linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Eryu Guan <eguan@redhat.com>
Cc: david@fromorbit.com, linux-xfs@vger.kernel.org
Subject: Re: [PATCH 39/39] xfs_scrub: create online filesystem scrub program
Date: Tue, 8 Nov 2016 00:37:07 -0800	[thread overview]
Message-ID: <20161108083706.GB20710@birch.djwong.org> (raw)
In-Reply-To: <20161105052256.GB27776@eguan.usersys.redhat.com>

On Sat, Nov 05, 2016 at 01:22:56PM +0800, Eryu Guan wrote:
> On Fri, Nov 04, 2016 at 05:28:44PM -0700, Darrick J. Wong wrote:
> > Create a filesystem scrubbing tool that walks the directory tree,
> > queries every file's extents, extended attributes, and stat data.  For
> > generic (non-XFS) filesystems this depends on the kernel to do nearly
> > all the validation.  Optionally, we can (try to) read all the file
> > data.
> > 
> > For XFS, we perform sequential scans of each AG's metadata, inodes,
> > extent maps, and file data.  Being XFS specific, we can work with
> > the in-kernel scrubbers to perform much stronger
> > metadata checking and cross-referencing.  We can also take advantage
> > of newer ioctls such as GETFSMAP to perform faster read verification.
> > 
> > In the future we will be able to take advantage of (still unwritten)
> > features such as parent directory pointers to fully validate all
> > metadata.  However, this tool /should/ work for most non-XFS
> > filesystems such as ext4 and btrfs.
> > 
> > Note also that the scrub tool can shut down the filesystem if errors
> > are found.  This is not a default option since scrubbing is very
> > immature at this time.  It can also ask the XFS driver in the kernel
> > to optimize or repair metadata, though this may not be successful.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> [snip]
> > --- /dev/null
> > +++ b/scrub/scrub.c
> > @@ -0,0 +1,1009 @@
> > +/*
> > + * Copyright (C) 2016 Oracle.  All Rights Reserved.
> > + *
> > + * Author: Darrick J. Wong <darrick.wong@oracle.com>
> > + *
> > + * This program is free software; you can redistribute it and/or
> > + * modify it under the terms of the GNU General Public License
> > + * as published by the Free Software Foundation; either version 2
> > + * of the License, or (at your option) any later version.
> > + *
> > + * This program is distributed in the hope that it would be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> > + * GNU General Public License for more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > + * along with this program; if not, write the Free Software Foundation,
> > + * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301, USA.
> > + */
> > +#include "libxfs.h"
> > +#include <stdio.h>
> > +#include <mntent.h>
> > +#include <unistd.h>
> > +#include <sys/types.h>
> > +#include <sys/stat.h>
> > +#include <sys/time.h>
> > +#include <sys/resource.h>
> > +#include <sys/statvfs.h>
> > +#include <sys/vfs.h>
> > +#include <fcntl.h>
> > +#include <dirent.h>
> > +#include "disk.h"
> > +#include "scrub.h"
> > +#include "../../repair/threads.h"
> 
> I have trouble compiling the djwong-devel branch, it failed to find
> "../../repair/threads.h", seems it should be "../repair/threads.h" here.

Doh.  Yeah.  I'll fix that in the morning.  Sorry about that.

In a nutshell: I build xfsprogs git for each arch in a build-$arch/ subdirs
containing symlinks to everything in the parent, which is why I never tripped
on this. :/

--D

> 
> Thanks,
> Eryu

      reply	other threads:[~2016-11-08  8:37 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-05  0:24 [PATCH v2 00/39] xfsprogs: online scrub/repair support Darrick J. Wong
2016-11-05  0:24 ` [PATCH 01/39] xfs: plumb in needed functions for range querying of the freespace btrees Darrick J. Wong
2016-11-05  0:24 ` [PATCH 02/39] xfs: provide a query_range function for " Darrick J. Wong
2016-11-05  0:24 ` [PATCH 03/39] xfs: create a function to query all records in a btree Darrick J. Wong
2016-11-05  0:24 ` [PATCH 04/39] xfs: introduce the XFS_IOC_GETFSMAP ioctl Darrick J. Wong
2016-11-05  0:25 ` [PATCH 05/39] xfs_io: support the new getfsmap ioctl Darrick J. Wong
2016-11-05  0:25 ` [PATCH 06/39] xfs: use GPF_NOFS when allocating btree cursors Darrick J. Wong
2016-11-05  0:25 ` [PATCH 07/39] xfs: add scrub tracepoints Darrick J. Wong
2016-11-05  0:25 ` [PATCH 08/39] xfs: create an ioctl to scrub AG metadata Darrick J. Wong
2016-11-05  0:25 ` [PATCH 09/39] xfs: generic functions to scrub metadata and btrees Darrick J. Wong
2016-11-05  0:25 ` [PATCH 10/39] xfs: scrub the backup superblocks Darrick J. Wong
2016-11-05  0:25 ` [PATCH 11/39] xfs: scrub AGF and AGFL Darrick J. Wong
2016-11-05  0:25 ` [PATCH 12/39] xfs: scrub the AGI Darrick J. Wong
2016-11-05  0:25 ` [PATCH 13/39] xfs: support scrubbing free space btrees Darrick J. Wong
2016-11-05  0:26 ` [PATCH 14/39] xfs: support scrubbing inode btrees Darrick J. Wong
2016-11-05  0:26 ` [PATCH 15/39] xfs: support scrubbing rmap btree Darrick J. Wong
2016-11-05  0:26 ` [PATCH 16/39] xfs: support scrubbing refcount btree Darrick J. Wong
2016-11-05  0:26 ` [PATCH 17/39] xfs: scrub inodes Darrick J. Wong
2016-11-05  0:26 ` [PATCH 18/39] xfs: scrub inode block mappings Darrick J. Wong
2016-11-05  0:26 ` [PATCH 19/39] xfs: scrub directory/attribute btrees Darrick J. Wong
2016-11-05  0:26 ` [PATCH 20/39] xfs: scrub directories Darrick J. Wong
2016-11-05  0:26 ` [PATCH 21/39] xfs: scrub extended attributes Darrick J. Wong
2016-11-05  0:26 ` [PATCH 22/39] xfs: scrub symbolic links Darrick J. Wong
2016-11-05  0:27 ` [PATCH 23/39] xfs: scrub realtime bitmap/summary Darrick J. Wong
2016-11-05  0:27 ` [PATCH 24/39] xfs: scrub should cross-reference with the bnobt Darrick J. Wong
2016-11-05  0:27 ` [PATCH 25/39] xfs: cross-reference bnobt records with cntbt Darrick J. Wong
2016-11-05  0:27 ` [PATCH 26/39] xfs: cross-reference inode btrees during scrub Darrick J. Wong
2016-11-05  0:27 ` [PATCH 27/39] xfs: cross-reference reverse-mapping btree Darrick J. Wong
2016-11-05  0:27 ` [PATCH 28/39] xfs: cross-reference refcount btree during scrub Darrick J. Wong
2016-11-05  0:27 ` [PATCH 29/39] xfs: scrub should cross-reference the realtime bitmap Darrick J. Wong
2016-11-05  0:27 ` [PATCH 30/39] xfs: add helper routines for the repair code Darrick J. Wong
2016-11-05  0:27 ` [PATCH 31/39] xfs: repair inode btrees Darrick J. Wong
2016-11-05  0:27 ` [PATCH 32/39] xfs: rebuild the rmapbt Darrick J. Wong
2016-11-05  0:28 ` [PATCH 33/39] xfs: repair refcount btrees Darrick J. Wong
2016-11-05  0:28 ` [PATCH 34/39] xfs: repair inode block maps Darrick J. Wong
2016-11-05  0:28 ` [PATCH 35/39] xfs: query the per-AG reservation counters Darrick J. Wong
2016-11-05  0:28 ` [PATCH 36/39] xfs_db: introduce fuzz command Darrick J. Wong
2016-11-05  0:28 ` [PATCH 37/39] xfs_db: print attribute remote value blocks Darrick J. Wong
2016-11-05  0:28 ` [PATCH 38/39] xfs_io: provide an interface to the scrub ioctls Darrick J. Wong
2016-11-05  0:28 ` [PATCH 39/39] xfs_scrub: create online filesystem scrub program Darrick J. Wong
2016-11-05  5:22   ` Eryu Guan
2016-11-08  8:37     ` Darrick J. Wong [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=20161108083706.GB20710@birch.djwong.org \
    --to=darrick.wong@oracle.com \
    --cc=david@fromorbit.com \
    --cc=eguan@redhat.com \
    --cc=linux-xfs@vger.kernel.org \
    /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).