://192.168.2.253/kickstart_files/centos53.ks

- centos53.ks looks like this
install
text
url --url http://192.168.2.253/centos/5.3/os/i386
lang en_GB
keyboard uk
network --device eth0 --bootproto static --ip 192.168.2.202 --netmask 255.255.255.0 --gateway 192.168.2.254 --nameserver 192.168.2.253 --hostname centos53.mylan.local
rootpw --iscrypted mypasswordhere
firewall --enabled --port=22:tcp
authconfig --enableshadow --enablemd5
selinux --disabled
timezone --utc Europe/London
bootloader --location=mbr --driveorder=sda
reboot
clearpart --drives=sda --all --initlabel
part /boot --fstype ext3 --ondisk sda --size=128
part /usr --fstype ext3 --ondisk sda --size=2048
part /var --fstype ext3 --ondisk sda --size=1024
part / --fstype ext3 --ondisk sda --size=512
part swap --ondisk sda --size=512
part /tmp --fstype ext3 --ondisk sda --size=128
part /home --fstype ext3 --ondisk sda --size=1 --grow

%packages --nobase
@core

I did attached the anacdump.txt but got error when posting, webserver unable to write to upload...

...and ignore this probable duplicate post. I've just read a lot further down a post I didnt see when searching the first time - crashdump possibly related to memory.

Having doubled the amount of memory from 256 to 512, the net install completes successfully.

Begs the question why this occurs though.

R