From: Jens Axboe <axboe@kernel.dk>
To: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>,
Mark Nelson <mark.nelson@inktank.com>,
Mark Nelson <mark.a.nelson@gmail.com>,
fio@vger.kernel.org
Cc: "xan.peng" <xanpeng@gmail.com>,
"ceph-devel@vger.kernel.org" <ceph-devel@vger.kernel.org>
Subject: Re: fio rbd completions (Was: fio rbd hang for block sizes > 1M)
Date: Mon, 27 Oct 2014 08:19:40 -0600 [thread overview]
Message-ID: <544E547C.30009@kernel.dk> (raw)
In-Reply-To: <544C2371.1020403@catalyst.net.nz>
On 10/25/2014 04:25 PM, Mark Kirkwood wrote:
> On 26/10/14 08:20, Jens Axboe wrote:
>> On 10/24/2014 10:50 PM, Mark Kirkwood wrote:
>>> On 25/10/14 16:47, Jens Axboe wrote:
>>>>
>>>> Since you're running rbd tests... Mind giving this patch a go? I don't
>>>> have an easy way to test it myself. It has nothing to do with this
>>>> issue, it's just a potentially faster way to do the rbd completions.
>>>>
>>>
>>> Sure - but note I'm testing this on my i7 workstation (4x osd's running
>>> on 2x Crucial M550) so not exactly server grade :-)
>>>
>>> With that in mind, I'm seeing slightly *slower* performance with the
>>> patch applied: e.g: for 128k blocks - 2 runs, 1 uncached and the next
>>> cached.
>>
>> Yeah, that doesn't look good. Mind trying this one out? I wonder if we
>> doubly wait on them - or perhaps rbd_aio_wait_for_complete() isn't
>> working correctly. If you try this one, we should know more...
>>
>> Goal is, I want to get rid of that usleep() in getevents.
>>
>
> Testing with v3 patch applied hangs. I did wonder if we had somehow hit
> a new variant of the cache issue - so reran with it disabled in
> ceph.conf. Result is the same:
>
> $ fio read-test.fio
> rbd_thread: (g=0): rw=read, bs=128K-128K/128K-128K/128K-128K,
> ioengine=rbd, iodepth=32
> fio-2.1.13-88-gb2ee7
> Starting 1 process
> rbd engine: RBD version: 0.1.8
> Jobs: 1 (f=1): [R(1)] [0.1% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta
> 01h:25m:15s]
There were, unfortunately, still two bugs left in -v3. I just posted an
updated one, please try that and see if it works for you.
--
Jens Axboe
next prev parent reply other threads:[~2014-10-27 14:19 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-24 2:38 fio rbd hang for block sizes > 1M Mark Kirkwood
2014-10-24 5:35 ` Jens Axboe
2014-10-24 6:17 ` Mark Kirkwood
2014-10-24 13:19 ` Mark Nelson
2014-10-24 14:09 ` Mark Nelson
2014-10-24 14:30 ` Jens Axboe
2014-10-24 22:45 ` Mark Kirkwood
2014-10-25 0:12 ` Mark Nelson
2014-10-25 0:37 ` Mark Kirkwood
2014-10-25 2:35 ` Mark Kirkwood
2014-10-25 3:47 ` Jens Axboe
2014-10-25 4:50 ` fio rbd completions (Was: fio rbd hang for block sizes > 1M) Mark Kirkwood
2014-10-25 19:20 ` Jens Axboe
2014-10-25 22:25 ` Mark Kirkwood
2014-10-27 9:27 ` Ketor D
2014-10-27 10:25 ` Ketor D
2014-10-27 14:19 ` Jens Axboe
2014-10-27 14:15 ` Jens Axboe
2014-10-27 14:19 ` Jens Axboe [this message]
2014-10-27 15:12 ` Ketor D
2014-10-27 15:22 ` Jens Axboe
2014-10-27 15:25 ` Jens Axboe
2014-10-27 15:29 ` Ketor D
2014-10-27 15:36 ` Jens Axboe
2014-10-27 15:45 ` Ketor D
2014-10-27 15:53 ` Jens Axboe
2014-10-27 16:20 ` Ketor D
2014-10-27 16:55 ` Jens Axboe
2014-10-27 21:59 ` Mark Kirkwood
2014-10-27 22:32 ` Jens Axboe
2014-10-27 23:21 ` Mark Kirkwood
2014-10-28 3:23 ` Ketor D
2014-10-28 4:01 ` Mark Kirkwood
2014-10-28 4:05 ` Jens Axboe
2014-10-28 4:49 ` Ketor D
2014-10-28 15:14 ` Jens Axboe
2014-10-28 15:49 ` Ketor D
2014-10-28 15:53 ` Jens Axboe
2014-10-28 17:09 ` Jens Axboe
2014-10-28 18:43 ` Ketor D
2014-10-29 7:15 ` Ketor D
2014-10-29 14:31 ` Jens Axboe
2014-10-30 2:50 ` Ketor D
2014-10-30 2:55 ` Jens Axboe
2014-10-30 5:29 ` Ketor D
2014-10-30 7:44 ` Mark Kirkwood
2014-10-30 8:04 ` Ketor D
2014-10-31 8:54 ` Mark Kirkwood
2014-10-24 22:30 ` fio rbd hang for block sizes > 1M Mark Kirkwood
2014-10-24 22:38 ` Mark Nelson
2014-10-24 14:11 ` Danny Al-Gaaf
2014-10-24 14:31 ` Jens Axboe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=544E547C.30009@kernel.dk \
--to=axboe@kernel.dk \
--cc=ceph-devel@vger.kernel.org \
--cc=fio@vger.kernel.org \
--cc=mark.a.nelson@gmail.com \
--cc=mark.kirkwood@catalyst.net.nz \
--cc=mark.nelson@inktank.com \
--cc=xanpeng@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox