Hi, one problem solved, but partitioning still fails: Uwe Gansert wrote

On Monday 04 September 2006 12:45, Frank Steiner wrote:

...

I'm trying to install SuSE 10.1 on a system with one SCSI disk. I used a profile to keep all existing partitions:

what architecture do you use? On PPC for example, such a partition is added. On i586 it's not.

Architecture is i586 and it is added because YaST2 thinks that the bios of my host does not support lba, see my first mail:

Partitions.ycp: BootCycl lba_support false Partitions.ycp: BootCycl /proc/xen/capabilities $[] Partitions.ycp: BootCycl lba_support false autoinstall/autoaprt.ycp: boot added automagically...

I looked at Partitioning.ypc and it looks like the bios does not return the right value: list internal_bios = (list<map>) SCR::Read (.probe.bios); boolean lba = internal_bios[0, "lba_support"]:false; Due to the log file, lba obviously is set to false. I've no idea why that happens, the host is 2 years old and of course supports lba. Maybe it's because no IDE disk is in the system and the SCSI controller is not onboard, I don't know. Always worked fine with SuSE 9.0 and SLES9. I compared this to SLES9, and there the code was different. Especially one could set the "lba32" option for lilo and YaST2 would then set lba to true. In 10.1 the lba32-option is ignored, so it's once more that you are losing control and YaST2 just relies on auto-detection which fails. Just like with the insmod parameter... Anyway, what I did was: - patch the storage/src/modules/Partitions.ycp to global define integer BootCyl() ``{ + boot_cyl = 4*1024*1024*1024; if( boot_cyl == 0 ) so that boot_cyl is always large enough. Works here because all our hosts support lba32. - recompile yast2-storage package - put the compiled Partitions.ybc into the driverupdate file and installed my patched driverupdate Then AY works on the patched Partitions.ybc file and no /boot is created. However, the partitioning still fails. I've flipped the order for Partition 2 and 3 as you recommended (although I don't understand why this makes a difference when I use "partition_nr" for every partition), but now I end up with /dev/sda1 / ext3 /dev/sda2 swap swap /dev/sda3 extended partition /dev/sda5 /var ext3 /dev/sda6 /export reiserfs Although I specify four primary partitions (or does partition_nr not correspond to /dev/sda$partition_nr? Current profile is below. Do you have an idea what goes wrong here? cu, Frank <partitioning config:type="list"> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/</mount> <partition_nr config:type="integer">1</partition_nr> <size>20gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>swap</mount> <partition_nr config:type="integer">2</partition_nr> <size>2gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/var</mount> <partition_nr config:type="integer">3</partition_nr> <size>4gb</size> </partition> <partition> <create config:type="boolean">true</create> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/export</mount> <partition_nr config:type="integer">4</partition_nr> <size>max</size> </partition> </partitions> <use>all</use> </drive> </partitioning> -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

Uwe Gansert wrote

On Monday 04 September 2006 14:55, Frank Steiner wrote:

...

Architecture is i586 and it is added because YaST2 thinks that the bios of my host does not support lba, see my first mail:

ah, okay. I did not read that carefully enough.

I think some option to force AY to ignore the 1024 cylinders stuff would be useful, now that the lba32 option is ignored. Because hacking an ugly patch into driverupdates is not nice :-(

...

Although I specify four primary partitions (or does partition_nr not correspond to /dev/sda$partition_nr?

it does but it's only used when reusing partitions In your case, you need the partition_type

Ah, I missed that, sorry! Creation now works with the partition_type! Thanks! I guess the docu should be changed because I stepped on this sentence "You can force autoyast to create only primary partitions by configuring only partition numbers below 5." at http://www.suse.de/~ug/autoyast_doc/CreateProfile.Partitioning.html and so I though that partition_nr would work for creating partitions, too. Anyway, I'm still having problems :-) Now I created the partitions, and for testing, I used the same profile, just that now I keep all partitions and just format some. But then I get "Error while configuring partitions". fdisk -l shows Disk /dev/sda: 146.7 GB, 146771896320 bytes 255 heads, 63 sectors/track, 17844 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sda1 1 1914 15374173+ 83 Linux /dev/sda2 1915 2553 5132767+ 82 Linux swap / Solaris /dev/sda3 2554 2685 1060290 83 Linux /dev/sda4 2686 17844 121764667+ 83 Linux and my profile is below. Is there anything obviously wrong in the profile? I've looked into y2log and don't understand the algorithm in autopart.ycp good enough to see why it is failing. The y2log.gz is here if you want to take a look: http://www.bio.ifi.lmu.de/~steiner/y2log.gz cu, Frank <partitioning config:type="list"> <drive> <device>/dev/sda</device> <partitions config:type="list"> <partition> <create config:type="boolean">false</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/</mount> <partition_nr config:type="integer">1</partition_nr> <partition_type>primary</partition_type> </partition> <partition> <create config:type="boolean">false</create> <filesystem config:type="symbol">swap</filesystem> <format config:type="boolean">false</format> <fstopt>defaults</fstopt> <mount>swap</mount> <partition_nr config:type="integer">2</partition_nr> <partition_type>primary</partition_type> </partition> <partition> <create config:type="boolean">false</create> <filesystem config:type="symbol">ext3</filesystem> <format config:type="boolean">true</format> <fstopt>defaults</fstopt> <mount>/var</mount> <partition_nr config:type="integer">3</partition_nr> <partition_type>primary</partition_type> </partition> <partition> <create config:type="boolean">false</create> <filesystem config:type="symbol">reiser</filesystem> <format config:type="boolean">false</format> <fstopt>defaults</fstopt> <mount>/export</mount> <partition_nr config:type="integer">4</partition_nr> <partition_type>primary</partition_type> </partition> </partitions> </drive> </partitioning> -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

Uwe Gansert wrote

that's doable but actually I would like to know why the lba stuff is detected wrong.

Can I do sth. to create a more detailed log? I don't see the values coming from the bios in y2log...

...

Thanks! I guess the docu should be changed because I stepped on this sentence

I have changed the code already, so it fits the documentation again :)

Ok, that's fine too :-)

...

Anyway, I'm still having problems :-) Now I created the partitions, and for testing, I used the same profile, just that now I keep all partitions and just format some. But then I get "Error while configuring partitions". fdisk -l shows

interesting. That's a bug that seems to be triggered when reusing all 4 primary partitions. I found the place where the bug is in the code but I'm not sure about the sideeffects if I change that (it's pre-Uwe-Gansert Code). I need to take a closer look to the fix and do some tests.

That would be great, because I expect to run into the same problem with SLES 10 on 30 IBM xSeries that have the same configuration. It would be great if I didn't have to install all of them manually :-) If you like, I can beta-test your pre-Uwe-Gansert-Code on my system, I really don't mind doing that. If you want to send me some .ybc (or .ycp files that I can recompile) I can use them via driverupdate and give you some feedback. It's a test system, so it doesn't matter if a sideeffect blows up the whole disk :-) cu, Frank -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *

Uwe Gansert wrote

I have attached a patch for autoinst.ycp After you have applied the patch, you need to recompile AutoinstStorage.ycp

Yep, that works :-) I couldn't detect problems with other configurations (with extended partitions), but we always specify everything and don't let anything be chosen automatically, so I can't tell about possible side-effects in that case. Thanks a lot! Now we can handle our servers :-) cu, Frank -- Dipl.-Inform. Frank Steiner Web: http://www.bio.ifi.lmu.de/~steiner/ Lehrstuhl f. Bioinformatik Mail: http://www.bio.ifi.lmu.de/~steiner/m/ LMU, Amalienstr. 17 Phone: +49 89 2180-4049 80333 Muenchen, Germany Fax: +49 89 2180-99-4049 * Rekursion kann man erst verstehen, wenn man Rekursion verstanden hat. *