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

Re: btrfs question



On Sat, 2013-02-23 at 21:06 -0600, Robert G. (Doc) Savage wrote:
> The man page for mkfs.btrfs says:
> 
>        -m, --metadata profile
>               Specify  how  metadata must be spanned across the devices speci-
>               fied. Valid values are raid0, raid1, raid10 or single.
> 
> I'm not quite sure how literally to read this. I have a software RAID5
> array /dev/md127p1 that I'm trying to decide whether or not to reformat
> from ext4 to btrfs. Since the RAID5 array already exists, would I just
> include "-m single" in the command line?

OK... This worked for CentOS 6.3:

        [root@lion] ~
        # mkfs.btrfs -d single -L pub -m single /dev/md127p1
        
        WARNING! - Btrfs Btrfs v0.19 IS EXPERIMENTAL
        WARNING! - see http://btrfs.wiki.kernel.org before using
        
        fs created label pub on /dev/md127p1
        	nodesize 4096 leafsize 4096 sectorsize 4096 size 2.00TB
        Btrfs Btrfs v0.19

What was surprising, in the time it took to hit Enter the reformat to
btrfs was done. The resulting volume has no "lost+found" directory.

I added this line to /etc/fstab:

        /dev/md127p1	/pub	btrfs	defaults    1 2

and mounted it with a simple:

        # mount /pub

The result reported by df -h is:

        /dev/md127p1    2.0T   28K  2.0T   1% /pub

I'm guessing the 28K shown allocated by df is the btrfs metadata.

--Doc Savage
  Fairview Heights, IL


-
To unsubscribe, send email to majordomo@silug.org with
"unsubscribe silug-discuss" in the body.