Redis master-slave replication principle

I do not like too much text (too faint), directly

First synchronization process (full sync):

650) this.width=650; "src=" Http://s3.51cto.com/wyfs02/M01/6F/B7/wKioL1WmTEiiBXrGAAFzSHHtTOA173.jpg "title=" 2a5d24eb-042c-44f0-911d-152ff179ca3c.png "alt=" Wkiol1wmteiibxrgaafzshhttoa173.jpg "/>

Synchronization during the run (partial resynchronization), no partial resynchronization prior to version 2.8:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M00/6F/B7/wKioL1WmTNCRKq9GAAHSL9fze8s533.jpg "style=" float: none; "title=" 2.png "alt=" Wkiol1wmtncrkq9gaahsl9fze8s533.jpg "/>

During Redis operation, partial resynchronization is interrupted, the whole process is as follows, introducing Runid, offset:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/6F/B7/wKioL1WmTNGgM1uEAALnl4fIid4250.jpg "style=" float: none; "title=" 3.png "alt=" Wkiol1wmtnggm1ueaalnl4fiid4250.jpg "/>

The default size of the replication backlog buffer is 1M, which can be customized as required. Algorithm: Second * write_size_per_second,

Second: Average time of master/slave disconnection;

Write_size_per_second: Refers to the command data generated by the master-slave average per second.

Runid: Consists of 40 random 16-based characters

Redis Master-slave replication principle