Re: [ceph-users] Cephfs IO monitoring

2017-08-10 Thread John Spray
On Thu, Aug 10, 2017 at 5:19 AM, Brady Deetz  wrote:
> Curious if there is a method way I could see in near real-time the io
> patters for an fs. For instance, what files are currently being read/written
> and the block sizes. I suspect this is a big ask. The only thing I know of
> that can provide that level of detail for a filesystem is dtrace with zfs.

Not at present, but something along these lines might get added in the
not-too-distant future.  The equivalent functionality for RBD is a
frequent request, and while for RBD this can sort of be implemented
either client or server side, I'm a fan of doing it on the server side
so that we can generalise it and cover CephFS at the same time.

Currently the closest you get is to set "debug mds = 4" and tail the
MDS logs to see which (metadata) operations are happening.

John

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


[ceph-users] Cephfs IO monitoring

2017-08-09 Thread Brady Deetz
Curious if there is a method way I could see in near real-time the io
patters for an fs. For instance, what files are currently being
read/written and the block sizes. I suspect this is a big ask. The only
thing I know of that can provide that level of detail for a filesystem is
dtrace with zfs.
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com