linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Liu Bo <bo.li.liu@oracle.com>
To: kreijack@inwind.it
Cc: Josef Bacik <josef@toxicpanda.com>,
	Paulo Dias <paulo.miguel.dias@gmail.com>,
	linux-btrfs@vger.kernel.org
Subject: Re: qcow2 images make scrub believe the filesystem is corrupted.
Date: Fri, 18 Aug 2017 10:59:39 -0700	[thread overview]
Message-ID: <20170818175939.GA25986@lim.localdomain> (raw)
In-Reply-To: <ceb36a9a-b73f-96c1-cb58-24f149b24001@libero.it>

On Fri, Aug 18, 2017 at 06:23:18PM +0200, Goffredo Baroncelli wrote:
> On 08/18/2017 01:39 AM, Josef Bacik wrote:
> [...]
> > This is happening because the app (the guest OS in this case, we saw this a lot
> > with windows guests) is changing the pages while they are in flight.  We
> > calculate the checksum of the page before it's written, so if it changes while
> > in flight we'll end up with a csum mismatch.
> > 
> > To fix this change kvm to not use O_DIRECT or set NODATASUM on your qcow2 image.
> > You'll have to re-create the image because NODATASUM won't apply to the already
> > invalid checksums.  Thanks,
> 
> Hi Josef,
> 
> could you elaborate: do you are saying that using O_DIRECT is incompatible with DATASUM ?
>

They're compatible, but applications need to be careful.

O_DIRECT takes userspace page, and it works like
DIO write
  p = get_user_page();
  add p to bio
  #btrfs submits this bio
  calc_checksum(bio);
  submit_bio();

There's a chance that page p got changed between calc_checksum() and
submit_bio(), which then causes the mismatch.

For buffered IO, dirty page cache pages is synchronized with page
fault by page lock and page writeback bit.

thanks,
-liubo

  parent reply	other threads:[~2017-08-18 18:02 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-16  1:12 qcow2 images make scrub believe the filesystem is corrupted Paulo Dias
2017-08-16  1:40 ` Qu Wenruo
2017-08-16  1:51   ` Paulo Dias
2017-08-16  2:28     ` Qu Wenruo
2017-08-16  2:46       ` Qu Wenruo
2017-08-16  7:47       ` Qu Wenruo
2017-08-16 23:32 ` Chris Murphy
2017-08-17  8:04   ` Duncan
2017-08-17 19:10     ` Chris Murphy
2017-08-17 20:17       ` Paulo Dias
2017-08-17 20:58         ` Chris Murphy
2017-08-17 23:39 ` Josef Bacik
2017-08-18 16:23   ` Goffredo Baroncelli
2017-08-18 17:43     ` Josef Bacik
2017-08-18 22:19       ` Goffredo Baroncelli
2017-08-19 13:08         ` Goffredo Baroncelli
2017-08-18 23:29       ` Qu Wenruo
2017-08-18 17:59     ` Liu Bo [this message]
2017-08-18 18:25       ` Paulo Dias

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=20170818175939.GA25986@lim.localdomain \
    --to=bo.li.liu@oracle.com \
    --cc=josef@toxicpanda.com \
    --cc=kreijack@inwind.it \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=paulo.miguel.dias@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;
as well as URLs for NNTP newsgroup(s).