Resilient implementation of client file operations and replication

A computing device comprising:

• a memory comprising machine executable code;
• and
• a processor coupled to the memory, the processor configured to execute the machine executable code to cause the processor to:
• assign a sequence number to a storage operation received from a client;
• insert the sequence number into a queue for pending storage operations that are to be locally implemented by a first node upon first storage and remotely replicated to a second node for implementation upon second storage based upon a replication relationship between the first storage and the second storage;
• generate a replication operation that is a replica of the storage operation;
• assign the sequence number to the replication operation;
• locally implementing, by the first node, the storage operation upon the first storage according to the sequence number;
• send the replication operation to the second node for remote implementation upon the second storage according to the sequence number;
• in response to not receiving an acknowledgement from the second node for the replication operation, generate and send a retry replication operation that is a retry of the replication operation, wherein the retry replication operation is assigned the sequence number that is also assigned to the storage operation and the replication operation;
• and
• in response to a protocol timer timing out for the sequence number, flush the queue and taking a synchronous replication relationship between the first node and the second node out of sync.

A non-transitory machine readable medium comprising instructions for performing a method, which when executed by a machine, causes the machine to:

• assign a sequence number to a storage operation received from a client;
• insert the sequence number into a queue for pending storage operations that are to be locally implemented by a first node upon first storage and remotely replicated to a second node for implementation upon second storage based upon a replication relationship between the first storage and the second storage;
• generate a replication operation that is a replica of the storage operation;
• assign the sequence number to the replication operation;
• locally implementing, by the first node, the storage operation upon the first storage according to the sequence number;
• send the replication operation to the second node for remote implementation upon the second storage according to the sequence number;
• in response to not receiving an acknowledgement from the second node for the replication operation, generate and send a retry replication operation that is a retry of the replication operation, wherein the retry replication operation is assigned the sequence number that is also assigned to the storage operation and the replication operation;
• and
• in response to a protocol timer timing out for the sequence number, flush the queue and taking a synchronous replication relationship between the first node and the second node out of sync.