September 2006 - openSUSE Autoinstall - openSUSE Mailing Lists

linuxrc - netdevice=<hwaddress>
by John Willis 18 Dec '12

18 Dec '12

Does anyone know if any of the SLES releases 9 or 10 support passing the linuxrc option netdevice=<hardware mac address> ? I noticed Red Hat has had this for sometime and it seems totally missing in SuSE which seems really odd. I'm still scratching my head if we really are still passing only by kernel interface name. I read in the change log for linuxrc-1.8.6-3 a change that "sounds" like this feature. * Mon Mar 14 2005 - snwint(a)suse.de - s390: fix segfault in net_list_s390_devs() - extra root image was sometimes not used - allow netdevices to be specified by hwaddress But it looks like linuxrc-1.8.6-3 is only available in SuSE Professional 9.3. "any" commentary would be sincerely appreciated the feature also seems to be fairly minor since any delimiter seperating the octets in the hwaddr string could be used to identify the difference between an interface name and interface hw address if i could figure out a generic method of overriding the linuxrc with a custom script and daisy chain it back into linuxrc after setting up the interface name, I would.. and have a slight idea on how to do that.. but i'd rather find out i'm just missing something and the feature is already there Thanks!

3 2

RE: [suse-autoinstall] SLES 10 RC2 remove-packages section?
by Matt Gillard 05 Mar '08

05 Mar '08

I dont have kinternet installed. This is what doesnt seem to work: <software> <patterns config:type="list"> <pattern>default</pattern> <pattern>Basis-Devel</pattern> <pattern>x11</pattern> </patterns> <packages config:type="list"> <packages>perl-libwww-perl</packages> <packages>net-snmp</packages> <packages>kernel-source</packages> </packages> <remove-packages config:type="list"> <packages>pptp</packages> <packages>smpppd</packages> <packages>ppp</packages> <packages>wvdial</packages> <packages>mutt</packages> </remove-packages> </software> In y2log it mentions that those packages should be removed, but it goes ahead and installs them anyway. I can do it manually on the command line: # rpm -e pptp smpppd ppp wvdial mutt and that works fine so I know I have no dependency problems. matt > -----Original Message----- > From: Uwe Gansert [mailto:ug@suse.de] > Sent: Friday, 30 June 2006 12:50 AM > To: suse-autoinstall(a)suse.com > Subject: Re: [suse-autoinstall] SLES 10 RC2 remove-packages section? > > > On Thursday 29 June 2006 16:45, Manfred Hollstein wrote: > > > If that fails, you have kinternet installed; add that to > the list and it > > should work. > > good point. > <remove-packages> can not break dependencies and so it's very > likely that > there is still a package that needs those packages. > I did a very small installation with no X - so it worked for me. > > -- > ciao, Uwe Gansert > > Uwe Gansert, Server Technologies Team > SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany > Fax: +49-(0)911-74053-476, Web: http://www.suse.de/~ug > now playing Covenant - Wasteland > > -- > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > > This email and any attachments may contain privileged and confidential information and are intended for the named addressee only. If you have received this e-mail in error, please notify the sender and delete this e-mail immediately. Any confidentiality, privilege or copyright is not waived or lost because this e-mail has been sent to you in error. It is your responsibility to check this e-mail and any attachments for viruses. No warranty is made that this material is free from computer virus or any other defect or error. Any loss/damage incurred by using this material is not the sender's responsibility. The sender's entire liability will be limited to resupplying the material.

4 6

Scripting dirinstall?
by Teun 25 Feb '07

25 Feb '07

Does anyone know if it is possible to use the 'dirinstall' from the command line? For instance i would like to be able to do something like: # yast dirinstall -d /someroot -s mypackages.sel If i do yast -h it says: "Usage: /sbin/yast [OPTIONS] [MODULE] [module specific parameters]" But is there a way to find out what 'module specific parameters' are available for a specific module? Thanks, Teun

3 2

Kernel Options
by discip＠pjm.com 06 Oct '06

06 Oct '06

Is there an easy way to automatically append an option to the default kernel that is installed? Or do I need to hack at this with some perl? Thanks, Paul

2 2

RE: [suse-autoinstall] autoyast freezes
by Marlier, Ian 29 Sep '06

29 Sep '06

> I'm doing an autoyast install with a SuSE 10.0 DVD. > First I have a problem with booting it up. It stops > at setting dhcp. I have three NICs in the box. > It will not work with the Intel(e1000) and couple of > RTL based. > > The only way I managed to boot thing up is trough an > RTL based NIC and by NFS based install. I chose > NFS server and dir, put the autoyast line and system > boots but only after it fails to dhcp on intel NIC. I'm assuming that you're using PXE to pass the install initrd and everything? Assuming so, you might try pausing the PXE process when the Intel NIC is trying to get an IP address, by hitting the "Pause/Break" key on your keyboard. Let it sit for about 30-60 seconds, and then hit space to start the process back up. (This situation -- where the NIC is on a network with DHCP, but doesn't get an address -- can be an indication that negotiation between the NIC and the switch to which it's network is taking a while -- and that the PXE process times out before the link is negotiated and brought up. It's happened with some of my machines, though only in cases where the networking hardware behind them is Cisco)

4 8

RE: [suse-autoinstall] Kernel Options
by GUNREBEN, PETER (Peter) 29 Sep '06

29 Sep '06

Paul, discip(a)pjm.com wrote: > discip(a)pjm.com wrote: > > Is there an easy way to automatically append an option to the > > default kernel that is installed? > > Or do I need to hack at this with some perl?> > > [...] > > Perfect! If autoyast actually created a valid document that > I could validate, or if there was a DTD available, I would > not have these issues. 1) You can use "yast2 autoyast" to generate or edit the XML-Files. Actually the XML-Code I've provided was generated using "yast2 autoyast". 2) There are DTD files available, see /usr/share/autoinstall/dtd/profile.dtd in RPM-Package: autoyast2-installation-2.13.56-3 Peter.

2 1

autoyast freezes
by Babarovic Ivica 28 Sep '06

28 Sep '06

Hi! I'm doing an autoyast install with a SuSE 10.0 DVD. First I have a problem with booting it up. It stops at setting dhcp. I have three NICs in the box. It will not work with the Intel(e1000) and couple of RTL based. The only way I managed to boot thing up is trough an RTL based NIC and by NFS based install. I chose NFS server and dir, put the autoyast line and system boots but only after it fails to dhcp on intel NIC. Ok ... I've managed to somehow install the system with hopefully correct xml setup file. The way I got this file is I first manually installed the system, exported the xml file with autoyast's tools menu and then edited that file untill it finally passed trough. Anyway ... now installation stops at first stage and box just freezes up. It will not reboot. After manually rebooting I get stuck at the grub prompt. I can see that menu.lst file is almost empty. What might be causing this? Best regards, I.B.

1 0

RE: [suse-autoinstall] Kernel Options
by GUNREBEN, PETER (Peter) 28 Sep '06

28 Sep '06

Paul, discip(a)pjm.com wrote: > Is there an easy way to automatically append an option to the > default kernel that is installed? > Or do I need to hack at this with some perl? What about: ================ <bootloader> <activate config:type="boolean">true</activate> <global> <default></default> <embed_stage1.5 config:type="boolean">false</embed_stage1.5> <gfxmenu>/boot/message</gfxmenu> <prompt>1</prompt> <timeout config:type="integer">8</timeout> </global> <loader_device></loader_device> <loader_type>grub</loader_type> <location>mbr</location> <repl_mbr config:type="boolean">true</repl_mbr> <sections config:type="list"> <section> <append> splash=silent showopts ***MYOPTIONS*** </append> <initrd>/boot/initrd</initrd> <kernel>/boot/vmlinuz</kernel> <name>SuSE 10.1</name> <original_name>linux</original_name> <root></root> <type>image</type> <vga></vga> </section> </sections> </bootloader> ================ You can create this also with "yast2 autoyast", of course. Peter.

2 1

Online Update not functioning after autoinstall
by Marlier, Ian 27 Sep '06

27 Sep '06

Hey, all -- Working on autobuilding with SUSE 10.1. I've got a local install source, rsynced from mirrors.kernel.org (I think it was). The installation works fine, but after install, Online Updates doesn't function. The error that I get when trying to configure Online Update is "No products to register". It looks like that error is thrown when the products returned by `/usr/lib64/zmd/query-pool products` don't intersect with Novell's list of supported SUSE products. And yet... [root@(fts0/dbopteron5):~]# /usr/lib64/zmd/query-pool products |product|SUSE-Linux-10.1-FTP|10.1-0|x86_64 [root@(fts0/dbopteron5):~]# and SUSE-Linux-10.1-FTP is one of the products returned by novell's registration server... This does NOT happen if I install manually, using the same local source (that's why I'm sending this here...) Has anyone: - Seen this? - Fixed this? If not, is it a bug with autoyast, or a bug more generally with zmd?

1 0

FW: Autoyast fails when trying to mount /boot with other EVMS Partitions
by Niall Calvert 27 Sep '06

27 Sep '06

I've being trying to identify a solution to an autoinstall problem for several weeks now without success and would appreciate some assistance. I'm trying to install sles10 from media located on a hard-drive partition, which was copied from the installation DVD. The problem does not occur when installing from the DVD using the same autoinst.xml file. I get the following error message after the volumes and boot partition has been formatted: Failure occurred during following action: Mount /dev/sda1 to /boot System error code was -3003 mount -t reiserfs /dev/sda1 /mnt/boot mount: /dev/sda1 already mounted or /mnt/boot busy >From spare console I see that nothing is mounted to /boot when the error occurs. However sda1 shows under /dev/evms so I suspect that sda1 is being used by EVMS in error. I tried applying the 'sles10_root_fix_v0.2' that I found on the Autoyast download area however this causes a different error. Thanks in advance for any suggestions on how to fix the problem. Best regards Niall Logs etc below. Grub Boot Config kernel (hd1,0)/boot/x86_64/loader/linux splash=verbose showopts ip=dhcp ramdisk_size=65536 install=hd://sdb3/software/sles10/sles10dvd/. autoyast=device://sdb1/autoinst.xml initrd (hd1,0)/boot/x86_64/loader/initrd Tail of Y2log file 2006-09-10 03:04:53,555 INFO libstorage(3024) - Volume.cc(checkDevice):962 checkDevice:/dev/sda1 ret:0 2006-09-10 03:04:53,555 INFO libstorage(3024) - SystemCmd.cc(SystemCmd):44 Konstruktor SystemCmd Nr:0 2006-09-10 03:04:53,555 INFO libstorage(3024) - Volume.cc(mount):1621 device:/dev/sda1 mp:/mnt/boot 2006-09-10 03:04:53,555 INFO libstorage(3024) - Volume.cc(mount):1626 device:/dev/sda1 mp:/mnt/boot 2006-09-10 03:04:53,555 INFO libstorage(3024) - SystemCmd.cc(execute):160 SystemCmd Executing:"modprobe reiserfs" 2006-09-10 03:04:53,563 INFO libstorage(3024) - SystemCmd.cc(doExecute):305 system() Returns:0 2006-09-10 03:04:53,563 INFO libstorage(3024) - SystemCmd.cc(execute):160 SystemCmd Executing:"mount -t reiserfs /dev/sda1 /mnt/boot" 2006-09-10 03:04:53,572 INFO libstorage(3024) - SystemCmd.cc(addLine):625 Adding Line 1 "mount: /dev/sda1 already mounted or /mnt/boot busy" 2006-09-10 03:04:53,572 INFO libstorage(3024) - SystemCmd.cc(getUntilEOF):591 pid:3902 added lines:1 stderr:1 2006-09-10 03:04:53,572 INFO libstorage(3024) - SystemCmd.cc(doExecute):305 system() Returns:32 2006-09-10 03:04:53,572 INFO libstorage(3024) - SystemCmd.cc(logOutput):636 stderr:mount: /dev/sda1 already mounted or /mnt/boot busy 2006-09-10 03:04:53,572 INFO libstorage(3024) - Volume.cc(mount):1667 ret:-3003 2006-09-10 03:04:53,572 INFO libstorage(3024) - Volume.cc(doMount):1011 ret:-3003 2006-09-10 03:04:53,572 INFO libstorage(3024) - Container.cc(commitChanges):148 ret:-3003 2006-09-10 03:04:53,572 INFO libstorage(3024) - Disk.cc(commitChanges):1618 ret:-3003 2006-09-10 03:04:53,572 INFO libstorage(3024) - Storage.cc(commitPair):3251 err at stage:3 type:1 cont:0 dest:0 name:sda1 2006-09-10 03:04:53,572 INFO libstorage(3024) - Storage.cc(ignoreError):3180 ret:0 2006-09-10 03:04:53,572 INFO libstorage(3024) - Storage.cc(commitPair):3259 stage:3 evms_activate:0 new_pair:0 2006-09-10 03:04:53,573 INFO libstorage(3024) - Storage.cc(commitPair):3272 ret:-3003 2006-09-10 03:04:53,573 INFO libstorage(3024) - Storage.cc(commit):3150 empty:1 2006-09-10 03:04:53,573 INFO libstorage(3024) - Storage.cc(commit):3158 ret:-3003 2006-09-10 03:04:53 <3> linux(3024) [YCP] Storage.ycp:3016 CommitChanges sint ret:-3003 2006-09-10 03:04:53 <1> linux(3024) [YCP] clients/inst_prepdisk.ycp:97 CommitChanges ret:-3003 Partitions / Volumes Sda1 - boot / and Swap lvm2 volumes in same evms container

2 2

SLES 10 and disk to disk autoyast installation
by Wenk, Moritz 27 Sep '06

27 Sep '06

Hello, I am trying to autoinstall SLES 10 in HD mode (disk to disk). It worked for SLES 9 but does not any longer with SLES 10. The problem is that YaST tries to mount my source partition when formating the other partitions and this fails because the source partition is alerady mounted. This is my partition schema: sda1: ext3 16GB sda2: swap sda3: fat32 16BG The third partition does contain the installation source. This is the autoyast partition plan: <partitioning config:type="list"> <drive> <device>/dev/sda</device> <initialize config:type="boolean">false</initialize> <is_evms_vg config:type="boolean">false</is_evms_vg> <is_lvm_vg config:type="boolean">false</is_lvm_vg> <use>1,2</use> <partitions config:type="list"> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">ext3</filesystem> <fsopt>defaults,acl</fsopt> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <partition_id config:type="integer">131</partition_id> <partition_nr config:type="integer">1</partition_nr> <size>16gb</size> </partition> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">swap</filesystem> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <partition_nr config:type="integer">2</partition_nr> <size>512mb</size> </partition> </partitions> </drive> </partitioning> And the kernel command line for autoyast: install=hd://sda3/suse AutoYaST=file:///autoyast.xml The autoyast file (autoyast.xml) is contained in the initrd. Kernel and initrd are booted from sda3. This worked for SLES 9 with a little trick (unmounting /mounts/extra in a pre-install script). Any ideas how to get the disk to disk autoyast installation working for SLES 10? Regards, i. A. Moritz Wenk Specialist Product Engineering & Development Corporate Technology Software MAXDATA Systeme GmbH Elbestraße 12-16 D-45768 Marl Telefon: +49 (0) 2365 952-2715 Telefax: +49 (0) 2365 952-2279 E-Mail : moritz.wenk(a)maxdata.de http://www.maxdata.com

2 2

merge scripts
by Arvid Walter 27 Sep '06

27 Sep '06

Hi, I cannot merge scripts: --snip-- <?xml version=1.0?> <!DOCTYPE profile SYSTEM /usr/share/autoinstall/dtd/profile.dtd> <profile xmlns=http://www.suse.com/1.0/yast2ns xmlns:config=http://www.suse.com/1.0/configns> <classes config:type=list> <class> <class_name>Repositories</class_name> <configuration>apt-suse10.xml</configuration> </class> </classes> <scripts> <init-scripts config:type=list> <script> <source> <![CDATA[#!/bin/sh apt-get install nagios-plugins ]]> </source> </script> </init-scripts> </scripts> </profile> --snap-- --snip-- <?xml version=1.0?> <!DOCTYPE profile SYSTEM /usr/share/autoinstall/dtd/profile.dtd> <profile xmlns=http://www.suse.com/1.0/yast2ns xmlns:config=http://www.suse.com/1.0/configns> <general> <signature-handling> <accept_unsigned_file config:type=boolean>true</accept_unsigned_file> </signature-handling> </general> <add-on> <add_on_products config:type=list> <listentry> <media_url>nfs://10.x.x.x/usr/local/Unattended/vmware/suse/</media_url> <product>VMware-server</product> <product_dir>/</product_dir> </listentry> </add_on_products> </add-on> <software> <packages config:type=list> <package>VMware-server</package> <package>gcc</package> <package>gcc-c++</package> <package>xorg-x11-libs</package> <package>fontconfig</package> <package>freetype2</package> <package>compat</package> <package>kernel-source</package> <package>xinetd</package> <package>wget</package> <package>patch</package> </packages> </software> <scripts> <init-scripts config:type=list> <script> <network_needed config:type=boolean>true</network_needed> <source> <![CDATA[#!/bin/sh echo -n Switching multicast dns off if [ ! $(grep mdns /etc/host.conf) ]; then echo mdns off >> /etc/host.conf; fi echo done echo -n Changing service number 902 to 904 cp /etc/services /etc/services.backup awk '{res=gsub(^902/, \t904/, $2); print $0 }' /etc/services.backup > /etc/services echo done echo -n Configure vmware server ... /usr/bin/vmware-config.pl default EULA_AGREED=yes rm -f /etc/vmware/not_configured echo done echo Installing the vmware web interface cd /tmp/ wget http://10.x.x.x/vmware/VMware-mui-current.tar.gz tar zxvf VMware-mui-current.tar.gz cd vmware-mui-distrib echo -n Applying patches wget http://10.x.x.x/vmware/vmware-install.pl.p0 patch -p0 < vmware-install.pl.p0 echo done echo -n Installing and configuring vmware server web interface ... perl vmware-install.pl --default EULA_AGREED=yes echo done echo -n Preparing xinetd insserv xinetd echo done]]> </source> </script> </init-scripts> </scripts> </profile> --snap-- Both xmls are beeing merged using the nagios part only. How can i use more than one script.

2 1

RE: [suse-autoinstall] Disk detection order
by Marlier, Ian 26 Sep '06

26 Sep '06

> -----Original Message----- > From: Karsten Künne [mailto:kuenne@rentec.com] > Sent: Friday, September 22, 2006 10:55 AM > To: suse-autoinstall(a)suse.com > Cc: Marlier, Ian > Subject: Re: [suse-autoinstall] Disk detection order > > On Friday 22 September 2006 10:11, Marlier, Ian wrote: > > Hey, all -- > > > > This is returning to an old question for me, as I've started creating > > 10.1 installation profiles now that the zypp/zmd issues seem to be > > fixed. > > > > I'm trying to install onto a server with the following: > > - 2 x 80GB SATA, on a Silicon Image controller, exported as JBOD. > > - 3 x 250GB SATA, on a 3ware 9550 controller, exported as a single RAID > > 5 > > > > I'm creating a software RAID 1 on the 80GB drives, and using that as the > > root partition. The RAID5 being partitioned for LVM, though not > > allocated yet. > > > > When autoyast starts the install, it autodetects the installed hardware > > (in, I think, alphabetical order?). The 3ware RAID array is detected as > > /dev/sda, and the two SiI disks as /dev/sdb and /dev/sdc > > > > When the install is complete and the system reboots, this order has > > switched; the SiI disks are /dev/sda and /dev/sdb, and the 3ware RAID is > > /dev/sdc. > > > > You have to load the modules for your controllers in the right order. Put > them > into the <initrd_modules> section in your profile in the proper order, the > 3ware module first and then the other one. Well, I'm hoping to have the 3ware RAID as /dev/sdc, so I have my initrd_modules section set up this way: <initrd_modules config:type="list"> <initrd_module> <module>amd74xx</module> </initrd_module> <initrd_module> <module>sata_sil</module> </initrd_module> <initrd_module> <module>3w_9xxx</module> </initrd_module> <initrd_module> <module>processor</module> </initrd_module> <initrd_module> <module>thermal</module> </initrd_module> <initrd_module> <module>fan</module> </initrd_module> <initrd_module> <module>jbd</module> </initrd_module> <initrd_module> <module>ext3</module> </initrd_module> <initrd_module> <module>raid1</module> </initrd_module> </initrd_modules> But it looks like it doesn't apply to the autoinstall process; the detection order is correct after the first installed boot, but not during the install process itself...

3 2

Trouble on OpenPBS server for Sles10.0 x86_64
by hp9 25 Sep '06

25 Sep '06

Hello, I installed bother server and nodes. But I can not configure OpenPBS server. The trouble is the follow: ------------------------------------------------------------- funmat:~ # qmgr -c "set server scheduling=true" qmgr obj= svr=default: Unauthorized Request funmat:~ # qmgr Max open servers: 4 Qmgr: create queue qdefault qmgr obj=qdefault svr=default: Unauthorized Request ----------------------------------------------------------- Anyone has experience to configure OpenPBS in SUSE? Thanks, hongbin

1 0

Troubles installing SLES10 from a FreeBSD Server
by Beat Rubischon 25 Sep '06

25 Sep '06

Hello! I try to set up SLES10 from a NFS installation source based on a Panasas storage system. From the software point of view, the Panasas is a FreeBSD 4 NFS server and offers the following RPC services: program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100005 3 udp 1009 mountd 100005 3 tcp 1023 mountd 100005 1 udp 1009 mountd 100005 1 tcp 1023 mountd 100003 3 udp 2049 nfs 100003 3 tcp 2049 nfs 100021 1 udp 996 nlockmgr 100021 3 udp 996 nlockmgr 100021 4 udp 996 nlockmgr 100021 1 tcp 1021 nlockmgr 100021 3 tcp 1021 nlockmgr 100021 4 tcp 1021 nlockmgr 100024 1 udp 988 status 100024 1 tcp 1020 status An final installed SLES10 is able to mount the shared filesystems, but not the installer. In the initrd is a /lbin/mount (Busybox?) which tries the wrong protocol (NFSv4?) and fails: mount("172.16.9.1:/cluster", "/mnt", "nfs", MS_MGC_VAL|MS_RDONLY, "\4") = -1 EPROTONOSUPPORT (Protocol not supported) Surprisingly, the installation from a stoneage RedHat only supporting NFSv2/v3 over UDP just works. Is there any possibility to replace the mount command in the initrd? Or have a look at the sources to get further information? Beat -- \|/ Beat Rubischon <beat(a)0x1b.ch> ( 0^0 ) http://www.0x1b.ch/~beat/ oOO--(_)--OOo--------------------------------------------------- Meine Erlebnisse, Gedanken und Traeume: http://www.0x1b.ch/blog/

1 0

AW: [suse-autoinstall] SLES 10 and disk to disk autoyast installation
by Wenk, Moritz 25 Sep '06

25 Sep '06

Hi, it is the same error code, but for me it is the partition containing the sources (the SLES CDs) that autoyast tries to mount. -----Ursprüngliche Nachricht----- Von: Niall Calvert [mailto:niall@calvert.cn] Gesendet: Freitag, 22. September 2006 18:29 An: Wenk, Moritz; suse-autoinstall(a)suse.com Betreff: RE: [suse-autoinstall] SLES 10 and disk to disk autoyast installation I've also been trying also to get disk-to-disk install working, also with EVMS volumes; do you get a similar error message to the following: Failure occurred during following action: Mount /dev/sda1 to /boot System error code was -3003 mount -t reiserfs /dev/sda1 /mnt/boot mount: /dev/sda1 already mounted or /mnt/boot busy Best regards Niall -----Original Message----- From: Wenk, Moritz [mailto:Moritz.Wenk@maxdata.com] Sent: 22 September 2006 14:15 To: suse-autoinstall(a)suse.com Subject: [suse-autoinstall] SLES 10 and disk to disk autoyast installation Hello, I am trying to autoinstall SLES 10 in HD mode (disk to disk). It worked for SLES 9 but does not any longer with SLES 10. The problem is that YaST tries to mount my source partition when formating the other partitions and this fails because the source partition is alerady mounted. This is my partition schema: sda1: ext3 16GB sda2: swap sda3: fat32 16BG The third partition does contain the installation source. This is the autoyast partition plan: <partitioning config:type="list"> <drive> <device>/dev/sda</device> <initialize config:type="boolean">false</initialize> <is_evms_vg config:type="boolean">false</is_evms_vg> <is_lvm_vg config:type="boolean">false</is_lvm_vg> <use>1,2</use> <partitions config:type="list"> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">ext3</filesystem> <fsopt>defaults,acl</fsopt> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <partition_id config:type="integer">131</partition_id> <partition_nr config:type="integer">1</partition_nr> <size>16gb</size> </partition> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">swap</filesystem> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <partition_nr config:type="integer">2</partition_nr> <size>512mb</size> </partition> </partitions> </drive> </partitioning> And the kernel command line for autoyast: install=hd://sda3/suse AutoYaST=file:///autoyast.xml The autoyast file (autoyast.xml) is contained in the initrd. Kernel and initrd are booted from sda3. This worked for SLES 9 with a little trick (unmounting /mounts/extra in a pre-install script). Any ideas how to get the disk to disk autoyast installation working for SLES 10? Regards, i. A. Moritz Wenk Specialist Product Engineering & Development Corporate Technology Software MAXDATA Systeme GmbH Elbestraße 12-16 D-45768 Marl Telefon: +49 (0) 2365 952-2715 Telefax: +49 (0) 2365 952-2279 E-Mail : moritz.wenk(a)maxdata.de http://www.maxdata.com -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

RE: [suse-autoinstall] SLES 10 and disk to disk autoyast installation
by Niall Calvert 22 Sep '06

22 Sep '06

I've also been trying also to get disk-to-disk install working, also with EVMS volumes; do you get a similar error message to the following: Failure occurred during following action: Mount /dev/sda1 to /boot System error code was -3003 mount -t reiserfs /dev/sda1 /mnt/boot mount: /dev/sda1 already mounted or /mnt/boot busy Best regards Niall -----Original Message----- From: Wenk, Moritz [mailto:Moritz.Wenk@maxdata.com] Sent: 22 September 2006 14:15 To: suse-autoinstall(a)suse.com Subject: [suse-autoinstall] SLES 10 and disk to disk autoyast installation Hello, I am trying to autoinstall SLES 10 in HD mode (disk to disk). It worked for SLES 9 but does not any longer with SLES 10. The problem is that YaST tries to mount my source partition when formating the other partitions and this fails because the source partition is alerady mounted. This is my partition schema: sda1: ext3 16GB sda2: swap sda3: fat32 16BG The third partition does contain the installation source. This is the autoyast partition plan: <partitioning config:type="list"> <drive> <device>/dev/sda</device> <initialize config:type="boolean">false</initialize> <is_evms_vg config:type="boolean">false</is_evms_vg> <is_lvm_vg config:type="boolean">false</is_lvm_vg> <use>1,2</use> <partitions config:type="list"> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">ext3</filesystem> <fsopt>defaults,acl</fsopt> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <partition_id config:type="integer">131</partition_id> <partition_nr config:type="integer">1</partition_nr> <size>16gb</size> </partition> <partition> <create config:type="boolean">false</create> <format config:type="boolean">true</format> <filesystem config:type="symbol">swap</filesystem> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <partition_nr config:type="integer">2</partition_nr> <size>512mb</size> </partition> </partitions> </drive> </partitioning> And the kernel command line for autoyast: install=hd://sda3/suse AutoYaST=file:///autoyast.xml The autoyast file (autoyast.xml) is contained in the initrd. Kernel and initrd are booted from sda3. This worked for SLES 9 with a little trick (unmounting /mounts/extra in a pre-install script). Any ideas how to get the disk to disk autoyast installation working for SLES 10? Regards, i. A. Moritz Wenk Specialist Product Engineering & Development Corporate Technology Software MAXDATA Systeme GmbH Elbestraße 12-16 D-45768 Marl Telefon: +49 (0) 2365 952-2715 Telefax: +49 (0) 2365 952-2279 E-Mail : moritz.wenk(a)maxdata.de http://www.maxdata.com -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

RE: [suse-autoinstall] Auto installation SLES9 SP2
by Lewandowski, Kelly 22 Sep '06

22 Sep '06

Hello again, I have fixed the version mismatch for the bcraid.ko problem. I now see the disks even though they are not seen correctly. I created an autoyast file to see if it works. Every thing works for the automated install. The problem: The disk drives are seen as individual drives vs the raid 1 or raid 5 that is set via the Broadcom cards bios. When I initiate the install, modules are loaded up, SLES loads sata_svw.ko, then loads bcraid.ko. ** In the autoyast file I have taken out the sata_svw module ** Autoinstall starts and installs SLES on an individual drive instead of the raid. I did a manual install (cd's and floppy) watching the loading of the modules. The Broadcom is loaded at the top. When the manual install is done we change the (using YaST /etc/sysconfig editor) System>kernel>INITRD_MODULES by deleting the sata-svw. Question: How do I change / modify the boot sequence to load the bcraid.ko module before or instead of the sata_svw.ko module. If this involves remaking the initrd file (witch I think it might not sure) can someone enlighten me on the procedures for this. Thanks, Kelly Lewandowski -----Original Message----- From: Lewandowski, Kelly [mailto:KLewandowski@breecehill.com] Sent: Wednesday, September 20, 2006 8:48 AM To: suse-autoinstall(a)suse.com Subject: RE: [suse-autoinstall] Auto installation SLES9 SP2 Thanks Yan, We are slated to upgrade when we have enough cycles to thoroughly test a new version against our platform. Our server also has a tape library, with an LTO-3 tape drive, and 10 tape cartridges. Front half of the box is a library, back half is a server with sata drives, and a DVD-RW. All cards in the server are half height cards to keep it at 4U. When I use the driver image on a manual install (CD's and usb floppy) the drivers that I have work fine. I will run through another manual install and verify. Thanks, Kelly Lewandowski -----Original Message----- From: Yan Fitterer [mailto:yfitterer@novell.com] Sent: Wednesday, September 20, 2006 1:52 AM To: Lewandowski, Kelly; suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] Auto installation SLES9 SP2 The overall process / setup seems sound. Looks like a classic case of version mismatch between bcraid.ko and running kernel. The binary module must be built for exactly the kernel version that it is used with. Either the kernel that runs during install isn't the one you expect, or the kernel module hasn't been built for quite the same version. Check the exact version of the kernel that's running during install (making sure it _is_ indeed the SP2 version, in case of install source problem). Then verify that the bcraid module you're using has been build for that very same version. Why are you using SLES9 SP2 anyway? That's old, outdated, and insecure - given the number of security updates that have been released since. Do verify that the bcraid driver hasn't been built for an SP3 (or later...) kernel. There should be no incompatibility issues with updating to SP3 (+ latest patches). Yan >>> On Tue, Sep 19, 2006 at 11:23 PM, in message <FEF375A836207347A500D2F7F65DA0665A21B9(a)exchange2k3.BHT.local>, "Lewandowski, Kelly" <KLewandowski(a)breecehill.com> wrote: > Hello all, > > > > I have been tasked by my manager to create a SLES auto installation > process for our 2 server products. > > > > We are using SLES 9 SP2. > > > > The 2 servers: > > > > 7210 Intel motherboard running a P4 processor with a BroadCom BC4852 > SATA Raid card. (8 SATA drives) > > 7520 Intel Motherboard running dual Xeon processors with a BroadCom > BC4852 SATA RAID card. (8 SATA drives) > > > > I have tried many of the suggestions on this list but nothing seems to > fully work. > > So any suggestions would be appreciated. The steps and setup of what I > have done are listed below. > > > > This is for the 7520 Xeon only. I am trying to get one system to work > first. > > > > 1. Setup > > a. 1 windows 2003 server running DHCP > b. 1 SLES9 SP2 2.6.5- 7.191- default running the tftp server > and nfs server. > c. 1 new Xeon server ready to image. 35 G RAID 5 across all > 8 drives. > > 2. Steps > > a. SLES server set up for tftp boot. > > /tftpboot Files are: initrd linux pxelinux.0 ./pxelinux.cfg/default > > default file: > > > > label linux > > kernel linux > > append initrd=initrd install=nfs://xxx.xxx.xxx.xxx/install/SLES9x64 > vga=0x0314 > > > > b. SLES server set up for NFS. 64 bit cd's plus server > pack cd's created via YaST > > /install/SLES9x64 > > c. BroadCom driver image dd'd to floppy then copied over > into the sp2/cd1/linux/suse/x86_64/install/update.tar.gz. Also copied > over ./modules/bcraid.ko > d. Control file is the one right off the cd's. Have not > used YaST to create one as of yet as I needed to make sure everything > loaded manual method first. > > 2. Problems > > a. new system boots. Gets its ip and location of the boot > file from the DHCP server. Installation begins. Accept the agreements. > Accept the languages. When it gets to the checking for hard disks it > comes back with none. > b. Errors on tty3 are as follows. > > Dud 100: Service Pack 2 > > insmod - f /modules/bcraid.ko > > insmod: error inserting '/modules/bcraid.ko': - 1 invalid module format > > c. Errors on tty4 are as follows. > > kernel: bcraid: disagrees about version of symbol struct_module > > > > Questions: > > I have tried using both the - smp- 64 and the - default- 64 from Broadcom. > Both show the same errors as above. > > > > Thanks, > > > > Kelly Lewandowski > > > > -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

Disk detection order
by Marlier, Ian 22 Sep '06

22 Sep '06

Hey, all -- This is returning to an old question for me, as I've started creating 10.1 installation profiles now that the zypp/zmd issues seem to be fixed. I'm trying to install onto a server with the following: - 2 x 80GB SATA, on a Silicon Image controller, exported as JBOD. - 3 x 250GB SATA, on a 3ware 9550 controller, exported as a single RAID 5 I'm creating a software RAID 1 on the 80GB drives, and using that as the root partition. The RAID5 being partitioned for LVM, though not allocated yet. When autoyast starts the install, it autodetects the installed hardware (in, I think, alphabetical order?). The 3ware RAID array is detected as /dev/sda, and the two SiI disks as /dev/sdb and /dev/sdc When the install is complete and the system reboots, this order has switched; the SiI disks are /dev/sda and /dev/sdb, and the 3ware RAID is /dev/sdc. I would like for the behavior to be consistent, one way or the other, but I can't seem to figure out how to force it to be. (In an ideal world, the 3ware RAID would be /dev/sdc during install and after initial boot.) Anyone have any suggestions, or pointers to documentation that I may have missed? I've been trying to keep track through Uwe's (many and wonderful!) improvements to autoyast, but may have missed something along the way... TIA! - Ian

2 1

firewall config vs apache config
by Rich 21 Sep '06

21 Sep '06

i am performing an autoinstall of sles 10 with a control file. it seems that firewall config in http-server conflicts with firewall settings that instruct to open access for http service. if <firewall> section has http as an allowed service, but <http-server> has <firewall config:type="boolean"> set to false (or missing), port 80 is not opened. is this the intended behaviour or have i missed something ? -- Rich

1 0

RE: [suse-autoinstall] Auto installation SLES9 SP2
by Lewandowski, Kelly 20 Sep '06

20 Sep '06

Thanks Yan, We are slated to upgrade when we have enough cycles to thoroughly test a new version against our platform. Our server also has a tape library, with an LTO-3 tape drive, and 10 tape cartridges. Front half of the box is a library, back half is a server with sata drives, and a DVD-RW. All cards in the server are half height cards to keep it at 4U. When I use the driver image on a manual install (CD's and usb floppy) the drivers that I have work fine. I will run through another manual install and verify. Thanks, Kelly Lewandowski -----Original Message----- From: Yan Fitterer [mailto:yfitterer@novell.com] Sent: Wednesday, September 20, 2006 1:52 AM To: Lewandowski, Kelly; suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] Auto installation SLES9 SP2 The overall process / setup seems sound. Looks like a classic case of version mismatch between bcraid.ko and running kernel. The binary module must be built for exactly the kernel version that it is used with. Either the kernel that runs during install isn't the one you expect, or the kernel module hasn't been built for quite the same version. Check the exact version of the kernel that's running during install (making sure it _is_ indeed the SP2 version, in case of install source problem). Then verify that the bcraid module you're using has been build for that very same version. Why are you using SLES9 SP2 anyway? That's old, outdated, and insecure - given the number of security updates that have been released since. Do verify that the bcraid driver hasn't been built for an SP3 (or later...) kernel. There should be no incompatibility issues with updating to SP3 (+ latest patches). Yan >>> On Tue, Sep 19, 2006 at 11:23 PM, in message <FEF375A836207347A500D2F7F65DA0665A21B9(a)exchange2k3.BHT.local>, "Lewandowski, Kelly" <KLewandowski(a)breecehill.com> wrote: > Hello all, > > > > I have been tasked by my manager to create a SLES auto installation > process for our 2 server products. > > > > We are using SLES 9 SP2. > > > > The 2 servers: > > > > 7210 Intel motherboard running a P4 processor with a BroadCom BC4852 > SATA Raid card. (8 SATA drives) > > 7520 Intel Motherboard running dual Xeon processors with a BroadCom > BC4852 SATA RAID card. (8 SATA drives) > > > > I have tried many of the suggestions on this list but nothing seems to > fully work. > > So any suggestions would be appreciated. The steps and setup of what I > have done are listed below. > > > > This is for the 7520 Xeon only. I am trying to get one system to work > first. > > > > 1. Setup > > a. 1 windows 2003 server running DHCP > b. 1 SLES9 SP2 2.6.5- 7.191- default running the tftp server > and nfs server. > c. 1 new Xeon server ready to image. 35 G RAID 5 across all > 8 drives. > > 2. Steps > > a. SLES server set up for tftp boot. > > /tftpboot Files are: initrd linux pxelinux.0 ./pxelinux.cfg/default > > default file: > > > > label linux > > kernel linux > > append initrd=initrd install=nfs://xxx.xxx.xxx.xxx/install/SLES9x64 > vga=0x0314 > > > > b. SLES server set up for NFS. 64 bit cd's plus server > pack cd's created via YaST > > /install/SLES9x64 > > c. BroadCom driver image dd'd to floppy then copied over > into the sp2/cd1/linux/suse/x86_64/install/update.tar.gz. Also copied > over ./modules/bcraid.ko > d. Control file is the one right off the cd's. Have not > used YaST to create one as of yet as I needed to make sure everything > loaded manual method first. > > 2. Problems > > a. new system boots. Gets its ip and location of the boot > file from the DHCP server. Installation begins. Accept the agreements. > Accept the languages. When it gets to the checking for hard disks it > comes back with none. > b. Errors on tty3 are as follows. > > Dud 100: Service Pack 2 > > insmod - f /modules/bcraid.ko > > insmod: error inserting '/modules/bcraid.ko': - 1 invalid module format > > c. Errors on tty4 are as follows. > > kernel: bcraid: disagrees about version of symbol struct_module > > > > Questions: > > I have tried using both the - smp- 64 and the - default- 64 from Broadcom. > Both show the same errors as above. > > > > Thanks, > > > > Kelly Lewandowski > > > >

1 0

installing from driver update package
by jaiber john 20 Sep '06

20 Sep '06

I have a driver update disk in .img format which has a device driver created according to the instructions on chapter 3 of http://mirror.mcs.anl.gov/suse-people/hvogel/Update-Media-HOWTO/Update-Medi…. I want to automate the driver installation for SLES9 SP2. How do I do that? I can re-master the initrd that comes with SLES9 SP2 CD1, insert this .img file into it. What other settings should I follow so that Yast detects it and installs it automatically? If i add driverupdate=1 as a boot parameter, it searches in floppy not what is inside initrd. Thanks -- <Jaiber John>

1 0

Auto installation SLES9 SP2
by Lewandowski, Kelly 20 Sep '06

20 Sep '06

Hello all, I have been tasked by my manager to create a SLES auto installation process for our 2 server products. We are using SLES 9 SP2. The 2 servers: 7210 Intel motherboard running a P4 processor with a BroadCom BC4852 SATA Raid card. (8 SATA drives) 7520 Intel Motherboard running dual Xeon processors with a BroadCom BC4852 SATA RAID card. (8 SATA drives) I have tried many of the suggestions on this list but nothing seems to fully work. So any suggestions would be appreciated. The steps and setup of what I have done are listed below. This is for the 7520 Xeon only. I am trying to get one system to work first. 1. Setup a. 1 windows 2003 server running DHCP b. 1 SLES9 SP2 2.6.5-7.191-default running the tftp server and nfs server. c. 1 new Xeon server ready to image. 35 G RAID 5 across all 8 drives. 2. Steps a. SLES server set up for tftp boot. /tftpboot Files are: initrd linux pxelinux.0 ./pxelinux.cfg/default default file: label linux kernel linux append initrd=initrd install=nfs://xxx.xxx.xxx.xxx/install/SLES9x64 vga=0x0314 b. SLES server set up for NFS. 64 bit cd's plus server pack cd's created via YaST /install/SLES9x64 c. BroadCom driver image dd'd to floppy then copied over into the sp2/cd1/linux/suse/x86_64/install/update.tar.gz. Also copied over ./modules/bcraid.ko d. Control file is the one right off the cd's. Have not used YaST to create one as of yet as I needed to make sure everything loaded manual method first. 2. Problems a. new system boots. Gets its ip and location of the boot file from the DHCP server. Installation begins. Accept the agreements. Accept the languages. When it gets to the checking for hard disks it comes back with none. b. Errors on tty3 are as follows. Dud 100: Service Pack 2 insmod -f /modules/bcraid.ko insmod: error inserting '/modules/bcraid.ko': -1 invalid module format c. Errors on tty4 are as follows. kernel: bcraid: disagrees about version of symbol struct_module Questions: I have tried using both the -smp-64 and the -default-64 from Broadcom. Both show the same errors as above. Thanks, Kelly Lewandowski

2 1

SDK license agreement/SLES10
by Frank Steiner 20 Sep '06

20 Sep '06

Hi, is there a way to get rid of that license agreement popping up for SDK on SLES 10? I configured it as add_on_product and now it gets on my nerves with that agreement stopping the autoinstall :-( cu, Frank -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

2 2

RE: [suse-autoinstall] FW: Autoyast fails when trying to mount /boot with other EVMS Partitions
by Niall Calvert 19 Sep '06

19 Sep '06

Hi Uwe, > which partition is that? media is on partition sdb3 formatted as FAT32 (a USB disk) >can you post your partitioning section of the profile? <partitioning config:type="list"> <drive> <device>/dev/system</device> <is_evms_vg config:type="boolean">true</is_evms_vg> <lvm2 config:type="boolean">true</lvm2> <partitions config:type="list"> <partition> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <lv_name>swap</lv_name> <mount>swap</mount> <size>2684354560</size> </partition> <partition> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <lv_name>sys_lx1</lv_name> <mount>/</mount> <size>8053063680</size> </partition> </partitions> <pesize>4M</pesize> <type config:type="symbol">CT_EVMS</type> <use>all</use> </drive> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <filesystem config:type="symbol">reiser</filesystem> <filesystem_id config:type="integer">131</filesystem_id> <format config:type="boolean">true</format> <mount>/boot</mount> <partition_id config:type="integer">131</partition_id> <partition_nr config:type="integer">1</partition_nr> <partition_type>primary</partition_type> <size>320753664</size> </partition> <partition> <evms_group>system</evms_group> <filesystem_id config:type="integer">142</filesystem_id> <partition_id config:type="integer">142</partition_id> <partition_nr config:type="integer">2</partition_nr> <partition_type>primary</partition_type> <size>10742215680</size> </partition> </partitions> <type config:type="symbol">CT_DISK</type> <use>all</use> </drive> </partitioning> Many thanks Niall -----Original Message----- From: Uwe Gansert [mailto:ug@suse.de] Sent: 18 September 2006 08:18 To: suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] FW: Autoyast fails when trying to mount /boot with other EVMS Partitions On Wednesday 13 September 2006 01:08, Niall Calvert wrote: > I'm trying to install sles10 from media located on a hard-drive > partition, which partition is that? > which was copied from the installation DVD. The problem does > not occur when installing from the DVD using the same autoinst.xml file. can you post your partitioning section of the profile? -- ciao, Uwe Gansert Uwe Gansert, Server Technologies Team SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany Business: http://www.suse.de/~ug -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

Anyone has experence on x86_64 autoyast installation?
by hp9 19 Sep '06

19 Sep '06

I have very bad experience to install SLES10.0 x86_64 on 64bits machine. Though I can do it manually. But I can not install it by AUTOYAST, because AUTOYAST always install i586 but x86_64! Anyone has experence on x86_64 autoyast installation?

3 2

RE: [suse-autoinstall] SuSE 10.1: Could not find the SUSE Linux Installation Source
by GUNREBEN, PETER (Peter) 19 Sep '06

19 Sep '06

GUNREBEN, PETER wrote: > > I have problems to get the autoyast installation (SuSE 10.1) > up and running with some new computers (K8N-DRE, 2xOpteron > 275). The main error message is "Could not find the SUSE > Linux Installation Source". The onboard NICs are Broadcom > 5721and I'm using the tg3 driver. I can see on console 3 that > there is a problem with the nfs mount. However, I can mount > this directory via nfs (from console 9) without any problems. > Any help or idea is highly appreciated. I'm desperately > searching for a solution. I just got a hint from Lee per E-Mail to add netwait=5 to the kernel command line. This solved the problem. Many thanks to Lee. Peter.

1 0

SuSE 10.1: Could not find the SUSE Linux Installation Source
by GUNREBEN, PETER (Peter) 19 Sep '06

19 Sep '06

Hi, I have problems to get the autoyast installation (SuSE 10.1) up and running with some new computers (K8N-DRE, 2xOpteron 275). The main error message is "Could not find the SUSE Linux Installation Source". The onboard NICs are Broadcom 5721and I'm using the tg3 driver. I can see on console 3 that there is a problem with the nfs mount. However, I can mount this directory via nfs (from console 9) without any problems. Any help or idea is highly appreciated. I'm desperately searching for a solution. ======== Console 3 ======== got info from file:/info hostname: 135.246.30.241 netmask: 255.255.254.0 broadcast: 135.246.31.255 gateway: 135.246.30.1 server: 135.246.75.102 nameserver: 135.246.31.80 Ok, that seems to have worked. :-) Ok, that seems to have worked. :-) Ok, that seems to have worked. :-) Looking for a network server... Trying to activate eth0 setting up localhost...done eth0 activated nfs: trying to mount 135.246.75.102:/raid0/SuSE/10.1 Starting portmap. mount: nfs mount failed, server says: unknown error -1 Ok, that didn't work; see if we can activate another network device... No further network cards found; giving up. Automatic setup not possible. ======== Console 3 ======== Peter.

1 0

RE: [suse-autoinstall] SLES10 Archive Selection: My autoyast alway install i586 but x86_64
by hp9 18 Sep '06

18 Sep '06

My autoyast alway install i586 but x86_64 in its process. Who know the reason about that? Thank you lots. The follow is my scripts: ------------------------------ <?xml version="1.0"?> <!DOCTYPE profile SYSTEM "/usr/share/autoinstall/dtd/profile.dtd"> <profile xmlns="http://www.suse.com/1.0/yast2ns" xmlns:config="http://www.suse.com/1.0/configns"> <bootloader> <activate config:type="boolean">true</activate> <global> <embed_stage1.5 config:type="boolean">true</embed_stage1.5> <lines_cache_id>0</lines_cache_id> <prompt>1</prompt> <stage1_dev>/dev/sda1</stage1_dev> <timeout config:type="integer">8</timeout> </global> <initrd_modules config:type="list"> <initrd_module> <module>piix</module> </initrd_module> <initrd_module> <module>ata_piix</module> </initrd_module> <initrd_module> <module>processor</module> </initrd_module> <initrd_module> <module>thermal</module> </initrd_module> <initrd_module> <module>fan</module> </initrd_module> </initrd_modules> <loader_device>/dev/sda1</loader_device> <loader_type>grub</loader_type> <location>boot</location> <repl_mbr config:type="boolean">false</repl_mbr> <sections config:type="list"> <section> <append>showopts</append> <initial>1</initial> <initrd>/boot/initrd</initrd> <kernel>/boot/vmlinuz</kernel> <lines_cache_id>1</lines_cache_id> <name>SUSE Linux 10.0</name> <original_name>linux</original_name> <root>/dev/sda1</root> <type>image</type> <vga>0x31a</vga> </section> <section> <append>showopts ide=nodma apm=off acpi=off noresume edd=off 3</append> <initrd>/boot/initrd</initrd> <kernel>/boot/vmlinuz</kernel> <lines_cache_id>2</lines_cache_id> <name>Failsafe -- SUSE Linux 10.0</name> <original_name>failsafe</original_name> <root>/dev/sda1</root> <type>image</type> <vga>normal</vga> </section> </sections> </bootloader>

1 0

RE: [suse-autoinstall] SLES10 Archive Selection: My autoyast alway install i586 but x86_64
by hp9 18 Sep '06

18 Sep '06

My autoyast alway install i586 but x86_64 in its process. Who know the reason about that? Thank you lots.

1 0

problem including customer packages: add-on product vs multi-product media
by Oliver Schweikert 18 Sep '06

18 Sep '06

Hello, I want to create a single installation DVD, that contains SLES10 and customer packages. The DVD shall also act as network installation source, when it is mounted. When using SLES8 I put the customer packages among the SLES packages and update the package description, and all works fine. Now i try to separate SUSE and customer rpms, but cannot get it to work (all while automatic installation process): add-on product approach: YaST expects an URL in the add_on_product file, but i can only provide a relative path, because if the media is used from a drive the protocol is cd://, but when acting as network source it is ftp://. Is there a way to specify the add-on product path relative to the media root? multi-product approach: When I append my product to /media.1/products (*asc/*key removed and <signature-handling> adjusted to accept it), the meta data are read from YaST myprod/setup/descr/packages,...), but if i define a package from this product YaST messes up the package selection and the installation fails: Instead of the 1.3 Gbyte from SLES sources, YaST claims to have only about 47 Mbyte to install and the first package (grub) fails to install at the %post script. Any idea, why YaST gets confused? Oliver

4 4

SLES10 Package Selection
by Rajeev Mishra 18 Sep '06

18 Sep '06

I would like to setup a network based "Interactive" and "autoyast" install of SLES10. My question is regarding package selection for interactive install. We generally do not install all the packages but make a list of all the packages, In sles9 there is an option of exporting the package list as a .sel file and then you can import the same file in your future installs. We then keep a copy of this file in sles9.0 install repository so if you are doing an interactive install of a machine then you can browse the user.sel and it imports your package selection. I do not see an option of importing/exporting of package selection in SLES10. I manually installed a SLES10 machine with all the packages we want. and then created a autoyast.xml file. Now, What I want is if I do an autoyast install of SLES10 then it should ask me all the install questions but all the packages should be selected from the autoyast.xml file I created . Thanks - Rajeev

2 1

[suse-autoinstall] SLES10 Kernel Selection
by hp9 18 Sep '06

18 Sep '06

I am installing SLES10 by autoyast. It is strange that when I install dual cores machine MANUALLY, the Kernel x86_64 had been installed. WHEN I install macine by using autoyast, the installed Kernel is i586 ! What's wrong? Any one had ecperience like that? Any good point to solve the problem?

1 0

Adding ZYPP repositories at install time
by Ian Grant 15 Sep '06

15 Sep '06

This is a little OT but I don't know where else to ask. In an autoyast script, how can I add a trusted key that rug/yast will use to verify signed ZYPP repositories? I am trying to add ZYPP repositories in an autoinstall script. (I can do it manually using YaST 'Installation Sources' and it asks me if I want to trust the key used to sign the repository. But I don't want to have to do this manually every time I install a machine.) The repository is one I have created myself using createrepo and signed as described here: http://en.opensuse.org/Secure_Installation_Sources#The_.22repomd.22_or_.22Y… To sign the repository I 'cd' to the top-level directory and ran gpg -a -u rpm-packager --detach-sign repodata/repomd.xml gpg -a --export rpm-packager > repodata/repomd.xml.key So I have the key in repodata/repomd.xml.key On the new machine (in an autoyast post-init script) I want to add the key to the rug keyring. I have tried gpg --homedir /root/.gnupg \ --import repodata/repomd.xml.key to add it to root's keyring, and I have tried adding the key to the zmd whitelist using rug ka packager 68C0D40964D6D632 "6E55 E6C6 8A0D 5557 C10D 571B 68C0 D409 64D6 D632" But neither work. When I run rug --quiet sa -t zypp \ nfs://repo.host/path/suse-10.1/extras \ SUSE-10.1-extras I get an error 'parsing metadata: Untrusted key rpm-packager ... (I am paraphrasing, I don't have a copy of the message and can't reproduce it without re-installing.) However, if I delete an installation source added using YaST 'Installation Source' graphical tool and then run the above 'rug sa' command it works, presumably because the key has been kept.

1 0

Problems with bootloader configuration in autoyast SLES10 RC3
by Matt Gillard 13 Sep '06

13 Sep '06

I have logged SR#10273247971 for this but I thought I would get peoples thoughts here as well. Problem is with the <default> global parameter in the <bootloader> section of autoyast. This works fine: <bootloader> <loader_type>lilo</loader_type> <location>mbr</location> </bootloader> Default kernel is SLES_10 as expected. This doesnt work, during boot loader installation an error is given that there is no default value given in lilo.conf. This occurs if I put an entry in the <default> tag or not. <bootloader> <global> <default>SLES_10</default> <gfxmenu></gfxmenu> <prompt>1</prompt> <timeout config:type="integer">5</timeout> </global> <loader_type>lilo</loader_type> <location>mbr</location> </bootloader> ie: This also gives the same error for the default tag. It should default to SLES_10 - I should not have to specify it. <bootloader> <global> <default></default> <gfxmenu></gfxmenu> <prompt>1</prompt> <timeout config:type="integer">5</timeout> </global> <loader_type>lilo</loader_type> <location>mbr</location> </bootloader> And this also the same error: <bootloader> <global> <gfxmenu></gfxmenu> <prompt>1</prompt> <timeout config:type="integer">5</timeout> </global> <loader_type>lilo</loader_type> <location>mbr</location> </bootloader> The reason I am doing this is because I do *not* want a graphical menu so I want to specify nothing for <gfxmenu> which seems to work ok. The text of the error during the bootloader config/installation phase is: "Value expected for 'default' at or above line 11 in file /etc/lilo.conf." And sure enough in /mnt/etc/lilo.conf at that point of the installation default has no value. thanks -- Matt Gillard UNIX Systems Engineer Coles Myer Ltd | Lvl 4 236 Bourke St | +61 3 963 51926 This email and any attachments may contain privileged and confidential information and are intended for the named addressee only. If you have received this e-mail in error, please notify the sender and delete this e-mail immediately. Any confidentiality, privilege or copyright is not waived or lost because this e-mail has been sent to you in error. It is your responsibility to check this e-mail and any attachments for viruses. No warranty is made that this material is free from computer virus or any other defect or error. Any loss/damage incurred by using this material is not the sender's responsibility. The sender's entire liability will be limited to resupplying the material.

4 13

hostname not set during autoinstall
by Hartmut Gottwald 13 Sep '06

13 Sep '06

Hi, I like to get the hostname passed from DHCP to the client during autoinstall. But whatever I do the hostname is set to "linux". However the IP was set correctly. Platform is IA64 and OS version SLES 10 A snippet from dhcpd.conf: ddns-update-style none; ignore unknown-clients; group { use-host-decl-names on; filename "elilo.efi"; subnet 192.168.72.0 netmask 255.255.255.0 { next-server 192.168.72.72; } host client02 { hardware ethernet 08:00:69:14:04:5E; fixed-address client02; } } ------ and from autoinstall.xml: <networking> <dns> <dhcp_hostname config:type="boolean">true</dhcp_hostname> <dhcp_resolv config:type="boolean">true</dhcp_resolv> <domain></domain> <hostname></hostname> </dns> ... ----- Tried several variations with dhcp_hostname/dhcp_resolv also set to false and <hostname></hostname> removed without success. Can't figure out what's wrong... can't even find documentation for dhcp_hostname/dhcp_resolv tags. Any ideas? Hartmut

3 5

How to make login service,rlogin, rsh work?
by hp9 11 Sep '06

11 Sep '06

I am installing SUSE 10.1 for a small cluster with autoyast. How to make login service,rlogin, rsh work without using yast1 manually? Hongbin

1 0

SLES10.1 autoinstall hangs at creating initrd
by hp9 11 Sep '06

11 Sep '06

Hi, I am trying to install SUSE 10.1 by network method and NFS and autoyast2 are fine. But SLES10.1 autoinstall hangs at creating initrd. THere is anyone had experience as that? It will be appreciated if anyone can provide ressolution for it. hongbin

3 2

bootloader configuration
by Karsten Künne 08 Sep '06

08 Sep '06

Hi, I try to configure the bootloader for SUSE 10.1 on a Sunfire X4100 with the following disk configuration: 2 SAS drives combined as a mirror via the SAS controller on SCSI id 0 2 other SAS drives on SCSI ids 2 and 3 In my autoyast profile I have the following section: <bootloader> <activate config:type="boolean">true</activate> <global config:type="list"> .... </global> <initrd_modules config:type="list"> <initrd_module> <module>mptsas</module> </initrd_module> <initrd_module> <module>scsi_mod</module> </initrd_module> .... </initrd_modules> <loader_device>/dev/sdc</loader_device> <loader_type>grub</loader_type> <location>mbr</location> <sections config:type="list"> <section config:type="list"> <section_entry> <key>title</key> <value>Linux</value> </section_entry> <section_entry> <key>kernel</key> <value>(hd2,4)/vmlinuz root=/dev/sdc7 vga=normal console=ttyS0,9600 console=tty0 noresume showopts</value> </section_entry> <section_entry> <key>initrd</key> <value>(hd2,4)/initrd</value> </section_entry> </section> <section config:type="list"> <section_entry> <key>title</key> <value>Failsafe</value> </section_entry> <section_entry> <key>kernel</key> <value>(hd2,4)/vmlinuz root=/dev/sdc7 vga=normal console=ttyS0,9600 console=tty0 noresume showopts ide=nodma apm=off acpi=off edd=off nosmp noapic maxcpus=0 3</value> </section_entry> <section_entry> <key>initrd</key> <value>(hd2,4)/initrd</value> </section_entry> </section> </sections> </bootloader> I want to install everything on the mirror which for whatever reason always appears as the last drive, even if it is SCSI id 0. But this seems to be a driver or controller issue and I can live with that. In my case the mirror is /dev/sdc resp. for grub it's (hd2). In the BIOS I explicitely set it to boot ONLY from the mirror and don't even look on the other drives. Now what happens is that Yast installs everything fine on the mirror but after the reboot the system doesn't come up anymore. The reason is that Yast never puts grub into the MBR on the mirror but instead puts it into the MBR of the first non-mirrored drive (which doesn't even have a valid partition table). This is the y2log_bootloader: GNU GRUB version 0.97 (640K lower / 3072K upper memory) [ Minimal BASH-like line editing is supported. For the first word, TAB lists possible command completions. Anywhere else TAB lists the possible completions of a device/filename. ] grub> root (hd2,4) Filesystem type is ext2fs, partition type 0x83 grub> install --stage2=/boot/grub/stage2 /grub/stage1 d (hd2,4) /grub/stage2 0x8000 (hd2,4)/grub/menu.lst grub> install --stage2=/boot/grub/stage2 /grub/stage1 d (hd0) /grub/stage2 0x8000 (hd2,4)/grub/menu.lst grub> quit Is that intentional behaviour and how can I get around it? Or is it just broken, in which case I can file a bug on bugzilla? Karsten. -- "The last time somebody said, `I find I can write much better with a word processor.', I replied, `They used to say the same thing about drugs.' -- Roy Blount, Jr.

1 1

Default installation encoding ISO-8895-1
by Klaus Jaensch 08 Sep '06

08 Sep '06

Hello, I want to set the default encoding of the installation to ISO-8859-1. I tried the following: <language>de_DE.ISO-8859-1</language> and <sysconfig config:type="list"> <sysconfig_entry> <sysconfig_key>RC_LANG</sysconfig_key> <sysconfig_path>/etc/sysconfig/language</sysconfig_path> <sysconfig_value>de_DE.ISO-8859-1</sysconfig_value> </sysconfig_entry> </sysconfig> but the encoding is finally always set to "de_DE.UTF-8". Any suggestions ? Klaus

1 0

WG: [suse-autoinstall] reboot after install
by Maurer, Andreas 08 Sep '06

08 Sep '06

Hello Simon, forceboot only forces the boot :) the reboot is done by: <reboot config:type="boolean">true</reboot> andreas ________________________________ Von: Simon Crute [mailto:scrute@novell.com] Gesendet: Do 07.09.2006 14:12 An: suse-autoinstall(a)suse.com Betreff: [suse-autoinstall] reboot after install Hi, I'm trying to get my autoyast setup to reboot after the install has finished, and then run the initscript scripts after the first boot. I've got the following section. <general> <mode> <confirm config:type="boolean">true</confirm> <forceboot config:type="boolean">true</forceboot> <second_stage config:type="boolean">true</second_stage> </mode> <mouse/> ... </general> but it's still not rebooting after an install. This causes me problems because the install kernel is a -default one, but I'm building -smp systems, so the initscrupt (which configures the sound card, and the display) fails cos there are no -default modules there, only -smp ones -- Simon Crute IS&T Bracknell Novell UK Ltd -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

4 12

Partitioning problem: /boot auto-added
by Frank Steiner 08 Sep '06

08 Sep '06

Hi, I'm trying to install SuSE 10.1 on a system with one SCSI disk. I used a profile to keep all existing partitions: /dev/sda1 / ext3 /dev/sda2 swap swap /dev/sda3 /var ext3 /dev/sda4 /export reiserfs The profile worked with SLES9, now fails with 10.1. While debugging I tried the same profile but toll AY to create all 4 partitions, just to see what comes out of it. Amazingly, AY creates this partitioning scheme: /dev/sda1 / ext3 /dev/sda2 /var ext3 /dev/sda3 extended partition /dev/sda5 swap swap /dev/sda6 /export reiserfs /dev/sda7 /boot ext2 A complete mess, of course (note that also swap and /var are swapped with the partition no.). Looking at y2log I find this: Partitions.ycp: BootCycl lba_support false Partitions.ycp: BootCycl /proc/xen/capabilities $[] Partitions.ycp: BootCycl lba_support false autoinstall/autoaprt.ycp: boot added automagically... So it looks like AY thinks that I need a special boot partition, which is nosense, of course. What can I do? Is this some known bug, or do I need to send the full y2log? The partitioning profile is below... cu, Frank <partitioning config:type="list"> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/</mount> <partition_nr config:type="integer">1</partition_nr> <size>20gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/var</mount> <partition_nr config:type="integer">3</partition_nr> <size>4gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>swap</mount> <partition_nr config:type="integer">2</partition_nr> <size>2gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/export</mount> <partition_nr config:type="integer">4</partition_nr> <size>max</size> </partition> </partitions> <use>all</use> </drive> </partitioning> -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

2 8

reboot after install
by Simon Crute 07 Sep '06

07 Sep '06

Hi, I'm trying to get my autoyast setup to reboot after the install has finished, and then run the initscript scripts after the first boot. I've got the following section. <general> <mode> <confirm config:type="boolean">true</confirm> <forceboot config:type="boolean">true</forceboot> <second_stage config:type="boolean">true</second_stage> </mode> <mouse/> ... </general> but it's still not rebooting after an install. This causes me problems because the install kernel is a -default one, but I'm building -smp systems, so the initscrupt (which configures the sound card, and the display) fails cos there are no -default modules there, only -smp ones -- Simon Crute IS&T Bracknell Novell UK Ltd

1 0

problems with language settings.
by Simon Crute 07 Sep '06

07 Sep '06

HI All, I'm having problems setting the language settings. I'm using this fragment of XML that's being pulled in via a rule. (basically, import a custom file depending on which domain the client is in) This fragment was created by yast2 autoyast. I just added the <general>tags round it. <?xml version="1.0"?> <!DOCTYPE profile SYSTEM "/usr/share/autoinstall/dtd/profile.dtd"> <profile xmlns="http://www.suse.com/1.0/yast2ns" xmlns:config="http://www.suse.com/1.0/configns"> <general> <keyboard> <keyboard_values> <capslock config:type="boolean">false</capslock> <delay></delay> <discaps config:type="boolean">false</discaps> <numlock></numlock> <rate></rate> <scrlock config:type="boolean">false</scrlock> <tty>tty1 tty2 tty3 tty4 tty5 tty6 tty8 tty9 tty10 tty11 tty12 tty13 tty14 tty15 tty16 tty17 tty18 tty19 tty20</tty> </keyboard_values> <keymap>english-uk</keymap> </keyboard> <language> <language>en_GB</language> <languages>en_US,en_GB</languages> </language> <timezone> <hwclock>UTC</hwclock> <timezone>Europe/London</timezone> </timezone> </general> </profile> However, the compleated system ends up with a US keyboard.... Interestingly, shortly after loading the xml fragment, the log file has 2006-08-17 17:01:55 <3> 192.168.21.139(3453) [Interpreter] Profile.ycp:118 Can't convert value '$["language":"en_GB", "languag es":"en_US,en_GB"]' to type 'string' in it. so, is it a bug, or user error ? ;-) -- Simon Crute IS&T Bracknell Novell UK Ltd

2 1

AutoYaST IP and partitioning problems
by Peter Ludikovsky 07 Sep '06

07 Sep '06

Good morning! I have some problems using AutoYaST to test the automatic installation I made from within VMware. But first things first. My target is to generate a bootable DVD that should be used to "reset" testing servers. The installation should run as quietly and automatic as possible from inserting the DVD and powering up the system to a complete system with Samba & Oracle. The configuration files can be retrieved at http://www.badongo.com/file/1308770 While the used linuxrc.config & info files should tell AutoYaST to use a static IP address (using VMwares emulated AMD PCnet32 card) it always tries to fetch information from a DHCP server, which is not present, or skips this step all together without BOOTP fallback. This is a major problem since the autoinstall directives should be fetched by HTTP. As soon as the installation proceeds past this it stops short at partitioning. I already tried using fresh disks (remember, still in VMware) and disks that had all required partitions written but to no avail. Is this a general problem with VMware or did I find a glitch? I would really like to be able to continue my test within VMware since I don't want to stress out the hardware now and have it break too soon. Regards, Peter

2 4

RAID1 md-device with "missing"-device
by Tob_Sch＠gmx.de 07 Sep '06

07 Sep '06

Hi, which entries in autoyast.xml do I need to configure a raid1 device with 2 devices, where one of these should be (at the beginning / installation time) a missing device. Sometimes, after the installation, I want to add a second HD and make a "mdadm -a". => so the functionality of "mdadm -C /dev/md0 -l raid1 -n 2 /dev/hda1 missing" and "mdadm -C /dev/md1 -l raid1 -n 2 /dev/hda2 missing" in autoyast is required Because, I didn't find anything about missing devices in the autoyast documentation, I tried to only change "/dev/hdb" to "missing", but that's not working... (see below) : : : <partitioning config:type="list"> <drive> <device>/dev/hda</device> <initialize config:type="boolean">true</initialize> <partitions config:type="list"> <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_type>primary</partition_type> <raid_name>/dev/md0</raid_name> <size>256M</size> </partition> <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_type>primary</partition_type> <raid_name>/dev/md1</raid_name> <size>max</size> </partition> </partitions> <use>all</use> </drive> <drive> <device>missing</device> <initialize config:type="boolean">true</initialize> <partitions config:type="list"> <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_type>primary</partition_type> <raid_name>/dev/md0</raid_name> <size>256M</size> </partition> <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_type>primary</partition_type> <raid_name>/dev/md1</raid_name> <size>max</size> </partition> </partitions> <use>all</use> </drive> </partitioning> <raid config:type="list"> <raid_device> <device_name>/dev/md0</device_name> <parity_algorithm>left-asymmetric</parity_algorithm> <persistent_superblock config:type="boolean">true</persistent_superblock> <raid_type>raid1</raid_type> <filesystem_id config:type="integer">131</filesystem_id> <chunk_size>4</chunk_size> <mount>/boot</mount> <format config:type="boolean">true</format> <filesystem config:type="symbol">xfs</filesystem> </raid_device> <raid_device> <device_name>/dev/md1</device_name> <lvm_group>vg00</lvm_group> <parity_algorithm>left-asymmetric</parity_algorithm> <persistent_superblock config:type="boolean">true</persistent_superblock> <raid_type>raid1</raid_type> <filesystem_id config:type="integer">142</filesystem_id> <chunk_size>4</chunk_size> </raid_device> </raid> : : : Thanks in advance. -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer

2 1

error in create_update_source.sh + fix
by Stan Novogroudski 06 Sep '06

06 Sep '06

Please fix it: in create_update_source.sh line 101 use if ! [[ "$VALUE" -eq "SHA1" ]]; then instead of if ! [[ "$VALUE" =~ "SHA1" ]]; then Operator =~ is used in Perl, not in shell. After this fix is create_update_source.sh working. Best regards. Stan

3 2

init-scripts without network config
by Paul DiSciascio 06 Sep '06

06 Sep '06

init-scripts don't seem to work if there is no network config in SLES 10. So i added a network section to my autoinst.xml. The problem with this is that the default for eth0 is dhcp, which is not allowed on my network, so i set bootproto to "none", which had absolutely no affect. So I removed the /etc/sysconfig/network/ifcfg-eth0-* file in my init-script and rebooted. The problem with this solution is that /etc/hosts and /etc/resolv.conf are already modified by the installer by this point (i would like that to not happen.) Is there a better way besides setting bootproto to "none" that will make the network subsystem come up, but with no devices configured? Or failing that, is there a way to make init-scripts work without the network dependency? Thanks, Paul

2 1

Preload modules before hardware detection? (wrong scsi order)
by Frank Steiner 05 Sep '06

05 Sep '06

Hi, similar post was sent in July with title "Problem with the order of disks", where Michael Schulz had the same problem we have: SCSI controllers are detected in the wrong order, so that the disk that should be /dev/sda ends up as /dev/sde. This used to work with SuSE 9.0 where we loaded the module for the first controller with "insmod gdth" passed to linuxrc (we are using pxeboot). Now, with 10.1, the init process of YA starts a "Hardware detection" *before* loading the modules defined in the info file or passed to the command line. This makes the insmod option for linuxrc more or less superfluous and removes any control over the order in which modules are loaded during the AY boot. Here, it makes the fc modules load before the scsi modules, thus moving our scsi disk to /dev/sde. Is it possible to disable automatic hardware detection or load the modules before the hardware detection? Or could I try thinks like renaming the init binary in the initrd and install my own script there that's just modprobe gdth exec init.suse Of course, I would prefer a method without changing the initrd if possible... cu, Frank -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

3 8

OpenSUSE 10.1 Online update in autoyast
by Ian Grant 01 Sep '06

01 Sep '06

Does anyone have XML for OpenSUSE 10.1 to cause autoyast to do an online update at install time? I am hoping this is better than the farce we have to go through at the moment to run it after the autoyast install: it has to be run, interactively, four times to get the security/recommended updates done. Failing that, does anyone know how to run it from the command line so it doesn't need somoeone to press keys?

6 7

RE: [suse-autoinstall] Preload modules before hardware detection? (wrong scsi order)
by Marlier, Ian 01 Sep '06

01 Sep '06

> -----Original Message----- > From: Frank Steiner [mailto:fsteiner-mail1@bio.ifi.lmu.de] > Sent: Friday, September 01, 2006 8:39 AM > To: suse-autoinstall(a)suse.com > Subject: [suse-autoinstall] Preload modules before hardware detection? > (wrong scsi order) > > Hi, > > similar post was sent in July with title "Problem with the order of disks", > where Michael Schulz had the same problem we have: > > SCSI controllers are detected in the wrong order, so that the disk that > should be /dev/sda ends up as /dev/sde. > > This used to work with SuSE 9.0 where we loaded the module for the first > controller with "insmod gdth" passed to linuxrc (we are using pxeboot). > Now, with 10.1, the init process of YA starts a "Hardware detection" > *before* loading the modules defined in the info file or passed to the > command line. > > This makes the insmod option for linuxrc more or less superfluous and > removes any control over the order in which modules are loaded during > the AY boot. Here, it makes the fc modules load before the scsi modules, > thus moving our scsi disk to /dev/sde. > > Is it possible to disable automatic hardware detection or load the modules > before the hardware detection? Or could I try thinks like renaming the > init binary in the initrd and install my own script there that's just > > modprobe gdth > exec init.suse > > Of course, I would prefer a method without changing the initrd if > possible... > > cu, > Frank I spent a bunch of time going around and around with this basic problem -- a machine with two SATA controllers, and the wrong one was being detected first. (System array is on a Silicon Image controller, Storage array is on a 3ware card -- and detection happens in alphabetical order). In the end, I found it easier to make a chroot-script that basically just does this: `sed -i "s/3w_9xxx sata_sil/sata_sil 3w_9xxx/g" /etc/sysconfig/kernel` `mkinitrd` (replace the module names, of course, with the ones relevant to your environment.) This just takes advantage of the fact that, by the time chroot scripts run, the sysconfig/kernel file is already populated (having just been used to build the "broken" initrd). The trick, here, is that you're installing to /dev/sde, but it's going to be /dev/sda when the system boots. The bootloader section, at least, needs to be adjusted to handle that case. I'm a little fortunate in that the sed regex is always the same in my environment. A more generalized version, though, could be done with something like `sed -i "s/3w_9xxx$.*$sata_sil/sata_sil\13w_9xxx/g" /etc/sysconfig/kernel` Anyway...your mileage with my hacks may vary, but hope this helps a little... - Ian

3 2

RE: [suse-autoinstall] OpenSUSE 10.1 Online update in autoyast
by hp9 01 Sep '06

01 Sep '06

Man! I also had very bad experience as you. It seems that SUSE Linux 10.1 autoyast does not work. It failed in partition and software Package Selections. Anyone has succeeded about online installation with that? >===== Original Message From gordon_schumacher(a)maxtor.com ===== >Ian Grant <Ian.Grant(a)cl.cam.ac.uk> wrote on 08/31/2006 03:33:14 AM: > ># Does anyone have XML for OpenSUSE 10.1 to cause autoyast to do an ># online update at install time? I am hoping this is better than the ># farce we have to go through at the moment to run it after the ># autoyast install: it has to be run, interactively, four times to get ># the security/recommended updates done. > > <script> > <filename>3-start_you.sh</filename> > <interpreter>shell</interpreter> > <source><![CDATA[ >#!/bin/sh >/bin/chvt 9 >exec < /dev/tty9 > /dev/tty9 >echo Performing online update... >online_update -l en -u http://my-system/srv/suseupdate -V >SuSEconfig >echo Rebooting the computer now to use the new kernel. >sleep 5 >/bin/chvt 1 >shutdown -r now >]]> > </source> > </script> > </init-scripts> > </scripts> > > >-- >To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com >For additional commands, e-mail: suse-autoinstall-help(a)suse.com

2 1

Not understanding rules.xml and autoinst.xml interaction
by John Comitas 01 Sep '06

01 Sep '06

I am running Suse Enterprise 9-1 I'm trying to use rules.xml to create partitions, but it never uses it, only uses autoinst.xml, which has nothing in the partition declaration. John Comitas Sr. Network Engineer

2 1