Re: Cant find sunit and swidth on luks encrypted raid6 device

[Eric Sandeen <sandeen@sandeen.net>]

Jochen K. wrote:

> [output with encryption]
> 
> mkfs.xfs -f /dev/mapper/raid6
> 
> meta-data=/dev/mapper/raid6      isize=256    agcount=32, agsize=53412581
> blks
>          =                       sectsz=512   attr=0
> data     =                       bsize=4096   blocks=1709202592, imaxpct=25
>          =                       sunit=0      swidth=0 blks, unwritten=1
> naming   =version 2              bsize=4096
> log      =internal log           bsize=4096   blocks=32768, version=1
>          =                       sectsz=512   sunit=0 blks, lazy-count=0
> realtime =none                   extsz=4096   blocks=0, rtextents=0
> b7f57ad0: Badness in key lookup (length)
> bp=(bno 13673620728, len 131072 bytes) key=(bno 13673620728, len 4096 bytes)

What generates the last two lines in the above message, and do you know
what it indicates?

For one thing, I notice your filesystem comes out as different sizes,
1709202880 without encryption, 1709202592 with encryption (i.e.,
smaller)  (these are 4k block units)

I assume the error message is trying to read or write 131072 bytes from
the offset 13673620728 * 512, or:

   7000893812736 -> 7000893943807

The filesystem claims to end at:

   1709202592 * 4096 = 7000893816832

which appears to be in the middle of the above range.

131072 is the amount that mkfs.xfs tries to zero at the end of a block
device.  (you could confirm that it's this write by stracing mkfs,
perhaps?)  It looks like somebody has the size wrong...

What does "blockdev --getsize64" and "blockdev --getsize" and/or
/proc/partitions say about how big /dev/mapper/raid6 is?

Can you use dd to write to the last 128k of the device, as reported by
blockdev --getsize64?

-Eric