FAILURE RECOVERY IN AN ASYNCHRONOUS REMOTE MIRRORING PROCESS

摘要

A method for generating a remote replicate of a logical entity, the method may include: calculating, by a local site, a replication cycle content that is associated with a replication cycle; wherein the replication cycle content represents differences in contents of logical entity portions between a point in time associated with the replication cycle and a point in time associated with an adjacent replication cycle that preceded the replication cycle; attempting to transmit messages that include the replication cycle content to a remote site and according to a predefined order that is responsive to addresses of the logical entity portions; monitoring a successful reception of the messages that include the replication cycle content by the remote site; wherein when finding that a first part of the replication cycle content was successfully received by the remote site and a second part of the replication cycle content was not successfully received by the remote site due to the failure then transmitting, after recovering from the failure, messages that include the second part; wherein the second part is detected using (a) at least one replication cycle identifier that identifies the replication cycle, and (b) at least one address of at least one logical entity portion.

主权项

1. A method for generating a remote replicate of a logical entity, the method comprises:
calculating, by a local site, a replication cycle content that is associated with a replication cycle; wherein the replication cycle content represents differences in contents of logical entity portions between a point in time associated with the replication cycle and a point in time associated with an adjacent replication cycle that preceded the replication cycle; attempting to transmit messages that include the replication cycle content to a remote site and according to a predefined order that is responsive to addresses of the logical entity portions; monitoring a successful reception of the messages that include the replication cycle content by the remote site; and wherein when finding that a first part of the replication cycle content was successfully received by the remote site and a second part of the replication cycle content was not successfully received by the remote site due to the failure then transmitting, after recovering from the failure, messages that include the second part; wherein the second part is detected using (a) at least one replication cycle identifier that identifies the replication cycle, and (b) at least one address of at least one logical entity portion.