...and it is, remounting the OSD's with a fixed allocsize of 128M e.g:
$ mount
...
/dev/vdb1 on /var/lib/ceph/osd/ceph-0 type xfs (rw,allocsize=128M)
prevents the previously observed transient 2x space utilization.
Reading
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?i
Yeah, I was looking at preallocation as the likely cause, but your link
is way better than anything I'd found (especially with the likely commit
- speculative preallocation - mentioned
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=055388a3188f56676c21e92962fc366
XFS aggressively preallocates blocks to prevent fragmentation.
Try du --apparent-size. You can tune preallocation for xfs filesystem with
'allocsize' mount option.
Check out this
http://serverfault.com/questions/406069/why-are-my-xfs-filesystems-suddenly-consuming-more-space-and-full-of-sparse-fil
On 12/04/14 05:42, Gregory Farnum wrote:
On Wed, Apr 9, 2014 at 8:41 PM, Mark Kirkwood
Looks like the actual object has twice the disk footprint. Interestingly,
comparing du vs ls info for it at that point shows:
$ ls -l
total 2097088
-rw-r--r-- 1 root root 1073741824 Apr 10 15:33 file__head_2
On Wed, Apr 9, 2014 at 8:41 PM, Mark Kirkwood
wrote:
> Redoing (attached, 1st file is for 2x space, 2nd for normal). I'm seeing:
>
> $ diff osd-du.0.txt osd-du.1.txt
> 924,925c924,925
> < 2048 /var/lib/ceph/osd/ceph-1/current/5.1a_head/file__head_2E6FB49A__5
> < 2048/var/lib/ceph/osd/ceph-1/cu
-boun...@lists.ceph.com
[mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Mark Kirkwood
Sent: Thursday, April 10, 2014 7:21 PM
To: Udo Lembke; ceph-users@lists.ceph.com
Subject: Re: [ceph-users] OSD space usage 2x object size after rados put
On 11/04/14 06:35, Udo Lembke wrote:
> Hi,
>
> On 10
On 11/04/14 06:35, Udo Lembke wrote:
Hi,
On 10.04.2014 20:03, Russell E. Glaue wrote:
I am seeing the same thing, and was wondering the same.
We have 16 OSDs on 4 hosts. The File system is Xfs. The OS is CentOS 6.4. ceph
version 0.72.2
I am importing a 3.3TB disk image into a rbd image.
At 2
n I would
expect 2x disk utilization, which accounts for replication.
-RG
- Original Message -
From: "Udo Lembke"
To: ceph-users@lists.ceph.com
Sent: Thursday, April 10, 2014 1:35:51 PM
Subject: Re: [ceph-users] OSD space usage 2x object size after rados put
Hi,
On 10.04.2014
Hi,
On 10.04.2014 20:03, Russell E. Glaue wrote:
> I am seeing the same thing, and was wondering the same.
>
> We have 16 OSDs on 4 hosts. The File system is Xfs. The OS is CentOS 6.4.
> ceph version 0.72.2
>
> I am importing a 3.3TB disk image into a rbd image.
> At 2.6TB, and still importing, 5
-
From: "Mark Kirkwood"
To: "Gregory Farnum"
Cc: ceph-users@lists.ceph.com
Sent: Thursday, April 10, 2014 5:36:11 AM
Subject: Re: [ceph-users] OSD space usage 2x object size after rados put
Some more checking:
- re-deploying the cluster and testing again - same resul
Some more checking:
- re-deploying the cluster and testing again - same result (initial 2x
space usage).
- re-deploying with ext4 for OSD's (instead of default xfs)...*no* 2x
space usage observed. Retested several times.
So looks like some combination of xfs/kernel/os version (Ubuntu 13.10)
Ah right - sorry, I didn't realize that my 'du' was missing the files! I
will retest and post updated output.
Cheers
Mark
On 10/04/14 15:04, Gregory Farnum wrote:
Right, but I'm interested in the space allocation within the PG. The
best guess I can come up with without trawling through the co
Right, but I'm interested in the space allocation within the PG. The
best guess I can come up with without trawling through the code is
that some layer in the stack is preallocated and then trimmed the
objects back down once writing stops, but I'd like some more data
points before I dig.
-Greg
Soft
It is only that single pg using the space (see attached) - but essentially:
$ du -m /var/lib/ceph/osd/ceph-1
...
2048/var/lib/ceph/osd/ceph-1/current/5.1a_head
2053/var/lib/ceph/osd/ceph-1/current
2053/var/lib/ceph/osd/ceph-1/
Which is resized to 1025 soon after. Interestingly I am n
I don't think the backing store should be seeing any effects like
that. What are the filenames which are using up that space inside the
folders?
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Wed, Apr 9, 2014 at 1:58 AM, Mark Kirkwood
wrote:
> Hi all,
>
> I've noticed that
15 matches
Mail list logo
