[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ceph-devel
Subject:    Re: [PATCH v3] libceph: add osd op counter metric support
From:       Jeff Layton <jlayton () kernel ! org>
Date:       2021-04-27 18:42:41
Message-ID: 9930d9e82776a3a3d314275a2c57d4b1890fa9cc.camel () kernel ! org
[Download RAW message or body]

On Mon, 2021-04-26 at 22:33 +0200, Ilya Dryomov wrote:
> On Mon, Apr 26, 2021 at 7:56 PM Jeff Layton <jlayton@kernel.org> wrote:
> > 
> > On Wed, 2020-11-11 at 09:32 +0800, Xiubo Li wrote:
> > > On 2020/11/10 23:44, Ilya Dryomov wrote:
> > > > On Tue, Nov 10, 2020 at 3:19 PM <xiubli@redhat.com> wrote:
> > > > > From: Xiubo Li <xiubli@redhat.com>
> > > > > 
> > > > > The logic is the same with osdc/Objecter.cc in ceph in user space.
> > > > > 
> > > > > URL: https://tracker.ceph.com/issues/48053
> > > > > Signed-off-by: Xiubo Li <xiubli@redhat.com>
> > > > > ---
> > > > > 
> > > > > V3:
> > > > > - typo fixing about oring the _WRITE
> > > > > 
> > > > >   include/linux/ceph/osd_client.h |  9 ++++++
> > > > >   net/ceph/debugfs.c              | 13 ++++++++
> > > > >   net/ceph/osd_client.c           | 56 +++++++++++++++++++++++++++++++++
> > > > >   3 files changed, 78 insertions(+)
> > > > > 
> > > > > diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h
> > > > > index 83fa08a06507..24301513b186 100644
> > > > > --- a/include/linux/ceph/osd_client.h
> > > > > +++ b/include/linux/ceph/osd_client.h
> > > > > @@ -339,6 +339,13 @@ struct ceph_osd_backoff {
> > > > >          struct ceph_hobject_id *end;
> > > > >   };
> > > > > 
> > > > > +struct ceph_osd_metric {
> > > > > +       struct percpu_counter op_ops;
> > > > > +       struct percpu_counter op_rmw;
> > > > > +       struct percpu_counter op_r;
> > > > > +       struct percpu_counter op_w;
> > > > > +};
> > > > OK, so only reads and writes are really needed.  Why not expose them
> > > > through the existing metrics framework in fs/ceph?  Wouldn't "fs top"
> > > > want to display them?  Exposing latency information without exposing
> > > > overall counts seems rather weird to me anyway.
> > > 
> > > Okay, I just thought in future this may also be needed by rbd :-)
> > > 
> > > 
> > > > The fundamental problem is that debugfs output format is not stable.
> > > > The tracker mentions test_readahead -- updating some teuthology test
> > > > cases from time to time is not a big deal, but if a user facing tool
> > > > such as "fs top" starts relying on these, it would be bad.
> > > 
> > > No problem, let me move it to fs existing metric framework.
> > > 
> > 
> > Hi Xiubo/Ilya/Patrick :
> > 
> > Mea culpa...I had intended to drop this patch from testing branch after
> > this discussion, but got sidetracked and forgot to do so. I've now done
> > that though.
> 
> On the subject of metrics, I think Xiubo's I/O size metrics patches
> need a look -- he reposted the two that were skipped a while ago.
> 

Thanks for reminding me. I saw that he sent those when I was OOTO, and I
forgot to revisit them. In the future, if I do that, ping me about them
and I'll try to get to them sooner.
-- 
Jeff Layton <jlayton@kernel.org>

[prev in list] [next in list] [prev in thread] [next in thread]