Preserve dir default_file_layout in encoded inode

2012-08-15 Thread Alexandre Oliva
Store dir layouts not only in the journal, but also in inode information for the directory, so that they aren't dropped on the floor when dirs are evicted from the MDS cache and dropped from the journal. Also restore dir layouts when fetching dirs back into the MDS cache. This is supposed to fix

Re: Ceph setup on single node : prob with "

2012-08-15 Thread Mark Kirkwood
On 16/08/12 16:44, hemant surale wrote: Hello Tommi, Ceph community I did mkdir the directory. Infact I have created a new partition by the same name and formatted using ext3. I also executed the following command for the partition/directory: "mount -o remount,user_xattr /" Still I am getting

Re: Ceph setup on single node : prob with "

2012-08-15 Thread hemant surale
Hello Tommi, Ceph community I did mkdir the directory. Infact I have created a new partition by the same name and formatted using ext3. I also executed the following command for the partition/directory: "mount -o remount,user_xattr /" Still I am getting the same error: >> 2012-08-14 14:31:29.390

Re: ceph 0.50 - error compiling on arch

2012-08-15 Thread Samuel Just
master/next/testing should now have the intrusive_ptr errors fixed. -Sam On Wed, Aug 15, 2012 at 1:12 PM, alphe salas wrote: > > > > /usr/include/boost/smart_ptr/intrusive_ptr.hpp: In instantiation of > 'boost::intrusive_ptr::intrusive_ptr(T*, bool) [with T = MOSDPGLog]': > osd/PG.h:871:26: req

Re: regression with poll(2)?

2012-08-15 Thread Sage Weil
On Wed, 15 Aug 2012, Atchley, Scott wrote: > On Aug 15, 2012, at 3:46 PM, Sage Weil wrote: > > > I'm experiencing a stall with Ceph daemons communicating over TCP that > > occurs reliably with 3.6-rc1 (and linus/master) but not 3.5. The basic > > situation is: > > > > - the socket is two proce

Re: regression with poll(2)?

2012-08-15 Thread Atchley, Scott
On Aug 15, 2012, at 3:46 PM, Sage Weil wrote: > I'm experiencing a stall with Ceph daemons communicating over TCP that > occurs reliably with 3.6-rc1 (and linus/master) but not 3.5. The basic > situation is: > > - the socket is two processes communicating over TCP on the same host, e.g. > >

regression with poll(2)?

2012-08-15 Thread Sage Weil
I'm experiencing a stall with Ceph daemons communicating over TCP that occurs reliably with 3.6-rc1 (and linus/master) but not 3.5. The basic situation is: - the socket is two processes communicating over TCP on the same host, e.g. tcp0 2164849 10.214.132.38:6801 10.214.132.38:5

Re: v0.48.1 argonaut stable update released

2012-08-15 Thread Oliver Francke
Well, On 08/14/2012 09:29 PM, Sage Weil wrote: On Tue, 14 Aug 2012, Oliver Francke wrote: Hi Sage, I just updated to debian-testing/0.50 this afternoon, after some hint: * osd: better tracking of recent slow operations This is actually about the admin socket command to dump operations in fli

Re: What's the default data dir for radosgw?

2012-08-15 Thread John Axel Eriksson
Yeah I actually figured it out through trial and error but thanks! On Wed, Aug 15, 2012 at 1:18 PM, Wido den Hollander wrote: > On 08/15/2012 12:21 PM, John Axel Eriksson wrote: >> >> I found somewhere that it's supposed to be >> /var/lib/ceph/radosgw/ceph-$id. Ok in my case I guess that would me

Re: What's the default data dir for radosgw?

2012-08-15 Thread Wido den Hollander
On 08/15/2012 12:21 PM, John Axel Eriksson wrote: I found somewhere that it's supposed to be /var/lib/ceph/radosgw/ceph-$id. Ok in my case I guess that would mean: /var/lib/ceph/radosgw/ceph-client.radosgw.gateway would that be correct? Since I need to store the keyring in that directory for exam

What's the default data dir for radosgw?

2012-08-15 Thread John Axel Eriksson
I found somewhere that it's supposed to be /var/lib/ceph/radosgw/ceph-$id. Ok in my case I guess that would mean: /var/lib/ceph/radosgw/ceph-client.radosgw.gateway would that be correct? Since I need to store the keyring in that directory for example and I want to use the defaults. John -- To unsu

[PATCH] docs: Add EU mirror for getting Ceph source and packages

2012-08-15 Thread Wido den Hollander
Signed-off-by: Wido den Hollander --- doc/install/debian.rst|8 doc/source/build-packages.rst |4 doc/source/get-tarballs.rst |3 ++- 3 files changed, 14 insertions(+), 1 deletion(-) diff --git a/doc/install/debian.rst b/doc/install/debian.rst index be39827..

[PATCH] libceph: Assign different global_id for each ceph_auth_client.

2012-08-15 Thread Yan, Zheng
From: "Yan, Zheng" The global_id for ceph_auth_client is always zero. It prevents the kernel from creating more than one ceph_clients that are connected to the same cluster. (ceph_debugfs_client_init() fails to create debugfs directory) Without this patch, I can't using rbd and cephfs on the sam

Re: No lock on RBD allow several mount on different servers...

2012-08-15 Thread Sébastien Han
Hi guys, Thank you for the tremendous answers :D How far are we to see this feature in the stable branch? Part of the 0.48.x or far away from that? Cheers! On Mon, Aug 13, 2012 at 7:49 PM, Yehuda Sadeh wrote: > On Mon, Aug 13, 2012 at 10:22 AM, Josh Durgin wrote: >> On 08/13/2012 09:55 AM, Gre