Granted Patent B2

US 12,436,707 · App. 18/421,649 · Granted Oct 7, 2025

Inventors: Krishna Murthy Chandraiah Setty Narasingarayanapeta (Bangalore, IN); Akhil Kaushik (Karnataka, IN); Nagaraj Lalsangi (Cary, NC)

Assignee: NetApp, Inc.

G06F3/065G06F3/0613G06F3/0619G06F3/0631G06F3/0653G06F3/0659G06F3/067

Abstract

Claims (45)

1. A computer implemented method performed by one or more processing resources of a distributed storage system, the method comprising:

initiating a resynchronization process due to a failure or loss of connectivity of a storage object of a primary storage cluster with the storage object becoming out of sync as an internal state while maintaining an external state as in sync for external entities in order to provide time for reestablishing synchronous replication between the storage object and a replicated storage object of a secondary storage cluster within duration of an operation (Op) timeout period; and

performing the resynchronization process between the storage object and the replicated storage object based on using a first Op log of the primary storage cluster and a second Op log of the secondary storage cluster.

2. The computer implemented method of claim 1 , wherein the resynchronization process is based on reestablishing a Sync Data Path between a data copy of the storage object and a mirrored data copy of the replicated storage object.

3. The computer implemented method of claim 1 , wherein the resynchronization resumes zero recovery point objective (RPO) protection.

4. The computer implemented method of claim 1 , further comprising:

initiating persistent inflight tracking and reconciliation of I/O operations based on the first Op log of the primary storage cluster and the second Op log of the secondary storage cluster;

replaying the I/O operations from one of the first and second Op logs; and

reconciling the I/O operations in the first Op log with the I/O operations in the second Op log by determining whether an Op being considered is present or absent in both of the first and second Op logs and skipping the Op if the Op is present or absent in both of the first and second Op logs from the primary and secondary storage clusters.

5. The computer implemented method of claim 1 , further comprising:

transitioning a consistency group (CG) for a node of the storage object to out of sync (OOS) when the resynchronization experiences an error or failure resulting in expiration of the Op timeout period.

6. The computer implemented method of claim 1 , wherein the I/O operations for the storage object are temporarily disallowed until resumption of synchronous replication and this avoids dependent write order inconsistencies.

7. A non-transitory computer-readable storage medium embodying a set of instructions, which when executed by one or more processing resources of a multi-site distributed storage system cause the one or more processing resources to:

initiate a resynchronization process due to a failure or loss of connectivity of a storage object of a node of a primary storage cluster with the storage object of the node becoming out of sync as an internal state while maintaining an external state as in sync for external entities in order to provide time for reestablishing synchronous replication between the storage object and a replicated storage object of a secondary storage cluster within duration of an operation (Op) timeout period; and

perform the resynchronization process between the storage object and the replicated storage object based on using a first Op log of the primary storage cluster and a second Op log of the secondary storage cluster.

8. The non-transitory computer-readable storage medium of claim 7 , wherein the resynchronization is based on reestablishing a Sync Data Path between a data copy of the storage object and a mirrored data copy of the replicated storage object.

9. The non-transitory computer-readable storage medium of claim 7 , wherein the resynchronization resumes zero recovery point objective (RPO) protection.

10. The non-transitory computer-readable storage medium of claim 7 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

initiating persistent inflight tracking and reconciliation of I/O operations based on the first Op log of the primary storage cluster and the second Op log of the secondary storage cluster;

replaying the I/O operations from one of the first and second Op logs; and

reconciling the I/O operations in the first Op log with the I/O operations in the second Op log by determining whether an Op being considered is present or absent in both of the first and second Op logs and skipping the Op if the Op is present or absent in both of the first and second Op logs from the primary and secondary storage clusters.

11. The non-transitory computer-readable storage medium of claim 7 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

transition a consistency group (CG) for the node of the storage object to out of sync (OOS) state when the resynchronization experiences an error or failure resulting in expiration of the Op timeout period.

12. The non-transitory computer-readable storage medium of claim 7 , wherein the input/output (I/O) operations for the storage object are temporarily disallowed until resumption of synchronous replication and this avoids dependent write order inconsistencies.

13. A multi-site distributed storage system having a primary storage site with a primary storage cluster and a secondary storage site with a secondary storage cluster, comprising:

one or more processing resources; and

a non-transitory computer-readable medium coupled to the one or more processing resources, having stored therein instructions, which when executed by the one or more processing resources cause the one or more processing resources to:

maintain a first operation (Op) Log for a first volume and a second Op Log for a second volume to be capable of facilitating resynchronization between a dataset stored on the first volume of the primary storage cluster of the distributed storage system and a mirror copy of the dataset stored on the second volume of the secondary storage cluster of the distributed storage system;

receive an operation associated with a file within the dataset;

determine whether the operation is a modifying operation; and

create an operation (Op) log entry with one of the first and second Op Logs if the operation is a modifying operation to cause a modification of the file.

14. The multi-site distributed storage system of claim 13 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

determine whether a storage node that is responsible for executing the operation is associated with the primary storage cluster; and

determine whether the operation is a write operation or not, wherein the Op log entry contains a sequence number assigned to the operation, and wherein the Op log entry is not required to contain data associated with the operation when the operation comprises a write operation.

15. The multi-site distributed storage system of claim 14 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

store a sequence number, a file identifier, and an Op range including offset and length for the operation when the operation is a write operation, wherein pay load data of the write operation is not stored.

16. The multi-site distributed storage system of claim 14 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

store data for an Op log entry including a sequence number, a file identifier, an Op range including offset and length, and pay load data of the operation when the operation is not a write Op and the storage node that is responsible for executing the operation is associated with the primary storage cluster.

17. The multi-site distributed storage system of claim 14 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

store data for an Op log entry including a sequence number and no pay load data when the operation is not a write operation and the storage node that is responsible for executing the operation is associated with the secondary storage cluster.

18. The multi-site distributed storage system of claim 13 , wherein different data is stored in a header of the Op Log entry for metadata operations, deallocate space operations, and write operations.

19. The multi-site distributed storage system of claim 13 , wherein the entry is populated during a modify phase of an Op handler.

20. The multi-site distributed storage system of claim 13 , wherein the instructions when executed by the one or more processing resources cause the one or more processing resources to:

execute a write operation on the mirror copy by reading data from the file; or

undoing, by the secondary storage cluster, the write operation on the mirror copy by replacing the data stored as a result of prior execution of the write operation on mirror copy with old data read from the file.

Priority Claims (2)

Continuity (2)

References Cited (139)