AutoYaST IP and partitioning problems
Good morning! I have some problems using AutoYaST to test the automatic installation I made from within VMware. But first things first. My target is to generate a bootable DVD that should be used to "reset" testing servers. The installation should run as quietly and automatic as possible from inserting the DVD and powering up the system to a complete system with Samba & Oracle. The configuration files can be retrieved at http://www.badongo.com/file/1308770 While the used linuxrc.config & info files should tell AutoYaST to use a static IP address (using VMwares emulated AMD PCnet32 card) it always tries to fetch information from a DHCP server, which is not present, or skips this step all together without BOOTP fallback. This is a major problem since the autoinstall directives should be fetched by HTTP. As soon as the installation proceeds past this it stops short at partitioning. I already tried using fresh disks (remember, still in VMware) and disks that had all required partitions written but to no avail. Is this a general problem with VMware or did I find a glitch? I would really like to be able to continue my test within VMware since I don't want to stress out the hardware now and have it break too soon. Regards, Peter
On Tuesday 29 August 2006 09:40, Peter Ludikovsky wrote:
While the used linuxrc.config & info files should tell AutoYaST to use a static IP address (using VMwares emulated AMD PCnet32 card) it always tries to fetch information from a DHCP server, which is not present, or skips this step all together without BOOTP fallback. This is a major problem since the autoinstall directives should be fetched by HTTP.
are we talking about SLES10? I just tried it in vmware and for me it works fine with hostip=1.2.3.4 netmask=255.255.0.0 insmod=pcnet32 autoyast=http://... install=cd no DHCP request is sent then. You can try to add netsetup=-dhcp but I did not need it
As soon as the installation proceeds past this it stops short at partitioning. I already tried using fresh disks (remember, still in VMware) and disks that had all required partitions written but to no avail. Is this a general problem with VMware or did I find a glitch?
your partitioning is very complex. I suggest you to start with sda only and then add drive by drive, partition by partition until it fails.
I would really like to be able to continue my test within VMware since I don't want to stress out the hardware now and have it break too soon.
vmware should work. -- ciao, Uwe Gansert Uwe Gansert, Server Technologies Team SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany Business: http://www.suse.de/~ug now playing Funker Vogt - The International Killer
On Tuesday 29 August 2006 09:40, Peter Ludikovsky wrote:
While the used linuxrc.config & info files should tell AutoYaST to use a static IP address (using VMwares emulated AMD PCnet32 card) it always tries to fetch information from a DHCP server, which is not present, or skips this step all together without BOOTP fallback. This is a major problem since the autoinstall directives should be fetched by HTTP.
are we talking about SLES10? I just tried it in vmware and for me it works fine with
hostip=1.2.3.4 netmask=255.255.0.0 insmod=pcnet32 autoyast=http://... install=cd
no DHCP request is sent then. You can try to add netsetup=-dhcp but I did not need it isoinfo tells me the application ID is SuSE-Linux-10.1-DVD-i386-10.1#0 When specifying these options on the isolinux command line they work fine for me too. But they are ignored as soon as they should be read from /info or /linuxrc.config which worked when I ran the last setup test 2 weeks ago. Since then the only changes made were at the
2006/8/29, Uwe Gansert <ug@suse.de>: post-install scripts.
As soon as the installation proceeds past this it stops short at partitioning. I already tried using fresh disks (remember, still in VMware) and disks that had all required partitions written but to no avail. Is this a general problem with VMware or did I find a glitch?
your partitioning is very complex. I suggest you to start with sda only and then add drive by drive, partition by partition until it fails. Tried that. Thing is, like the problem with configuring the IP, it worked before. For reference, the partition layout is as follows: Partition FS Mount Size sda1 ext2 /boot auto sda2 swap max / fixed sda5 reiserfs / fixed sda6 reiserfs /usr fixed sda7 reiserfs /var fixed sd{b,c,d,e,f}1 raid5 max sd{b,c,d,e,f}2 swap fixed
I would really like to be able to continue my test within VMware since I don't want to stress out the hardware now and have it break too soon.
vmware should work. Good to know.
Regards, Peter
2006/8/29, Peter Ludikovsky <peter.ludikovsky@gmail.com>:
Good morning!
I have some problems using AutoYaST to test the automatic installation I made from within VMware. But first things first.
My target is to generate a bootable DVD that should be used to "reset" testing servers. The installation should run as quietly and automatic as possible from inserting the DVD and powering up the system to a complete system with Samba & Oracle.
The configuration files can be retrieved at http://www.badongo.com/file/1308770
While the used linuxrc.config & info files should tell AutoYaST to use a static IP address (using VMwares emulated AMD PCnet32 card) it always tries to fetch information from a DHCP server, which is not present, or skips this step all together without BOOTP fallback. This is a major problem since the autoinstall directives should be fetched by HTTP.
As soon as the installation proceeds past this it stops short at partitioning. I already tried using fresh disks (remember, still in VMware) and disks that had all required partitions written but to no avail. Is this a general problem with VMware or did I find a glitch?
I would really like to be able to continue my test within VMware since I don't want to stress out the hardware now and have it break too soon.
Regards, Peter
Update: I got both partitioning & static IP running. For static IP I added netsetup: -dhcp to linuxrc.config For partitioning I had to disable all scripts. As soon as I re-enable one of them partitioning fails again and yes, this error is reproduceable. If anyone knows how to get them working (esp. init-scripts) mail me. Regards, Peter
On Friday 01 September 2006 11:22, Peter Ludikovsky wrote:
Update: I got both partitioning & static IP running.
For static IP I added netsetup: -dhcp to linuxrc.config For partitioning I had to disable all scripts. As soon as I re-enable one of them partitioning fails again and yes, this error is reproduceable. If anyone knows how to get them working (esp. init-scripts) mail me.
that's very unsual and I don't see any connection between the scripts section and the partitioning section. Do you mixup <install> <configure> and section outside of both? Either use <install> and <configure> for the whole profile or leave those elements out completely (they are not needed on 10.1/SLES10 anymore). For SP1, 10.2 the compatibility code is a lot more fault-tolerant in that way but on 10.1 and SLES10 you simply should not use <install> and <configure> at all. -- ciao, Uwe Gansert Uwe Gansert, Server Technologies Team SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany Business: http://www.suse.de/~ug now playing Funker Vogt - Tragic Hero
participants (2)
-
Peter Ludikovsky -
Uwe Gansert