Flexible I/O Tester development
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Hyukjoong Kim <wangmir@gmail.com>, fio@vger.kernel.org
Subject: Re: Problem on using fio replay function with fio iolog trace
Date: Tue, 27 Oct 2015 18:50:14 +0900	[thread overview]
Message-ID: <562F48D6.8090603@kernel.dk> (raw)
In-Reply-To: <CAAkw5-PAt2O01H-i4WrvPaBPuMX7zPDyg8u9jS3KyA4PrNu=2A@mail.gmail.com>

On 10/27/2015 02:33 PM, Hyukjoong Kim wrote:
> Hi,
>
> I'm trying to replay iolog using fio iolog, but i failed with that.
>
> simply, I check the example trace with write_iolog option, and I got
> trace like this,
> --------------
> fio version 2 iolog
> /mnt/test/rand-write.1.0 add
> /mnt/test/rand-write.1.0 open
> /mnt/test/rand-write.1.0 read 64757760 4096
> /mnt/test/rand-write.1.0 read 794873856 4096
> /mnt/test/rand-write.1.0 read 905355264 4096
> /mnt/test/rand-write.1.0 write 504672256 4096
> -------------
> So, I make a new trace with similar format like this
> -------------
> fio version 2 iolog
> /dev/sdb open
> /dev/sdb write 4096 4096
> ------------
> However, I failed to replay both with different error.
>
> Firstly, I ran extracted file (fio --name test --read_iolog=trace
> --replay_redirect=/dev/sdb), but error was like this,
> --------------------------------------------
> test: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
> fio-2.0.10
> Starting 1 process
> bad iolog2: fio version 2 iolog
> fio: pid=14839, got signal=11
>
>
> Run status group 0 (all jobs):
> --------------------------------------------
>
> and if I delete the first line than, fio said don't support iolog 1
>
>
> Secondly, I ran the file I made with same command, and error was
> different from former,
>
> ----------------------------------------
> test: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
> fio-2.0.10
> Starting 1 process
> fio: pid=12868, got signal=11
>
>
> Run status group 0 (all jobs):
> --------------------------------------------------
>
> I think fio was killed by someone, but I don't know why.
>
> So,, how can I fix this problem? I tried this behavior with apt
> repository fio, and current git, and fio 2.0.10 also, but error was
> same.

Are you sure you tried a recent version? I just tried your trace file 
with current git:

axboe@xps13:/home/axboe/git/fio $ ./fio --name=replay 
--read_iolog=replay.log
replay: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=sync, iodepth=1
fio-2.2.10-26-g3d57
Starting 1 process
[...]
Run status group 0 (all jobs):
    READ: io=12KB, aggrb=222KB/s, minb=222KB/s, maxb=222KB/s, 
mint=54msec, maxt=54msec
   WRITE: io=4KB, aggrb=74KB/s, minb=74KB/s, maxb=74KB/s, mint=54msec, 
maxt=54msec

and it's working fine. 2.0.10 is pretty old.

Assuming current -git, does it work without replay_redirect? Does it 
work with that option?

-- 
Jens Axboe



      reply	other threads:[~2015-10-27  9:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-27  5:33 Problem on using fio replay function with fio iolog trace Hyukjoong Kim
2015-10-27  9:50 ` Jens Axboe [this message]

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=562F48D6.8090603@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=fio@vger.kernel.org \
    --cc=wangmir@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