Method and apparatus for splitting a replicated volume

• US 7,805,401 B2
• Filed: 10/31/2006
• Issued: 09/28/2010
• Est. Priority Date: 04/14/2003
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Toner, Stephen G.
• Primary Examiner(s): Vo, Tim T.
• Assistant Examiner(s): OWYANG, MICHELLE N
• Application Number:
  US11/555,105
• Publication Number:
  US 20080104132A1
• Time in Patent Office: 1,428 Days
• Field of Search: None
• US Class Current: 707/610
• CPC Class Codes: G06F 16/119 Details of migration of file systems migration mechanisms in storage systems G06F3/0647

Abstract

At least two replicated instances of a source volume are split while allowing clients to access data moved during the split. Clients are redirected to the first replicated instance of the source volume. The first replicated instance is split by first moving files in a split path from the first replicated instance to the target volume. Then, after the files in the split path have been successfully moved to the target volume, a junction is inserted at the split directory to redirect clients to the target volume. After the first replicated instance is split, a second junction replaces the split path on the replicated instance of the first replicated instance.

Claims

1. A system to move a subdirectory tree, comprising:

a computer;

a first replicated instance of a source volume;

a directory tree on the first replicated instance of the source volume, the directory tree including a split path, the split path including a split directory;

a second replicated instance of the source volume, the second replicated instance including a copy of the directory tree, the copy of the directory tree including a copy of the split path, the copy of the split path including a split directory;

a target volume;

a volume location database to store corresponding volume names and locations;

a volume manager to redirect a client from the second replicated instance of the source volume to the first replicated instance of the source volume using the volume location database;

a file verifier to verify that each file in the copy of the split path on the second replicated instance of the source volume is closed and to notify the volume manager when each file in the copy of the split path on the second replicated instance of the source volume is closed; and

a subdirectory mover to move each file in the split path on the first replicated instance of the source volume to the target volume responsive to verification that each file in the copy of the split path on the second replicated instance of the source volume is closed while allowing the client to access each file in the split path on the first replicated instance of the source volume, wherein the volume manager is operative to insert a junction in the second replicated instance of the source volume pointing to the target volume after the subdirectory mover moves each file in the split path on the first replicated instance of the source volume to the target volume and to delete each file in the split path on the first replicated instance of the source volume.

2. The system according to claim 1, wherein the volume manager includes a junction creator to insert a temporary junction at the split directory on the second replicated instance of the source volume pointing to a temporary distributed file system globally unique identifier (DFS GUID) assigned to the first replicated instance of the source volume before the subdirectory mover moves each file in the split path on the first replicated instance of the source volume to the target volume.

3. The system according to claim 2, further comprising a DFS GUID creator to create the temporary DFS GUID assigned to the first replicated instance of the source volume.

4. The system according to claim 2, further comprising a junction remover to remove from the second replicated instance of the source volume the temporary junction pointing to the temporary DFS GUID assigned to the first replicated instance of the source volume.

5. The system according to claim 1, wherein the volume manager includes a junction creator to insert a first junction at the split directory on the first replicated instance of the source volume pointing to a DFS GUID assigned to the target volume.

6. The system according to claim 5, wherein the volume manager is operative to insert a second junction at the split directory on the second replicated instance of the source volume pointing to the DFS GUID assigned to the target volume and to delete each file in the copy of the split path on the second replicated instance of the source volume.

7. The system according to claim 6, further comprising a replication manager to propagate a copy of the second junction to the second replicated instance of the source volume at the split directory on the second replicated instance of the source volume and to replicate a deletion of each file in the copy of the split path on the second replicated instance of the source volume.

8. A computer-implemented method to move a subdirectory tree on a first replicated instance of a source volume to a target volume, comprising:

assigning to the first replicated instance of the source volume a first distributed file system globally unique identifier (DFS GUID);

inserting at a second replicated instance of the source volume a first junction pointing to the first DFS GUID;

verifying each file in a split path on the second replicated instance of the source volume is closed including verifying that each file in the split path on the second replicated instance of the source volume is not being currently accessed by a user, wherein the split path includes a split directory;

notifying a volume manager that each file in the split path on the second replicated instance of the source volume is closed;

copying each file in a corresponding split path on the first replicated instance of the source volume to the target volume responsive to the notifying that each file in the split path on the second replicated instance of the source volume is closed;

assigning to the target volume a second DFS GUID;

inserting at the split directory on the first replicated instance of the source volume a second junction pointing to the second DFS GUID;

deleting each file in the split path on the first replicated instance of the source volume;

removing from the second replicated instance of the source volume the first junction pointing to the first DFS GUID assigned to the first replicated instance of the source volume; and

updating the second replicated instance of the source volume.

9. The method according to claim 8, wherein updating the second replicated instance of the source volume includes:

inserting at the split directory on the second replicated instance of the source volume a copy of the second junction to the second DFS GUID pointing to the target volume; and

deleting each file in the split path on the second replicated instance of the source volume.

10. The method according to claim 8, wherein:

assigning to the first replicated instance of the source volume a first DFS GUID includes storing an assignment of the first DFS GUID to the first replicated instance of the source volume in a volume locator database (VLDB); and

assigning to the target volume a second DFS GUID includes storing an assignment of the second DFS GUID to the target volume in the VLDB.

11. The method according to claim 8 further comprising un-assigning the first DFS GUID from the first replicated instance of the source volume.

12. The method according to claim 11, wherein un-assigning the first DFS GUID from the first replicated instance of the source volume includes removing an assignment of the first DFS GUID to the first replicated instance of the source volume in a volume location database (VLDB).

13. A computer apparatus to move a subdirectory tree from a replicated storage volume to a target storage volume, comprising:

a volume locator database (VLDB) to store a first entry including a first assignment of a first distributed file system globally unique identifier (DFS GUID) to a first replicated instance of a source storage volume, a second entry including a second assignment of the first DFS GUID to a second replicated instance of the source storage volume, and a third entry including a third assignment of a second DFS GUID to the target storage volume;

a DFS GUID creator to create a fourth entry in the VLDB including a fourth assignment of a temporary DFS GUID to the first replicated instance of the source storage volume;

a volume manager including a junction creator to insert a first junction at the second replicated instance of the source storage volume pointing to the temporary DFS GUID assigned to the first replicated instance of the source storage volume; and

a file verifier to verify that each file in a split path on the second replicated instance of the source storage volume is closed and to notify the volume manager when each file in the split path on the second replicated instance of the source storage volume is closed.

14. The apparatus according to claim 13, further comprising a subdirectory mover to move each file in the split path on the first replicated instance of the source storage volume to the target storage volume while allowing a client to access a file in the split path of the first replicated instance of the source storage volume.

15. The apparatus according to claim 14, wherein the volume manager is operative to insert at a split directory on the first replicated instance of the source storage volume a second junction pointing to the second DFS GUID assigned to the target storage volume.

16. The apparatus according to claim 15, further comprising a junction remover to remove the first junction from the second replicated instance of the source storage volume.

17. The apparatus according to claim 15, wherein the volume manager is operative to insert at the split path on the second replicated instance of the source storage volume a third junction pointing to the second DFS GUID assigned to the target storage volume and to delete each file in the split path on the second replicated instance of the source storage volume.

18. The apparatus according to claim 15, further comprising a replication manager to propagate the second junction to the second replicated instance of the source storage volume at the split path on the second replicated instance of the source storage volume and to replicate a deletion of each file in the split path on the second replicated instance of the second replicated instance of the source storage volume.

19. An article, comprising a storage medium, said storage medium having stored thereon instructions, that, when executed by a machine, result in:

assigning to a first replicated instance of a source volume a first distributed file system globally unique identifier (DFS GUID);

inserting at a split directory of a second replicated instance of the source volume a first junction pointing to the first DFS GUID;

verifying each file in a split path on the second replicated instance of the source volume is closed including verifying each file in the split path on the second replicated instance of the source volume is not being currently accessed by a user;

notifying a volume manager that each file in the split path on the second replicated instance of the source volume is closed;

copying each file in the split path on the first replicated instance of the source volume to a target volume responsive to the notifying that each file in the split path on the second replicated instance of the source volume is closed;

assigning to the target volume a second DFS GUID;

inserting at the split directory on the first replicated instance of the source volume a second junction pointing to the second DFS GUID;

deleting each file in the split path on the first replicated instance of the source volume;

removing from the second replicated instance of the source volume the first junction pointing to the first DFS GUID assigned to the first replicated instance of the source volume; and

updating the second replicated instance of the source volume.

20. The article according to claim 19, wherein updating the second replicated instance of the source volume the move of the files in the split path from the first replicated instance of the source volume to the target volume includes:

inserting at the split directory on the second replicated instance of the source volume a copy of the second junction to the second DFS GUID pointing to the target volume; and

deleting each file in the split directory on the second replicated instance of the source volume.

21. The article according to claim 19, wherein:

assigning to the first replicated instance of the source volume a first DFS GUID includes storing an assignment of the first DFS GUID to the first replicated instance of the source volume in a volume locator database (VLDB); and

assigning to the target volume a second DFS GUID includes storing an assignment of the second DFS GUID to the target volume in the VLDB.

22. The article according to claim 19 further comprising un-assigning the first DFS GUID from the first replicated instance of the source volume.

23. The article according to claim 22, wherein un-assigning the first DFS GUID from the first replicated instance of the source volume includes removing an assignment of the first DFS GUID to the first replicated instance of the source volume in a volume location database (VLDB).