October 2006 - openSUSE Autoinstall - openSUSE Mailing Lists

Repository in DVD
by Jonathan Meller 23 Oct '06

23 Oct '06

Hellos! Can I copy the repository tree to a DVD and install from it? -- Jonathan Meller jonathan.meller(a)tteng.com.br ======================================================= Tools & Technologies - T&T Engenheiros Associados Ltda. Projetos de SW - Desenvolvendo soluções. Rua Riachuelo, 1098 Conj. 901 Centro - Porto Alegre - CEP 90010-272 Fone DDR: (51)3220-3260 Fone/Fax: (51)3220-3220 / 3220-3206 http://www.tteng.com.br/ tet(a)tteng.com.br =======================================================

1 0

lilo on ppc/IBM pSeries fails
by Frank Steiner 17 Oct '06

17 Oct '06

Hi, after I fixed that problem in my profile with the prep partition, Installation of SLES10 ppc on an IBM pSeries ran through, but setting up lilo fails. AY runs through, no error message, but the system didn't boot (different error messages on boot: prompt). I set up a chroot postscript with a sleep command so that I could login into the installing system with ssh, and found the following /mnt/etc/lilo.conf: # Modified by YaST2. Last modification on Mo Okt 16 10:37:54 UTC 2006 menu-scheme = Wb:kw:Wb:Wb timeout = 80 lba32 change-rules reset read-only prompt default = SLES_10 boot = /dev/sda image = /boot/vmlinux initrd = /boot/initrd ###Don't change this comment - YaST2 identifier: Original name: linux### label = SLES_10 root = /dev/sda3 image = /boot/vmlinux ###Don't change this comment - YaST2 identifier: Original name: failsafe### label = Failsafe append = " 3" initrd = /boot/initrd root = /dev/sda3 This can't work because there are many parameters that are not known on ppc. Indeed calling lilo in a chroot env returns: 141:/ # lilo ERROR: !!!!!!!!!! unkown option menu-scheme !!!!!!!!!!!!! and so on. After I removed all unknown options, it warned me that the "activate" entry was missing: 141:/ # lilo running on chrp guessing the chrp boot device Boot target is /dev/sda1 Warning: Extend PReP boot partition /dev/sda1 to hold yaboot 1+0 records in 1+0 records out 4096 bytes (4.1 kB) copied, 0.012297 seconds, 333 kB/s Installing /lib/lilo/chrp/yaboot.chrp onto /dev/sda1 Converted /etc/lilo.conf to /etc/yaboot.conf Warning: boot path not set to boot linux as requested Warning: please add '/pci@3fffdf0a000/pci@2/pci1069,b166@1/scsi@0/sd@8,0' somewhere to your boot path manually It looks like AY sets up a lilo.conf for an x86 system and not for PPC, but since no error message is returned in AY (it should if calling lilo fails, shouldn't it?), the system reboots into an unbootable state. Is this known and is there a fix for it already? You can find the log here: http://www.bio.ifi.lmu.de/~steiner/y2log.ppc2.gz 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 5

Hi,
by Jochen Schaefer 16 Oct '06

16 Oct '06

Hi, I'm trying to do a network installation of SLEDx86_64 to a vmware 64Bit-instance. The installation source is originated from a September 2006 EVAL DVD. Unfortunately it does not work for x86_64: access.log: "GET /inst/streams/sled10/x86_64/boot/i386/root HTTP/1.0" 404 1044 "-" "-" The directory boot/i386 does not exists neither on the x86_64 DVD nor on my instserver instead and like expected boot/x86_64 directory exists Creating a Link i386->x86_64 didn't work too. The process stoped some GET-Requests later. Wenn installing from i386 source everything is fine: "GET /inst/streams/sled10/i386/boot/i386/root HTTP/1.0" 200 84094976 "-" "-" Installing from DVD worked for both architectures. Is this a broken DVD-Set? Jochen -- Jochen Schaefer Senior Consultant Integrated Services Novell GmbH Mobil: +49(0)1735876969 E-mail: jschaef(a)novell.com Address: Duesseldorf , 40470, Germany Noerdlicher Zubringer 9-11 http://www.novell.com

1 0

RE: [suse-autoinstall] can not find installation source
by GUNREBEN, PETER (Peter) 13 Oct '06

13 Oct '06

Lisa, Lisa Dong wrote: > On Wed, 11 Oct 2006, GUNREBEN, PETER (Peter) wrote: > > Lisa Dong wrote: > > > > > when I boot up the system, I see the the system request > > > IP address from the dhcp server,and dhcp offer the > > > IP address to that mac address, it then load the kernel, > > > .....start hardware detect, usb device detect, then > > > try to load info from eth0 eth1 > > > then stop right here, a red screen appear, can not find > > > installation source [...] > > Additionally, I had to add the kernel module for the NIC > > (in my case: "insmod=tg3") and also some SATA and SCSI > > modules to get the installation working. You may also > > try to set "netdevice=eth0". > > > > BTW, when the red screen appears, you may press > > <Ctrl>-<Alt>-F9 and try to mount the NFS installation > > source manually, just to see whether this works. > I did try netwait=30, netdevice=eth0, and tried manually > mount of installation source, none of them work. In the <Ctrl>-<Alt>-F9 console: Have you tried to "insmod" your NIC driver? What does "ifconfig" tell you? Can you "ping" any computer in your subnet? Peter.

3 2

SLES 10 autoinstall not working with serial console
by Stefan Hinker - Sun Solution Center - Benchmark Center Germany 13 Oct '06

13 Oct '06

Hi All, I have a working configuration of autoyast/autoinstall for SLES10. This configuration breaks (stops with the "An error occurred during the installation" message) if I include the option "console=ttyS0" in the linuxrc parameters. The last error message I see on syslog is: yast2[3711]: <5> [ui-ncurses] YNCursesUI.cc(YNCursesUI):101 NCurses: (-1) newterm() failed Has anyone seen this before? Is there a fix/workaround available? Should I log a bug against this? Any reply appreciated. Config files and syslog output available on request. stefan -- Stefan Hinker Sun Solution Center - Benchmarking Tel: +49 6103 752-300 Sun Microsystems GmbH Fax: +49 6103 752-299 Amperestr. 6 Stefan.Hinker(a)Sun.COM D-63225 Langen http://www.sun.com/solutioncenters/

2 4

RE: [suse-autoinstall] can not find installation source
by GUNREBEN, PETER (Peter) 11 Oct '06

11 Oct '06

Lisa, Lisa Dong wrote: > when I boot up the system, I see the the system request IP > address from the dhcp server,and dhcp offer the IP address to > that mac address, it then load the kernel, .....start > hardware detect, usb device detect, then try to load info from eth0 > eth1 > > then stop right here, a red screen appear, can not find > installation source > > Here is my info file > > default linux > serial 0,9600n8 > label linux > kernel 10.1kernel > append console=tty0 initrd=10.1initrd.img insmod=mptsas > autoyast=http://10.128.128.11/garibaldi_node.xml netmask=255.255.240.0 > install=nfs://10.128.128.11/export/dist/SuSe10.1 ^^^^ This is not your info file, is it? For my Broadcom NIC, the netwait=5 was not enough. I'm actually using netwait=10. Additionally, I had to add the kernel module for the NIC (in my case: "insmod=tg3") and also some SATA and SCSI modules to get the installation working. You may also try to set "netdevice=eth0". BTW, when the red screen appears, you may press <Ctrl>-<Alt>-F9 and try to mount the NFS installation source manually, just to see whether this works. Peter.

2 1

can not find installation source
by Lisa Dong 11 Oct '06

11 Oct '06

my install server server(dhcp server, tftp server and nfs server) has two network interface, one for public, one for internal, the autoinstallation for public network works, but the installation for internal network doesn't work, the error messages are "can not find installation source", when I tried to do the internal network autoinstall, I change dhcp server and control file in ~pxelinux.cfg to the internal network settings(using internal IP....) the contents in the /etc/exports file is export *(ro,all_squash,async,insecure) I can mount this nfs file from another internal machine, but it can't mount for internal autointall. Any suggestion is appreciated. Lisa

2 3

Partitioning issues with pSeries
by Frank Steiner 10 Oct '06

10 Oct '06

Hi, I'm trying to install SLES10 on an IBM pSeries p655. I use the same partitioning scheme that I used successfully for SLES10 in i586, just that I need a special prep-type partition as /dev/sda1. However, partitioning fails. Looking at y2log, it seems that AY creates an additional primary partition with ext2 or sth. The log file is here: http://www.bio.ifi.lmu.de/~steiner/y2log.ppc.gz And I used the following profile (derived from SLES9 and the working SLES10 i586 profiles). Any ideas what I'm doing wrong? cu, Frank <?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> <global> </global> <loader_device>/dev/sda</loader_device> <loader_type>lilo</loader_type> <location>custom</location> </bootloader> <partitioning config:type="list"> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">prep</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>prep</mount> <partition_nr config:type="integer">1</partition_nr> <partition_type>primary</partition_type> <size>1</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> <partition_type>primary</partition_type> <size>2gb</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>/</mount> <partition_nr config:type="integer">3</partition_nr> <partition_type>primary</partition_type> <size>15gb</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>/local</mount> <partition_nr config:type="integer">5</partition_nr> <size>8gb</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">6</partition_nr> <size>5gb</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>/usr/local/storage</mount> <partition_nr config:type="integer">7</partition_nr> <size>max</size> </partition> </partitions> <use>all</use> </drive> </partitioning> </profile> -- 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 4

how to test without yast
by Arvid Walter 10 Oct '06

10 Oct '06

Hi, I have still problems understanding the results of merging mutible configurations. As far as I know combining classes is supposed to be a merge between different classes of only one member of a class each. merge Software/nagios.xml <-------> Network/primergy.xml I use a class 'Software' and have several xmls, nagios.xml, vmware.xml, base.xml e.g. residing in the corresponding directory. Unfortunately the packages and scripts section will be merged althought I use the dont_merge element. To understand the way of merging it is helpfull to test without a complete installation process. How can I achive that using xsltproc e.g. ? p.s. I dont want to use yast for a test

2 1

AutoYaST and LVM hassles?
by David J N Begley 09 Oct '06

09 Oct '06

Is anyone using AutoYaST to build/rebuild machines using LVM when those machines may already have LVs/VGs on the existing hard disks? I'm currently hitting an annoying problem with 9.2 and was wondering if anyone using a newer version of SuSE can verify whether or not this particular problem has been fixed. Essentially, it's the same problem previously reported here: http://lists.suse.com/archive/suse-autoinstall/2006-Mar/0005.html Unfortunately the recommended fix (using a pre-script instead of initialize=true) isn't working for me; I've tried using parted to "mklabel msdos" on the disks, I've even tried manually "lvremove" followed by "vgremove" and then parted's "mklabel msdos" (all according to the logs, have been successful) - all these in a pre-script, but the dreaded "lvremove" error keeps reappearing. It's as if AutoYaST has learnt the LVs very early on in the boot process and remembers them up to that point (despite the fact that the pre-script has removed them by this time), then insists on trying to remove them _after_ having already removed all the partitions (rather than the reverse). I've managed to work around just about every other bug/problem I've so far hit with AutoYaST, except this one. If someone can confirm that this is fixed in newer versions then I'll just sit tight 'til 10.2 is released. :-) Thanks..

2 2