System and method for system state replication

• US 8,001,079 B2
• Filed: 02/29/2008
• Issued: 08/16/2011
• Est. Priority Date: 02/29/2008
• Status: Active Grant

• Current Assignee: Carbonite LLC (Open Text Corporation)
• Original Assignee: Double-Take Software Incorporated (Open Text Corporation)
• Inventors: Lu, Xiaofeng, Demlow, David J., Beeler, Rob
• Primary Examiner(s): LE, DEBBIE M
• Application Number:
  US12/040,289
• Publication Number:
  US 20090222498A1
• Time in Patent Office: 1,264 Days
• Field of Search: 707/204, 709/246, 714/15
• US Class Current: 707/616
• CPC Class Codes: G06F 16/10 File systems; File servers

Abstract

The invention generally describes a system and method for data backup and recovery, in particular for backing up an entire server (i.e., system state and data) in substantially real-time. The invention provides high availability by failing a source server over to a target server if the source server fails, where the target server is reconfigured nearly identically to the source server. The invention does not require the source and target servers to have identical hardware or operating system versions. The invention further allows for rapid, automatic, restoration of a source server.

Claims

1. A method for backing up a first server, comprising the steps of:

providing a disk storage connected to a second server, said disk storage comprising a staging area;

replicating changes to data files and system state files of the first server to the staging area connected to the second server;

wherein the replicated changes to the files of the first server are made to a corresponding copy of the files stored in the staging area of the second server;

wherein the staging area of the second server stores replicated versions of files that are locked to avoid overwriting of live files on the second server; and

upon detection of a failure of the first server, failing over the first server to the second server by merging the replicated system state files of the first server into a system state of the second server.

2. The method of claim 1, further comprising an initial step of copying the data files and system state files of the first server to the second server.

3. The method of claim 1, where the step of replicating changes is performed asynchronously.

4. The method of claim 1, where the step of merging the replicated system state files from the first server into a system state of the second server further comprises the steps of:

merging a registry of the second server with a registry of the first server and saving the merged registry to a file in a temporary storage area on the second server;

copying a non-conflicting subset of the replicated system state files of the first server to the temporary storage area, where the non-conflicting subset is constructed so as to retain hardware-dependent configuration of the second server; and

moving the system state files from the temporary storage area so as to overwrite corresponding system state files on the second server.

5. The method of claim 4, where the step of merging the replicated system state files from the first server into a system state of the second server further comprises the final step of rebooting the second server.

6. The method of claim 1, further comprising, after failover and restart of the first server, resynchronizing the data files of the second server to the first server.

7. The method of claim 6, further comprising the steps of:

after resynchronizing the data files of the second server to the first server, copying the system state files from the staging area on the second server to the first server; and

merging the copied system state files into a system state of the first server.

8. The method of claim 1, further comprising the step of replicating file changes to system state files of the second server to the first server, where replicated file changes are made to a corresponding copy of the file on the first server, copies of the system state files being located in a staging area on local storage of the first server.

9. The method of claim 8, further comprising, after failover and restart of the first server, merging the replicated system state files of the second server into a system state of the first server.

10. The method of claim 8, further comprising the steps of:

detecting a failure of the second server;

after restart of the second server, copying the system state files from the staging area of the first server to the second server; and

merging the copied system state files into a system state of the second server.

11. The method of claim 1, where the first server includes a plurality of source servers.

12. The method of claim 1, where the second server includes a plurality of target servers.

13. The method of claim 1, further comprising capturing image snapshots of the files on the first server at a plurality of times, and, upon detection of a failure at the first server, giving a user an option to restore the second server to any one of said captured image snapshots.

14. The method of claim 1, wherein the step of replicating changes to data files and system state files of the first server to the second server additionally comprises intercepting file change requests at an input output level and mirroring those file change requests to the second server so that system state files from the first server are replicated in near real-time onto the second server.

15. The method of claim 1, comprising the steps of directing the second server to rescue the first server by restoring the first server'"'"'s file integrity by rebuilding the first server'"'"'s data and system files by transferring the data and system files from the second server back to the first server.

16. The method of claim 1, comprising the steps of directing the second server to reverse roles with the first server by:

directing the second server to assume the role of the first server, transferring system and state files which relate to the second server'"'"'s role as the second server to the first server so that the first server can assume the role of the second server.

17. A method for backing up a first server comprising the steps of:

providing a disk storage connected to a second server, said disk storage comprising a staging area;

replicating changes to data files and system state files of the first server to the staging area connected to the second server;

wherein the replicated changes to the files of the first server are made to a corresponding copy of the files stored in the staging area of the second server;

wherein the staging area of the second server stores replicated versions of files that are locked to avoid overwriting of live files on the second server;

upon detection of the failure of the first server, failing over the first server to a third server by the steps of;

copying the replicated data files of the first server from the second server to the third server;

copying the replicated system state files of the first server from the second server to the third server, where the replicated system state files of the first server are copied to a staging area on the third server; and

merging the copied system state files into a system state of the third server.

18. The method of claim 17, where the step of replicating changes is performed asynchronously.

19. The method of claim 17, where the step of failing over the first server to a third server further comprises selecting the third server from a list of servers, where the list is stored on the second server.

20. The method of claim 17, where the method of failing over the first server to a third server further comprises, prior to copying the replicated system state files of the first server from the second server to the third server, validating the system state of the third server for compatibility with a system state of the first server.

21. The method of claim 17, further comprising an initial step of copying the data files and system state files of the first server to the second server.

22. The method of claim 17, further comprising, after failover and restart of the first server, resynchronizing the data files of the third server to the first server.

23. The method of claim 22, further comprising the steps of:

after resynchronizing the data files of the third server to the first server, copying the system state files from the staging area on the second server to the first server; and

merging the copied system state files into a system state of the first server.

24. The method of claim 22, further comprising the steps of:

after resynchronizing the data files of the third server to the first server, copying the system state files from the staging area on the third server to the first server; and

merging the copied system state files into a system state of the first server.

25. The method of claim 17, further comprising capturing image snapshots of the files on the first server at a plurality of times, and, upon detection of a failure at the first server, giving a user an option to restore the data files of said third server to reflect any one of said captured image snapshots.

26. A system for backup of a source server, comprising:

the source server, further including local storage and a connection to a network;

a target server, including local storage and a connection to the network;

a disk storage connected to the target server, said disk storage comprising a staging area;

a replication component executing on the source and target servers, where the replication component replicates changes to data files and system state files of the source server to the staging area connected to the target server;

wherein the replicated changes to the files of the source server are made to a corresponding copy of the files stored in the staging area of the target server;

wherein the staging area of the target server stores replicated versions of files that are locked to avoid overwriting of live files on the target server;

a failover component executing on the source and target servers, where the failover component on the target server;

detects a failure of the source server, and

upon detection of the failure of the source server, fails over the source server to the target server by merging the replicated system state files of the source server into a system state of the target server.

27. The system of claim 26, wherein the replication component performs an initial step of copying the data files and system state files of the source server to the target server.

28. The system of claim 26, wherein the replication component replicates changes asynchronously.

29. The system of claim 26, wherein replicated changes to a file of the source server are made to a corresponding copy of the file on the target server, copies of the system state files being located in a staging area on the local storage of the target server.

30. The system of claim 26, wherein the step of merging the replicated system state files from the source server into a system state of the target server further comprises the steps of:

merging a registry of the target server with a registry of the source server and saving the merged registry to a file in a temporary storage area on the target server;

copying a non-conflicting subset of the replicated system state files of the source server to the temporary storage area, where the non-conflicting subset is constructed so as to retain hardware-dependent configuration of the target server; and

moving the files from the temporary storage area so as to overwrite corresponding system state files on the target server.

31. The system of claim 30, wherein the step of merging the replicated system state files from the source server into a system state of the target server further comprises the final step of rebooting the second server.

32. The system of claim 29, wherein the failover component further performs the step of, after failover and restart of the source server, resynchronizing the data files of the target server to the source server.

33. The system of claim 32, wherein the failover component further performs the steps of:

after resynchronizing the data files of the target server to the source server, copying the system state files from the staging area on the target server to the source server; and

merging the copied system state files into a system state of the source server.

34. The system of claim 26, wherein the replication component further performs the step of replicating file changes to system state files of the target server to the source server, where replicated file changes are made to a corresponding copy of the file on the source server, copies of the system state files being located in a staging area on the local storage of the target server.

35. The system of claim 34, wherein the failover component further performs the step of, after failover and restart of the source server, merging the replicated system state files of the target server into a system state of the source server.

36. The system of claim 34, wherein the failover component further performs the steps of:

detecting a failure of the target server;

after restart of the target server, copying the system state files from the staging area of the source server to the target server; and

merging the copied system state files into a system state of the target server.

37. The system of claim 26, wherein image snapshots of the files on the source server are captured by an image snapshot application operated by said image replication component, and wherein said failover component is configured to, upon detection of a failure at the source server, give a user an option to restore the target server to reflect any one of said captured image snapshots.

38. The system of claim 26, wherein the replication component intercepts file change requests at an input output level and mirrors those file change requests to the target server so that system state files from the source server are replicated in near real-time onto the target server.

39. The system of claim 29, wherein the staging area of the target server stores replicated versions of files that are locked or open on the target server to avoid overwriting of live files on the target server.

40. The system of claim 26, wherein the failover component of the target server rescues the source server by restoring the source server'"'"'s file integrity by rebuilding the source server'"'"'s data and system files by transferring the data and system files from the target server back to the source server.

41. The system of claim 26, wherein the failover component of the target server causes the target server to reverses roles with the source server by directing the target server to assume the role of the source server, and transferring system and state files which relate to the target server'"'"'s role as the target server to the source server so that the source server can assume the role of the target server.

42. A system for backing up a source server comprising:

the source server, further including local storage and a connection to a network;

a backup server, including local storage and a connection to the network;

at least one target server, including local storage and a connection to the network;

a disk storage connected to a target server, said disk storage comprising a staging area;

a replication component executing on the source and backup servers, where the replication component replicates changes to data files and system state files of the source server to the staging area connected to the target server;

wherein the replicated changes to the files of the source server are made to a corresponding copy of the files stored in the staging area of the target server;

wherein the staging area of the target server stores replicated versions of files that are locked to avoid overwriting of live files on the target server;

a failover component executing on the source, backup, and target servers, where the failover component on the backup server;

detects a failure of the source server, and

upon detection of the failure of the source server, fails over the source server to a selected target server from the at least one target servers by the steps of;

copying the replicated data files of the source server from the backup server to the selected target server;

copying the replicated system state files of the source server from the backup server to the selected target server, where the replicated system state files of the source server are copied to a staging area on the selected target server; and

merging the copied system state files into a system state of the selected target server.

43. The system of claim 42, wherein the replication component replicates changes asynchronously.

44. The system of claim 42, wherein the backup server selects the selected target server from a list of servers, where the list is stored on the backup server.

45. The system of claim 42, wherein the failover component further performs the step of, prior to copying the replicated system state files of the source server from the backup server to the selected target server, validating the system state of the selected target server for compatibility with a system state of the source server.

46. The system of claim 42, wherein the replication component further performs an initial step of copying the data files and system state files of the source server to the backup server.

47. The system of claim 42, wherein the failover component further performs the step of, after failover and restart of the source server, resynchronizing the data files of the selected target server to the source server.

48. The system of claim 47, wherein the failover component further performs the steps of:

after resynchronizing the data files of the selected target server to the source server, copying the system state files from the staging area on the backup server to the source server; and

merging the copied system state files into a system state of the source server.

49. The system of claim 47, wherein the failover component further performs the steps of:

after resynchronizing the data files of the selected target server to the source server, copying the system state files from the staging area on the selected target server to the source server; and

merging the copied system state files into a system state of the source server.

50. A computer-readable medium encoding a program for backing up a first server, where the program comprises a set of instructions for performing the steps of:

replicating changes to data files and system state files of the first server to a staging area stored in a disk storage connected to a second server;

wherein the replicated changes to the files of the first server are made to a corresponding copy of the files stored in the staging area of the second server;

wherein the staging area of the second server stores replicated versions of files that are locked to avoid overwriting of live files on the second server;

detecting a failure of the first server; and

upon detection of the failure of the first server, failing over the first server to the second server by merging the replicated system state files of the first server into a system state of the second server.

51. The computer-readable medium of claim 50, wherein the program further comprises instructions for performing an initial step of copying the data files and system state files of the first server to the second server.

52. The computer-readable medium of claim 50, wherein the step of replicating changes is performed asynchronously.

53. The computer-readable medium of claim 50, wherein the step of merging the copied system state files from the first server into the system state of the second server further comprises the steps of:

merging a registry of the second server with a registry of the first server and saving the merged registry to a file in a temporary storage area on the second server;

copying a non-conflicting subset of the replicated system state files of the first server to the temporary storage area, where the non-conflicting subset is constructed so as to retain hardware-dependent configuration of the second server; and

moving the system state files from the temporary storage area so as to overwrite corresponding system state files on the second server.

54. The computer-readable medium of claim 53, wherein the step of merging the copied system state files from the first server into the system state of the second server further comprises the final step of rebooting the second server.

55. The computer-readable medium of claim 50, wherein the program further comprises instructions for performing the step of, after failover and restart of the first server, resynchronizing the data files of the second server to the first server.

56. The computer-readable medium of claim 55, wherein the program further comprises instructions for performing the steps of:

after resynchronizing the data files of the second server to the first server, copying the system state files from the staging area on the second server to the first server; and

merging the copied system state files into a system state of the first server.

57. The computer-readable medium of claim 50, wherein the program further comprises instructions for performing the step of replicating file changes to system state files of the second server to the first server, where replicated file changes are made to a corresponding copy of the file on the first server, copies of the system state files being located in a staging area on local storage of the first server.

58. The computer-readable medium of claim 57, wherein the program further comprises instructions for performing the step of, after failover and restart of the first server, merging the replicated system state files of the second server into a system state of the first server.

59. The computer-readable medium of claim 57, wherein the program further comprises instructions for performing the steps of:

detecting a failure of the second server;

after restart of the second server, copying the system state files from the staging area of the first server to the second server; and

merging the copied system state files into a system state of the second server.

60. The computer-readable medium of claim 50, wherein the first server is a plurality of source servers.

61. The computer-readable medium of claim 50, wherein the second server is a plurality of target servers.

62. A computer-readable medium encoding a program for backing up a first server, where the program comprises a set of instructions for performing the steps of:

replicating changes to data files and system state files of the first server to replicating changes to data files and system state files of the first server to a staging area stored in a disk storage connected to a second server;

wherein the replicated changes to the files of the first server are made to a corresponding copy of the files stored in the staging area of the second server;

wherein the staging area of the second server stores replicated versions of files that are locked to avoid overwriting of live files on the second server;

detecting a failure of the first server; and

upon detection of the failure of the first server, failing over the first server to the third server by the steps of;

copying the replicated data files of the first server from the second server to the third server;

copying the replicated system state files of the first server from the second server to the third server, where the replicated system state files of the first server are copied to a staging area on the third server; and

merging the copied system state files into a system state of the third server.

63. The computer-readable medium of claim 62, wherein the step of replicating changes is performed asynchronously.

64. The computer-readable medium of claim 62, wherein the step of failing over the first server to a third server further comprises selecting the third server from a list of servers, where the list is stored on the second server.

65. The computer-readable medium of claim 62, wherein the method of failing over the first server to a third server further comprises, prior to copying the replicated system state files of the first server from the second server to the third server, validating the system state of the third server for compatibility with a system state of the first server.

66. The computer-readable medium of claim 62, wherein the program further comprises instructions for performing an initial step of copying the data files and system state files of the first server to the second server.

67. The computer-readable medium of claim 62, wherein the program further comprises instructions for performing the step of, after failover and restart of the first server, resynchronizing the data files of the third server to the first server.

68. The computer-readable medium of claim 67, wherein the program further comprises instructions for performing the steps of:

after resynchronizing the data files of the third server to the first server, copying the system state files from the staging area on the second server to the first server; and

merging the copied system state files into a system state of the first server.

69. The computer-readable medium of claim 67, wherein the program further comprises instructions for performing the steps of:

after resynchronizing the data files of the third server to the first server, copying the system state files from the staging area on the third server to the first server; and

merging the copied system state files into a system state of the first server.