Re: block size issue when re-syncing RAID1 with ext2fs on top

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



* Neil Brown <neilb@xxxxxxx> [2006-01-16 23:38]:
> On Monday January 16, lists@xxxxxxxxxxxxxxxxxxx wrote:
> > Hi,
> > 
> > I'm experiencing a problem on a 2.2.16C37_III driven Cobalt RaQ4
> > after I add a new 2nd disk to a RAID1.
> 
> 2.2.16 .... that's old, isn't it!

No question :-) I don't really use it voluntarily ;-)

> raid1 was only every available as external patches for 2.2 kernels...

Ah, hmm. That is a heavily patched Cobalt kernel, indeed.

> > I'm uncertain whether this is a RAID, ext2fs or even a hardware
> > issue, that's why I'm writing both to ext2-devel and linux-raid.
> > 
> > Setup:
> > - /dev/md1(hd[ab]1): ext2: /
> > - /dev/md3(hd[ab]3): ext2: /var
> > - /dev/md4(hd[ab]4): ext2: /home
> > - /dev/md6(hd[ab]6): swap: -
> > 
> > When I boot up the system with only hda, everything is fine (except
> > that the RAIDs run degraded, but that's expected :P), even e2fsck
> > doesn't complain when checking md[134].
> > 
> > Now, when I connect hdb and boot the machine, the Cobalt init scripts
> > detect a broken RAID, clone the hda partition table to hdb and then
> > add the newly created hdb partitions to the corresponding md.
> > That's the point where it breaks:
> > 
> > --8<--[ lots of ]--8<--
> > ll_rw_block: device 03:01: only 4096-char blocks implemented (1024)
> > EXT2-fs error (device ide0(3,1)): ext2_write_inode: unable to read \
> > 	inode block [...]
> > --8<--
> 
> The fact that EXT2 is reporting an error with ide0(3,1) is a worry.
> Ext2 shouldn't even see the ide drives.  It should just see md devices
> which should hide all the underlying details.
> 
> So I suspect the problem is that you are mounting the root filesystem
> of hda1 rather than md1
> 
> The resync code will set the blocksize for hda1 to 4096, and the ext2
> code will try to read 1024byte blocks, and it fails with the message
> you see.

Ah! That enlightens the dark. I guess it's due to the fact that the
Cobalt boot loader seems to pass "root=/dev/hda1" instead of
"root=/dev/md1" to the kernel.

Thanks a lot!
-- 
Wolfram Schlich
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Linux RAID Wiki]     [ATA RAID]     [Linux SCSI Target Infrastructure]     [Linux Block]     [Linux IDE]     [Linux SCSI]     [Linux Hams]     [Device Mapper]     [Device Mapper Cryptographics]     [Kernel]     [Linux Admin]     [Linux Net]     [GFS]     [RPM]     [git]     [Yosemite Forum]


  Powered by Linux