Am Sonntag, 30. Dezember 2012, 21:38:35 schrieb Andrey Borzenkov:

В Sun, 30 Dec 2012 14:57:34 +0100

Markus Koßmann <mkossmann_ml1@gmx.de> пишет:

...

Everytime when grub2-branding-openSUSE is update, I'am getting:

( 6/32 ) Installation von: grub2-branding-openSUSE-12.3-1.16 ..............................[fertig] Zusätzliche rpm-Ausgabe: Generating grub.cfg ... Found theme: /boot/grub2/themes/openSUSE/theme.txt Found linux image: /boot/vmlinuz-3.7.0-1-desktop Found initrd image: /boot/initrd-3.7.0-1-desktop Found linux image: /boot/vmlinuz-3.7.0-1-default Found initrd image: /boot/initrd-3.7.0-1-default Found linux image: /boot/vmlinuz-3.6.3-1-desktop Found initrd image: /boot/initrd-3.6.3-1-desktop Found linux image: /boot/vmlinuz-3.6.3-1-default Found initrd image: /boot/initrd-3.6.3-1-default ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 Found Windows 7 (loader) on /dev/sda1 Found openSUSE 11.4 (x86_64) on /dev/sda6

Running fsck -f on that ext4 partition doesn't find any problem however. But on a clone of that partition done with dd that error also shows up. Even if it is not mounted. Very strange. Does anyone have an idea, what might cause that problem, how to fix it and ( if needed) to file a bug against what ?

The error itself is emitted by dmraid. It is called as part of building grub2.cfg by os-prober script. So you should file it against dmraid.

Try "dmraid -sa -c" or "dmraid -r -c" whether it emits the same error.

Neither I'am running any dmraid nor there is a sil SATA controler in that system. It's allready caused by the discovery because a dmraid -vvvvr says: NOTICE: /dev/dm-8: sil discovering WARN: sil: major version 65280 in area 4; format handler tested on version 2 only ERROR: sil: invalid metadata checksum in area 4 on /dev/dm-8 NOTICE: /dev/dm-8: via discovering Looking into the sources of dmraid i found the following code in .../ataraid/sil.c: is_sil(struct sil *sil) { return SIL_MAGIC_OK(sil) && sil->disk_number < 8; } static int sil_valid(struct lib_context *lc, struct dev_info *di, void *meta, unsigned int area) { struct sil *sil = meta; if (!is_sil(sil)) return 0; if (sil->major_ver != 2) log_warn(lc, "%s: major version %u in area %u; format handler " "tested on version 2 only", handler, sil->major_ver, area); if (!checksum(sil)) LOG_ERR(lc, 0, "%s: invalid metadata checksum in area %u on %s", handler, area, di->path); if (di->sectors < sil->thisdisk_sectors) LOG_ERR(lc, 0, "%s: invalid disk size in metadata area %u on %s", handler, area, di->path); return 1; } and sil.h defines #define SIL_MAGIC 0x3000000 #define SIL_MAGIC_OK(sil) ((sil->magic & 0x3ffffff) == SIL_MAGIC) Obviously by bad luck is_sil() returns true on that partition and that can only be fixed if someone knows a more safe way to discover a sil ataraid. To check that I zeroed out the clone with dd , made a fresh ext4 filesystem and then copied the data over with tar ( instead of dd) . And that copy doesn't show the message any more. I will send a mail to the ataraid list about this issue. But I don't think that it makes sense to open a SUSE bugzilla here. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org