From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.suse.de ([195.135.220.15]:41369 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750858AbeC2M4V (ORCPT ); Thu, 29 Mar 2018 08:56:21 -0400 Subject: Re: [PATCH v2 4/7] btrfs: check if the fsid in the primary sb and copy sb are same To: Anand Jain , linux-btrfs@vger.kernel.org References: <20180329103705.13566-1-anand.jain@oracle.com> <20180329103705.13566-5-anand.jain@oracle.com> From: Nikolay Borisov Message-ID: <36e7f186-0b9e-8dde-7be8-0afc7f279147@suse.com> Date: Thu, 29 Mar 2018 15:56:18 +0300 MIME-Version: 1.0 In-Reply-To: <20180329103705.13566-5-anand.jain@oracle.com> Content-Type: text/plain; charset=utf-8 Sender: linux-btrfs-owner@vger.kernel.org List-ID: On 29.03.2018 13:37, Anand Jain wrote: > During the btrfs dev scan make sure that other copies of superblock > contain the same fsid as the primary SB. So that we bring to the > user notice if the superblock has been overwritten. > > mkfs.btrfs -fq /dev/sdc > mkfs.btrfs -fq /dev/sdb > dd if=/dev/sdb of=/dev/sdc count=4K skip=64K seek=64K obs=1 ibs=1 > mount /dev/sdc /btrfs > > Caveat: Pls note that older btrfs-progs do not wipe the non-overwriting > stale superblock like copy2 if a smaller mkfs.btrfs -b is created. > Thus this patch in the kernel will report error. The workaround is to wipe > the superblock manually, like > dd if=/dev/zero of= seek=274877906944 ibs=1 obs=1 count4K > OR apply the btrfs-progs patch > btrfs-progs: wipe copies of the stale superblock beyond -b size > which shall find and wipe the non overwriting superblock during mkfs. > > Signed-off-by: Anand Jain > v1->v2: > Do an explicit read for primary superblock. Drop kzalloc(). > Fix split pr_err(). Same thing about the change log as in 2/7. Otherwise : Reviewed-by: Nikolay Borisov > --- > fs/btrfs/volumes.c | 48 ++++++++++++++++++++++++++++++++++++++---------- > 1 file changed, 38 insertions(+), 10 deletions(-) > > diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c > index e63723f23227..035affa447fa 100644 > --- a/fs/btrfs/volumes.c > +++ b/fs/btrfs/volumes.c > @@ -1198,39 +1198,67 @@ static int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr, > int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder, > struct btrfs_fs_devices **fs_devices_ret) > { > + struct btrfs_super_block *disk_super_primary; > struct btrfs_super_block *disk_super; > struct btrfs_device *device; > struct block_device *bdev; > + struct page *page_primary; > struct page *page; > int ret = 0; > u64 bytenr; > + int i; > > - /* > - * we would like to check all the supers, but that would make > - * a btrfs mount succeed after a mkfs from a different FS. > - * So, we need to add a special mount option to scan for > - * later supers, using BTRFS_SUPER_MIRROR_MAX instead > - */ > - bytenr = btrfs_sb_offset(0); > flags |= FMODE_EXCL; > > bdev = blkdev_get_by_path(path, flags, holder); > if (IS_ERR(bdev)) > return PTR_ERR(bdev); > > - ret = btrfs_read_disk_super(bdev, bytenr, &page, &disk_super); > + /* > + * We would like to check all the supers and use one good copy, > + * but that would make a btrfs mount succeed after a mkfs from > + * a different FS. > + * So, we need to add a special mount option to scan for > + * later supers, using BTRFS_SUPER_MIRROR_MAX instead. > + * So, just validate if all copies of the superblocks are ok > + * and have the same fsid. > + */ > + bytenr = btrfs_sb_offset(0); > + ret = btrfs_read_disk_super(bdev, bytenr, &page_primary, > + &disk_super_primary); > if (ret < 0) > goto error_bdev_put; > > + for (i = 1; i < BTRFS_SUPER_MIRROR_MAX; i++) { > + bytenr = btrfs_sb_offset(i); > + ret = btrfs_read_disk_super(bdev, bytenr, &page, &disk_super); > + if (ret < 0) { > + ret = 0; > + continue; > + } > + > + if (memcmp(disk_super_primary->fsid, disk_super->fsid, > + BTRFS_FSID_SIZE)) { > + pr_err("BTRFS (device %pg): superblock fsid mismatch, primary %pU copy%d %pU", > + bdev, disk_super_primary->fsid, i, > + disk_super->fsid); > + ret = -EINVAL; > + btrfs_release_disk_super(page); > + goto error_rel_primary; > + } > + btrfs_release_disk_super(page); > + } > + > mutex_lock(&uuid_mutex); > - device = device_list_add(path, disk_super); > + device = device_list_add(path, disk_super_primary); > if (IS_ERR(device)) > ret = PTR_ERR(device); > else > *fs_devices_ret = device->fs_devices; > mutex_unlock(&uuid_mutex); > > - btrfs_release_disk_super(page); > +error_rel_primary: > + btrfs_release_disk_super(page_primary); > > error_bdev_put: > blkdev_put(bdev, flags); >