xfs: write unmount record for ro mounts
authorEric Sandeen <sandeen@sandeen.net>
Wed, 9 Aug 2017 01:19:47 +0000 (18:19 -0700)
committerDarrick J. Wong <darrick.wong@oracle.com>
Tue, 22 Aug 2017 16:22:23 +0000 (09:22 -0700)
There are dueling comments in the xfs code about intent
for log writes when unmounting a readonly filesystem.

In xfs_mountfs, we see the intent:

/*
 * Now the log is fully replayed, we can transition to full read-only
 * mode for read-only mounts. This will sync all the metadata and clean
 * the log so that the recovery we just performed does not have to be
 * replayed again on the next mount.
 */

and it calls xfs_quiesce_attr(), but by the time we get to
xfs_log_unmount_write(), it returns early for a RDONLY mount:

 * Don't write out unmount record on read-only mounts.

Because of this, sequential ro mounts of a filesystem with
a dirty log will replay the log each time, which seems odd.

Fix this by writing an unmount record even for RO mounts, as long
as norecovery wasn't specified (don't write a clean log record
if a dirty log may still be there!) and the log device is
writable.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
fs/xfs/xfs_log.c

index 4ebd0bafc914ce7b7f1e4d35134c9dedb09a23b5..972eda87db2bae174317f2a8ba90fc0db2462de4 100644 (file)
@@ -812,11 +812,14 @@ xfs_log_unmount_write(xfs_mount_t *mp)
        int              error;
 
        /*
-        * Don't write out unmount record on read-only mounts.
+        * Don't write out unmount record on norecovery mounts or ro devices.
         * Or, if we are doing a forced umount (typically because of IO errors).
         */
-       if (mp->m_flags & XFS_MOUNT_RDONLY)
+       if (mp->m_flags & XFS_MOUNT_NORECOVERY ||
+           xfs_readonly_buftarg(log->l_mp->m_logdev_targp)) {
+               ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
                return 0;
+       }
 
        error = _xfs_log_force(mp, XFS_LOG_SYNC, NULL);
        ASSERT(error || !(XLOG_FORCED_SHUTDOWN(log)));