Re: [PATCH RFC] btrfs: self heal from SB fail

["Austin S. Hemmelgarn" <ahferroin7@gmail.com>]

On 2017-12-08 07:59, Qu Wenruo wrote:
> 
> 
> On 2017年12月08日 20:51, Austin S. Hemmelgarn wrote:
>> On 2017-12-08 02:57, Anand Jain wrote:
>>> -EXPERIMENTAL-
>>> As of now when primary SB fails we won't self heal and would fail mount,
>>> this is an experimental patch which thinks why not go and read backup
>>> copy.
>> I like the concept, and actually think this should be default behavior
>> on a filesystem that's already mounted (we fix other errors, why not
>> SB's), but I don't think it should be default behavior at mount time for
>> the reasons Qu has outlined (picking up old BTRFS SB's after
>> reformatting is bad).  However, I do think it's useful to be able to ask
>> for this behavior on mount, so that you don't need to fight with the
>> programs to get a filesystem to mount when the first SB is missing
>> (perhaps add a 'usebackupsb' option to mirror 'usebackuproot'?).
> 
> Yeah, I also like the idea of 'usebackupsuper'/'usebackupsuper=n' mount
> option than do it automatically.
I still think there should be an option to do automatic detection (it's 
not particularly hard, and it's not likely to do the wrong thing in most 
cases), but being able to explicitly specify a particular superblock for 
a mount is definitely a step in the right direction.
> 
> Thanks,
> Qu
> 
>>>
>>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>>> ---
>>>    fs/btrfs/disk-io.c |  8 +++++++-
>>>    fs/btrfs/volumes.c | 10 +++++++---
>>>    2 files changed, 14 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
>>> index 9b20c1f3563b..a791b8dfe8a8 100644
>>> --- a/fs/btrfs/disk-io.c
>>> +++ b/fs/btrfs/disk-io.c
>>> @@ -3190,7 +3190,7 @@ struct buffer_head *btrfs_read_dev_super(struct
>>> block_device *bdev)
>>>         * So, we need to add a special mount option to scan for
>>>         * later supers, using BTRFS_SUPER_MIRROR_MAX instead
>>>         */
>>> -    for (i = 0; i < 1; i++) {
>>> +    for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
>>>            ret = btrfs_read_dev_one_super(bdev, i, &bh);
>>>            if (ret)
>>>                continue;
>>> @@ -4015,11 +4015,17 @@ static int btrfs_check_super_valid(struct
>>> btrfs_fs_info *fs_info)
>>>            ret = -EINVAL;
>>>        }
>>>    +#if 0
>>> +    /*
>>> +     * Need a way to check for any copy of SB, as its not a
>>> +     * strong check, just ignore this for now.
>>> +     */
>>>        if (btrfs_super_bytenr(sb) != BTRFS_SUPER_INFO_OFFSET) {
>>>            btrfs_err(fs_info, "super offset mismatch %llu != %u",
>>>                  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
>>>            ret = -EINVAL;
>>>        }
>>> +#endif
>>>          /*
>>>         * Obvious sys_chunk_array corruptions, it must hold at least
>>> one key
>>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>>> index 9fa2539a8493..f368db94d62b 100644
>>> --- a/fs/btrfs/volumes.c
>>> +++ b/fs/btrfs/volumes.c
>>> @@ -1369,7 +1369,7 @@ int btrfs_scan_one_device(const char *path,
>>> fmode_t flags, void *holder,
>>>    {
>>>        struct btrfs_super_block *disk_super;
>>>        struct block_device *bdev;
>>> -    struct page *page;
>>> +    struct buffer_head *sb_bh;
>>>        int ret = -EINVAL;
>>>        u64 devid;
>>>        u64 transid;
>>> @@ -1392,8 +1392,12 @@ int btrfs_scan_one_device(const char *path,
>>> fmode_t flags, void *holder,
>>>            goto error;
>>>        }
>>>    -    if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
>>> +    sb_bh = btrfs_read_dev_super(bdev);
>>> +    if (IS_ERR(sb_bh)) {
>>> +        ret = PTR_ERR(sb_bh);
>>>            goto error_bdev_put;
>>> +    }
>>> +    disk_super = (struct btrfs_super_block *) sb_bh->b_data;
>>>          devid = btrfs_stack_device_id(&disk_super->dev_item);
>>>        transid = btrfs_super_generation(disk_super);
>>> @@ -1413,7 +1417,7 @@ int btrfs_scan_one_device(const char *path,
>>> fmode_t flags, void *holder,
>>>        if (!ret && fs_devices_ret)
>>>            (*fs_devices_ret)->total_devices = total_devices;
>>>    -    btrfs_release_disk_super(page);
>>> +    brelse(sb_bh);
>>>      error_bdev_put:
>>>        blkdev_put(bdev, flags);
>>>