1and1 rolls out new server packages

[Highland]

Just a general heads up. 1and1 has new server packages and the big notes are

-All servers are now 64 bit Athlons
-No more FTP backup. All servers have RAID 1
-A lot more HDD storage, RAM and bandwidth (ES2 has 8GB of RAM)
-Still using FC4/PSA8 as default image

Something tells me CentOS will not be so easy to install on them anymore. Not to mention backing up data offsite.

[scott]

I thought I heard they supported CentOS on reimage now. Not that I'm sure they didn't fuck with it of course.

[Highland]

It's not showing up as a reimage option for me but I'm still using my older servers too. I know that these new servers are hosted out of their new KS datacenter but I'm uncertain if that changes anything as far as the control panel goes.

[fergie]

Just got a new 64-bit SATA Business Root Server I and trying to get CentOS on it... Has anyone had any joy? Am fiddling the AOOI script but not sure what I'm doing really. Any change of an update Scott?

[scott]

Sure, want to give me access to the box? Its probably because the ServerCD Im pulling is i386. I need to work in an architecture detection routine.

[fergie]

No probs, I'll just get it re-imaged to its virgin FC4 state and pm you the details.

[fergie]

Well I had a play while waiting for you to open my pm and I got a bit further... Here are the commands I used:

Code: Select all

. /etc/sysconfig/network-scripts/ifcfg-eth0

LOADER=/etc/lilo.conf
DNS=`grep nameserver /etc/resolv.conf |head -n 1 | awk '{print $2}'`

yum -y install dosfstools

swapoff -a
/sbin/sfdisk --force --change-id /dev/sda 2 7
/sbin/sfdisk --force --change-id /dev/sdb 2 7
/sbin/mkfs.vfat /dev/sda2
/sbin/mkfs.vfat /dev/sdb2
/bin/mount /dev/sda2 /mnt/
cd /mnt

wget http://www.gtlib.gatech.edu/pub/centos/4.4/isos/x86_64/CentOS-4.4.ServerCD-x86_64.iso

mkdir 1
/bin/mount -o loop CentOS-4.4.ServerCD-x86_64.iso 1/
cp 1/isolinux/vmlinuz /boot/vmlinuz.atomic
cp 1/isolinux/initrd.img /boot/

cd /
wget http://www.dfinternet.net/tests/ks-1and1.cfg >/dev/null 2>&1

cat >> $LOADER << EOF
image=/boot/vmlinuz.atomic
        label=atomic
        initrd=/boot/initrd.img
        append="ks=hd:md1:/ks-1and1.cfg console=ttyS0,57600n8 dns=$DNS"
EOF

lilo
lilo -R atomic

Unfortunately, I think I got something wrong on the config (append="ks=hd:md1:/ks-1and1.cfg console=ttyS0,57600n8 line maybe) as when I rebooted and watched on the serial console, I was prompted for all sorts of info already specified in my ks.cfg so I think the loader missed it. I also think I screwed up the partitions and RAID

In the process of re-imaging from 1and1 again and I promise to leave it alone for a while. I'll give you until the end of play tomorrow to have the server and play around, although I do need to get on towards the end of the week and might be prepared to live with FC4 if this is going to be a nightmare...

Let me know how you get on!

[scott]

Its probably because you've got to change some things around for a x86_64 ks.cfg. What you can do is go through the install via serial, and then send me your /root/anaconda-ks.cfg

[fergie]

Changed a few things to suit, but unsure of the Disk / Partition config, even through the interactive install! Have never worked on RAID systems so don't know where to start, sorry...

Will you get chance to have a play with my access details? (Check your Private Messages).

My half-edited ks.cfg is here:

Code: Select all

# Kickstart file automatically generated by anaconda.

install
harddrive --partition=sda2 --dir=
lang en_US.UTF-8
langsupport --default=en_US.UTF-8 en_US.UTF-8
keyboard us
xconfig --card "Intel 915" --videoram 16384 --hsync 31.5-37.9 --vsync 50-70 --resolution 800x600 --depth 16
network --device eth0 --bootproto dhcp
rootpw --iscrypted $1$wjaV0A/I$qZ./GcXKO4w2UyGKx5ibr/
firewall --disabled
selinux --disabled
authconfig --enableshadow --enablemd5
timezone Europe/Berlin
bootloader --location=mbr --append="console=ttyS0,57600n8"
# The following is the partition information you requested
# Note that any partitions you deleted are not expressed
# here so unless you clear all partitions first, this is
# not guaranteed to work
clearpart --linux
part /boot --fstype ext3 --size=512
part / --fstype ext3 --size=1 --grow

reboot

%packages
@ web-server
@ mail-server
@ dns-server
@ server-cfg
@ dialup
@ admin-tools
@ printing
@ ftp-server
@ smb-server
kernel-smp
e2fsprogs
grub

%post

echo "/sbin/route add -net 10.255.255.1 netmask 255.255.255.255 dev eth0" >> /etc/rc.d/rc.local
echo "/sbin/route add default gw 10.255.255.1" >> /etc/rc.d/rc.local

sfdisk --change-id /dev/sda 2 82
sfdisk --change-id /dev/sdb 2 82

mkswap /dev/sda2
mkswap /dev/sdb2

echo "/dev/sda2       none            swap    sw" >> /etc/fstab
echo "/dev/sdb2       none            swap    sw" >> /etc/fstab

[fergie]

Got as far as installing and rebooting and then got this on the console:

Code: Select all

LILO 22.3.4 boot:
Loading lxserEBDA too big

Any ideas?

[scott]

Hmm, that lxserEBDA is one of the 1and1 kernels isn't it?

[fergie]

Think so:

Code: Select all

#more /etc/lilo.conf
boot=/dev/sda
root=/dev/md1

install=/boot/boot.b
vga=normal
timeout=36
prompt
lba32

read-only

default=lxser

serial=0,57600n8
append="console=ttyS0,57600 console=tty0 panic=30"

image=/boot/vmlinuz
        label=lxser
        append="console=tty0 console=ttyS0,57600 panic=30"

image=/boot/vmlinuz
        label=lx

image=/boot/vmlinuz.old
        label=old
        append="console=tty0 console=ttyS0,57600 panic=30"

Have done some newsgroup digging but not got an answer yet... Back to a virgin image if you want to have a go?

[scott]

Yeah, Im fighting with the new build tree right now so I wont be able to get to anything else until I figure it out.

[fergie]

Started again with the interactive install and set up the partitions and the RAIDs but got this error during formatting:

Code: Select all

Traceback (most recent call last):
  File "/usr/bin/anaconda", line 1181, in ?
    intf.run(id, dispatch, configFileData)
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/text.py", line 510, in run
    dispatch.gotoNext()
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/dispatch.py", line 171, in gotoNext
    self.moveStep()
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/dispatch.py", line 239, in moveStep
    rc = apply(func, self.bindArgs(args))
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/packages.py", line 743, in doPreInstall
    method.mergeFullHeaders(id.grpset.hdrlist)
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/harddrive.py", line 172, in mergeFullHeaders
    self.mountMedia(1)
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/harddrive.py", line 52, in mountMedia
    self.mountDirectory()
  File "/var/tmp/anaconda-10.1.1.46//usr/lib/anaconda/harddrive.py", line 90, in mountDirectory
    readOnly = 1);
  File "/usr/lib/anaconda/isys.py", line 189, in mount
    rc = _isys.mount(fstype, device, location, readOnly, bindMount, remount)
SystemError: (16, 'Device or resource busy')

> /usr/lib/anaconda/isys.py(189)mount()
-> rc = _isys.mount(fstype, device, location, readOnly, bindMount, remount)
(Pdb)

Think I might have tried to format the CD image... Also, not too sure I got the bootloader settings right. I was encouraged to use GRUB. Is there a way to use LILO?

[eWebtricity]

Any update on the 1and1 Athlon/Opteron install with the new raid setups?