System and method for network file system server replication using reverse path lookup

• US 9,552,367 B2
• Filed: 10/26/2012
• Issued: 01/24/2017
• Est. Priority Date: 09/16/2011
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Zhou, Dongmei, Shang, Guoxian, Chang, Baojian
• Primary Examiner(s): Lee, Wilson
• Application Number:
  US13/661,881
• Publication Number:
  US 20140122428A1
• Time in Patent Office: 1,551 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 11/202 where processing functionality is redundant redundant communication control functionality G06F11/2005, redundant storage control functionality G06F11/2089
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2048 where the redundant components share neither address space nor persistent storage
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 16/1794 Details of file format conversion
• CPC Class Codes: G06F 16/184 implemented as replicated file system

Abstract

The system and method described herein may use reverse path lookup to build mappings between file handles that represent network file system objects and full path names associated therewith and distinguish hard links between different file system objects having the same identifier with different parents or file names. The mappings and information distinguishing the hard links may then be cached to enable replicating changes to the file system. For example, a server may search the cached information using a file handle associated with a changed file system object to obtain the file name and full path name associated therewith. The server may then send the file name and full path name and metadata describing how the file system object was changed to a replica host, which may then replicate the change to the file system object.

Claims

1. A system configured to replicate a network file system, the system comprising:

a master file system;

a master host having a hardware processor system configured to;

receive a request to apply a change to the master file system, and receive a file handle that represents an object associated with the change to the master file system;

interpret the file handle to determine the object that the file handle represents, apply the change to the object within the master file system, and determine a full path name that the object has within the master file system based on parent data that the object has within the master file system; and

transmit metadata that describes the change to the master file system and a mapping between an inode number that identifies the object within the master file system and the full path name associated with the object, to a replica host configured to replicate the change using the mapping and the metadata.

2. The system of claim 1, further comprising a cache configured to store the mapping between the inode number and the full path name, wherein the processor system is configured to reference the cached mapping to enable replication of the change to the object.

3. The system of claim 1, wherein the processor system is further configured to:

inspect an inode corresponding to the inode number that identifies the object to determine whether the object represents a directory or a file within the master file system; and

read an entry in the inode that points to a parent directory associated with the inode in response to determining that the object represents a directory, wherein the parent directory represents the parent data that the directory object has within the master file system.

4. The system of claim 1, wherein the processor system is further configured to:

inspect an inode corresponding to the inode number that identifies the object to determine whether the object represents a directory or a file within the master file system; and

read an entry in the inode that points to a current directory that contains the inode in response to determining that the object represents a file, wherein the current directory represents the parent data that the file object has within the master file system.

5. The system of claim 1, wherein the processor system is configured to retrieve the mapping between the inode number and the full path name from a device-specific mapping file based on a device identifier or file system identifier associated with the file handle.

6. The system of claim 1, wherein the processor system is configured to store the mapping between the inode number and the full path name in a binary balance tree.

7. The system of claim 1, wherein the processor system is further configured to distinguish a hard link associated with the object in response to determining that another object having different parent data or a different file name, from the object shares the inode number that identifies the object.

8. The system of claim 7, wherein the processor system is configured to use the mapping between the inode number and the full path name associated with the object to distinguish the hard link if the object and the other object reside in different directories.

9. The system of claim 7, wherein the processor system is configured to add the file names associated with the object and the other object to the mapping between the inode number and the full path name associated with the object to distinguish the hard link if the object and the other object reside in a common directory.

10. A method to replicate a network file system, the method comprising:

receiving, at a master host, a request to apply a change to a master file system, wherein the request includes a file handle that represents an object associated with the change to the master file system;

interpreting the file handle to determine the object that the file handle represents at the master host, wherein the master host interprets the file handle to apply the change to the object within the master file system and determine a full path name that the object has within the master file system based on parent data that the object has within the master file system; and

transmitting metadata that describes the change to the master file system and a mapping between an inode number that identifies the object within the master file system and the full path name associated with the object, from the master host to a replica host, wherein the replica host replicates the change using the transmitted mapping and the transmitted metadata.

11. The method of claim 10, wherein the mapping between the inode number and the full path name is cached and a replication client executing on the master host references the cached mapping to enable replication of the change to the object.

12. The method of claim 10, further comprising;

inspecting an inode corresponding to the inode number that identifies the object to determine whether the object represents a directory or a file within the master file system; and

reading an entry in the inode that points to a parent directory associated with the inode in response to determining that the object represents a directory, wherein the parent directory represents the parent data that the directory object has within the master file system.

13. The method of claim 10, further comprising:

inspecting an inode corresponding to the inode number that identifies the object to determine whether the object represents a directory or a file within the master file system; and

reading an entry in the inode that points to a current directory that contains the inode in response to determining that the object represents a file, wherein the current directory represents the parent data that the file object has within the master file system.

14. The method of claim 10, further comprising retrieving the mapping between the inode number and the full path name from a device-specific mapping file based on a device identifier or file system identifier associated with the file handle.

15. The method of claim 10, wherein the mapping between the inode number and the full path name is stored in a binary balance tree.

16. The method of claim 10, further comprising distinguishing a hard link associated with the object in response to the master host determining that another object having different parent or a different file name, from the object shares the inode number that identifies the object.

17. The method of claim 16, wherein the master host uses the mapping between the inode number and the full path name associated with the object to distinguish the hard link if the object and the other object reside in different directories.

18. The method of claim 16, wherein the master host adds the file names associated with the object and the other object to the mapping between the inode number and the full path name associated with the object to distinguish the hard link if the object and the other object reside in a common directory.

19. A computer program product comprising:

a non-transitory computer readable storage medium comprising computer-readable program code executable by a hardware processor embodied therewith to replicate a network file system, the computer readable program code comprising;

computer readable program code configured to receive, at a master host, a request to apply a change to a master file system, wherein the request includes a file handle that represents an object associated with the change to the master file system;

computer readable program code configured to interpret the file handle to determine the object that the file handle represents at the master host, wherein the master host interprets the file handle to apply the change to the object within the master file system and determines a full path name that the object has within the master file system based on parent data that the object has within the master file system; and

computer readable program code configured to transmit metadata that describes the change to the master file system and a mapping between an inode number that identifies the object within the master file system and the full path name associated with the object, from the master host to a replica host, wherein the replica host replicates the change using the transmitted mapping and the transmitted metadata.

20. The computer program product of claim 19, wherein the computer readable program code is further configured to build, at the master host, the mapping between the inode number that identifies the object within the master file system and the full path name associated with the object.