weird disk_config file?
Hi, Attached is a weird disk_config file is my disk_config file for a certain class. For some reason I get these errors: Calling task_install Calling task_partition Partitioning local harddisks setup_harddisks version 0.42fai Probing disks: /dev/sda /dev/sdb Disks found: sda sdb sfdisk: ERROR: sector 0 does not have an msdos signature /dev/sdb: unrecognized partition table type No partitions found Using config file: /var/lib/fai/config/disk_config/ICTO_MYSQL Mapping disk name disk1 to sda ERROR: Mountpoints / extendedsda swap1 /var /tmp /usr /home do not fit. setup_harddisks version 0.42fai Probing disks: /dev/sda /dev/sdb Disks found: sda sdb I now wrote a whole new disk_config file and the installation works. However I can't see what's wrong with the attached diskconfig. Any ideas? TIA -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur ICT Department, Infrastructure office Groep SystemenSystems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- # generic disk configuration for one small disk # disk size from 500Mb up to what you can buy today # # type mountpoint size in mb [mount options] [;extra options] disk_config disk1 primary / 500-1 rw,errors=remount-ro ; -c -j ext3 logical swap 12288 rw logical /var 6- rw ; -m 5 -j ext3 logical /tmp 5000-1 rw ; -m 0 -j ext3 logical /usr 16000-3 rw ; -j ext3 logical /home 5000-1 rw,nosuid; -m 1 -j ext3
Re: locales settings
Tim Cutts wrote: On 6 Jun 2007, at 8:05 am, Rudy Gevaert wrote: Hi, Since I started installing etch I get this error (when something uses perl): perl: warning: Setting locale failed. perl: warning: Please check that your locale settings: LANGUAGE = (unset), LC_ALL = (unset), LANG = en_US.UTF-8 are supported and installed on your system. perl: warning: Falling back to the standard locale (C). locale: Cannot set LC_CTYPE to default locale: No such file or directory locale: Cannot set LC_MESSAGES to default locale: No such file or directory locale: Cannot set LC_ALL to default locale: No such file or directory I haven't got a thing in my fai config space saying a thing about locales. I can let the messages go away by installing the locales packages and setting that locale as installed. But that doesn't solve my question. Where is it set that the locale must be en_US.UTF-8 ? Set the locale you want in the FAI debconf settings. For example, we don't set a default locale here, so /srv/fai/config/debconf/FAIBASE contains: locales locales/default_environment_locale select None If I do debconf-get-selections I already get: locales locales/default_environment_locale select None (I haven't defined anything in my config, so it's the default) -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur ICT Department, Infrastructure office Groep SystemenSystems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: locales settings
Tim Cutts wrote: On 6 Jun 2007, at 10:59 am, Rudy Gevaert wrote: If I do debconf-get-selections I already get: locales locales/default_environment_locale select None (I haven't defined anything in my config, so it's the default) Is that in the installed system, or in the FAI NFS chroot? On the installed system. The system default locale is defined by the file /etc/default/locale. This is modified with the update-locale program, which is in turn called by the locales package postinst script. thrace:~# apt-cache policy locales locales: Installed: 2.3.6.ds1-13 Candidate: 2.3.6.ds1-13 Version table: *** 2.3.6.ds1-13 0 500 http://ftp.be.debian.org etch/main Packages 100 /var/lib/dpkg/status thrace:~# dpkg -L locales | grep default thrace:~# dpkg -L locales | grep etc thrace:~# I'm using 64bit arch though. -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur ICT Department, Infrastructure office Groep SystemenSystems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
sources.list per host
Hi, I there a way to have a sources.list during the installation (and when the system is installed) for some specific hosts. E.g. we have rolled our own multipath packages and I don't want to include that repository on each machine. I could use a script that appends that repository and then do an update; and install using that repository. But then I can't use the package_config/CLASS for adding software. An ideal way would be to be to say if class X is defined use sources.list.X instead of sources.list. Is there already away (or work arround) to achieve this? Thanks in advance, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: sources.list per host
Michael Tautschnig wrote: Would it suffice if you fcopy'd sources.list at the right time? We're doing that to some extend, using the hook updatebase.DEFAULT: Ah yes the hooks! I tried putting a cfengine script as a hook but fai.log says: Calling hook: updatebase.DEFAULT oeral: (CFINPUTS is set to nothing) oeral: Can't open file cfengine.conf updatebase.DEFAULT FAILED with exit code 1. I think I should set some variables, or is it not possible to use cfengine here? Thanks in advance, Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
Michael Tautschnig wrote: So it is telling me the same, but nevertheless continues to work; actually it is to be expected that there is not partition table on a new set of harddisks... For now I created the partions by doing cfdisk /dev/mapper/disk and then created a hook to create a filesystem on the partions. I'm a but overloaded with work so I'll look into this later on (I hope next week). Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: apt.conf.d/*
Holger Levsen wrote: Hi, On Thursday 15 June 2006 12:47, Michael Tautschnig wrote: I'm not sure if this apt.conf will be installed to the new system. This may be a bug if it's done. I think it is (or at least it was like this in 2.8.4 IIRC). So we should file a bug I just did that -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
xen and fai
Hi, I can install a xen virtual machine with FAI when I export my partitions to the virtual machine. I would like to export a disk to my virtual machine and install (and partition) it with FAI. When I boot with FAI it sees the disk as /dev/sda. But FAI reports /dev/sda: cannot get geometry If then do cfdisk /dev/sda I can partition that disk. Does anybody have any clues? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
Michael Tautschnig wrote: Hi, I can install a xen virtual machine with FAI when I export my partitions to the virtual machine. I would like to export a disk to my virtual machine and install (and partition) it with FAI. When I boot with FAI it sees the disk as /dev/sda. But FAI reports /dev/sda: cannot get geometry If then do cfdisk /dev/sda I can partition that disk. Does anybody have any clues? Is it possible at all to export a disk -- and partition it -- in a domain other than domain 0? I don't think so... In domain 0 I have my images (that are on a san). I export in domain 0 the following: disk=['phy:mapper/itext,sda,w'] Thanks for your reply! -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
Michael Tautschnig wrote: So - what? What kind of thing is mapper/itext - a partition, the whole SAN? mapper/itext is a LUN that is exported by our SAN. TIA -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
Michael Tautschnig wrote: Michael Tautschnig wrote: So - what? What kind of thing is mapper/itext - a partition, the whole SAN? mapper/itext is a LUN that is exported by our SAN. Thanks; do you have any pointers since when Xen should support partitioning from within the dom-U? AFAIK that simply isn't supported... We are using a recent xen Package: xen-hypervisor-3.0-amd64 Source: xen-3.0 I can manually use fdisk in my virtual domain to partition the disk. This the error I get in my fai.log: sfdisk: ERROR: sector 0 does not have an msdos signature /dev/sda: unrecognized partition table type No partitions found -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
64bit nfsroot
Hi, I collegue ask me why I set up a 64 bit nfsroot, next to mine 32 bit nfsroot. His point is that if you boot with a 64bit kernel, and with 32 bit userspace (the nfsroot) you can install a 64 bit system. And that I'm doing on my install server: 64bit kernel with 32 bit emulation, with a 64bit chroot where I create my 64bit fai nfsroot. Any comments on this? Cheers, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
apt.conf.d/*
Hi,
I can understand that FAI has to set:
// Options for apt-get for FAI
APT
{
// Options for apt-get
Get
{
Assume-Yes true;
Fix-Missing true;
Show-Upgraded true;
Purge true; // really purge! Also removes config files
List-Cleanup true;
ReInstall false;
AllowUnauthenticated true;
};
};
But when you then manually remove a package it purges, and this is not
the default way of doing things...
The above config should only be used when doing an installation (and
then the config file is in the nfsroot) and when doing a softupdate.
For doing the softupdate, apt can be called with a specifiek config file.
Or am I wrong?
I would like to here some comments about this...
Wkr,
--
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734
Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept.
Groep Systemen Systems group
Universiteit Gent Ghent University
Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: apt.conf.d/*
Thomas Lange wrote:
On Wed, 14 Jun 2006 11:32:50 +0200, Rudy Gevaert [EMAIL PROTECTED] said:
APT
{
// Options for apt-get
Get
{
Assume-Yes true;
Fix-Missing true;
Show-Upgraded true;
Purge true; // really purge! Also removes config
files
The above config should only be used when doing an installation (and
then the config file is in the nfsroot)
Yep, you're right.
and when doing a softupdate.
No, must not use this config during softupdate.
For doing the softupdate, apt can be called with a specifiek config file.
I'm not sure if this apt.conf will be installed to the new
system. This may be a bug if it's done.
I did a fresh install now and :
topolino:~# ls /etc/apt/apt.conf.d/*
/etc/apt/apt.conf.d/10fai /etc/apt/apt.conf.d/70debconf
/etc/apt/apt.conf.d/90fai
I grept trought my configspace and can't find any files like that.
Nor a script that copies them...
--
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734
Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept.
Groep Systemen Systems group
Universiteit Gent Ghent University
Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: Trouble Building NFSROOT
Toni Mueller wrote: I have all fai 2.10.1 packages that I can think of, and run from a local apt-proxy mirror. Anyway, building the NFSROOT fails like this: # debug=1 make-fai-nfsroot -v ... lots of stuff, then: Building Dependency Tree... E: Couldn't find package fai-nfsroot But of course the system is able to find that package: # apt-get -f install fai-nfsroot Reading Package Lists... Done Building Dependency Tree... Done The following NEW packages will be installed: fai-nfsroot 0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded. ... So, at least the package is there. What gives? It is in the nfsroot that it needs the fai-nfsroot (I think). Did you add the FAI repository to /etc/fai/sources.list? Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
boot parameters not set correctly ( bug in get-boot-info ?)
Hi, I think I found litte bug in the get-boot-info script. At the end of the script you will see: dmesg | grep -q Sending DHCP requests get_dhcp_info dmesg | grep -q Sending BOOTP requests get_bootp_info if [ $boot -eq 0 ]; then get_fixed_info fi I think it means that FAI checks which method was used when it got its nfsroot. In my case, when I have with ip=eth1:dhcp in my pxeconfig file for that host, it always executed the get_fixed_info subroutine. This is because dmesg contains the following string. Sending DHCP and RARP requests ., OK Grep can't find Sending DHCP requests ... When it executes the get_fixed_info subroutine my boot.log looks like this. topolino:/var/log/fai/current# cat boot.log netdevices_all=eth0 eth1 eth2 eth3 netdevices_up=eth1 netdevices=eth1 IPADDR=eth1 SERVER=dhcp NETMASK=dhcp GATEWAYS=dhcp BROADCAST=eth1.255 And thus it breaks the rest of the installation on several ways: - dns could be set up incorrectly (if resolv.conf-install-server is different than the resolv.conf for the install). - can't copy to the log host: it copies to the server 'dhcp'. - installation of software As a fix I changed get-boot-info like this: dmesg | grep -q Sending DHCPget_dhcp_info #dmesg | grep -q Sending BOOTP requests get_bootp_info But this breaks BOOTP. I'm booting my install with a 2.6.16.20 kernel. For AMD64. I'm trying to install sarge 64 bit. Is the above normal, or did I configure something wrong? (I created my 64 bit nfsroot in a chroot, and last week I had forgotten to copy the pxe kernel outside of the chroot to my tftpd server. So it is possible I've forgotten something else too...) Kind regards, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: boot parameters not set correctly ( bug in get-boot-info ?)
Hi Thomas, Thomas Lange wrote: On Mon, 12 Jun 2006 10:51:10 +0200, Rudy Gevaert [EMAIL PROTECTED] said: In my case, when I have with ip=eth1:dhcp in my pxeconfig file for that host, it always executed the get_fixed_info subroutine. This is because dmesg contains the following string. Sending DHCP and RARP requests ., OK I think you are not using the default fai-kernel. Your kernel seems to have the CONFIG_IP_PNP_RARP set, which is not enablerd in the fai-kernels. Yes it's my own kernel. I've recompiled without the option you said. CONFIG_IP_PNP=y CONFIG_IP_PNP_DHCP=y CONFIG_IP_PNP_BOOTP=y # CONFIG_IP_PNP_RARP is not set The error remains the same. I think that in later kernels the message has changed. I'm booting my install with a 2.6.16.20 kernel. For AMD64. I'm trying to install sarge 64 bit. Please try the default kernel from fai-kernels. Are there any drivers missing in this kernel? I can't seem to find any fai 64 bit kernel. IMO FAI use the generic amd64 bit kernel (or any other I define). For this specific I need mptsas. It's compiled as module in the generic kernel. But because my root file system is on those disks I'm force to use an own compiled kernel. Right? Btw in the i386 fai kernel (2.6.16) hasn't got CONFIG_MEGARAID_SAS set. Also I'm not really that keen on compiling modules for hard disk drivers, and then having trouble with initrd images... The same counts for network drivers. What do you think? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: boot parameters not set correctly ( bug in get-boot-info ?)
Rudy Gevaert wrote: The error remains the same. I think that in later kernels the message has changed. This is fixed. I again forgot to copy the image for pxe booting (again). I would still appreciate your opinion about my kernel problem. Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
fai and 64 bit
Hi, These are my first steps in 64bit land and have a problem with setting up FAI I've installed a 64 bit kernel on a 32 bit system (Dell PE1850). I then created a 64 bit chroot on the system. In the chroot I then compiled a 64 bit kernel. In the same chroot I have then created a fai nfsroot with with that kernel. Now when booting from that nfsroot it stops on the following error: request_module: runaway loop modprobe binfmt-464c I'm using a 2.6.16.20 kernel in that nfsroot. What can I do to solve this problem? Thanks in advance, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit
Michael Tautschnig wrote:
Hi,
These are my first steps in 64bit land and have a problem with setting
up FAI
I've installed a 64 bit kernel on a 32 bit system (Dell PE1850). I then
created a 64 bit chroot on the system. In the chroot I then compiled a
64 bit kernel.
In the same chroot I have then created a fai nfsroot with with that kernel.
Now when booting from that nfsroot it stops on the following error:
request_module: runaway loop modprobe binfmt-464c
I'm using a 2.6.16.20 kernel in that nfsroot.
What can I do to solve this problem?
What about this thread:
http://lists.debian.org/debian-amd64/2004/01/msg00129.html
The thread suggests:
- ... another module was necessary (System V IPC in my case) ...
- Kernel support for ELF binaries
- IA32 bit emulation
= are all compiled in my kernel
If I install that kernel on an other 64 bit system, it boots fine. Only
when it's installed in the FAI nfsroot it doesn't work.
I'm going to try this now
# function make() { /usr/bin/make ARCH=x86_64 HOSTCC=gcc -m32 $@; }
# export -f make
# export UNAME_MACHINE=x86_64
I'll post my results...
--
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734
Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept.
Groep Systemen Systems group
Universiteit Gent Ghent University
Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit
Rudy Gevaert wrote:
when it's installed in the FAI nfsroot it doesn't work.
I'm going to try this now
# function make() { /usr/bin/make ARCH=x86_64 HOSTCC=gcc -m32 $@; }
# export -f make
# export UNAME_MACHINE=x86_64
I'll post my results...
And that didn't work.
:(
--
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734
Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept.
Groep Systemen Systems group
Universiteit Gent Ghent University
Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit
Michael Tautschnig wrote: [...] What about this thread: http://lists.debian.org/debian-amd64/2004/01/msg00129.html The thread suggests: - ... another module was necessary (System V IPC in my case) ... - Kernel support for ELF binaries - IA32 bit emulation = are all compiled in my kernel If I install that kernel on an other 64 bit system, it boots fine. Only when it's installed in the FAI nfsroot it doesn't work. In some other threads I found - incorrect binformat. (i.e. you try to load a module which was compiled for a different architecture). (http://www.busybox.net/lists/busybox/2005-October/016713.html) - what about your modprobe format, i.e. what does file modprobe report in your nfsroot vs. file modprobe on the native 64bit system? Do you really have a 64bit nfsroot or are you using your 32bit nfsroot? This is in the FAI nfsroot: (chroot /ugdisk/pure64 and then chroot /fai/nfsroot-amd64) wopr:/# file /sbin/modprobe /sbin/modprobe: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for GNU/Linux 2.6.0, dynamically linked (uses shared libs), stripped On the native machine 64bit machine: /sbin/modprobe: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for GNU/Linux 2.6.0, dynamically linked (uses shared libs), stripped I'm using the 64 bit nfsroot PS.: What about an older kernel? I'm going to try it. -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit
Michael Tautschnig wrote: [...] I'm using the 64 bit nfsroot PS.: What about an older kernel? I'm going to try it. Just another thought: Have you cross-compiled (at least 32/64) your kernel or has it been compiled on a native 64bit system? If it has been cross-compiled, what about building the same kernel(-configuration) on a native system? I've only compiled my kernels in the 64 bit chroot on the 32 bit userspace running a 64bit kernel. I'll try to compile on the native 64 bit platform later this afternoon thanks for the help! -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit
Michael Tautschnig wrote: [...] I'm using the 64 bit nfsroot PS.: What about an older kernel? I'm going to try it. Just another thought: Have you cross-compiled (at least 32/64) your kernel or has it been compiled on a native 64bit system? If it has been cross-compiled, what about building the same kernel(-configuration) on a native system? I build the kernel now on the 64 bit system. And copied it to the 64bit chroot. And then did a fai setup. Booting that nfsroot then didn't work. It gives the same error. I also ran fai-setup on the 64 bit system and then copied the nfsroot to my fai install server. When I then booted it also gave me the same error. I'm now going to try a different kernel. -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: fai and 64 bit (solved partially)
A co-worker pointed out that that I didn't update the kernel that pxe boots. And indeed, pxe still booted with a 32 bit kernel. I have booted now into FAI. I have some problems left, but that's for tomorrow. I'm going to write this up and put it on the wiki, tomorrow or next week. Thanks for all the help! Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
make nfsroot in 64bit chroot
Hi, I'm trying to create a fai nfsroot in a 64bit chroot on a system with a 64 bit kernel with 32bit userland. These are the steps I already took: - install 32bit sarge - install 64bit kernel - create chroot with # cdebootstrap --arch=amd64 sarge /ugdisk/pure64 \ http://ftp.belnet.be/debian-amd64/debian/ # cp /etc/hostname /etc/resolv.conf /etc/network/interfaces \ /ugdisk/pure64/etc/ # chroot /ugdisk/pure64/ - set up fai: # apt-get install fai-server - configure fai... - call fai-setup: wopr:/# fai-setup Account $LOGUSER=fai already exists. Make sure that all install clients can log into this account without a password. Using interface eth1 to determine local hostname. Warning: cannot open /proc/net/dev (No such file or directory). Limited output. and then it hangs... Running make-fai-nfsroot works fine... Is this normal? Thanks in advance, Rudy (I haven't tested the nfsroot because I'm having some kernel problems that I have to fix first) -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: FAI and RAID ?
Paul Lussier wrote: What's the current status of using SW RAID in FAI? Has it been added to FAI proper, or are the directions at: http://faiwiki.informatik.uni-koeln.de/index.php/User:Svamberg/Software_RAID still the way to go? If I recall correctly it hasn't been added to upstream. I thought there was some discussion about this some months ago. Have a look at the archives. Kind regards, Rudy
remove package exim4 from FAIBASE
Hi, I suggest to remove the package exim4 from the FAIBASE package_config file. When adding postfix to an other class (and also having FAIBASE defined) the install fails to install a lot of packages. E.g. grub. So when rebooting it fails to boot :) Could this be considered? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: remove package exim4 from FAIBASE
Thomas Lange wrote: On Fri, 02 Jun 2006 10:00:25 +0200, Rudy Gevaert [EMAIL PROTECTED] said: I suggest to remove the package exim4 from the FAIBASE package_config file. When adding postfix to an other class (and also having FAIBASE defined) the install fails to install a lot of packages. E.g. grub. Just add the line exim4- to your class. Then it should not be installed, even if your client belongs to the class FAIBASE. Yep it worked :) (I tried putting exim- before but ofcourse that didn't work) -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: network interfaces exchanged
Andreas Sindermann wrote: Hi, during the fai-installation the nfsroot-kernel sees the two network interfaces eth0 and eth1 in a different order than the kernel used during normal operation. The effect is that when booting from harddisk no network is available. How does one care about this usually? I think you can change that with changing the order of your loaded modules. Or you can install a correct /etc/network/interfaces that sends all your internet traffic to the correct interface. Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: network interfaces exchanged
Thomas Lange wrote: On Fri, 19 May 2006 10:49:11 +0200 (MEST), Andreas Sindermann [EMAIL PROTECTED] said: Hi, during the fai-installation the nfsroot-kernel sees the two network interfaces eth0 and eth1 in a different order than the kernel used during normal operation. The effect is that when booting from harddisk no network is available. How does one care about this usually? This is a known issue. I'm not sure if there's a proper fix. If you have different types of NIC you can add eth1 to the ip parameter in pxelinux.cfg. Read nfsroot.txt in the kernel source for more information or have a look at /etc/fai/menu.lst where we use this parameter ip=client-ip:server-ip:gw-ip:netmask:hostname:device:autoconf If you want I can provide a little patch that lets you give the -1 option to boot the installation over th1 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert [EMAIL PROTECTED] tel:+32 9 264 4734 Directie ICT, afd. Infrastructuur Direction ICT, Infrastructure dept. Groep Systemen Systems group Universiteit Gent Ghent University Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: make-fai-bootfloppy
Thomas Lange wrote: On Thu, 30 Mar 2006 16:47:34 +0200, Rudy Gevaert [EMAIL PROTECTED] said: Is there an other way to create the fai-boot-floppy without having a loopback device? My install server has a xen virtual machine and the loopback device isn't in the kernel ( :( ), rebuilding the kernel would be a pain so I would like to avoid it. If you really use a (real) floppy disk, this script does not need the loopback mount. Hi, The server is a virtual (xen) server. It doesn't have any floppy. I now have recompiled kernel with loopback device. But: xen4:~# make-fai-bootfloppy -f /root/fai_floppy.img 1440+0 records in 1440+0 records out 1474560 bytes transferred in 0.006242 seconds (236230513 bytes/sec) Creating an ext2 filesystem on floppy device /root/fai_floppy.img. Creating the boot floppy with grub. cp: writing `/media/floppy/vmlinuz-2.6.14.2-ugent-fai': No space left on device Is this because of the tmpfs that is to small? xen4:~# df Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda1 2015824 1855096140248 93% / tmpfs30284 4 30280 1% /dev/shm baviaan.private.ugent.be:/home/debiankernels 407387840 66537640 336711376 17% /debiankernels xen4:~# mount /dev/sda1 on / type ext3 (rw) proc on /proc type proc (rw) sysfs on /sys type sysfs (rw) devpts on /dev/pts type devpts (rw,gid=5,mode=620) tmpfs on /dev/shm type tmpfs (rw) baviaan.private.ugent.be:/home/debiankernels on /debiankernels type nfs (rw,intr,addr=10.1.39.27) Thanks in advance, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent / Ghent Universityfax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: make-fai-bootfloppy
Henning Glawe wrote: On Fri, Mar 31, 2006 at 05:19:49PM +0200, Rudy Gevaert wrote: 1440+0 records in 1440+0 records out 1474560 bytes transferred in 0.006242 seconds (236230513 bytes/sec) Creating an ext2 filesystem on floppy device /root/fai_floppy.img. Creating the boot floppy with grub. cp: writing `/media/floppy/vmlinuz-2.6.14.2-ugent-fai': No space left on device Is this because of the tmpfs that is to small? sounds more like your kernel is too big for a 1.44MB floppy. Damn, it is: xen4:~# ls -lh /boot/fai/vmlinuz-2.6.14.2-ugent-fai -rw-r--r-- 1 root root 1.6M Nov 24 23:09 /boot/fai/vmlinuz-2.6.14.2-ugent-fai I already tried to add an -F fai.iso option, but it still complains because dd creates a too small image. Can I set this to a bigger value? (And only use a cd to boot the machine) Thanks in advance, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent / Ghent Universityfax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
make-fai-bootfloppy
Hi, Is there an other way to create the fai-boot-floppy without having a loopback device? My install server has a xen virtual machine and the loopback device isn't in the kernel ( :( ), rebuilding the kernel would be a pain so I would like to avoid it. Thanks in advance, Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent / Ghent Universityfax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: debianmirror
Micha Beyer wrote: Hallo Rudy, at sunday 12 march 2006 21:48 you wrote: must I have a local debianmirror for fai-setup, or is the line in /etc/fai/fai.conf mirrorhost=ftp.debian.org enough, Any mirror is fine. Although it's better to a mirror close to you. Okay, I want using my local mirror, the repository with apt-proxy. while the problem is the package fai-nfsroot? I can't installed these packet. I use here a Debian testing. I think there isn't any fai-nfsroot package. But this: [EMAIL PROTECTED]:~$ apt-cache show fai-nfsroot Package: fai-nfsroot Priority: extra Section: admin Installed-Size: 152 Maintainer: Thomas Lange [EMAIL PROTECTED] Architecture: all Source: fai Version: 2.9.1 Depends: fai-client Filename: pool/main/f/fai/fai-nfsroot_2.9.1_all.deb Size: 30238 MD5sum: 5ce347fd06b8b6f47a4397c525300638 Description: Fully Automatic Installation nfsroot package FAI is a tool for doing Fully Automatic Installation of Debian(-based) Linux systems. This package contains the software that is only needed in the FAI nfsroot. Do not install this package on a normal machine. You do have to run make-fai-nfsroot. Yes of course, the message is The packet fai-nfsroot is not available. That's my problem. I'm doing a wild guess here, but it looks like you haven't installed the fai-nfsroot on the install server. Look for it in: http://www.informatik.uni-koeln.de/fai/download/ or add the correct apt repository to your sources.list on the install server. (This is explained in the manual.) (I'm not using the latest FAI package, so I could be wrong.) Which version and distribution you are using? I'm using 2.8.4. This was the last version where FAI was in one package. (Not counting the install kernel package). Kind regards, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent / Ghent Universityfax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: debianmirror
Micha Beyer wrote: Hallo there, must I have a local debianmirror for fai-setup, or is the line in /etc/fai/fai.conf mirrorhost=ftp.debian.org enough, Any mirror is fine. Although it's better to a mirror close to you. while the problem is the package fai-nfsroot? I can't installed these packet. I use here a Debian testing. I think there isn't any fai-nfsroot package. You do have to run make-fai-nfsroot. (I'm not using the latest FAI package, so I could be wrong.) Rudy
LVM
Hello, I'm trying to write a script that sets up LVM. First I partition a partition the normal way, and mount it onto /lvm. In my script I then umount /lvm. Then I initialize it for LVM `$rootcmd pvcreate -f /dev/hda6`; Then I create my volume group: `$rootcmd vgcreate ldisk /dev/hda6`; So far, so good: Then I just need to make my logical volumes: `$rootcmd lvcreate -L2000 -nlvol1 ldisk`; But this doesn't work: /proc/misc: No entry for device-mapper found Is device-mapper driver missing from kernel? Failure to communicate with kernel device-mapper driver. /proc/misc: No entry for device-mapper found Is device-mapper driver missing from kernel? Failure to communicate with kernel device-mapper driver. Incompatible libdevmapper 1.01.00-ioctl (2005-01-17)(compat) and kernel driver striped: Required device-mapper target(s) not detected in your kernel lvcreate: Create a logical volume Does anyone have any ideas? Thanks in advance, Rudy -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent fax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
Hello I'm still having troubles with FAI and XEN. I'm trying to use FAI to install a virtual machine. I made a virtual domain config file and filled in the nfsroot parameters. I also made a xenU kernel and included in the FAI nfsroot. When I create a domain it stops on the error that it can't find a hard disk. However when I boot with the option sysinfo I do see in the end that it finds sda1. But with the install it doesn't. I would appreciate some tips or help :) Also the Xen documentation says in the virtual kernel you should disable loadable module support. Does this have any consequences on FAI. Thanks in advance, -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Rudy Gevaert e-mail: [EMAIL PROTECTED] Directie ICT, Afdeling Infrastructuur Groep Systemen tel: +32 9 264 4734 Universiteit Gent fax: +32 9 264 4994 Krijgslaan 281, gebouw S9, 9000 Gent, Belgie www.UGent.be -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Re: xen and fai
On Tue, Aug 16, 2005 at 05:22:33PM +0200, Henning Glawe wrote: this looks like something may be missing in the dhcp configuration... could you try to boot the client with a static IP configuration on the kernel command line? If it doesn't work: how is you xen bridge set up? the name looks quite 'custom' to me... you can try to boot the minimal xen ttylinux and checking the network connectivity from there... In deed, my network setup for the domain was wrong. Thanks Rudy
xen and fai
Hello, I saw your email on the FAI mailinglist regarding Xen and FAI. I'm trying to use FAI to install a Xen virtual machine. I already have a working FAI install server. This is my xen domain config file: kernel=/home/xen/domU/vmlinuz-2.6.10-xenU builder='linux' memory = 128 name = zalm nics=1 vif = [ 'mac=aa:00:00:77:40:36, bridge=subnet10'] disk=['file:/home/xen/zalm-root,sda1,w'] dhcp=dhcp root = /dev/nfs nfs_server = '10.1.39.84' nfs_root = '/usr/lib/fai/nfsroot' When creating the Xen domain: ... IP-Config: Incomplete network configuration information. Looking up port of RPC 13/2 on 10.1.39.84 RPC: sendmsg returned error 101 portmap: RPC call returned error 101 Root-NFS: Unable to get nfsd port number from server, using default Looking up port of RPC 15/1 on 10.1.39.84 RPC: sendmsg returned error 101 portmap: RPC call returned error 101 Root-NFS: Unable to get mountd port number from server, using default RPC: sendmsg returned error 101 mount: RPC call returned error 101 Root-NFS: Server returned error -101 while mounting /usr/lib/fai/nfsroot VFS: Unable to mount root fs via NFS, trying floppy. VFS: Cannot open root device nfs or unknown-block(2,0) Please append a correct root= boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(2,0) Any idea's? Or should I go to the Xen mailinglists? Thanks in advance, Rudy
implementation of classes inclear
Hello, The following is not clear to me: The documentation says that if you want to use DHCPC you have to define the NETWORK class too. Ok... In classes_description.txt it says that the network class has some information in the scripts directory. I grep trough my fai config space and have no such files, or files installed mentioning that class The same goes for the SERVER. However, the examples provided with FAI have files relevant for the SERVER and NETWORK class. So could it just be that classes_descriptions just says: we have made several classes for you, go and look in the example directory and copy those files we mention here?. I removed the NETWORK class form my host config file and my installation system continues to work. Summary: it is unclear which classes are provided by FAI by default. (except LAST and DEFAULT). And where can you see what those predefined classes do. Thank you in advance, Rudy
postfix exim implies grub error
Hello, I have a working FAI setup for a host (for the moment the only host) called 'kip'. in class/kip I have: COMPILE NONIS SERVER NTP MBR NETWORK UGENT In package_config/UGENT I have: PACKAGES aptitude netbackup make PACKAGES aptitude sysstat bigsister-agent libunix-syslog-perl The file class/50-host-classes appends FAIBASE DHCPC to all default clients. Working with this setup, my client gets installed perfectly. Now *adding* to package_config/UGENT PACKAGES remove exim4 PACKAGES aptitude postfix (FAIBASE installs exim4, but I want postfix) gives the following error after install: fai.log:ERROR: 65280 65280 shell.log:chroot: cannot run command `/sbin/update-grub': No such file or directory software.log:ERROR: 65280 65280 And when rebooting I get the grub prompt. Any ideas what is going wrong? Thanks in advance, Rudy PS thanks to the people who replied to my previous post (the post is held in a queue because the from was wrong)
netboot strategie
Hello, I'm using netboot to install a system using FAI. Now what are the best strategies to reboot the system. I use fai-chboot -IBv hostname to install a client. This works. Then the install system disables that entrie, so on the next boot the client boots from the localdist. But the documentation says that the first bootdevice should be the NIC, and the second the hard disk. Thus when my client reboots after install it complains about not begin able to find the install kernel. A way around the problem is set netboot to boot from localhost. Am I missing anything here? Or is this the way to do it? Thank you in advance, Rudy
cfengine
Hello, After installing a computer with FAI I see that cfengine is also installed. I want now to use cfengine. I gather I can copy over the cfengine config files and everything will work? Or are there some build in functions in FAI? Thanks in advance, Rudy
Re: cfengine
Hi, On Saturday 05 March 2005 23:36, you wrote: Hello, After installing a computer with FAI I see that cfengine is also installed. I want now to use cfengine. I gather I can copy over the cfengine config files and everything will work? Or are there some build in functions in FAI? you can look at the script file /usr/share/doc/fai/examples/simple/scripts/DEFAULT/S20 for example. But, if I'm correct these scripts only get executed when installed. They aren't run periodically, right? And, sure, the cfengine documentation which is quite verbose, be sure not to get lost :) Thanks, although it's hard not to get lost ;) Rudy
kernel compile problem
Hello I'm trying to setup afs and am having problems with compiling the afs modules and a new kernel. Just following the afs debian documentation, making the afs modules gives a unresolved symbol when installing the afs modules. So I'm trying to do it an other way: just compile a new kernel: make oldconfig make-kpkg --append-to-version -fai --rev rev1 kernel_image make-kpkg --append-to-version -fai --rev rev1 modules_image And then installing antftp:/usr/src# dpkg -i *2.4.29*deb Selecting previously deselected package kernel-image-2.4.29-fai. (Reading database ... 16960 files and directories currently installed.) Unpacking kernel-image-2.4.29-fai (from kernel-image-2.4.29-fai_rev1_i386.deb) ... Selecting previously deselected package openafs-modules-2.4.29-fai. Unpacking openafs-modules-2.4.29-fai (from openafs-modules-2.4.29-fai_1.2.13-1 +rev1_i386.deb) ... Setting up kernel-image-2.4.29-fai (rev1) ... depmod: *** Unresolved symbols in /lib/modules/2.4.29-fai/kernel/drivers/ide/ ide-core.o Setting up openafs-modules-2.4.29-fai (1.2.13-1+rev1) ... Gives a unresolved symbol in ide-core.o What could the problem be? Thanks in advance, Rudy
Re: fcopy gives permission denied
On Tuesday 15 February 2005 17:12, Holger Levsen wrote: if the file in /fai/files also has permissions of root,root,0600 then it's not readable when exported (try it on the client). You have to loose the permissions which might introduce a security breach. Yes this works. Thanks
fcopy gives permission denied
Hello, I have the following script: gevaert:/home/fai/antldap/last# cat /ed/fai_configdir/scripts/LDAPS #!/bin/sh fcopy -ir -m root,root,600 /etc/ldap/slapd.conf the shell.log reports: shell.log:fcopy: copy /fai/files/etc/ldap/slapd.conf/LDAPS to /tmp/target/etc/ldap/slapd.conf failed. Permission denied Why can't I copy that file? I don't see any problem in the original file have permissions 600. And then being replace with a newer file. Thanks in advance, Rudy
Re: kerberos
Hello Ezra, On Monday 17 January 2005 19:50, Ezra Kissel wrote: I found this HOWTO helpful: http://www.cryptnet.net/fdp/admin/kerby-infra/en/kerby-infra.html On Debian systems, you'll want the krb5-admin-server and krb5-kdc packages. Debconf should help you setup an initial realm, but just start reading man pages on all of the krb5 and related k commands. Manage principals locally with kadmin.local You'll want to get the kadmin server configured correctly to perform admin tasks remotely. I guess I wasn't clear enough. I would like to set up a fully functioning KDC with users in it. I now found that it is possible to add password on the command line when adding principals. It would be then possible to add a script that inserts the principals. This would be however not secure. An other way, and I think a better way, would be to make dumps om the KDC database, and then load them. Rudy
kerberos
Hello, Does anybody have any pointers on setting up kerberos. I want to set up a KDC during installation. The installtion is easy I think. But how does one do the configuration? E.g. settup up the principals? Making the new realm? Thanks in advance, -- Rudy Gevaert[EMAIL PROTECTED] Web pagehttp://www.webworm.org GNU/Linux user and Savannah hacker http://savannah.gnu.org
Re: error in DEFAULT/S01
On Thursday 13 January 2005 13:05, you wrote: IIRC some of the simple example scripts have a last command like ifclass DEMO do_something() So if DEMO is not defined, the scripts end with an error. It would be nice if you could put an echo -n or something like this as the last line of these scripts... Yes you are correct, when removing that (DEMO isn't defined here) line the script doesn't end with an error. However it still reports: rcS.log:Rename for 0A14 failed.No such file or directory -- Rudy Gevaert[EMAIL PROTECTED] Web pagehttp://www.webworm.org GNU/Linux user and Savannah hacker http://savannah.gnu.org
error in DEFAULT/S01
Hi My installations always give me the following: rcS.log:DEFAULT/S01 FAILED with exit code 1. rcS.log:Rename for 0A14 failed.No such file or directory shell.log:DEFAULT/S01 FAILED with exit code 1. I can't see that is wrong in the scripts/DEFAULT/S01 file. Any ideas? Thanks in advance, Rudy
Re: nis nfs netgroup problem
On Tue, Nov 30, 2004 at 01:52:58PM +0100, Frank Lenaerts wrote: Could you try another test? Export /home to testbak.gevaert and try to mount that. That works without a problem... braindumpgateway gets a mount request, gets the netgroup via nis, gets testbak and gateway for the testbakken group, tries to verify if testbak corresponds to the IP address it gets the mount request from (maybe hostname resolving is not working correctly)/braindump I can resolve testbak.gevaert on the server, and can resolve gateway.gevaert on testbak. Thanks in advance, Rudy
Re: nis nfs netgroup problem
Hi Frank, I changed my setup this week, so I could exclude a gateway/firewall problem. Now my NIS server is running on the machine testbak.gevaert The NFS server is running ont he machine gateway.gevaert On Mon, Nov 29, 2004 at 09:46:49AM +0100, Frank Lenaerts wrote: What's the output of a showmount -e gateway.gevaert? testbak:/var/yp# showmount -e gateway.gevaert Export list for gateway.gevaert: /home @testbakken /ed/nfsroot 10.0.0.0/255.255.255.0 /ed/debmirror 10.0.0.0/255.255.255.0 /ed/fai_configdir 10.0.0.0/255.255.255.0 Looks ok but I've never seen fully qualified hostnames in NIS context. AFAIK, NIS is flat i.e. does not use scoping as in DNS. Could you replace testbakken (testbak.gevaert,,) with testbakken (testbak,,) and retry it? Ok, I did so: (on the nis server:) testbak:~# cat /etc/netgroup lappie (duvel,,) testbak (testbak,,) gateway (gateway,,) ant01 (ant01,,) ant02 (ant02,,) ant03 (ant03,,) testbakken gateway testbak ants ant01 ant02 ant03 faiclients ants testbakken gateway (on the gateway:) gevaert:~# cat /etc/nsswitch.conf passwd: nis compat group: nis compat shadow: nis compat hosts: files dns networks: files protocols: db files services: db files ethers: db files rpc:db files netgroup: nis (on the nis server:) testbak:~# cat /etc/nsswitch.conf passwd: compat group: compat shadow: compat hosts: files dns networks: files protocols: db files services: db files ethers: db files rpc:db files netgroup: nis Using the program netgroup (out of package ng-utils): testbak:~# netgroup -h testbakken testbak gateway gevaert:~# netgroup -h testbakken testbak gateway (Export file on the NFS server (gateway.gevaert)) gevaert:~# cat /etc/exports /ed/debmirror 10.0.0.0/255.255.255.0(ro,sync) /ed/fai_configdir 10.0.0.0/255.255.255.0(async,ro) /ed/nfsroot 10.0.0.0/255.255.255.0(async,ro,no_root_squash) /home @testbakken(rw,sync) #/home 10.0.0.0/255.255.255.0(async,rw) This is the same showmount outputs (see before) Running: testbak:~# mount gateway.gevaert:/home /mnt/ mount: gateway.gevaert:/home failed, reason given by server: Permission denied Gives in the NFS server's syslog: Nov 29 15:28:01 gevaert rpc.mountd: refused mount request from 10.0.0.250 for /home (/): no export entry I'm puzzled... Thanks in advance! Echt waar Rudy
Re: nsswitch.conf not copied
On Sun, Nov 28, 2004 at 10:00:58AM +1100, Nicolas Triantafillou wrote: during the install, do you see NIS defined in the list of classes? Yes I have a script in my scripts/ directory with: That did the trick! Thanks Rudy
nis nfs netgroup problem
Hello, I'm trying to set up NIS with NFS. I've posted this mail to the debian user list, but haven't received any response to it :( That's why I'm posting it here. And I need NIS with NFS for fai to work like intented. The server running NIS and NFS is a nat box. I'm using the package ipmasq for the masquerading. Only boxen behind the nat box are using NIS and NFS. NFS works when I export a directory on the server like this: /home 10.0.0.0/255.255.255.0(rw,sync) However when I want to use a netgroup, it doesn't work: /home @testbakken(async,rw,no_root_squash) testbak:~# mount gateway.gevaert:/home /home mount: gateway.gevaert:/home failed, reason given by server: Permission denied syslog reports: Nov 27 10:52:16 gevaert rpc.mountd: refused mount request from 10.0.0.250 for /home (/): no export entry Nov 27 10:52:27 gevaert ypserv[27491]: refused connect from 157.193.206.67:606 to procedure ypproc_domain (gevaert,;0) Nov 27 10:52:47 gevaert ypserv[27491]: refused connect from 157.193.206.67:606 to procedure ypproc_domain (gevaert,;0) 157.193.206.67 is the external ip from the natbox. I don't think the last two lines have anything to do with the mount request because they show up when I am not working on the machine too. This is the entry in /etc/exports /home @testbakken(async,rw,no_root_squash) and this is the contents of /etc/netgroup lappie (duvel.gevaert,,) testbakken (testbak.gevaert,,) ant01 (ant01.gevaert,,) ant02 (ant02.gevaert,,) ant03 (ant03.gevaert,,) ants ant01 ant02 ant03 faiclients ants testbakken When I run netgroup on testbak machine testbak:~# netgroup -h testbakken testbak.gevaert I see that nis and netgroups are working. What could be the problem? NIS is setup correctly because I can login onto the a NIS client machine (testbak) when using NIS. Thanks in advance, Rudy
nsswitch.conf not copied
Hi, I defined the NIS, and put the following in /ed/fai_configdir/files/etc/nsswitch.conf/NIS: passwd: nis files group: nis files shadow: nis files hosts: files dns networks: files protocols: db files services: db files ethers: db files rpc:db files netgroup: nis However, these files don't get copied when the install is done. What could it be? Thanks in advance, Rudy
Re: questions about install
Hello, First thank you for the reply. On Mon, Nov 22, 2004 at 11:09:55PM +0100, Henning Sprang wrote: otherwise it didn't install. Is this right? Or do I have to define it somewhere else? In the class directory? There are several ways to configure the FAI_ACTION. The above is one. Another one is to do it in the class directory, the FAI examples show how this is set in LAST.var, based on hostnames, but i think you could Yes that did it. this is the way to define classes for singles host based on their hostname, see 01alias.var for examples of other ways to define classes based on parts of hostname. Yep, I saw, read and used it. Also I've found much more files in /usr/local/share/fai/* , are these now used? Can I move them to my FAI_CONFIGDIR ? If actions you've taken in /ed/fai_configdir lead to visible results, i don't think /usr/local/share/fai is used, but i didn't play around with this so i am not sure. I did some looking around and it seems that /usr/local/share/fai/ was wrom a previous attempt to use fai. Anyway I reread the manual twice and it has become clearer to me now. Thanks for you reply! Rudy
questions about install
Hi, After some troubles I finaly managed to get an unattended install running. I can't get netgroups running, but I got arround that for now. I do have two questions: I made a bootfloppy with: make-fai-bootfloppy -vF FAI_ACTION=install otherwise it didn't install. Is this right? Or do I have to define it somewhere else? In the class directory? Also I put FAI_CONFIGDIR=/ed/fai_configdir , and copied the files in /usr/share/doc/fai/examples/simple/* to that directory. I then renamed the file demo to ant01 (my test machine). It then installed. Do I have to make such a file for each machine? I don't think so. Also I've found much more files in /usr/local/share/fai/* , are these now used? Can I move them to my FAI_CONFIGDIR ? Thanks in advance, Rudy
problem with setting up mirror
Hello I upgraded my system to unstable to be able to use the latest FAI. And I have problems with installing a mirror of unstable. gevaert:/ed/debmirror# ./mkdebmirror -- create mirror for debian -- -- create mirror for debian-non-US -- Won't mirror without dists/unstable/non-US/main/binary-i386/Packages.gz signature in Release at /usr/bin/debmirror line 1094. releasing 1 pending lock... at /usr/lib/perl5/LockFile/Simple.pm line 182. -- create mirror for debian -- -- create mirror for debian-non-US -- Won't mirror without dists/unstable/non-US/main/binary-i386/Packages.gz signature in Release at /usr/bin/debmirror line 1094. releasing 1 pending lock... at /usr/lib/perl5/LockFile/Simple.pm line 182. -- create mirror for debian-security -- Won't mirror without dists/sarge/updates/main/debian-installer/binary-i386/Packages.gz signature in Release at /usr/bin/debmirror line 1094. releasing 1 pending lock... at /usr/lib/perl5/LockFile/Simple.pm line 182. receiving file list ... done client: nothing to do: perhaps you need to specify some filenames or the --recursive option? rsync error: some files could not be transferred (code 23) at main.c(723) This is what I put in mkdebmirror debug=$@ arch=i386 dist=unstable destdir=/ed/debmirror # first sync from a mirror near to you update_from ftp.uni-koeln.de # since this mirror is not always up to date, sync the rest from an official mirror update_from ftp.de.debian.org # even one should not mirror the security site, but I will do it. echo -- create mirror for debian-security -- debmirror $destdir/debian-security $allopt --host=security.debian.org -r debian-security -d sarge/updates,woody/updates # get installation disk images via rsync rsync_disks unstable ftp2.de.debian.org I installed debmirror from unstable. Thanks in advance, Rudy
Re: make nfsroot
Hi Henning, On Thu, Oct 14, 2004 at 07:57:03PM +0200, Henning Glawe wrote: On Thu, Oct 14, 2004 at 05:53:05PM +0200, Rudy Gevaert wrote: I'm trying to set up fai. I'm using debian stable and would like to set up other stable systems with fai. [...] E: Couldn't find package module-init-tools What could the problem be? what FAI version are you using ? fai_2.6.3 I use the following deb sources: #FAI deb http://www.informatik.uni-koeln.de/fai/ download/ it sounds like you are trying to produce an nfsroot for sarge or sid: module-init-tools is needed for 2.6 kernels and only available there FAI_DEBOOTSTRAP=woody file:/ed/debmirror/debian maybe try to remove the offending package(s) from $packages in /etc/fai/make-fai-nfsroot.conf or use the latest woody-targetted version of FAI (2.6 and newer are targeted at sarge). Ah, I see. I'll remove my deb source above and use the standard woody package. Thanks, Rudy
make nfsroot
Hello, I'm trying to set up fai. I'm using debian stable and would like to set up other stable systems with fai. I used the mkdebmirror script and after running fai-setup I got the following error: ... Adding additional packages to /ed/nfsroot: module-init-tools dhcp3-client ssh file rdate hwinfo portmap bootpc rsync wget rsh-client less dump reiserfsprogs usbutils dpkg-dev ext2resize hdparm smartmontools parted raidtools2 lvm2 dnsutils ntpdate dosfstools cfengine cvs jove xfsprogs xfsdump sysutils dialog discover mdetect libnet-perl netcat libapt-pkg-perl grub lilo read-edid kudzu hwtools dmidecode E: Couldn't find package module-init-tools What could the problem be? Thanks in advance, Rudy
