Error recovery for SRDF assist

• US 7,237,040 B1
• Filed: 09/28/2001
• Issued: 06/26/2007
• Est. Priority Date: 08/28/2001
• Status: Expired due to Term

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Kamvysselis, Peter
• Primary Examiner(s): Dharia; Rupal
• Assistant Examiner(s): SHINGLES, KRISTIE D
• Application Number:
  US09/965,926
• Time in Patent Office: 2,097 Days
• Field of Search: 709/232, 709/237, 709/248, 709/211, 714/6, 714/7, 711/162, 711/170, 707/204, 707/203, 707/1
• US Class Current: 709/248
• CPC Class Codes: H04L 63/0428 wherein the data content is protected, e.g. by encrypting or encapsulating the payload
• CPC Class Codes: H04L 67/06 specially adapted for file transfer, e.g. file transfer protocol [FTP]
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y10S 707/99938 Concurrency, e.g. lock management in shared database
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

Transmitting data from a source to a destination, includes receiving the data from the source and providing the data to the destination using a network, where the data is acknowledged to the source as being successfully received at the destination prior to all of the data being provided to the network. The source may be a primary storage device. The destination may be a secondary storage device. The data may be provided from the source in a first format and may be provided to the network in a second format that is different from the first format where the first format may be RDF and the second format may be TCP/IP or UDP. The destination may receive the data in a first format different from a second format used to transmit the data over the network. The first format may be RDF and the second format may be TCP/IP or UDP.

Claims

1. A method for performing data recovery in a computer system, comprising:

sending data from a first storage device to at least one other secondary storage device, said data being sent in a plurality of data packets, each of said plurality of packets being associated with a sequence number having a first predetermined value;

upon determining that the data has been successfully stored on all of said at least one other secondary storage device, deleting journal entries in a sender corresponding to said data; and

upon determining a failure in connection with sending data between a first storage device and at least one other secondary storage device, deleting journal entries in each of said at least one other secondary storage device, and resending unsynchronized journal entries from the sender by sending a plurality of data packets all having a same sequence number lower than sequence numbers associated with other unsent packets and then sending any remaining data packets having a next higher sequence number, wherein data packets having the same sequence number are sent in an order that is independent of an order in which the data packets were created and wherein data packets having the same sequence number represent different data that is independently generated from a plurality of different sources,

wherein in response to the sequence number in the sender becoming equal to a second predetermined value different from the first predetermined value, acknowledging receipt of the blocks of data corresponding to the packets of data that are assigned the first predetermined value as the sequence number and sending the packets of data that are assigned the first predetermined value as the sequence number to said at least one other secondary storage device.

2. The method of claim 1, wherein said sender is a WAN blade coupled to said first storage device.

3. The method of claim 2, wherein when a failure is determined, journal entries in each of said secondary storage device are determined to be unsynchronized.

4. The method of claim 3, wherein a failure prevents a consistency group of storage devices from synchronizing data, said first storage device and said at least one other secondary storage device being included in said consistency group.

5. The method if claim 4, wherein, upon occurrence of a failure, the consistency group triggers so that the secondary storage devices are deemed not ready by the first storage device and any subsequent updates on the first storage device are indicated as invalid on the secondary storage devices.

6. The method of claim 5, wherein in response to said failure, data marked as invalid on the secondary devices is copied from the first to the other secondary storage devices as part of data restoration.

7. The method of claim 1, wherein said failure is a link failure occurring when at least one communication link fails.

8. The method of claim 7, further comprising:

detecting a link failure by failure of a linked device to response to a direct inquiry.

9. The method of claim 8, wherein said link failure is a failed link between two WAN blades connected by a network.

10. The method of claim 8, wherein said link failure is a failed link between a WAN blade coupled to a primary storage device.

11. The method of claim 10, further comprising:

in response to detecting said failed link, journaling writes to the WAN blade rather than the primary storage device, said WAN blade acting as a buffer to compensate for said failed link.

12. The method of claim 11, wherein, upon said WAN blade having a journal that overflows, said WAN blade not acknowledging write operations by the primary storage device.

13. The method of claim 12, further comprising:

clearing said journal on said WAN blade in response to said journal overflowing by sending messages to the primary storage device to invalidate previously written and acknowledged data.

14. The method of claim 1, wherein said acknowledging includes sending an acknowledgement to a host in the computer system sending data to the first storage device prior to said data actually being transferred to the at least one secondary storage device.

15. A computer program product for performing data recovery in a computer system, comprising:

machine executable code that sends data from a first storage device to at least one other secondary storage device, said data being sent in a plurality of data packets, each of said plurality of packets being associated with a sequence number having a first predetermined value;

machine executable code that, upon determining that the data has been successfully stored on all of said at least one other secondary storage device, deletes journal entries in a sender corresponding to said data;

machine executable code that, upon determining a failure in connection with sending data between a first storage device and at least one other secondary storage device, deletes journal entries in each of said at least one other secondary storage device, and resends unsynchronized journal entries from the sender by sending a plurality of data packets all having a same sequence number lower than sequence numbers associated with other unsent packets and then sending any remaining data packets having a next higher sequence number, wherein data packets having the same sequence number are sent in an order that is independent of an order in which the data packets were created and wherein data packets having the same sequence number represent different data that is independently generated from a plurality of different sources; and

machine executable code that, in response to the sequence number in the sender becoming equal to a second predetermined value different from the first predetermined value, acknowledges receipt of the blocks of data corresponding to the packets of data that are assigned the first predetermined value as the sequence number and sending the packets of data that are assigned the first predetermined value as the sequence number to said at least one other secondary storage device.

16. The computer program product of claim 15, wherein said sender is a WAN blade coupled to said first storage device.

17. The computer program product of claim 16, further comprising machine executable code that, when a failure is determined, determines journal entries in each of said secondary storage device to be unsynchronized.

18. The computer program product of claim 17, further comprising machine executable code that determines failure prevents a consistency group of storage devices from synchronizing data, said first storage device and said at least one other secondary storage device being included in said consistency group.

19. The computer program product claim 18, further comprising machine executable code that, upon occurrence of a failure, triggers a consistency group in that the secondary storage devices are deemed not ready by the first storage device and any subsequent updates on the first storage device are indicated as invalid on the secondary storage devices.

20. The computer program product of claim 19, further comprising machine executable code that, in response to said failure, marks data as invalid on the secondary devices and copies said data from the first storage device to the secondary storage devices as part of data restoration.

21. The computer program product of claim 15, further comprising machine executable code that determines said failure is a link failure occurring when at least one communication link fails.

22. The computer program product of claim 21, further comprising:

machine executable code that detects a link failure by failure of a linked device to response to a direct inquiry.

23. The computer program of claim 22, further comprising machine executable code that determines said link failure is a failed link between two WAN blades connected by a network.

24. The computer program product of claim 22, further comprising machine executable code that determines said link failure is a failed link between a WAN blade coupled to a primary storage device.

25. The computer program product of claim 24, further comprising:

machine executable code that, in response to detecting said failed link, journals writes to the WAN blade rather than the primary storage device, said WAN blade acting as a buffer to compensate for said failed link.

26. The computer program product of claim 25, further comprising machine executable code that, upon said WAN blade having a journal that overflows, causes said WAN blade not to acknowledge write operations by the first storage device.

27. The computer program product of claim 26, further comprising:

machine executable code that clears said journal on said WAN blade in response to said journal overflowing by sending messages to the primary storage device to invalidate previously written and acknowledged data.

28. The computer program product of claim 15, wherein said machine executable code that acknowledges includes machine executable code that sends an acknowledgement to a host in the computer system sending data to the first storage device prior to said data actually being transferred to the at least one secondary storage device.