February 2006 - openSUSE Autoinstall - openSUSE Mailing Lists

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

Problem with AutoInstallRules.ycp and merge.xslt
by Luzian Scherrer 16 Mar '06

16 Mar '06

Hello, I think there is a typo in AutoInstallRules.ycp (version 1.34.4.9 that comes with SLES9 SP2): the xsltproc to merge classes is called with the following param there (line 733): --param dontmerge \"'package, addon'\" and I'm quite sure that it sould be: --param dontmerge \"'package'\" --param dontmerge2 \"'addon'\" The result of this typo is that neither <package> nor <addon> are correctly aggregated, which is the desired effect I believe and I also hope. Quick demonstration: idhpb02:/ # cat /tmp/a.xml <?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"> <install> <software> <addons config:type="list"> <addon>addon-A</addon> </addons> <packages config:type="list"> <package>package-A</package> </packages> </software> </install> </profile> idhpb02:/ # cat /tmp/b.xml <?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"> <install> <software> <addons config:type="list"> <addon>addon-B</addon> </addons> <packages config:type="list"> <package>package-B</package> </packages> </software> </install> </profile> Calling it the wrong way: idhpb02:/ # xsltproc --novalid --param replace "'false'" --param dontmerge "'package, addon'" --param with "'/tmp/b.xml'" /usr/share/ autoinstall/xslt/merge.xslt /tmp/a.xml <?xml version="1.0"?> <ns1:profile xmlns:ns1="http://www.suse.com/1.0/yast2ns" xmlns:config="http://www.suse.com/1.0/configns" xmlns="http:// www.suse.com/1.0/yast2ns"> <ns1:install> <ns1:software> <ns1:addons config:type="list"> <ns1:addon>addon-B</ns1:addon> </ns1:addons> <ns1:packages config:type="list"> <ns1:package>package-B</ns1:package> </ns1:packages> </ns1:software> </ns1:install> </ns1:profile> Calling it the (as I think) right way: idhpb02:/ # xsltproc --novalid --param replace "'false'" --param dontmerge "'package'" --param dontmerge2 "'addon'" --param with "'/ tmp/b.xml'" /usr/share/autoinstall/xslt/merge.xslt /tmp/a.xml <?xml version="1.0"?> <ns1:profile xmlns:ns1="http://www.suse.com/1.0/yast2ns" xmlns:config="http://www.suse.com/1.0/configns" xmlns="http:// www.suse.com/1.0/yast2ns"> <ns1:install> <ns1:software> <ns1:addons config:type="list"> <ns1:addon>addon-A</ns1:addon><ns1:addon>addon-B</ns1:addon> </ns1:addons> <ns1:packages config:type="list"> <ns1:package>package-A</ns1:package><ns1:package>package-B</ ns1:package> </ns1:packages> </ns1:software> </ns1:install> </ns1:profile> Could this be fixed? Or is it already known and possibly a workaround in existance? Thanks in advance, -Luzian

4 6

Changing Kernel in pxe install
by Reeves, Matthew J. 01 Mar '06

01 Mar '06

Classification: UNCLASSIFIED I'm trying to set up a pxe environment to install SuSE 9.0 on some Dell Poweredge 1850s. The kernel that comes with 9.0 doesn't support the raid controllers. I want to change the installing kernel to 2.4.32 so that the raid works. I can get the systems booting and linuxrc runs but it asks for a module disk then goes into manual mode. In the manual mode I can kick off a ftp install but after root is downloaded the system just goes back to the linuxrc menu. Is there an easy way to change the kernel? Thanks, Matt

3 2

Standard software installation using autoyast.xml
by Trey Sizemore 28 Feb '06

28 Feb '06

I'm wondering if an autoyast.xml file could be used solely for designating software packages to be installed. In other words, for a minimal server install, I want the same packages installed on those servers regardless of the hardware present (one may be a Proliant server and another an Optiplex). But I *do* want packages X, Y, and Z installed on both. Is this possible? Partitioning wouldn't appear to work because the disk sizes might be different in the machines, but I want to update the process a much as possible. -- Cheers, Trey ---- Help fight continental drift. Linux chameleon 2.6.13-15-default i686 GNU/Linux 12:30pm up 4 days 18:39, 7 users, load average: 0.02, 0.35, 0.45

2 1

SuSE 10.0 yast2 Autoinstall Modul
by Folkhard Mueller 28 Feb '06

28 Feb '06

Dear Listmembers, Has anybody out there used the builtin autoinstall tool in yast2 SuSE Professional Vers. 10.0, unpatched, and created some useful operable control Files with it? I did it under Suse Version 9.1 successfully (after some hacking), but under Suse 10.0 all xmls created are garbage more or less! Especially for Package Selection nothing works really! Any comments are welcome! Beste Regards F.Mueller -- |\ _,,,---,,_ ZZZzz /,`.-'`' -. ;-;;,_ _______|,4- ) )-,_. ,\ (____-'_________________________________________ '---''(_/--' `-'\_) Dipl. Phys. F.Mueller EPCOS AG Surface Acoustic Wave Components Division SAW RD CAD P.O. Box 80 17 09 Phone: ++49 89 636 23316 D-81617 Munich Fax: ++49 89 636 27304 Germany mailto: Folkhard.Mueller(a)epcos.com Web: http://www.epcos.com

2 1

AW: [suse-autoinstall] SuSE 10.0 yast2 Autoinstall Modul
by KOEHLER, SVEN - PFHO 28 Feb '06

28 Feb '06

hi folkhard, > [...]and created some useful operable control Files with it? i did so. at least during my first tries. i had to edit the control file as there were some inconviniences regarding my actual tasks but i think the software selection was one of the few corners, i never had problems with. if you need help from this list you may wan't to concretize your problems greetz sven

1 0

forcing reboot after autoyast installation stage 1
by Kai Dusek 27 Feb '06

27 Feb '06

Dear all, I want to install a SLES9 SP2 or SP3 Linux with autoyast on a z/VM guest without any manual intervention. Unfortunately the installation is always stopping after stage 1 (package installation, preparing bootloader, ...) with an unclear error message on my 3270. I have to re-IPL my system manually to continue with stage 2 (post installation, suse-config etc.). That's output to 3270: [...] umount: /y2update: not mounted umount: /var/adm/mount: not mounted Feb 15 19:48:19 suse sshd¬322|: Received signal 15; terminating. [...] bash: no job control in this shell SuSE Instsys boeid120:/ # [...] My parmfile looks like: ramdisk_size=65536 root=/dev/ram1 ro init=/linuxrc dasd=150,151 TERM=dumb AUTOYAST=http://<IP>/autoinst.xml AUTOINSTALL=yes IP_INTERFACE=qeth IP_HOST=<hostname> IP_ADDR=x.x.x.x IP_NETMASK=x.x.x.x IP_BROADCAST=x.x.x.x IP_GATEWAY=x.x.x.x IP_DNS=x.x.x.x IP_SEARCH=x.x.x.x READ_DEVNO=7000 WRITE_DEVNO=7001 DATA_DEVNO=7002 PORTNAME=xxxxxx INST_INFO=ftp INST_IP_ADDR=x.x.x.x INST_IP_DIR=/SLES/sles9_s390_sp3 FTPUSER=<user> FTPPASSWORD=<pw> INST_SCREEN=VNC VNC_PASSWORD=<pw> IP_MTU=1492 INST_PASSWORD=<pw> The forceboot option is set to true. Kind regards Kai Dusek

1 0

RE: [suse-autoinstall] suse 10.0 nfs install
by Gunreben, Peter (Peter) 27 Feb '06

27 Feb '06

Roger, Roger Oberholtzer wrote: > On Wed, 2006-02-22 at 17:51 +0100, Gunreben, Peter (Peter) wrote: > > Roger, > > > > Roger Oberholtzer wrote: > > > > > > The install kernel boots. At some point it wants to get a > DHCP address > > > for the ethernet card. This 'should' be working as it got > an address > > > when it first booted. I see this in the server's > > > /var/log/messages. But > > > this seems to timeout. My dhcp entry is: > > > > > > host jboc1 { > > > > > > filename "/suseInstall/pxelinux.0"; > > > fixed-address 192.30.105.204; > > > hardware ethernet 00:30:48:56:38:B2; > > > next-server 192.30.105.201; > > > option host-name "jboc1"; > > > } > > > > > > 192.30.105.201 is the server for everything (dhcp, pxe, > suse install) > > > > > > After this, I get a message that the SUSE installation source > > > cannot be > > > found. When trying to set it up by hand (nice it does that), > > > I see that > > > the default values it offers are the ones I expect (despite > > > the probable > > > DHCP oddness). But when it tries to mount the nfs install > source, it > > > fails with 'unknown error -1'. > > > > Have you checked whether the nfs-export works properly, e.g. > > by mounting it manually? I had a similar problem, where > > I tried to export a loop-mounted directory tree. Only, > > a userspace-nfsserver can do that. > > I can mount it from other systems. Using > > mount -t nfs source:/vol1/distro/OpenSUSE10.0/suse10.0 /tmp/vv > > I can read files and go into directories. It just got back to my mind that I had an additional problem. For any reason I still do not know, I had to add insmod=<my_NIC_kernelmodule> to the kernel parameters in order to get everything work. Of course, you need to replace <my_NIC_kernelmodule> with the one you need. May be it's worth to give it a try. Peter.

6 17

how to autoinstall RAID spares disks ?
by rejaine＠bhz.jamef.com.br 24 Feb '06

24 Feb '06

Hi list, I need to install 4 RAID disks and 1 SPARE disk (backup disk) using autoinstall, but I don't know how ... My /etc/raidtab wil have to be like this: raiddev /dev/md0 raid-level 5 nr-raid-disks 4 ----> nr-spare-disks 1 persistent-superblock 1 parity-algorithm left-asymmetric chunk-size 128 device /dev/sda5 raid-disk 0 device /dev/sdb5 raid-disk 1 device /dev/sdc5 raid-disk 2 device /dev/sdd5 raid-disk 3 ---> device /dev/sde5 ---> spare-disk 0 raiddev /dev/md1 raid-level 5 nr-raid-disks 4 ----> nr-spare-disks 1 persistent-superblock 1 parity-algorithm left-asymmetric chunk-size 128 device /dev/sda5 raid-disk 0 device /dev/sdb5 raid-disk 1 device /dev/sdc5 raid-disk 2 device /dev/sdd5 raid-disk 3 ---> device /dev/sde5 ---> spare-disk 1 etc... I try to modify <raid_type>spare</raid_type) instead of <raid_type>raid</raid_type> like this: <drive> <device>/dev/sde</device> ... snip... <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_nr config:type="integer">5</partition_nr> -----> <raid_name>/dev/md0</raid_name> -----> <raid_type>spare</raid_type> <size>30G</size> </partition> <partition> <format config:type="boolean">false</format> <partition_id config:type="integer">253</partition_id> <partition_nr config:type="integer">6</partition_nr> ----> <raid_name>/dev/md1</raid_name> ----> <raid_type>spare</raid_type> <size>21G</size> </partition> etc.. But don't work... I'm using SLES9 (whith service pack2) and Suse Linux 10 How can I configure this??? (sorry by bad english)

1 0

How to make blocking init-scripts?
by T. Ribbrock 23 Feb '06

23 Feb '06

Hi all, I just ran into some annoyance with the initial scripts that the autoinstall can perform: One of the tasks I do via such a script is the installation of all current updates and of some additional packages from an additional source. This does take a good while (depending on the machine I'm installing on - we still have some PIIIs running here...). However, I noticed that the machine will happily display the login screen even though this task has not finished yet. I find this undesirable, as i want the updates and installations to be finished before anything is started by a user (or root). Is it possible to write an init-script in a way that will make the machine wait for the script to finish before displaying the login screen? Cheerio, Thomas

4 6

adding custom packages - could not set add-on selection error
by Dave 23 Feb '06

23 Feb '06

Hello, Please excuse me if this question has been answered before, but I've read the documentation, and after following the steps about adding custom packages, we are still having difficulties. I also read the FAQ but didn't see this problem addressed there. Unfortunately, the list archives are not searchable so I'm not really able to go through them all to see if this has been answered. Here's the goal: + Have AutoYAST install additional RPM packages that I specify in an ".sel" file. Here's what we tried: according to http://www.suse.de/~nashif/autoinstallation/9.1/html/CreateProfile.Software… + created custom-9.3-83.noarch.sel in /9.3/suse/setup/descr and added package names to "+Ins:" section + added custom-9.3-83.noarch.sel entry to selections file in /9.3/suse/setup/descr + added custom-9.3-83.noarch.sel entry to directory.yast file in /9.3/suse/setup/descr + updated MD5SUMS file in /9.3/suse/setup/descr to reflect new MD5's for files directory.yast, selections, and custom-9.3-83.noarch.sel + modified control file to included an addon directive <addon>custom</addon> Here's the error we get during the software selection phase of AutoInstall: "Could not set add-on selection: custom" I can send the files along as attachments if that is helpful in diagnosing the problem. If someone knows the solution, we would love to hear it! Thanks, Dave __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

3 3

Bug? "Adding complete configurations"
by KOEHLER, SVEN - PFHO 23 Feb '06

23 Feb '06

List, the configuration files added in the control file get stored in /var/adm/files/ but if i'm adding two php.ini's (/etc/php5/apache2/php.ini and /etc/php5/cli/php.ini) only one appears in a fully installed system. does the first of them get overwritten? as far as i can see is the content of the files in etc correct. greetz sven

1 0

Re: [suse-autoinstall] autoyast - defining additional installation sources
by Yan Fitterer 23 Feb '06

23 Feb '06

> If I am correct the yast/order and yast/instorder files list directories > relative to the network boot location - still searching for a description of Yes. The yast directory needs to be in the directory you specify as the installation source. > how yast/order and yast/instorder are treated. Here's a typical content: # cat yast/instorder /SUSE-SLES-9-Service-Pack-Version-1/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 # cat yast/order /SUSE-SLES-9-Service-Pack-Version-1/CD1 /SUSE-SLES-9-Service-Pack-Version-1/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 /SUSE-CORE-Version-9/CD1 Yan

2 1

Re: [suse-autoinstall] adding custom packages - could not set add -on selection error
by KOEHLER, SVEN - PFHO 23 Feb '06

23 Feb '06

hi dave, just a hint: the list is searchable through google. you just have to add an "site:http://lists.suse.com/archive/suse-autoinstall/" after your search keywords. this works with this list as well as with all other sites indexed by google. unfortunately i can't tell anything regarding your problem greetz sven -----Ursprüngliche Nachricht----- Von: Dave [mailto:davo_muc@yahoo.com] Gesendet: Donnerstag, 23. Februar 2006 12:19 An: suse-autoinstall(a)suse.com Betreff: [suse-autoinstall] adding custom packages - could not set add-on selection error Hello, Please excuse me if this question has been answered before, but I've read the documentation, and after following the steps about adding custom packages, we are still having difficulties. I also read the FAQ but didn't see this problem addressed there. Unfortunately, the list archives are not searchable so I'm not really able to go through them all to see if this has been answered. Here's the goal: + Have AutoYAST install additional RPM packages that I specify in an + ".sel" file. Here's what we tried: according to http://www.suse.de/~nashif/autoinstallation/9.1/html/CreateProfile.Software. html#id2515639 + created custom-9.3-83.noarch.sel in /9.3/suse/setup/descr and added + package names to "+Ins:" section + added custom-9.3-83.noarch.sel entry to selections file in + /9.3/suse/setup/descr added custom-9.3-83.noarch.sel entry to + directory.yast file in /9.3/suse/setup/descr updated MD5SUMS file in + /9.3/suse/setup/descr to reflect new MD5's for files directory.yast, selections, and custom-9.3-83.noarch.sel + modified control file to included an addon directive + <addon>custom</addon> Here's the error we get during the software selection phase of AutoInstall: "Could not set add-on selection: custom" I can send the files along as attachments if that is helpful in diagnosing the problem. If someone knows the solution, we would love to hear it! Thanks, Dave __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

2 1

RE: [suse-autoinstall] suse 10.0 nfs install
by Gunreben, Peter (Peter) 22 Feb '06

22 Feb '06

Roger, Roger Oberholtzer wrote: > > The install kernel boots. At some point it wants to get a DHCP address > for the ethernet card. This 'should' be working as it got an address > when it first booted. I see this in the server's > /var/log/messages. But > this seems to timeout. My dhcp entry is: > > host jboc1 { > > filename "/suseInstall/pxelinux.0"; > fixed-address 192.30.105.204; > hardware ethernet 00:30:48:56:38:B2; > next-server 192.30.105.201; > option host-name "jboc1"; > } > > 192.30.105.201 is the server for everything (dhcp, pxe, suse install) > > After this, I get a message that the SUSE installation source > cannot be > found. When trying to set it up by hand (nice it does that), > I see that > the default values it offers are the ones I expect (despite > the probable > DHCP oddness). But when it tries to mount the nfs install source, it > fails with 'unknown error -1'. Have you checked whether the nfs-export works properly, e.g. by mounting it manually? I had a similar problem, where I tried to export a loop-mounted directory tree. Only, a userspace-nfsserver can do that. Peter.

2 1

suse 10.0 nfs install
by Roger Oberholtzer 22 Feb '06

22 Feb '06

Are there any known issues with using the OpenSUSE 10.0 original distro to do an NFS install to a PXE-booted client? Any show stopping bugs that require an update? I have set up dhcp ok, pxelinux.0 runs and uses my default file. All options as described in the SUSE online docs have been used. But I do not quite get the install to start: The install kernel boots. At some point it wants to get a DHCP address for the ethernet card. This 'should' be working as it got an address when it first booted. I see this in the server's /var/log/messages. But this seems to timeout. My dhcp entry is: host jboc1 { filename "/suseInstall/pxelinux.0"; fixed-address 192.30.105.204; hardware ethernet 00:30:48:56:38:B2; next-server 192.30.105.201; option host-name "jboc1"; } 192.30.105.201 is the server for everything (dhcp, pxe, suse install) After this, I get a message that the SUSE installation source cannot be found. When trying to set it up by hand (nice it does that), I see that the default values it offers are the ones I expect (despite the probable DHCP oddness). But when it tries to mount the nfs install source, it fails with 'unknown error -1'. Why is it looking for a DHCP server when I add the hostip= command to the kernel command line? This option seems to be ignored. -- Roger Oberholtzer OPQ Systems AB Ramböll Sverige AB Kapellgränd 7 P.O. Box 4205 SE-102 65 Stockholm, Sweden Tel: Int +46 8-615 60 53 Fax: Int +46 8-31 42 23

2 3

autoyast - defining additional installation sources
by Andrew Bramble 22 Feb '06

22 Feb '06

Hello , Firsttime poster. I'm super impressed by yast autoinstall - having got it operating nicely with pxe in almost no time. One thing I cannot discover though is how to define additional sources via an autoinstall.xml profile. Reading the DTD's has not enlightened me either. Is there any way to define more installation sources for autoyast? AB

2 4

RE: [suse-autoinstall] Installing multiple SuSE versions over NFS using Autoyast
by Gunreben, Peter (Peter) 22 Feb '06

22 Feb '06

Richard, Richard Hobbs wrote: > We currently have one DHCP server, one TFTP server and one > NFS server. Using > these, we install 64-bit SuSE 10.0 onto all of our 64-bit machines. > > However, we will soon need to start installing 64-bit SuSE onto some > machines and 32-bit SuSE onto other machines. > > As I understand it, the installation source (in this case the > 64-bit DVD > files) is specified in "/tftpboot/pxelinux.cfg/default", and > while I can > change this to the 32-bit DVD, I can't see an obvious way of > configuring > both. The DHCP Server specifies the TFTP server to boot from, > and specified > the kernel to boot from (in this case > "tftpserver:/tftpboot/pxelinux.0"). Hmmm, I'm not sure whether I really got your problem. When the DHCP server knows the MAC-adresses of the machines it can also use a specific tftp-server and/or kernel/initrd. Just boot the 32-bit computers with kernel/initrd and the 64-bit computers with kernel64/initrd64. In your case you also need to specify the installation source, e.g. by putting an info-file into the initrd. See for example: http://marc.theaimsgroup.com/?l=suse-autoinstall&m=113042352724245&w=2 If you are using the combined 32-bit/64-bit DVD, you only need to boot the specific kernel/initrd for 32/64-bit installation. I hope I hit the point. Peter.

1 0

Installing multiple SuSE versions over NFS using Autoyast
by Richard Hobbs 22 Feb '06

22 Feb '06

Hello, We currently have one DHCP server, one TFTP server and one NFS server. Using these, we install 64-bit SuSE 10.0 onto all of our 64-bit machines. However, we will soon need to start installing 64-bit SuSE onto some machines and 32-bit SuSE onto other machines. As I understand it, the installation source (in this case the 64-bit DVD files) is specified in "/tftpboot/pxelinux.cfg/default", and while I can change this to the 32-bit DVD, I can't see an obvious way of configuring both. The DHCP Server specifies the TFTP server to boot from, and specified the kernel to boot from (in this case "tftpserver:/tftpboot/pxelinux.0"). Is there an easy way of booting some machines into a 32-bit installation and other machines into a 64-bit installation? Thanks in advance! :-) Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0

Re: [suse-autoinstall] autoyast/info file problem
by KOEHLER, SVEN - PFHO 22 Feb '06

22 Feb '06

I discovered problems with the floppy protocol. linuxrc found everything right, but then yast couldn't find the file. autoyast=device://fd0/info (bootparameter) resp. autoyast: device://fd0/info (linuxrc) worked for me. @dave: did you intentionally leave the comments (# start_linuxrc_conf / # end_linuxrc_conf) out? greetz sven -----Ursprüngliche Nachricht----- Von: Yan Fitterer [mailto:yfitterer@novell.com] Gesendet: Dienstag, 21. Februar 2006 23:44 An: dbellis(a)mmm.com Cc: suse-autoinstall(a)suse.com Betreff: Re: [suse-autoinstall] autoyast/info file problem Actually - I've made a mistake myself. It should be: autoyast: floppy:///control.xml Note the extra / I read somewhere as well that there was a bug somewhere that prevented this from working (confirmed on SuSE 9.3). The workaround was to use: autoyast: device://fd0/control.xml As well, I don't believe you need the autoyast=floppy boot option. If the info file is found by linuxrc on the floppy (as it should), then that will be enough. Specifying the option on the command line may override what you put in the info file (not sure...). Yan >>> On 21/02/2006 at 22:12, in message >>> <43FB9052.6718.008C.0(a)novell.com>, "Yan Fitterer" <yfitterer(a)novell.com> wrote: > You have some redundancy, and a couple of mistakes. Try: > > install: nfs://169.10.63.215/instserv/9.0_x64 > hostip: 169.10.248.36 > netmask: 255.255.255.0 > gateway: 169.10.248.1 > nameserver: 169.10.8.4 > autoyast: floppy://control.xml > > I think "ip:" might be deprecated. Try "HostIP:" as above. > autoyast: floppy://file.xml needs to point to the actual path and name > for > the control file. > > See > > http://www.suse.com/~ug/autoyast_doc/invoking_autoinst.html#id2543177 > > and > > http://www.suse.com/~ug/autoyast_doc/info_file_format.html > > Yan > >>>> On 21/02/2006 at 21:41, in message > <OFB9BC1F12.CDE5BBF5-ON8625711C.00762869-8625711C.007720BB(a)mmm.com>, > <dbellis(a)mmm.com> wrote: >> I am new to SUSE and am trying to setup an AutoYaST install using the >> SUSE Linux CD1 to boot with the autoyast=floppy boot option. >> >> On the floppy disk I have the autoinst.xml file and an info file. >> >> The info file contains the following: >> >> install: nfs://169.10.63.215/instserv/9.0_x64 >> netdevice: eth0 >> server: 169.10.63.215 >> serverdir: /instserv/9.0_x64 >> ip: 169.10.248.36 >> netmask: 255.255.255.0 >> gateway: 169.10.248.1 >> nameserver: 169.10.8.4 >> autoyast: floppy >> >> I can mount the nfs share manually, but when I run the install >> process it fails with the following message: >> >> "Could not find the SUSE Linux Enterprise Server 9 Installation >> Source" >> >> Does anyone have an idea what I am doing wrong? >> >> Any help would be greatly appreciated. >> >> Thanks, >> Dave Bellis >> -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

autoyast/info file problem
by dbellis＠mmm.com 21 Feb '06

21 Feb '06

I am new to SUSE and am trying to setup an AutoYaST install using the SUSE Linux CD1 to boot with the autoyast=floppy boot option. On the floppy disk I have the autoinst.xml file and an info file. The info file contains the following: install: nfs://169.10.63.215/instserv/9.0_x64 netdevice: eth0 server: 169.10.63.215 serverdir: /instserv/9.0_x64 ip: 169.10.248.36 netmask: 255.255.255.0 gateway: 169.10.248.1 nameserver: 169.10.8.4 autoyast: floppy I can mount the nfs share manually, but when I run the install process it fails with the following message: "Could not find the SUSE Linux Enterprise Server 9 Installation Source" Does anyone have an idea what I am doing wrong? Any help would be greatly appreciated. Thanks, Dave Bellis

2 2

cant add gid to groups
by mailing list 21 Feb '06

21 Feb '06

hi, i want to add groups and set group ids for that groups in autoyast, if i create a xml with the gui it doesnt create the gid for some reason. How do i specify the group id? thanks

2 2

extended partition
by jean.manco＠bpm.it 21 Feb '06

21 Feb '06

hi, when i deploy a server with sles9 by pxe/nfs server i have the follow errors : .the first partition (/dev/cciss/c0d0p1) is extended and i don't want so. .when i want recovery my server after crash, with a recovery action by installation cd i don't view all partitions but only the extended how can i set the partitions normally in autoyast file? i pasted the lines that indicate the partitions. thanks a lot **************************************************** .................... <partitioning config:type="list"> <drive> <initialize config:type="boolean">true</initialize> <partitions config:type="list"> <partition> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fs_options> <opt_blocksize> <option_str>-b</option_str> <option_value>auto</option_value> </opt_blocksize> <opt_bytes_per_indoe> <option_str>-i</option_str> <option_value>auto</option_value> </opt_bytes_per_indoe> <opt_raid> <option_str>-Rstride=</option_str> <option_value>none</option_value> </opt_raid> <opt_reserved_blocks> <option_str>-m</option_str> <option_value>auto</option_value> </opt_reserved_blocks> </fs_options> <loop_fs config:type="boolean">false</loop_fs> <mount>/boot</mount> <mountby config:type="symbol">device</mountby> <noauto config:type="boolean">false</noauto> <partition_id config:type="integer">131</partition_id> <partition_type></partition_type> <size>100MB</size> </partition> <partition> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <partition_type></partition_type> <size>8GB</size> </partition> <partition> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/dati</mount> <partition_id config:type="integer">131</partition_id> <partition_type></partition_type> <size>4GB</size> </partition> <partition> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <partition_id config:type="integer">131</partition_id> <partition_type></partition_type> <size>auto</size> </partition> </partitions> <use>all</use> </drive> </partitioning> **************************************************** .................... byez Jean Manco

2 2

Emacs broken after auto-install, but working after manual install
by Richard Hobbs 20 Feb '06

20 Feb '06

Hello, If I install from DVD manually (i.e. creating partitions and selecting software by hand etc...) emacs within X works fine. However, if I use my autoinstall script to install SuSE over NFS, emacs starts ok in X, but each character is replaced with a square/rectangle. Also, the following is printed to either stdout or stderr: ============================================================ stg-opti2:~ # emacs Warning: Cannot convert string "-*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-*" to type FontStruct Warning: Cannot convert string "-*-helvetica-medium-r-*--*-120-*-*-*-*-iso8859-1" to type FontStruct stg-opti2:~ # ============================================================ As I said, emacs works fine if I install manually. Am I simply missing some packages, or has something more serious happened here? I have typed "xlsfonts | grep helv" and nothing is returned, but if I look in the Font Installer within YaST the fonts do appear to exist. Whether they are strictly the correct ones or not is a different matter though. Any clues are very much appreciated :-) Thanks again, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

2 4

unattempted autoyast installation on z/VM
by Kai Dusek 17 Feb '06

17 Feb '06

Dear all, I'm trying to setup an enviroment for installing z/VM guests (SLES9-SP2/3, 31bit) completely unattended. This is my parmfile: ramdisk_size=65536 root=/dev/ram1 ro init=/linuxrc dasd=150,151 TERM=dumb AUTOYAST=http://<IP>/autoinst.xml AUTOINSTALL=yes IP_INTERFACE=qeth IP_HOST=<hostname> IP_ADDR=x.x.x.x IP_NETMASK=x.x.x.x IP_BROADCAST=x.x.x.x IP_GATEWAY=x.x.x.x IP_DNS=x.x.x.x IP_SEARCH=x.x.x.x READ_DEVNO=7000 WRITE_DEVNO=7001 DATA_DEVNO=7002 PORTNAME=xxxxxx INST_INFO=ftp INST_IP_ADDR=x.x.x.x INST_IP_DIR=/SLES/sles9_s390_sp3 FTPUSER=<user> FTPPASSWORD=<pw> INST_SCREEN=VNC VNC_PASSWORD=<pw> IP_MTU=1492 INST_PASSWORD=<pw> The forceboot option in the autoyast control file is set to true, confirm is set to false. The installation itself is running unattended (even if I had some problem in the past with the environment setup and a bug in YaST), so the major part of the game is basically solved. But after disk formatting/partitioning and package installation I always have to "trigger" the second stage by IPLing the guest manually since the automated reboot doesn't take place. This is my output: [...] umount: /y2update: not mounted umount: /var/adm/mount: not mounted Feb 15 19:48:19 suse sshd¬322|: Received signal 15; terminating. You should be able to login via telnet/ssh now. To restart network setup, enter: netsetup To continue the installation, enter: ./inst_source bash: no job control in this shell SuSE Instsys boeid120:/ # [...] After a manual IPL the system is doing the second stage installation - and it looks like the system installation is done completely and correctly. What do I have to do to prevent re-IPLing the system manually? BTW: Is it possible to "re-direct" installation output to 3270 instead of VNC? Mit freundlichen Gruessen/Kind regards Kai Dusek

1 0

RE: [suse-autoinstall] SLES 9 SP2
by Kurt Chang 16 Feb '06

16 Feb '06

Here is what I have xianserv:/export/SLES9/yast # cat order /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 /SUSE-CORE-Version-9/CD1 xianserv:/export/SLES9/yast # cat instorder /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 xianserv:/export/SLES9/yast # What is /SUSE-Updates? Thanks Kurt -----Original Message----- From: Mike Marion [mailto:mmarion@qualcomm.com] Sent: Monday, February 13, 2006 8:35 PM To: Kurt Chang Cc: Karsten Petersen; suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] SLES 9 SP2 On Mon, Feb 13, 2006 at 01:04:44PM -0500, Kurt Chang wrote: > Thank you for your suggestion. My system looks like this What do your yast/order|instorder files have in them? My directories are similar and I have: cornholio sles9-sp2 {497}$ cat yast/order /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-Updates /SUSE-Updates /SUSE-SLES-Version-9/CD1 /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 /SUSE-CORE-Version-9/CD1 cornholio sles9-sp2 {498}$ cat yast/instorder /SUSE-SLES-9-Service-Pack-Version-2/CD1 /SUSE-Updates /SUSE-SLES-Version-9/CD1 /SUSE-CORE-Version-9/CD1 -- Mike Marion-Unix SysAdmin/Staff Engineer-http://www.qualcomm.com "The day that Microsoft makes something that doesn't suck is the day that they start making vacuum cleaners." - shaman(a)kos.net

2 1

Performing an online update from within an init script
by Richard Hobbs 16 Feb '06

16 Feb '06

Hello, Is it possible to run an online update using YaST2 from within an init script? If so, which command should I use, and how can I specify what types of packages I want to upgrade (e.g. security, recommended etc...)? Thanks in advance, people :-) Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

3 4

Keyboard is still US within X server
by Richard Hobbs 16 Feb '06

16 Feb '06

Hello, I have auto-installed a number of machines now, and when in a terminal outside of the X server (i.e. in runlevel 3, for example), the keyboard works fine. However, when I use runlevel 5, and use the "Konsole" or if I login and run "startx", the keyboard within X is still using the US layout. Here is the <general> part of the XML config, which I'm pretty sure is correct, but it's not setting the keyboard layout within X. ============================================================ <general> <clock> <hwclock>UTC</hwclock> <timezone>Etc/GMT</timezone> </clock> <keyboard> <keymap>english-uk</keymap> </keyboard> <language>en_GB</language> <mode> <confirm config:type="boolean">false</confirm> </mode> </general> ============================================================ Any ideas? Thanks in advance :-) Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0

RE: [suse-autoinstall] Performing an online update from within an init script
by Schumacher, Gordon 16 Feb '06

16 Feb '06

# From: Richard Hobbs [mailto:richard.hobbs@crl.toshiba.co.uk] # Sent: Wednesday, February 15, 2006 7:42 AM # To: suse-autoinstall(a)suse.com # Subject: [suse-autoinstall] Performing an online update from within an init script # # Hello, # # Is it possible to run an online update using YaST2 from within an init # script? # # If so, which command should I use, and how can I specify what types of # packages I want to upgrade (e.g. security, recommended etc...)? # # Thanks in advance, people :-) This should get you started - run this as an init_script: #!/bin/sh /bin/chvt 9 exec < /dev/tty9 > /dev/tty9 echo Performing online update... online_update -l en -u <<<<url_to_update_from>>>> -V SuSEconfig echo Rebooting the computer now to use the new kernel. sleep 5 /bin/chvt 1 shutdown -r now

2 1

Suse 9.3 panics in both suse professional & fail safe version
by kanji linux 15 Feb '06

15 Feb '06

Hi We have a Iwill dk88 system (dual core amd opteron 200 series, 16GB RAM, NVIDIA nforce4 professional 2000 chipset, ATI RageXL VGA controller with AMI BIOS). I was able to successfully install the system using NFS network installation. But while trying to boot from the hard disk, both the suse9.3 professional & failsafe versions goes into kernel panic mode. Now, this problem is similar to the links posted earlier: 1. http://lists.suse.com/archive/suse-amd64/2005-Sep/0043.html 2. http://lists.suse.com/archive/suse-amd64/2005-May/0095.html but no solution seems to be provided. Now there is no problem with the system as such because i was able to successfully install RHEL4 on it. I did notice the fact that boot options are the following : **************************************** - SUSE LINUX 9.3 kernel (hd0,0)/vmlinuz root=/dev/hda3 vga=ox31 selinux=0 splash=silent console=tty0 console=ttyS0 resume=/dev/hda2 showopts initrd (hd0,0)/initrd - Failsafe -- SUSE LINUX 9.3 kernel (hd0,0)/vmlinuz root=/dev/hda3 showopts ide=nodma apm=off acpi=off vga=normal noresume selinux=0 barrier=off iommu=noforce 3 initrd (hd0,0)/initrd ****************************************** if SUSE linux 9.3 is selected to be booted, the video goes blank and no output or response can be seen. If the "vga" parameter is set to "normal", then the system goes in to kernel panic mode. Failsafe - also goes into kernel panic mode. I salvaged the following stack trace displayed after kernel panic : ******************************************************************* Kernel Panic - not syncing. Aiee, Killing interrupt controller Badness in do_unblock_screen at driver/char/vt.c: 2840 Call Trace: <ffffffff80278710>{do_unblank_screen+80}<ffffffff80120d5c>{bust_spinlocks+28} <ffffffff80136f56>{panic+198}<ffffffff8013795d>{vprintk+479} .......{__upread+29}....... {_oops_end+}... *************************************************************** Any ideas on what could be the reason ? Thanks Kanji --------------------------------- Yahoo! Autos. Looking for a sweet ride? Get pricing, reviews, & more on new and used cars.

1 0

Simultaneous SuSE 10.0 Autoyast installs fail!
by Richard Hobbs 14 Feb '06

14 Feb '06

Hello, I have a working and fully tested DHCP server, TFTP server, NFS file server and XML config profile which I have used to build the first of many servers. The whole system has worked fine, but now I've hit a rather critical problem - doing more than one server at a time seems to break them all! I have rebuilt our first server perhaps 15 times due to testing the XML profile, and each time it worked perfectly. I now have 8 machines which I would like to install, so I've copied the 001422B0C61F XML file to a filename which matches the other MAC addresses, and I have told our DHCP server to give out the necessary MAC addresses as well. The only things I have changed in our XML config profile is the any reference to the hostname, and the MAC address and IP address. Everything else is the same. When I boot up all 8 machines from the network, they all boot into the boot loader from the TFTP server, and they all boot into autoyast. They all load their own XML config files, but part of the way through installing the packages, the machines hang. Ctrl + Alt + F[1-6] does nothing, so all I can do is reboot using Ctrl + Alt + Del. The TFTP Server, NFS Server and the server hosting the config profiles (also over NFS) is the same machine, which can easily keep up with the demand for files... So much so, in fact, that the system load hardly increases at all during an install of a remote machine. What could be stopping me doing multiple NFS autoyast installs?? Any suggestions are welcome. Thanks in advance, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

3 10

RE: [suse-autoinstall] wake on lan
by Frank van Tol 14 Feb '06

14 Feb '06

Good to know is also: - it will not reset a powered on client. - "very old" clients may not have a NIC or BIOS with WOL support at all. - WOL support must be enabled on the NIC every time (depends on the NIC) a client is started. Some OS's (windows) do this for you, some linux distributions too by means of scripts. See "ethtool eth0" and "etherwake". -----Original Message----- From: Thomas Mieslinger [mailto:Thomas.Mieslinger@gls-itservices.com] Sent: dinsdag 14 februari 2006 16:06 To: suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] wake on lan It's not really a protocol nor is it a API like PXE is. It is just a ethernet paket that contains a "magic" start sequence, 16 times the Mac adress of the client to wake up and the paket is send to Mac Adress ff:ff:ff:ff:ff:ff -> ie broadcasted. The NIC hardware in the client is permanently powered on and looks at every packet that comes over the wire. If it finds the magic start sequence and its Mac Adress the computer is started. -- GLS IT Services Thomas Mieslinger GLS Germany Str. 1-7 fon: +49 6677 17 463 36286 Neuenstein fax: +49 6677 17 111 Germany eMail: thomas.mieslinger(a)gls-itservices.com Roger Oberholtzer <roger(a)opq.se> schrieb am 14.02.2006 15:28:03: > On Tue, 2006-02-14 at 13:37 +0100, Thomas Mieslinger wrote: > > > > > > > > The trick you're looking for is in the pc's bios that you're going > > to install. For example ibm thinkcentre s50 desktop pcs know two > > boot orders: > > standard and automatic bootup. The automatic bootup order is used if the > > system has been woken up over lan. Else it uses standard boot order. > > Leave it to IBM. We are using SuperMicro. I will have to look a > comparable setting. I am not sure all our SuperMicro MBs support > wake-on-lan. > > I thought wake-on-lan was some other protocol than PXE. Not exactly > how that translates into a boot. But if it looked interesting, I could > invest time in finding out. Perhaps it is mainly useful in selecting > two different boot device lists as IBM have it. > > > > > > regards Thomas > > -- > > GLS IT Services > > Thomas Mieslinger > > GLS Germany Str. 1-7 fon: +49 6677 17 463 > > 36286 Neuenstein fax: +49 6677 17 111 > > Germany eMail: thomas.mieslinger(a)gls-itservices.com > > > > Roger Oberholtzer <roger(a)opq.se> schrieb am 14.02.2006 12:53:39: > > > > > > > > Has anyone done a network install of SUSE when the client is > > > started with wake-on-lan? Is this even possible? > > > > > > The reason I ask is I am unsure how you disable a machine from > > > being installed each time it boots. Do you have to turn on/off PXE > > > (or whatever you're using) in the BIOS when you want to change behavior? Or > > > modify the dhcp server's setup? I guess either would work. What > > > are people doing in practice? > > > > > > > > > -- > > > Roger Oberholtzer > > > > > > OPQ Systems AB > > > Ramböll Sverige AB > > > Kapellgränd 7 > > > P.O. Box 4205 > > > SE-102 65 Stockholm, Sweden > > > > > > Tel: Int +46 8-615 60 20 > > > Fax: Int +46 8-31 42 23 > > > > > > > > > -- > > > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > > > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > > > > > > > > -- > Roger Oberholtzer > > OPQ Systems AB > Ramböll Sverige AB > Kapellgränd 7 > P.O. Box 4205 > SE-102 65 Stockholm, Sweden > > Tel: Int +46 8-615 60 20 > Fax: Int +46 8-31 42 23 > > > -- > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > -- To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com For additional commands, e-mail: suse-autoinstall-help(a)suse.com

1 0

wake on lan
by Roger Oberholtzer 14 Feb '06

14 Feb '06

Has anyone done a network install of SUSE when the client is started with wake-on-lan? Is this even possible? The reason I ask is I am unsure how you disable a machine from being installed each time it boots. Do you have to turn on/off PXE (or whatever you're using) in the BIOS when you want to change behavior? Or modify the dhcp server's setup? I guess either would work. What are people doing in practice? -- Roger Oberholtzer OPQ Systems AB Ramböll Sverige AB Kapellgränd 7 P.O. Box 4205 SE-102 65 Stockholm, Sweden Tel: Int +46 8-615 60 20 Fax: Int +46 8-31 42 23

3 6

Fwd: Re: [suse-autoinstall] wake on lan
by Yan Fitterer 14 Feb '06

14 Feb '06

Maybe safer: dd of=/dev/sda if=/dev/zero bs=512 count=1 As far as I know, the core part of the MBR is only 512 bytes. There _is_ some free space afterwards (not sure how much), where things like GRUB can store some of their boot code. BTW - you _will_ destroy the partion table with this. The partitions shouldn't be touched though, so if you wanted (?) you could re-create them manually as long as you know the exact layout. Yan > >>> On 14/02/2006 at 12:14:24, in message <43F1C9A0.9040409(a)opencfd.co.uk>, Mattijs > Janssens <m.janssens(a)opencfd.co.uk> wrote: > > You set up the machine to boot from harddisk first and then lan. > > > > So initially it will not have an mbr and install from lan. Afterwards it > > will pick up the harddisk installation. > > > > If you need to reinstall just(!) wipe the MBR: > > > > dd of=/dev/sda if=/dev/zero bs=1024 count=1 > > > > (no guarantees on that one - don't complain if you wipe your whole disk) > > > > Roger Oberholtzer wrote: > >> Has anyone done a network install of SUSE when the client is started > >> with wake-on-lan? Is this even possible? > >> > >> The reason I ask is I am unsure how you disable a machine from being > >> installed each time it boots. Do you have to turn on/off PXE (or > >> whatever you're using) in the BIOS when you want to change behavior? Or > >> modify the dhcp server's setup? I guess either would work. What are > >> people doing in practice? > >> > >> > > >

2 1

RE: [suse-autoinstall] wake on lan
by Frank van Tol 14 Feb '06

14 Feb '06

Yes, I deploy desktops with Altiris. Altiris takes care of the correct PXE behaviour, i.e. only provide a client with an install when one is scheduled for it (MAC based job assignment). -----Original Message----- From: Roger Oberholtzer [mailto:roger@opq.se] Sent: dinsdag 14 februari 2006 12:54 To: suse-autoinstall(a)suse.com Subject: [suse-autoinstall] wake on lan Has anyone done a network install of SUSE when the client is started with wake-on-lan? Is this even possible? The reason I ask is I am unsure how you disable a machine from being installed each time it boots. Do you have to turn on/off PXE (or whatever you're using) in the BIOS when you want to change behavior? Or modify the dhcp server's setup? I guess either would work. What are people doing in practice? -- Roger Oberholtzer OPQ Systems AB Ramböll Sverige AB Kapellgränd 7 P.O. Box 4205 SE-102 65 Stockholm, Sweden Tel: Int +46 8-615 60 20 Fax: Int +46 8-31 42 23 -- 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] Simultaneous SuSE 10.0 Autoyastinstalls fail!
by Richard Hobbs 14 Feb '06

14 Feb '06

Hello, Not sure how much of this is relevant now I know the problem has changed, but I'll reply anyway :-) Mattijs: I don't specify the machine to give out the DHCP hostname in the XML file, because I don't like relying on remote servers unless I have to. The machine will have a static IP address eventually, so I'd like to specify that as soon as possible. 2) I don't use VNC. I have a KVM switch plugged into the back of every single machine. Hans-Joachim: That is correct. I do not have a rules.xml file anywhere, and autoyast points to the directory. As I said, it is picking up the correct XML file based on the MAC address, but it's hanging at 56% of "emacs-21.3". Thanks again, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 > -----Original Message----- > From: Hans-Joachim Ehlers [mailto:HansJoachim.Ehlers@eumetsat.int] > Sent: 14 February 2006 10:43 > To: Richard Hobbs; suse-autoinstall(a)suse.com > Subject: RE: [suse-autoinstall] Simultaneous SuSE 10.0 > Autoyastinstalls fail! > > So if you have for each machine a seperate config file then check if > autoyast points to a directory or file. > In your case it must point to a directory and no > ./rules/rules.xml must > exist. > > All from memory. > > hth > Hajo > > >>> "Richard Hobbs" <richard.hobbs(a)crl.toshiba.co.uk> > 14/02/06 11:34:36 > >>> > Hello, > > Mattijs: I can appreciate that if I boot all 8 machines > simultaneously, > they > might overload the DHCP Server, but if I boot one machine up first, > wait for > it to begin installing the packages, and then boot a second > machine up, > as > soon as the second machine starts it's installation, the first machine > (and > the second machine) hangs. > > And yes, the machines can see the XML file (or else they won't not > begin > installing) and they can also see the installation source (or they > would not > start installing the packages). > > Remember, it breaks after installing quite a few packages (maybe 25% > complete). > > Hans-Joachim: I have a different XML config file for each MAC address > of the > machine I am installing, see below: > > ============================================================ > nfsserver:/install/configProfiles # ls -l > total 220 > drwxr-xr-x 3 root root 400 Feb 14 10:14 . > drwxr-xr-x 6 root root 176 Feb 8 08:52 .. > -rw-r--r-- 1 root root 24032 Feb 14 10:12 000E0C062613 > -rw-r--r-- 1 root root 24021 Feb 13 16:21 001422B0351F > -rw-r--r-- 1 root root 24020 Feb 13 16:25 001422B0C567 > -rw-r--r-- 1 root root 24021 Feb 13 16:42 001422B0C624 > -rw-r--r-- 1 root root 24020 Feb 13 16:25 001422B0C633 > -rw-r--r-- 1 root root 24020 Feb 13 16:23 001422B0C666 > -rw-r--r-- 1 root root 24020 Feb 13 16:28 001422B0C726 > -rw-r--r-- 1 root root 24021 Feb 13 16:28 001422B0C993 > -rw-r--r-- 1 root root 24021 Feb 13 16:29 001422B0C9FC > drwxr-xr-x 2 root root 88 Feb 7 15:59 oldProfiles > -rw-r--r-- 1 root root 610 Feb 3 11:26 template.xml > nfsserver:/install/configProfiles # > ============================================================ > > So, if I want to install a new machine of identical hardware, I simply > copy > an existing profile, and modify the hostname and IP address. Why will > this > approach not work? > > Also, I am using DHCP to boot the machines, but the XML file has a > static IP > address (which is the same as the IP address given to the machine from > DHCP). > > Thanks again, > Richard. > > -- > Richard Hobbs (Systems Administrator) > Toshiba Research Europe Ltd. - Speech Technology Group > Web: http://www.toshiba-europe.com/research/ > Normal Email: richard.hobbs(a)crl.toshiba.co.uk > Mobile Email: mobile(a)mongeese.co.uk > Tel: +44 1223 376964 Mobile: +44 7811 803377 > > > -----Original Message----- > > From: Mattijs Janssens [mailto:m.janssens@opencfd.co.uk] > > Sent: 14 February 2006 09:59 > > Cc: suse-autoinstall(a)suse.com > > Subject: Re: [suse-autoinstall] Simultaneous SuSE 10.0 > > Autoyast installs fail! > > > > I don't know whether this is applicable to your situation but > > I have a > > cluster where if I boot them all at the same time some of > the machine > > > don't get their hostname set. Don't know whether the network > > isn't ready > > or maybe the dhcp server gets overloaded. I can imagine for > > your booting > > this might upset the installation. > > > > Another thing: the machines can see the xml file. Can they > > also see the > > installation source? > > > > Richard Hobbs wrote: > > > Hello, > > > > > > I have a working and fully tested DHCP server, TFTP server, > > NFS file server > > > and XML config profile which I have used to build the first > > of many servers. > > > > > > The whole system has worked fine, but now I've hit a rather > > critical problem > > > - doing more than one server at a time seems to break them all! > > > > > > I have rebuilt our first server perhaps 15 times due to > > testing the XML > > > profile, and each time it worked perfectly. I now have 8 > > machines which I > > > would like to install, so I've copied the 001422B0C61F XML file to > a > > > filename which matches the other MAC addresses, and I have > > told our DHCP > > > server to give out the necessary MAC addresses as well. > > > > > > The only things I have changed in our XML config profile is the > any > > > reference to the hostname, and the MAC address and IP > > address. Everything > > > else is the same. > > > > > > When I boot up all 8 machines from the network, they all > > boot into the boot > > > loader from the TFTP server, and they all boot into > > autoyast. They all load > > > their own XML config files, but part of the way through > > installing the > > > packages, the machines hang. Ctrl + Alt + F[1-6] does > > nothing, so all I can > > > do is reboot using Ctrl + Alt + Del. > > > > > > The TFTP Server, NFS Server and the server hosting the > > config profiles (also > > > over NFS) is the same machine, which can easily keep up > > with the demand for > > > files... So much so, in fact, that the system load hardly > > increases at all > > > during an install of a remote machine. > > > > > > What could be stopping me doing multiple NFS autoyast installs?? > > > > > > Any suggestions are welcome. > > > > > > Thanks in advance, > > > Richard. > > > > > > > > > -- > > Best regards, > > > > Mattijs > > > > Mattijs Janssens > > > > OpenCFD Ltd. > > The Mews, Picketts Lodge, > > Picketts Lane, Salfords, > > Surrey RH1 5RG. > > Tel: +44 (0)1293 821272 > > Email: M.Janssens(a)OpenCFD.co.uk > > URL: http://www.OpenCFD.co.uk > > > > -- > > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > > > > > > > _____________________________________________________________________ > > This e-mail has been scanned for viruses by Verizon Business > > Internet Managed Scanning Services - powered by MessageLabs. > > For further information visit http://www.mci.com > > > > > > > > _____________________________________________________________________ > This e-mail has been scanned for viruses by Verizon Business Internet > Managed Scanning Services - powered by MessageLabs. For further > information visit http://www.mci.com > > -- > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > > > _____________________________________________________________________ > This e-mail has been scanned for viruses by Verizon Business > Internet Managed Scanning Services - powered by MessageLabs. > For further information visit http://www.mci.com > > _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0

RE: [suse-autoinstall] Simultaneous SuSE 10.0 Autoyast installs fail!
by Hans-Joachim Ehlers 14 Feb '06

14 Feb '06

So if you have for each machine a seperate config file then check if autoyast points to a directory or file. In your case it must point to a directory and no ./rules/rules.xml must exist. All from memory. hth Hajo >>> "Richard Hobbs" <richard.hobbs(a)crl.toshiba.co.uk> 14/02/06 11:34:36 >>> Hello, Mattijs: I can appreciate that if I boot all 8 machines simultaneously, they might overload the DHCP Server, but if I boot one machine up first, wait for it to begin installing the packages, and then boot a second machine up, as soon as the second machine starts it's installation, the first machine (and the second machine) hangs. And yes, the machines can see the XML file (or else they won't not begin installing) and they can also see the installation source (or they would not start installing the packages). Remember, it breaks after installing quite a few packages (maybe 25% complete). Hans-Joachim: I have a different XML config file for each MAC address of the machine I am installing, see below: ============================================================ nfsserver:/install/configProfiles # ls -l total 220 drwxr-xr-x 3 root root 400 Feb 14 10:14 . drwxr-xr-x 6 root root 176 Feb 8 08:52 .. -rw-r--r-- 1 root root 24032 Feb 14 10:12 000E0C062613 -rw-r--r-- 1 root root 24021 Feb 13 16:21 001422B0351F -rw-r--r-- 1 root root 24020 Feb 13 16:25 001422B0C567 -rw-r--r-- 1 root root 24021 Feb 13 16:42 001422B0C624 -rw-r--r-- 1 root root 24020 Feb 13 16:25 001422B0C633 -rw-r--r-- 1 root root 24020 Feb 13 16:23 001422B0C666 -rw-r--r-- 1 root root 24020 Feb 13 16:28 001422B0C726 -rw-r--r-- 1 root root 24021 Feb 13 16:28 001422B0C993 -rw-r--r-- 1 root root 24021 Feb 13 16:29 001422B0C9FC drwxr-xr-x 2 root root 88 Feb 7 15:59 oldProfiles -rw-r--r-- 1 root root 610 Feb 3 11:26 template.xml nfsserver:/install/configProfiles # ============================================================ So, if I want to install a new machine of identical hardware, I simply copy an existing profile, and modify the hostname and IP address. Why will this approach not work? Also, I am using DHCP to boot the machines, but the XML file has a static IP address (which is the same as the IP address given to the machine from DHCP). Thanks again, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 > -----Original Message----- > From: Mattijs Janssens [mailto:m.janssens@opencfd.co.uk] > Sent: 14 February 2006 09:59 > Cc: suse-autoinstall(a)suse.com > Subject: Re: [suse-autoinstall] Simultaneous SuSE 10.0 > Autoyast installs fail! > > I don't know whether this is applicable to your situation but > I have a > cluster where if I boot them all at the same time some of the machine > don't get their hostname set. Don't know whether the network > isn't ready > or maybe the dhcp server gets overloaded. I can imagine for > your booting > this might upset the installation. > > Another thing: the machines can see the xml file. Can they > also see the > installation source? > > Richard Hobbs wrote: > > Hello, > > > > I have a working and fully tested DHCP server, TFTP server, > NFS file server > > and XML config profile which I have used to build the first > of many servers. > > > > The whole system has worked fine, but now I've hit a rather > critical problem > > - doing more than one server at a time seems to break them all! > > > > I have rebuilt our first server perhaps 15 times due to > testing the XML > > profile, and each time it worked perfectly. I now have 8 > machines which I > > would like to install, so I've copied the 001422B0C61F XML file to a > > filename which matches the other MAC addresses, and I have > told our DHCP > > server to give out the necessary MAC addresses as well. > > > > The only things I have changed in our XML config profile is the any > > reference to the hostname, and the MAC address and IP > address. Everything > > else is the same. > > > > When I boot up all 8 machines from the network, they all > boot into the boot > > loader from the TFTP server, and they all boot into > autoyast. They all load > > their own XML config files, but part of the way through > installing the > > packages, the machines hang. Ctrl + Alt + F[1-6] does > nothing, so all I can > > do is reboot using Ctrl + Alt + Del. > > > > The TFTP Server, NFS Server and the server hosting the > config profiles (also > > over NFS) is the same machine, which can easily keep up > with the demand for > > files... So much so, in fact, that the system load hardly > increases at all > > during an install of a remote machine. > > > > What could be stopping me doing multiple NFS autoyast installs?? > > > > Any suggestions are welcome. > > > > Thanks in advance, > > Richard. > > > > > -- > Best regards, > > Mattijs > > Mattijs Janssens > > OpenCFD Ltd. > The Mews, Picketts Lodge, > Picketts Lane, Salfords, > Surrey RH1 5RG. > Tel: +44 (0)1293 821272 > Email: M.Janssens(a)OpenCFD.co.uk > URL: http://www.OpenCFD.co.uk > > -- > To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com > For additional commands, e-mail: suse-autoinstall-help(a)suse.com > > > _____________________________________________________________________ > This e-mail has been scanned for viruses by Verizon Business > Internet Managed Scanning Services - powered by MessageLabs. > For further information visit http://www.mci.com > > _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.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] Simultaneous SuSE 10.0 Autoyast installs fail!
by Hans-Joachim Ehlers 14 Feb '06

14 Feb '06

>The only things I have changed in our XML config profile is the any >reference to the hostname, and the MAC address and IP address. Everything >else is the same. IMHO that will not work because the machine needs a hostname and IP adress entry otherwise the installation will fail. So you are going to use the rules.xml to set hostname and ip-adress or change your autoinst.xml in such a way that the client will be fully dhcp configured. Keep in mind that if you are using dchp that the clients depends on it during their lifetime. In case the client shall have fixed hostnames and ip numbers you have to go with the rules.xml hth Hajo >>> "Richard Hobbs" <richard.hobbs(a)crl.toshiba.co.uk> 14/02/06 09:50:36 >>> Hello, I have a working and fully tested DHCP server, TFTP server, NFS file server and XML config profile which I have used to build the first of many servers. The whole system has worked fine, but now I've hit a rather critical problem - doing more than one server at a time seems to break them all! I have rebuilt our first server perhaps 15 times due to testing the XML profile, and each time it worked perfectly. I now have 8 machines which I would like to install, so I've copied the 001422B0C61F XML file to a filename which matches the other MAC addresses, and I have told our DHCP server to give out the necessary MAC addresses as well. The only things I have changed in our XML config profile is the any reference to the hostname, and the MAC address and IP address. Everything else is the same. When I boot up all 8 machines from the network, they all boot into the boot loader from the TFTP server, and they all boot into autoyast. They all load their own XML config files, but part of the way through installing the packages, the machines hang. Ctrl + Alt + F[1-6] does nothing, so all I can do is reboot using Ctrl + Alt + Del. The TFTP Server, NFS Server and the server hosting the config profiles (also over NFS) is the same machine, which can easily keep up with the demand for files... So much so, in fact, that the system load hardly increases at all during an install of a remote machine. What could be stopping me doing multiple NFS autoyast installs?? Any suggestions are welcome. Thanks in advance, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.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] /info files on boot CDs
by Hans-Joachim Ehlers 14 Feb '06

14 Feb '06

Used directories ./x86 ( Contains all the suse stuff ) ./x86isoimage ( Contains the newly created iso image and auxillery files ) Contents of ./x86isoimage $ cat mkisoimage #!/bin/sh /usr/bin/mkisofs -o SuSE-93.iso \ -b boot/loader/isolinux.bin \ -c boot/boot.cat \ -no-emul-boot \ -boot-load-size 4 \ -boot-info-table \ -hide-joliet-trans-tbl \ -hide-rr-moved \ -allow-leading-dots \ -l -r -J \ -V "SuSE 9.3 DVD" \ -publisher "SuSE" \ -p "stock" \ -A "SuSE 9.3 DVD" \ -exclude-list mkisoimage.i386.excl \ ../x86/ $ cat mkisoimage.i386.excl *.x86_64.rpm *.src.rpm *.nosrc.rpm *.iso iso jonas-*.rpm rfc-*.rpm fillets-ng-*.rpm kde3-i18n-sv*.rpm FlightGear-*.rpm jboss-*.rpm */suse/src */suse/nosrc */suse/x86_64 */suse/noarch */suse/i586 */suse/i686 */boot/loader/initrd*orig* Ajust the excl to your needs. The above excl. file is for is a minimal x86 CD ( 300 MB) for network installation ( boot from CD, install from network ) hth Hajo >>> Ian Grant <Ian.Grant(a)cl.cam.ac.uk> 13/02/06 18:07:31 >>> Dear suse-autoinstall people, We have machines without floppies and we want to use an /info file. I understand I can make one and put it in the /initrd on the boot CD. I can probably manage this using the mkinitrd scripts etc, but how would I then create a boot CD ISO image? I am hoping there is a magic mkisofs command line that will fetch what I need out of the boot/loader directory of the SuSE 9.3 distribution and make me a new boot CD. Can anyone point me at more info? Thanks Ian -- 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 9 SP2
by Kurt Chang 14 Feb '06

14 Feb '06

Hello, Thank you for your suggestion. My system looks like this /export/SLES9 # ls -l total 0 drwxr-xr-x 6 root root 368 Feb 13 11:09 . drwxr-xr-x 7 root root 152 Feb 8 10:43 .. drwxr-xr-x 7 root root 168 Feb 13 10:35 SUSE-CORE-Version-9 drwxr-xr-x 5 root root 120 Feb 13 10:39 SUSE-SLES-9-Service-Pack-Version-2 drwxr-xr-x 3 root root 72 Feb 13 10:28 SUSE-SLES-Version-9 lrwxrwxrwx 1 root root 28 Feb 13 10:29 boot -> SUSE-SLES-Version-9/CD1/boot lrwxrwxrwx 1 root root 31 Feb 13 10:29 content -> SUSE-SLES-Version-9/CD1/content lrwxrwxrwx 1 root root 35 Feb 13 10:29 control.xml -> SUSE-SLES-Version-9/CD1/control.xml lrwxrwxrwx 1 root root 51 Feb 13 11:09 driverupdate -> SUSE-SLES-9-Service-Pack-Version-2/CD1/driverupdate lrwxrwxrwx 1 root root 44 Feb 13 11:09 linux -> SUSE-SLES-9-Service-Pack-Version-2/CD1/linux lrwxrwxrwx 1 root root 31 Feb 13 10:29 media.1 -> SUSE-SLES-Version-9/CD1/media.1 drwxr-xr-x 2 root root 104 Feb 13 10:36 yast xianserv:/export/SLES9 # It still does not work. Thanks Kurt -----Original Message----- From: Karsten Petersen [mailto:karsten.petersen@megware.com] Sent: Monday, February 13, 2006 12:57 PM To: Kurt Chang Cc: suse-autoinstall(a)suse.com Subject: Re: [suse-autoinstall] SLES 9 SP2 Hi, please make sure that you have created the following symlinks in your install directory, I saw error messages like those you describe with SP3 when they were not there: driverupdate -> sp3-x86_64/CD1/driverupdate linux -> sp3-x86_64/CD1/linux Best wishes, Karsten Kurt Chang schrieb: > Hello, > > > > I have a question in regards to the auto install procedure. Using YAST, > I set up a NFS share where I have loaded the 6 CD for SLES9 (64 bits) > and then the three SP2 CD's. When I installed a server via NFS, I got > the message, "Insert SUSE SLES 9 Service Pack 2 CD 1" right before SUSE > is ready to install. I have another NFS server for 32 bit distro and > I did not experience this problem. Any suggestion? Thank you very much > for your help > > > > Kurt > > -- HPC System Engineer Tel.: +49 3722 528 43 Fax: +49 3722 528 15 E-Mail: karsten.petersen(a)megware.com MEGWARE Computer GmbH Vertrieb und Service Nordstraße 19 09247 Chemnitz/Röhrsdorf Germany http://www.megware.com

2 1

SLES 9 SP2
by Kurt Chang 13 Feb '06

13 Feb '06

Hello, I have a question in regards to the auto install procedure. Using YAST, I set up a NFS share where I have loaded the 6 CD for SLES9 (64 bits) and then the three SP2 CD's. When I installed a server via NFS, I got the message, "Insert SUSE SLES 9 Service Pack 2 CD 1" right before SUSE is ready to install. I have another NFS server for 32 bit distro and I did not experience this problem. Any suggestion? Thank you very much for your help Kurt

1 0

/info files on boot CDs
by Ian Grant 13 Feb '06

13 Feb '06

Dear suse-autoinstall people, We have machines without floppies and we want to use an /info file. I understand I can make one and put it in the /initrd on the boot CD. I can probably manage this using the mkinitrd scripts etc, but how would I then create a boot CD ISO image? I am hoping there is a magic mkisofs command line that will fetch what I need out of the boot/loader directory of the SuSE 9.3 distribution and make me a new boot CD. Can anyone point me at more info? Thanks Ian

1 1

cant get Xinclude to work with suse 10
by mailing list 13 Feb '06

13 Feb '06

hi im having a problem getting my xinclude to work with suse 10. Heres are some parts of my autoinstall.xml <?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" xmlns:xinclude=" http://www.w3.org/2001/XInclude"> <bootloader> <activate config:type="boolean">false</activate> <loader_type>grub</loader_type> </bootloader> <files config:type="list"> <file> <file_contents><![CDATA[# $OpenLDAP: pkg/ldap/libraries/libldap/ldap.conf,v 1.9 2000/09/04 19:57:01 kurt Exp $ and a few lines down ... <general> <xinclude:include href=" http://192.168.100.19/autoyast/includes/hdd_prepare_scsi.xml"/> <clock> <hwclock>localtime</hwclock> <timezone>Europe/Berlin</timezone> </clock> <keyboard> <keymap>english-us</keymap> </keyboard> <language>en_US</language> <mode> <confirm config:type="boolean">false</confirm> <forceboot config:type="boolean">true</forceboot> </mode> <mouse> <id>21_imps2</id> </mouse> </general> yep ... that part seems to work ok .. since autoyast accepts it ... now to my hdd prepare_scsi.xml file: <partitioning config:type="list" xmlns="http://www.suse.com/1.0/yast2ns" xmlns:config="http://www.suse.com/1.0/configns" xmlns:xinclude="http://www.w3.org/2001/XInclude"> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <size>+1GB</size> </partition> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/boot</mount> <partition_id config:type="integer">131</partition_id> <size>+100M</size> </partition> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <partition_id config:type="integer">131</partition_id> <size>+12GB</size> </partition> <partition> <crypt>twofish256</crypt> <format config:type="boolean">false</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/data/UserData</mount> <partition_id config:type="integer">131</partition_id> <size>max</size> </partition> </partitions> <type config:type="symbol">CT_DISK</type> <use>free</use> </drive> </partitioning> i didnt find any docu related to suse 10. I guess a lot of the syntax changed since 9.0 (the docu i found) Autoyast tells me: ERROR Error while configuring partitions Try Again In the log (/var/log/YaST2/y2log) i can see (at the eof) that the SDA was detected fine .. but thats it .. im stuck.. thank in advance for any help Florian Zuegner.

3 2

Random time for automatic online-update
by Guido Kölsch 09 Feb '06

09 Feb '06

Hi, i have a question about the random update-time in autoyast suse 9.3. It seems not working for me, because after install every machine has the same time for automatic update. Any hints? -- ------------------------------------------------------------------------ Guido Kölsch Max-Planck-Institut fuer Radioastronomie Phone: +49 (0)228 525 164 Auf dem Huegel 69, D-53121 Bonn, Germany FAX : +49 (0)228 525 229 email: gkoelsch(a)mpifr-bonn.mpg.de ------------------------------------------------------------------------

2 2

Configuring remote NTP servers using xml config file
by Richard Hobbs 08 Feb '06

08 Feb '06

Hello, I am trying to configure some remote NTP servers in the XML config file to be used with autoyast, and while it appears to be working, I would like to know if I'm doing it the correct or best way. Here is the <ntp> section from the config: ============================================================ <ntp-client> <configure_dhcp config:type="boolean">false</configure_dhcp> <peers config:type="list"> <peer> <address>127.127.1.0</address> <comment>################################################################### ############# ## /etc/ntp.conf ## ## Sample NTP configuration file. ## See package 'ntp-doc' for documentation, Mini-HOWTO and FAQ. ## Copyright (c) 1998 S.u.S.E. GmbH Fuerth, Germany. ## ## Author: Michael Andres, <ma(a)suse.de> ## ############################################################################ #### ## ## Radio and modem clocks by convention have addresses in the ## form 127.127.t.u, where t is the clock type and u is a unit ## number in the range 0-3. ## ## Most of these clocks require support in the form of a ## serial port or special bus peripheral. The particular ## device is normally specified by adding a soft link ## /dev/device-u to the particular hardware device involved, ## where u correspond to the unit number above. ## ## Generic DCF77 clock on serial port (Conrad DCF77) ## Address: 127.127.8.u ## Serial Port: /dev/refclock-u ## ## (create soft link /dev/refclock-0 to the particular ttyS?) ## # server 127.127.8.0 mode 5 prefer ## ## Undisciplined Local Clock. This is a fake driver intended for backup ## and when no outside source of synchronized time is available. ## </comment> <fudge_comment></fudge_comment> <fudge_options> stratum 10 # LCL is unsynchronized</fudge_options> <initial_sync config:type="boolean">false</initial_sync> <options> # local clock (LCL)</options> <type>__clock</type> </peer> <peer> <address>/var/lib/ntp/drift/ntp.drift # path for drift file</address> <comment> ## ## Outside source of synchronized time ## ## server xx.xx.xx.xx # IP address of server ## ## Miscellaneous stuff ## </comment> <initial_sync config:type="boolean">false</initial_sync> <options></options> <type>driftfile</type> </peer> <peer> <address>/var/log/ntp # alternate log file</address> <comment> </comment> <initial_sync config:type="boolean">false</initial_sync> <options></options> <type>logfile</type> </peer> <peer> <address>192.168.3.2</address> <comment></comment> <initial_sync config:type="boolean">true</initial_sync> <options></options> <type>server</type> </peer> <peer> <address>192.168.3.86</address> <comment></comment> <initial_sync config:type="boolean">false</initial_sync> <options></options> <type>server</type> </peer> <peer> <address>uk.pool.ntp.org</address> <comment></comment> <initial_sync config:type="boolean">false</initial_sync> <options></options> <type>server</type> </peer> </peers> <start_at_boot config:type="boolean">true</start_at_boot> <start_in_chroot config:type="boolean">true</start_in_chroot> </ntp-client> ============================================================ Our intention is to time sync from one (or both) of our internal servers (192.168.3.x) or, if they are unavailable, sync from uk.pool.ntp.org. Ideally, we would like to sync from 192.168.3.2 and 192.168.3.86 and to take an average of the two, but I don't know if NTP works like this. If it only syncs from a single server at a time, then 192.168.3.2 should be attempted first, then 192.168.3.86, and then uk.pool.ntp.org. Is our config going to give us what we require? Thanks in advance to those who can help, it's very much appreciated :-) Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0

RE: [suse-autoinstall] Help with installing multiple machines over network
by Hans-Joachim Ehlers 07 Feb '06

07 Feb '06

Why not just using a simple post install script ? >>> "Richard Hobbs" <richard.hobbs(a)crl.toshiba.co.uk> 07/02/06 11:59:10 >>> Hello, .... This web page ( http://www.tuxmachines.org/node/4630 ) indicates that it's a known bug which has been fixed in SuSE 10.1b1, but is there any way around this when using SuSE 10.0? If not, I'll just have to modify it by hand after the install, right? Thanks again, Richard.

3 6

RE: [suse-autoinstall] Help with installing multiple machines over network
by Richard Hobbs 07 Feb '06

07 Feb '06

Hello, Thank you for all your help regarding this. For those who missed out on the solution (due to the conversation going away from the mailing list for a while), I basically had to remove the <install> tags as they have been deprecated! Everything I have tried so far is now working, except one small thing (which I think is a known bug). I have the runlevel specified in the autoyast config file, but it's being ignored. This web page ( http://www.tuxmachines.org/node/4630 ) indicates that it's a known bug which has been fixed in SuSE 10.1b1, but is there any way around this when using SuSE 10.0? If not, I'll just have to modify it by hand after the install, right? Thanks again, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 > -----Original Message----- > From: Uwe Gansert [mailto:ug@suse.de] > Sent: 06 February 2006 13:01 > To: Richard Hobbs > Subject: Re: [suse-autoinstall] Help with installing multiple > machines over network > > On Monday 06 February 2006 13:42, you wrote: > > > <mode> > > <confirm config:type="boolean">false</confirm> > > </mode> > > you might want to set this to "true" for the testing phase. > Autoyast stops > then before it is doing something with the harddisc. > > -- > ciao, Uwe Gansert > > Uwe Gansert, Server Technologies Team > SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany > e-mail: uwe.gansert(a)suse.de, Tel: +49-(0)911-74053-0, > Fax: +49-(0)911-74053-476, Web: http://www.suse.de > > _____________________________________________________________________ > This e-mail has been scanned for viruses by Verizon Business > Internet Managed Scanning Services - powered by MessageLabs. > For further information visit http://www.mci.com > > _____________________________________________________________________ This e-mail has been scanned for viruses by Verizon Business Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0

RE: [suse-autoinstall] Help with installing multiple machines ove r network
by Gunreben, Peter (Peter) 03 Feb '06

03 Feb '06

Richard, Richard Hobbs wrote: [...] > I have multiple machines, all of which are identical, onto > which I would like to install SuSE 10. > > I have already installed one machine and got it setup the way > I would like it, in terms of package selections, > partitioning etc... And I would now like to replicate that > across the remaining machines if possible. > > I have read the first section of the SuSE 10 manual, entitled > "Advanced Deployment Scenarios", but this only seems to document > how to boot the system into the manual installation routine > remotely. Ideally, I need to boot from the network and begin > a fully automated install which duplicates every single aspect > of the first system, except with a different hostname and IP > (which I don't mind entering manually during the install if > necessary). > > Can this be done? I'm sure it can, but I can't find any > guides or tutorials on how to do it, so if anyone can help, > that would be very much appreciated Yes, it can be done. Basically, you need two things: 1) The mechanism to boot from the network 2) The mechanism for the automatic installation ad 1) Regarding the boot from network, I'd recommend pxegrub: http://www.gnu.org/software/grub/manual/grub.html#Network Alternatively, you could also build a boot-CD, which contains bootloader, kernel and initrd and starts the network installation. However, this might not be your preferred way of installing the computers. ad 2) I'd recommend to use autoyast. The autoyast is the SuSE mechanism for automatic installation. Using "yast2 autoyast" you can generate an XML-File, which contains the entire configuration for the installation. To install using autoyast, you just need to boot the kernel load the Installation initrd (DVD:/boot/loader/initrd) and add the kernel-parameter install=nfs://<IP-Adressse>/<Pfad>/<XML-File> in your boot-loader configuration. Mailing-List: mailto:suse-autoinstall@suse.com Archive with search function: http://marc.theaimsgroup.com/?l=suse-autoinstall Documentation: http://www.suse.com/~ug/autoyast_doc Regards, Peter.

4 18

Problem with VGA.
by Rodrigo_Albert_Fedosi 03 Feb '06

03 Feb '06

Hi, I have a problem with the autoinstallation of the SuSE 10.0. In the end of installation the kdm isn't started, because the monitor stay out of the synchronization. The problem happens with some models of the monitor, for resolution of the problem I have that in text mode execute the command "sax2 --vesa 0:800x600@65". The models: Samsung 440b, 551v to the time. Thanks, Rodrigo Albert Fedosi Suporte TI Tel: (31) 3419-8854 Fax: (31) 3419-8803 rodrigo(a)bhz.jamef.com.br www.jamef.com.br

1 0

problem while creating partitions on 2 harddisks
by Michael D. 03 Feb '06

03 Feb '06

Hello! I found some very strange behaviour while trying to let autoyast create partitions on /dev/hda and /dev/hdb. The partitioning scheme I need is very simple: /dev/hda1 - /boot (primary, auto size, ext2) /dev/hda2 - swap (primary, auto size) /dev/hda3 - / (primary, max size, reiser) /dev/hdb1 - /home (primary, max size, reiser) Autoinstall was fine while I was using /dev/hda only. But after /dev/hdb1 was added, yast2 died while installing the RPMs - fried by the OOM-killer. I could strip down the problem to a possible bug in autoyast. The XML below contains only the partition info, it was created from scratch with autoyast from SUSE 10.0 (but SUSE 9.3 does the same). The partitioning in the first interactive install screen says only "create /dev/hdb1 with reiserfs", all info from /dev/hda ist lost. With my full autoinstall XML this screen was not shown, the installer has no real root, uses the ramdisk - and gets slayed some minutes later by the OOM-killer. Can anyone comfirm this problem, or am I misunderstanding some of the autoyast options (i.e.the "primary" or the "auto/max")? Thanks Michael ---- 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"> <partitioning config:type="list"> <drive> <device>/dev/hdb</device> <initialize config:type="boolean">true</initialize> <partitions config:type="list"> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <fstopt>noatime,acl,user_xattr</fstopt> <label></label> <loop_fs config:type="boolean">false</loop_fs> <mount>/home</mount> <mountby config:type="symbol">device</mountby> <noauto config:type="boolean">false</noauto> <partition_id config:type="integer">131</partition_id> <partition_type>primary</partition_type> <size>auto</size> </partition> </partitions> <use>all</use> </drive> <drive> <device>/dev/hda</device> <initialize config:type="boolean">true</initialize> <partitions config:type="list"> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">ext2</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>/boot</mount> <partition_id config:type="integer">131</partition_id> <partition_type>primary</partition_type> <size>auto</size> </partition> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <loop_fs config:type="boolean">false</loop_fs> <mount>swap</mount> <partition_id config:type="integer">130</partition_id> <partition_type>primary</partition_type> <size>auto</size> </partition> <partition> <crypt>twofish256</crypt> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <fstopt>acl,user_xattr</fstopt> <label></label> <loop_fs config:type="boolean">false</loop_fs> <mount>/</mount> <mountby config:type="symbol">device</mountby> <noauto config:type="boolean">false</noauto> <partition_id config:type="integer">131</partition_id> <partition_type>primary</partition_type> <size>max</size> </partition> </partitions> <use>all</use> </drive> </partitioning> </profile> ---- snip __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 2

RE: [suse-autoinstall] Help with installing multiple machines over network
by Schumacher, Gordon 02 Feb '06

02 Feb '06

# From: Richard Hobbs [mailto:richard.hobbs@crl.toshiba.co.uk] # Sent: Thursday, February 02, 2006 10:42 AM # To: 'Uwe Gansert'; suse-autoinstall(a)suse.com # Subject: RE: [suse-autoinstall] Help with installing multiple machines # over network [snip] # Also, I have read somewhere that I can use a configuration # server to store xml files for each machine I have, and given # the quantity of machines, this would be the ideal approach # for me because each one obviously needs a different IP # address and hostname, so I can't use the same XML file for # all the machines. I don't know whether this would work for you, but what I'm doing in my configuration is running DHCP, and launching a post-script to change the hostname. Then afterwards, I can go back and manually configure the system to where I want it: #!/bin/bash suffix=$(hexdump -n 4 -e '"%x"' /dev/urandom) echo myprefix-$suffix.your.domain.name >/etc/HOSTNAME cat >/etc/hosts <<@@@@End-Of-Hosts-File@@@@ # # hosts This file describes a number of hostname-to-address # mappings for the TCP/IP subsystem. It is mostly # used at boot time, when no name servers are running. # On small systems, this file can be used instead of a # "named" name server. # Syntax: # # IP-Address Full-Qualified-Hostname Short-Hostname # 127.0.0.1 localhost # special IPv6 addresses ::1 localhost ipv6-localhost ipv6-loopback fe00::0 ipv6-localnet ff00::0 ipv6-mcastprefix ff02::1 ipv6-allnodes ff02::2 ipv6-allrouters ff02::3 ipv6-allhosts 127.0.0.2 myprefix-$suffix.your.domain.name myprefix-$suffix @@@@End-Of-Hosts-File@@@@ ]]>

1 0

Help with installing multiple machines over network
by Richard Hobbs 02 Feb '06

02 Feb '06

Hello, I hope I have joined the correct mailing list here, but apologies if not. I have multiple machines, all of which are identical, onto which I would like to install SuSE 10. I have already installed one machine and got it setup the way I would like it, in terms of package selections, partitioning etc... And I would now like to replicate that across the remaining machines if possible. I have read the first section of the SuSE 10 manual, entitled "Advanced Deployment Scenarios", but this only seems to document how to boot the system into the manual installation routine remotely. Ideally, I need to boot from the network and begin a fully automated install which duplicates every single aspect of the first system, except with a different hostname and IP (which I don't mind entering manually during the install if necessary). Can this be done? I'm sure it can, but I can't find any guides or tutorials on how to do it, so if anyone can help, that would be very much appreciated :-) Thanks in advance, Richard. -- Richard Hobbs (Systems Administrator) Toshiba Research Europe Ltd. - Speech Technology Group Web: http://www.toshiba-europe.com/research/ Normal Email: richard.hobbs(a)crl.toshiba.co.uk Mobile Email: mobile(a)mongeese.co.uk Tel: +44 1223 376964 Mobile: +44 7811 803377 _____________________________________________________________________ This e-mail has been scanned for viruses by MCI's Internet Managed Scanning Services - powered by MessageLabs. For further information visit http://www.mci.com

1 0