Bug 245701 - grub uses by-ID but before used by-label
Summary: grub uses by-ID but before used by-label
Status: RESOLVED DUPLICATE of bug 246684
Alias: None
Product: openSUSE 10.3
Classification: openSUSE
Component: Update Problems (show other bugs)
Version: Alpha 1
Hardware: Other Other
: P5 - None : Normal (vote)
Target Milestone: ---
Assignee: Olaf Dabrunz
QA Contact: Jiri Srain
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-02-15 09:25 UTC by Andreas Jaeger
Modified: 2007-02-19 09:06 UTC (History)
1 user (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
Logfiles (713.83 KB, application/x-bzip2)
2007-02-15 09:26 UTC, Andreas Jaeger
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Andreas Jaeger 2007-02-15 09:25:20 UTC
* Installed 10.2 and added everything in fstab with LABELS.
* LABELS are used in grub.conf
* Update to 10.3 Alpha1: grub.conf uses root=/dev/by-id/ instead of /dev/by-label

During an update the method should not get changed.
Comment 1 Andreas Jaeger 2007-02-15 09:26:34 UTC
Created attachment 119333 [details]
Logfiles
Comment 2 Thomas Fehr 2007-02-15 10:22:08 UTC
I have neither to do with bootloader config nor with update code. 
Update is doing its own fstab reading so if it detects a certain mount-by 
method it needs to put this information into TargetMap, since bootloader 
gets its information from TargetMap.

The mount-by method is specified by the key "mount_by" in the map describing
a partition. This can have the values `device, `id, `path, `label or `uuid.
New default is `id since SLES10 SP1 and SL 10.3 (it was `device before).

Lukas, maybe you should get rid of doing your own fstab analysation in update
code. E.g. there is a function Storage::ReadFstab which gets a directory 
with fstab (and potentially cryptotab) as parameter and returns a list of maps
with read fstab entries. I use this function when doing the "Import existing
fstab" in expert partitoner. No idea if it currently fits you needs but maybe
you should look into using it. If you need any changes just tell me.

BTW this same problem will hit when doing an update from SLES10.
Comment 3 Lukas Ocilka 2007-02-19 08:56:37 UTC
This comes from y2log-1 (first stage):
  bootloader/routines/lib_iface.ycp:151 Storing bootloader sections [$["append":"splash=silent  showopts", "image":"/boot/vmlinuz", "initrd":"/boot/initrd", "key":"/boot/initrd", "name":"openSUSE FACTORY 10.3", "original_name":"linux", "root":"/dev/disk/by-id/ata-HITACHI_DK13FA-40B_3LU814-part1", "type":"image", "vgamode":"0x317"], $["append":"hwprobe=modules.pata resume=/dev/sda3 splash=silent showopts", "image":"(hd0,0)/boot/vmlinuz-2.6.20-2-default", "initial":"1", "initrd":"(hd0,0)/boot/initrd-2.6.20-2-default", "key":"(hd0,0)/boot/initrd-2.6.20-2-default", "lines_cache_id":"0", "name":"Kernel-2.6.20-2-default", "root":"/dev/disk/by-id/ata-HITACHI_DK13FA-40B_3LU814-part1", "type":"image", "vgamode":"0x317"], $["append":"showopts ide=nodma apm=off acpi=off noresume nosmp noapic maxcpus=0 edd=off 3", "image":"/boot/vmlinuz", "initrd":"/boot/initrd", "key":"/boot/initrd", "name":"Failsafe -- openSUSE FACTORY 10.3", "original_name":"failsafe", "root":"/dev/disk/by-id/ata-HITACHI_DK13FA-40B_3LU814-part1", "type":"image", "vgamode":"normal"], $["append":"   splash=silent showopts", "image":"/boot/vmlinuz-xen", "initrd":"/boot/initrd-xen", "key":"/boot/initrd-xen", "name":"XEN", "original_name":"xen", "root":"/dev/disk/by-id/ata-HITACHI_DK13FA-40B_3LU814-part1", "type":"xen", "vgamode":"0x317", "xen":"/boot/xen.gz", "xen_append":""]]

This comes from y2log (second stage):
  kernel /boot/vmlinuz-2.6.20-2-default root=/dev/disk/by-label/ROOT vga=0x317 hwprobe=modules.pata resume=/dev/sda3 splash=silent showopts

Reassigning to 'odabrunz', creating new separate bug #246684.
Comment 4 Stefan Fent 2007-02-19 09:06:55 UTC
So, I'll close this duplicate to bug #246684, as the bootloader correctly uses the mount_by value, which is wrong here.

*** This bug has been marked as a duplicate of bug 246684 ***