xfs
[Top] [All Lists]

Re: [PATCH] xfs: fix double free of trans in log recovery on I/O error

To: "Brian Foster" <bfoster@xxxxxxxxxx>
Subject: Re: [PATCH] xfs: fix double free of trans in log recovery on I/O error
From: "Alex Lyakas" <alex@xxxxxxxxxxxxxxxxx>
Date: Sun, 31 Aug 2014 11:50:52 +0300
Cc: <xfs@xxxxxxxxxxx>, "Dave Chinner" <david@xxxxxxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
Importance: Normal
In-reply-to: <20140825142025.GA10135@xxxxxxxxxxxxxxx>
References: <1408648692-15957-1-git-send-email-bfoster@xxxxxxxxxx> <CAOcd+r32+6gafv5+MCNyNTH1Ncas9CJHHrku+Y=sZefDzzS1Zw@xxxxxxxxxxxxxx> <20140825142025.GA10135@xxxxxxxxxxxxxxx>
Hi Brian, Dave,
I tested this patch on kernel 3.16, top commit:

commit 19583ca584d6f574384e17fe7613dfaeadcdc4a6
Author: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Date:   Sun Aug 3 15:25:02 2014 -0700

   Linux 3.16

and, yes, it appears to fix the issue.

Trouble is that our production kernel is 3.8.13, and we cannot upgrade to mainline kernel easily. Question is whether we can expect some patch suitable for our kernel, or, since our kernel is EOL and not a long-term one, we cannot?

Thanks for your help,
Alex.


-----Original Message----- From: Brian Foster
Sent: 25 August, 2014 5:20 PM
To: Alex Lyakas
Cc: xfs@xxxxxxxxxxx ; Dave Chinner
Subject: Re: [PATCH] xfs: fix double free of trans in log recovery on I/O error

On Sun, Aug 24, 2014 at 12:20:20PM +0300, Alex Lyakas wrote:
Hi Brian,

On Thu, Aug 21, 2014 at 10:18 PM, Brian Foster <bfoster@xxxxxxxxxx> wrote:
> XFS log recovery builds up an xlog_recover object as it passes through
> the log operations on the physical log. These structures are managed via
> a hash table and are allocated when a new transaction is encountered and
> freed once a commit operation for the transaction is encountered.
>
> This state machine for active transactions is implemented by a
> combination of xlog_do_recovery_pass(), which walks through the log
> buffers and xlog_recover_process_data() which processes log operations
> within each buffer. The latter function decides whether to allocate a
> new xlog_recover, add to it or commit and ultimately free it.  If an
> error occurs at any point during the lifecycle of a particular
> xlog_recover, xlog_recover_process_data() frees the object and returns
> an error.
>
> xlog_recover_commit_trans() handles the final processing of the
> transaction. It submits whatever I/O is required for the transaction and
> frees xlog_recover object along with the transaction items it tracks. If
> an error occurs at the final stages of the commit operation, such as I/O
> failure, both xlog_recover_commit_trans() and
> xlog_recover_process_data() attempt to free the trans object.
>
> Modify xlog_recover_commit_trans() to only free the trans object on
> successful completion of the trans, including any I/O errors that might
> occur when recovering the log.
>
> Signed-off-by: Brian Foster <bfoster@xxxxxxxxxx>
> ---
>
> Hi all,
>
> I found that the recent buffer I/O rework fixes didn't address the crash
> reproduced by the dm-flakey/log recovery test case I posted recently. I
> tracked the crash down to this, which allows the test to pass. This
> addresses the crash I saw when running the reproducer manually with the
> metadump that Alex posted as well.
>
> FWIW, I also went back and tested the xfs_buf_iowait() experiment in
> both scenarios (Alex's metadump and xfstests test) and they all
> reproduce the same crash for me. I think that either I'm still not
> reproducing the original problem, something else might have contaminated
> the original xfs_buf_iowait() test to give a false positive, or
> something else entirely is going on.
>
> Alex,
>
> If you have a chance, I think it might be interesting to see whether you
> reproduce any problems with this patch. It looks like this is a
> regression introduced by:
>
>         2a84108f xfs: free the list of recovery items on error
>
> ... but I have no idea if that's in whatever kernel you're running.
I am running kernel 3.8.13 with some changes (published at
https://github.com/zadarastorage/zadara-xfs-pushback), but this
problem also happens on pristine 3.8.13 from
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git,
branch linux-stable-3.8.y.

I do not have commit 2a84108f in this kernel. It was introduced in 3.14.
I applied your patch to 3.8.13, but it doesn't fix the issue. Same
problem happens when testing scenario that I described in
http://oss.sgi.com/pipermail/xfs/2014-August/037637.html.


Ok, thanks. Yeah, I don't see the double free regression in the 3.8.13
stable branch. I went back to that kernel to try and confirm some
things. I do reproduce the problem with your metadump as well as the
test case I put together. I tested Dave's buf hold across sync I/O patch
and that does indeed prevent the problem.

For whatever reason, neither the test case nor your metadump reproduce
the same problem on latest kernels. Instead, they reproduce this double
free regression. I suspect this is what you ran into when you reproduced
on a more recent kernel. If you'd like, feel free to try and verify that
by running your reproducer again on a recent kernel with this patch and
see if you can still reproduce a crash as with the 3.8.13 kernel.

Brian

Thanks,
Alex.

>
> Brian
>
>  fs/xfs/xfs_log_recover.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> index 176c4b3..daca9a6 100644
> --- a/fs/xfs/xfs_log_recover.c
> +++ b/fs/xfs/xfs_log_recover.c
> @@ -3528,10 +3528,15 @@ out:
>         if (!list_empty(&done_list))
>                 list_splice_init(&done_list, &trans->r_itemq);
>
> -       xlog_recover_free_trans(trans);
> -
>         error2 = xfs_buf_delwri_submit(&buffer_list);
> -       return error ? error : error2;
> +
> +       if (!error)
> +               error = error2;
> +       /* caller frees trans on error */
> +       if (!error)
> +               xlog_recover_free_trans(trans);
> +
> +       return error;
>  }
>
>  STATIC int
> --
> 1.8.3.1
>

<Prev in Thread] Current Thread [Next in Thread>