On 05/15/2012 03:43 PM, Richard W.M. Jones wrote:
On Tue, May 15, 2012 at 10:33:42AM +0800, Wanlong Gao wrote:
Add the new API btrfs-fsck to check the btrfs filesystem. Btrfs is currently under heavy development, and not suitable for any uses other than benchmarking and review. But it'll be useful in the near future.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com> --- daemon/btrfs.c | 17 +++++++++++++++++ generator/generator_actions.ml | 11 +++++++++++ src/MAX_PROC_NR | 2 +- 3 files changed, 29 insertions(+), 1 deletion(-)
diff --git a/daemon/btrfs.c b/daemon/btrfs.c index ab3babc..62f07f3 100644 --- a/daemon/btrfs.c +++ b/daemon/btrfs.c @@ -619,3 +619,20 @@ do_btrfs_set_seeding (const char *device, int svalue) free (err); return r; } + +int +do_btrfs_fsck (const char *device) +{ + char *err; + int r; + + r = commandr (NULL, &err, "btrfsck", device, NULL); + if (r == -1) { + reply_with_error ("%s: %s", device, err); + free (err); + return -1; + } + + free (err); + return r; +}
This worries me. Isn't btrfsck interactive? Ordinary fsck has a meaningful exit code, but btrfsck doesn't. Will it get one in future?
It may because btrfsck is a experimental tool and will change a lot in the future.
diff --git a/generator/generator_actions.ml b/generator/generator_actions.ml index fc2df42..33ffd6b 100644 --- a/generator/generator_actions.ml +++ b/generator/generator_actions.ml @@ -7232,6 +7232,17 @@ If C<devices> is an empty list, this does nothing."); Enable or disable the seeding feature of a device that contains a btrfs filesystem.");
+ ("btrfs_fsck", (RErr, [Device "device"], []), 332, [Optional "btrfs"], + [InitPartition, IfAvailable "btrfs", TestRun ( + [["mkfs_btrfs"; "/dev/sda1"; ""; ""; "NOARG"; ""; "NOARG"; "NOARG"; ""; ""]; + ["btrfs_fsck"; "/dev/sda1"]])], + "check a btrfs filesystem", + "\ +Used to check a btrfs filesystem, C<device> is the device file where the +filesystem is stored.
Put a blank line between paragraphs.
+NOTE: Btrfs is currently under heavy development, and not suitable for any +uses other than benchmarking and review.");
Instead of "NOTE:", it's better to use I<Note:> or B<Note:> (see perlpod(1)).
Will send V2. Thanks, Wanlong Gao
Rich.