On Tue, 4 Jun 2013 17:22:08 -0400
"J. Bruce Fields" wrote:
> On Fri, May 31, 2013 at 11:07:29PM -0400, Jeff Layton wrote:
> > Having a global lock that protects all of this code is a clear
> > scalability problem. Instead of doing that, move most of the code to be
> > protected by the i_lock inst
On Fri, May 31, 2013 at 11:07:31PM -0400, Jeff Layton wrote:
> Testing has shown that iterating over the blocked_list for deadlock
> detection turns out to be a bottleneck. In order to alleviate that,
> begin the process of turning it into a hashtable. We start by turning
> the fl_link into a hlist
On Fri, May 31, 2013 at 11:07:30PM -0400, Jeff Layton wrote:
> Currently, when there is a lot of lock contention the kernel spends an
> inordinate amount of time taking blocked locks off of the global
> blocked_list and then putting them right back on again. When all of this
> code was protected by
On Fri, May 31, 2013 at 11:07:29PM -0400, Jeff Layton wrote:
> Having a global lock that protects all of this code is a clear
> scalability problem. Instead of doing that, move most of the code to be
> protected by the i_lock instead.
>
> The exceptions are the global lists that file_lock->fl_link
Within a single log, the version must be always increasing. The
reason for the epoch is that you might have two logs:
a b
(1, 1) (1, 1)
(1, 2) (1, 2)
(1, 3)
(2, 3)
(2, 4)
If a dies after persisting (1, 3) but before sending it to b, b might
become primary and go on accep
Hi Sam,
When calling merge_log on the following : http://pastealacon.com/32449 ( where
the first column are the versions of the log entries and the second column are
the versions of the olog entries ) I expect 5,2 and 6,1 from log to be added to
divergent
https://github.com/ceph/ceph/blob/maste
On Fri, May 31, 2013 at 11:07:27PM -0400, Jeff Layton wrote:
> ...save 3 bytes of stack space.
>
> Signed-off-by: Jeff Layton
ACK.--b.
> ---
> fs/locks.c |9 +
> 1 files changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/fs/locks.c b/fs/locks.c
> index a7d2253..cef0e04 1006
On Fri, May 31, 2013 at 11:07:28PM -0400, Jeff Layton wrote:
> Move the fl_link list handling routines into a separate set of helpers.
> Also move the global list handling out of locks_insert_block, and into
> the caller that ends up triggering it as that allows us to eliminate the
> IS_POSIX check
On 06/04/2013 11:04 AM, Stefan Priebe wrote:
Am 04.06.2013 17:23, schrieb Sage Weil:
On Tue, 4 Jun 2013, Stefan Priebe - Profihost AG wrote:
Hi,
is there a way to flatten the rbd export-diff to a "new" image FILE. Or
do i always have to:
rbd import "OLD BASE IMAGE"
rbd import-diff diff1
rbd i
Am 04.06.2013 17:23, schrieb Sage Weil:
On Tue, 4 Jun 2013, Stefan Priebe - Profihost AG wrote:
Hi,
is there a way to flatten the rbd export-diff to a "new" image FILE. Or
do i always have to:
rbd import "OLD BASE IMAGE"
rbd import-diff diff1
rbd import-diff diff1-2
rbd import-diff diff2-3
rbd
Hello Yehuda,
Sorry this was actually directed to you (florian actually told me you were the
go to person for row).
I am not totally familiar with s3 how does a single namespace ensure a
accounts/user don't have access to the resource of the others ?
Glad to know you are tackling the multi-te
On Tue, Jun 4, 2013 at 3:26 AM, Roman Alekseev wrote:
> Hello,
>
> Please help me to understand op/s (operation per second) value in 'pgmap
> v71520: 3352 pgs: 3352 active+clean; 212 GB data, 429 GB used, 23444 GB /
> 23874 GB avail; 89237KB/s wr, 24op/s' line?
> What does "operation" mean?
What
Having RCU for modification mostly workloads never is a good idea, so
I don't think it makes sense to mention it here.
If you care about the overhead it's worth trying to use per-cpu lists,
though.
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message t
On Tue, 4 Jun 2013, Stefan Priebe - Profihost AG wrote:
> Hi,
>
> is there a way to flatten the rbd export-diff to a "new" image FILE. Or
> do i always have to:
>
> rbd import "OLD BASE IMAGE"
> rbd import-diff diff1
> rbd import-diff diff1-2
> rbd import-diff diff2-3
> rbd import-diff diff3-4
>
On Tue, 4 Jun 2013 10:53:22 -0400
"J. Bruce Fields" wrote:
> On Tue, Jun 04, 2013 at 07:46:40AM -0700, Christoph Hellwig wrote:
> > Having RCU for modification mostly workloads never is a good idea, so
> > I don't think it makes sense to mention it here.
> >
> > If you care about the overhead it
(resending due to formatting)
I'm not Florian, but I think I can help here.
The radosgw user-tenant model is (currently) different from the swift
one. It's more like the S3 model, where users live in a single
namespace. So the current Swift user mapping is not perfect. We
created 'subusers' to em
On Tue, 4 Jun 2013 07:46:40 -0700
Christoph Hellwig wrote:
> Having RCU for modification mostly workloads never is a good idea, so
> I don't think it makes sense to mention it here.
>
> If you care about the overhead it's worth trying to use per-cpu lists,
> though.
>
Yeah, I looked at those t
On Tue, Jun 04, 2013 at 07:46:40AM -0700, Christoph Hellwig wrote:
> Having RCU for modification mostly workloads never is a good idea, so
> I don't think it makes sense to mention it here.
>
> If you care about the overhead it's worth trying to use per-cpu lists,
> though.
Yes. The lock and unl
On Tue, 04 Jun 2013 16:19:53 +0200
"Stefan (metze) Metzmacher" wrote:
> Hi Jeff,
>
> > There's no reason we have to protect the blocked_hash and file_lock_list
> > with the same spinlock. With the tests I have, breaking it in two gives
> > a barely measurable performance benefit, but it seems re
Hi Jeff,
> There's no reason we have to protect the blocked_hash and file_lock_list
> with the same spinlock. With the tests I have, breaking it in two gives
> a barely measurable performance benefit, but it seems reasonable to make
> this locking as granular as possible.
as file_lock_{list,lock}
Hi,
is there a way to flatten the rbd export-diff to a "new" image FILE. Or
do i always have to:
rbd import "OLD BASE IMAGE"
rbd import-diff diff1
rbd import-diff diff1-2
rbd import-diff diff2-3
rbd import-diff diff3-4
rbd import-diff diff4-5
...
and so on?
I would like to apply the diffs on loc
Hi Ilya,
The changes on this branch originated in our Ganesha NFS driver for
Ceph, so I'm not sure where the gap is, if any.
I'll send an update to the list when we've finish re-integrating against
the libcephfs-wip merge branch.
Matt
- "Ilya Storozhilov" wrote:
> Hi Ceph developers,
>
>
On Tue, 4 Jun 2013 07:56:44 -0400
Jim Rees wrote:
> Jeff Layton wrote:
>
> > Might be nice to look at some profiles to confirm all of that. I'd also
> > be curious how much variation there was in the results above, as they're
> > pretty close.
> >
>
> The above is just a random re
Hi Sam,
Thanks for the explanation. I misread the third case, my description was
incorrect. I amended https://github.com/ceph/ceph/pull/340 accordingly.
Cheers
On 06/03/2013 10:28 PM, Samuel Just wrote:
> In all three cases, we know the authoritative log does not contain an
> entry for oe.soid,
Jeff Layton wrote:
> Might be nice to look at some profiles to confirm all of that. I'd also
> be curious how much variation there was in the results above, as they're
> pretty close.
>
The above is just a random representative sample. The results are
pretty close when running thi
On Mon, 3 Jun 2013 18:00:24 -0400
"J. Bruce Fields" wrote:
> On Fri, May 31, 2013 at 11:07:26PM -0400, Jeff Layton wrote:
> > Signed-off-by: Jeff Layton
> > ---
> > fs/locks.c | 24 +++-
> > include/linux/fs.h |6 ++
> > 2 files changed, 25 insertions(+), 5
On Mon, 3 Jun 2013 17:31:01 -0400
"J. Bruce Fields" wrote:
> On Fri, May 31, 2013 at 11:07:23PM -0400, Jeff Layton wrote:
> > Executive summary (tl;dr version): This patchset represents an overhaul
> > of the file locking code with an aim toward improving its scalability
> > and making the code a
Hello,
Please help me to understand op/s (operation per second) value in 'pgmap
v71520: 3352 pgs: 3352 active+clean; 212 GB data, 429 GB used, 23444 GB
/ 23874 GB avail; 89237KB/s wr, 24op/s' line?
What does "operation" mean?
Thanks
--
Kind regards,
R. Alekseev
--
To unsubscribe from this
Hi Ceph developers,
in order to represent NFS-frontend to CephFS data storage we are trying to use
innovative Open-By-Handle API from 'src/include/cephfs/libcephfs.h' file, which
is of 'wip-libcephfs' branch at the moment. API looks quite consistent and
useful but we couldn't find a method to g
Hello Florian,
I was wondering how the Keystone integration with ceph, I have been
reading the documentation of the way it shows how to configure the
keystone endpoints here :
http://ceph.com/docs/next/radosgw/config/
and I don't see how the part :
keystone endpoint-create --service-id --publi
Add tests for the advisory locking API calls.
Signed-off-by: Filippos Giannakos
---
src/Makefile.am |6 +
src/test/librados/lock.cc | 301 +
2 files changed, 307 insertions(+)
create mode 100644 src/test/librados/lock.cc
diff --git a/s
Hi team,
This set of patches export the RADOS advisory locking functionality to the C/C++
API.
They have been refactored to incorporate Josh suggestions from his review:
* Always set tag to "" for exclusive lock
* Add duration argument to lock_{exclusive, shared}
* Add lock flags to librados.h
Add functions to the librados C/C++ API, to take advantage and utilize the
advisory locking system offered by RADOS.
Signed-off-by: Filippos Giannakos
---
src/Makefile.am|5 +-
src/include/rados/librados.h | 102 ++-
src/include/rados/librados.hpp | 2
On Tue, Jun 4, 2013 at 2:20 PM, Chen, Xiaoxi wrote:
> Hi Greg,
> Yes, Thanks for your advice ,we do turn down the
> osd_client_message_size_cap to 100MB/OSD ,both Journal queue and filestore
> queue are set to 100MB also.
> That's 300MB/OSD in total, but from TOP we see:
>
34 matches
Mail list logo
