Skip to contentSkip to conversation

btrfs-tools backport for Debian Lenny

DISCUSSION Martin Steigerwald 3 participants 4 messages Feb 17, 2010 Btrfs file system

lore
Share & tools
lore
#0 lore raw

Cc'd backports-users mailing list, cc'd package maintainer Daniel Baumann as well.

Hi!

JFYI - most of you might prefer to run btrfs-tools from git -, I made a btrfs-tools 0.19 backport for Debian Lenny from the package in Debian Squeeze and put it online at:

http://people.teamix.net/~ms/debian/lenny-backports/btrfs-tools/

You can find discussion on sponsoring it here:

http://lists.backports.org/lurker-bpo/thread/20100217.143508.41736c9f.en.html

I think it won't work with the latest backports 2.6.30 kernel due to btrfs format changes, I test it with a self compiled 2.6.32.3. A 2.6.32 for backports org should be due soon and then my btrfs-tools backport might appear in the bpo archive as well.

It basically works, but I am wondering it the RAID options work correctly. I used:

mkfs.btrfs -d raid1 -m
raid1 /dev/mango1/homelokal1 /dev/mango2/homelokal2

on two 200 GiB logical volumes and get

mango:~# df -hT /mnt/zeit
Dateisystem   Typ    Größe Benut  Verf Ben% Eingehängt auf
/dev/mapper/mango1-homelokal1
             btrfs    400G  101M  400G   1% /mnt/zeit

Shouldn't that be 200GiB for a BTRFS Raid 1 setup?

Ciao,

Signature · 2 lines
Martin Steigerwald - team(ix) GmbH - http://www.teamix.de
gpg: 19E3 8D42 896F D004 08AC A0CA 1E10 C593 0399 AE90
#1 lore raw
Quoted 31 lines · Martin Steigerwald

Cc'd backports-users mailing list, cc'd package maintainer Daniel Baumann as
well.

Hi!

JFYI - most of you might prefer to run btrfs-tools from git -, I made a btrfs-tools 0.19 backport for Debian Lenny from the package in Debian Squeeze
and put it online at:

http://people.teamix.net/~ms/debian/lenny-backports/btrfs-tools/

You can find discussion on sponsoring it here:

http://lists.backports.org/lurker-bpo/thread/20100217.143508.41736c9f.en.html

I think it won't work with the latest backports 2.6.30 kernel due to btrfs format changes, I test it with a self compiled 2.6.32.3. A 2.6.32 for backports org should be due soon and then my btrfs-tools backport might appear in the bpo archive as well.

It basically works, but I am wondering it the RAID options work correctly. I
used:

mkfs.btrfs -d raid1 -m
raid1 /dev/mango1/homelokal1 /dev/mango2/homelokal2

Quoted 10 lines

on two 200 GiB logical volumes and get

mango:~# df -hT /mnt/zeit
Dateisystem   Typ    Größe Benut  Verf Ben% Eingehängt auf
/dev/mapper/mango1-homelokal1
            btrfs    400G  101M  400G   1% /mnt/zeit

Shouldn't that be 200GiB for a BTRFS Raid 1 setup?

this has been discussed several times on the list.

Since btrfs file level redundancy policy, the proper way for it to behave is to show full physical capacity of the devices.

Usage will be correctly accounted, so if you write a 100MB file in that raid1 FS tree
the used space will indicate 200MB of additional used space.

Regards,

Quoted 5 lines

Ciao,
--
Martin Steigerwald - team(ix) GmbH - http://www.teamix.de
gpg: 19E3 8D42 896F D004 08AC A0CA 1E10 C593 0399 AE90

Signature · 1 line
Miguel Sousa Filipe
#2 lore raw

On Wed, Feb 17, 2010 at 3:32 PM, Martin Steigerwald <[email protected]> wrot= e:

Quoted 2 lines

Cc'd backports-users mailing list, cc'd package maintainer Daniel Bau=

mann as

Quoted 6 lines

well.

Hi!

JFYI - most of you might prefer to run btrfs-tools from git -, I made=

a

Quoted 1 line

btrfs-tools 0.19 backport for Debian Lenny from the package in Debian=

Squeeze

Quoted 7 lines

and put it online at:

http://people.teamix.net/~ms/debian/lenny-backports/btrfs-tools/

You can find discussion on sponsoring it here:

http://lists.backports.org/lurker-bpo/thread/20100217.143508.41736c9f=

=2Een.html

Quoted 2 lines

I think it won't work with the latest backports 2.6.30 kernel due to =

btrfs

Quoted 2 lines

format changes, I test it with a self compiled 2.6.32.3. A 2.6.32 for backports org should be due soon and then my btrfs-tools backport mig=

ht

Quoted 3 lines

appear in the bpo archive as well.

It basically works, but I am wondering it the RAID options work corre=

ctly. I

Quoted 9 lines

used:

mkfs.btrfs -d raid1 -m
raid1 /dev/mango1/homelokal1 /dev/mango2/homelokal2

on two 200 GiB logical volumes and get

mango:~# df -hT /mnt/zeit
Dateisystem =C2=A0 Typ =C2=A0 =C2=A0Gr=C3=B6=C3=9Fe Benut =C2=A0Verf =

Ben% Eingeh=C3=A4ngt auf

Quoted 2 lines

/dev/mapper/mango1-homelokal1
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 btrfs =C2=A0 =C2=A0400G =C2=

=A0101M =C2=A0400G =C2=A0 1% /mnt/zeit

Quoted 3 lines

Shouldn't that be 200GiB for a BTRFS Raid 1 setup?

this has been discussed several times on the list. Since btrfs file level redundancy policy, the proper way for it to behave is to show full physical capacity of the devices. Usage will be correctly accounted, so if you write a 100MB file in that raid1 FS tree
the used space will indicate 200MB of additional used space. Regards,

--
Miguel Sousa Filipe
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" = in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw

Am Mittwoch 17 Februar 2010 schrieben Sie:

Quoted 1 line

On Wed, Feb 17, 2010 at 3:32 PM, Martin Steigerwald <[email protected]>

wrote:

Leaving Cc to backports-users mailinglist and Debian package maintainer dropped as its a technical BTRFS discussion.

Quoted 1 line
Quoted 1 line

Hi!

[...]

Quoted 8 lines
Quoted 8 lines

It basically works, but I am wondering it the RAID options work correctly. I used:

mkfs.btrfs -d raid1 -m
raid1 /dev/mango1/homelokal1 /dev/mango2/homelokal2

on two 200 GiB logical volumes and get

Quoted 8 lines
Quoted 6 lines
mango:~# df -hT /mnt/zeit
Dateisystem   Typ    Größe Benut  Verf Ben% Eingehängt auf
/dev/mapper/mango1-homelokal1
            btrfs    400G  101M  400G   1% /mnt/zeit

Shouldn't that be 200GiB for a BTRFS Raid 1 setup?

this has been discussed several times on the list.

Sorry, I admit I follow BTRFS irregularily and main point in my mail was to announce the backport.

Quoted 5 lines

Since btrfs file level redundancy policy, the proper way for it to behave is to show full physical capacity of the devices. Usage will be correctly accounted, so if you write a 100MB file in that raid1 FS tree
the used space will indicate 200MB of additional used space.

Hmmm, that still doesn't match, cause I wrote a 100MiB file to the partition:

mango:~# dd if=/dev/zero of=/mnt/zeit/ddimg bs=1M count=100
100+0 Datensätze ein
100+0 Datensätze aus
104857600 Bytes (105 MB) kopiert, 0,313594 s, 334 MB/s

Thus I should have get 200MiB used in df -h. Hmmm, maybe accurate size reporting needs a newer BTRFS than in 2.6.32.3?

Ciao,

Signature · 2 lines
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7