Re: [ceph-users] two mount points, two diffrent data

2015-01-18 Thread Rafał Michalak

 Because you are not using a cluster aware filesystem - the respective
 mounts
 don't know when changes are made to the underlying block device (rbd) by
 the
 other mount. What you are doing *will* lead to file corruption.

 Your need to use a distributed filesystem such as GFS2 or cephfs.

 CephFS would be probably be the easiest to setup.


Thanks for help
I use cephfs and it's working great !
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


Re: [ceph-users] two mount points, two diffrent data

2015-01-16 Thread Robert Sander
On 14.01.2015 14:20, Rafał Michalak wrote:
 
 #node1
 mount /dev/rbd/rbd/test /mnt
 
 #node2
 mount /dev/rbd/rbd/test /mnt

If you want to mount a filesystem on one block device onto multiple
clients, the filesystem has to be clustered, e.g. OCFS2.

A normal local filesystem like ext4 or XFS is not aware that other
clients may alter the underlying block device. This is a sure receipt
for data corruption and loss.

Maybe you should look at CephFS.

Regards
-- 
Robert Sander
Heinlein Support GmbH
Schwedter Str. 8/9b, 10119 Berlin

http://www.heinlein-support.de

Tel: 030 / 405051-43
Fax: 030 / 405051-19

Zwangsangaben lt. §35a GmbHG:
HRB 93818 B / Amtsgericht Berlin-Charlottenburg,
Geschäftsführer: Peer Heinlein -- Sitz: Berlin



signature.asc
Description: OpenPGP digital signature
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


Re: [ceph-users] two mount points, two diffrent data

2015-01-16 Thread Lindsay Mathieson
On Wed, 14 Jan 2015 02:20:21 PM Rafał Michalak wrote:
 Why data not replicating on mounting fs ?
 I try with filesystems ext4 and xfs
 The data is visible only when unmounted and mounted again


Because you are not using a cluster aware filesystem - the respective mounts 
don't know when changes are made to the underlying block device (rbd) by the 
other mount. What you are doing *will* lead to file corruption.

Your need to use a distributed filesystem such as GFS2 or cephfs.

CephFS would be probably be the easiest to setup.
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


Re: [ceph-users] two mount points, two diffrent data

2015-01-16 Thread Michael Kuriger
You’re using a file system on 2 hosts that is not cluster aware.  Metadata 
written on hosta is not sent to hostb in this case.  You may be interested in 
looking at cephfs for this use case.


Michael Kuriger
mk7...@yp.com
818-649-7235
MikeKuriger (IM)

From: Rafał Michalak rafa...@gmail.commailto:rafa...@gmail.com
Date: Wednesday, January 14, 2015 at 5:20 AM
To: ceph-users@lists.ceph.commailto:ceph-users@lists.ceph.com 
ceph-users@lists.ceph.commailto:ceph-users@lists.ceph.com
Subject: [ceph-users] two mount points, two diffrent data

Hello I have trouble with this situation

#node1
mount /dev/rbd/rbd/test /mnt
cd /mnt
touch test1
ls (i see test1, OK)

#node2
mount /dev/rbd/rbd/test /mnt
cd /mnt
(i see test1, OK)
touch test2
ls (i see test2, OK)

#node1
ls (i see test1, BAD)
touch test3
ls (i see test1, test3 BAD)

#node2
ls (i see test1, test2 BAD)

Why data not replicating on mounting fs ?
I try with filesystems ext4 and xfs
The data is visible only when unmounted and mounted again

I check health in ceph status and is HEALTH_OK

What I doing wrong ?
Thanks for any help.


My system
Ubuntu 14.04.01 LTS

#ceph --version
ceph version 0.80.7 (6c0127fcb58008793d3c8b62d925bc91963672a3)

#modinfo libceph
filename:   /lib/modules/3.13.0-44-generic/kernel/net/ceph/libceph.ko
license: GPL
description:Ceph filesystem for Linux
author:   Patience Warnick 
patie...@newdream.netmailto:patie...@newdream.net
author:   Yehuda Sadeh 
yeh...@hq.newdream.netmailto:yeh...@hq.newdream.net
author:   Sage Weil s...@newdream.netmailto:s...@newdream.net
srcversion: B8E83D4DFC53B113603CF52
depends:libcrc32c
intree:Y
vermagic:   3.13.0-44-generic SMP mod_unload modversions
signer:   Magrathea: Glacier signing key
sig_key:  50:8C:3B:4B:F1:08:ED:36:B6:06:2F:81:27:82:F7:7C:37:B9:85:37
sig_hashalgo:   sha512

___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


[ceph-users] two mount points, two diffrent data

2015-01-14 Thread Rafał Michalak
Hello I have trouble with this situation

#node1
mount /dev/rbd/rbd/test /mnt
cd /mnt
touch test1
ls (i see test1, OK)

#node2
mount /dev/rbd/rbd/test /mnt
cd /mnt
(i see test1, OK)
touch test2
ls (i see test2, OK)

#node1
ls (i see test1, BAD)
touch test3
ls (i see test1, test3 BAD)

#node2
ls (i see test1, test2 BAD)

Why data not replicating on mounting fs ?
I try with filesystems ext4 and xfs
The data is visible only when unmounted and mounted again

I check health in ceph status and is HEALTH_OK

What I doing wrong ?
Thanks for any help.


My system
Ubuntu 14.04.01 LTS

#ceph --version
ceph version 0.80.7 (6c0127fcb58008793d3c8b62d925bc91963672a3)

#modinfo libceph
filename:   /lib/modules/3.13.0-44-generic/kernel/net/ceph/libceph.ko
license: GPL
description:Ceph filesystem for Linux
author:   Patience Warnick patie...@newdream.net
author:   Yehuda Sadeh yeh...@hq.newdream.net
author:   Sage Weil s...@newdream.net
srcversion: B8E83D4DFC53B113603CF52
depends:libcrc32c
intree:Y
vermagic:   3.13.0-44-generic SMP mod_unload modversions
signer:   Magrathea: Glacier signing key
sig_key:
 50:8C:3B:4B:F1:08:ED:36:B6:06:2F:81:27:82:F7:7C:37:B9:85:37
sig_hashalgo:   sha512
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com