glusterd/snapshot: Change file-system uuid to file-system label

Problem: In XFS changing file-system UUID with xfs_admin is causing
too much delay with large file-system. The time taken by xfs_admin
tool to change UUID is directly proportional to the size of the file
system.

Cause: In XFS file-system UUID is stored in file-system superblock.
Therefore for chaning UUID all the superblock needs to be changed.

Fix: Instead of using file-system UUID use file-system label.

Change-Id: Ifb4c668fb29cfc1c89d9b221abc8d09dc09589ec
BUG: 1115107
Signed-off-by: Rajesh Joseph <rjoseph@redhat.com>
Reviewed-on: http://review.gluster.org/8215
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-by: Avra Sengupta <asengupt@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Tested-by: Vijay Bellur <vbellur@redhat.com>

1 files changed, 6 insertions, 6 deletions

diff --git a/xlators/mgmt/glusterd/src/glusterd-handshake.c b/xlators/mgmt/glusterd/src/glusterd-handshake.c
index 0fd60806da2..7d582d443f4 100644
--- a/xlators/mgmt/glusterd/src/glusterd-handshake.c
+++ b/xlators/mgmt/glusterd/src/glusterd-handshake.c
@@ -454,16 +454,16 @@ glusterd_create_missed_snap (glusterd_missed_snap_info *missed_snapinfo,
         }
 
         /* After the snapshot both the origin brick (LVM brick) and
-         * the snapshot brick will have the same file-system UUID. This
+         * the snapshot brick will have the same file-system label. This
          * will cause lot of problems at mount time. Therefore we must
-         * generate a new UUID for the snapshot brick
+         * generate a new label for the snapshot brick
          */
-        ret = glusterd_update_fs_uuid (brickinfo);
+        ret = glusterd_update_fs_label (brickinfo);
         if (ret) {
                 gf_log (this->name, GF_LOG_ERROR, "Failed to update "
-                        "file-system uuid for %s brick", brickinfo->path);
-                /* Failing to update UUID should not cause snapshot failure.
-                 * Currently UUID is updated only for XFS and ext2/ext3/ext4
+                        "file-system label for %s brick", brickinfo->path);
+                /* Failing to update label should not cause snapshot failure.
+                 * Currently label is updated only for XFS and ext2/ext3/ext4
                  * file-system.
                  */
         }