The Cluster-require-full-coverage of Redis-cluster

In the configuration of a Redis cluster, there is a configuration called

Cluster-require-full-coverage

Here is a detailed explanation for this configuration.

In this paper, the cluster environment of three nodes is demonstrated, and the clusters with more than three nodes need to be demonstrated again.

When Cluster-require-full-coverage is no, it indicates that the cluster is still available when the main library responsible for a slot is offline and there is no corresponding recovery from the library, as demonstrated in the following scenario.

1. View the configuration

650) this.width=650; "Src=" Http://s5.51cto.com/wyfs02/M02/8A/04/wKioL1gkJcSSyhE9AACG8fJ1hn0814.png-wh_500x0-wm_3 -wmp_4-s_1499807311.png "style=" Float:none; "title=" Configuration View "alt=" Wkiol1gkjcssyhe9aacg8fj1hn0814.png-wh_50 "/>

2. All downtime from the vault, no failure recovery, the cluster is still in service.

650) this.width=650; "Src=" Http://s3.51cto.com/wyfs02/M01/8A/08/wKiom1gkJciymp3kAAUMxot9NKo812.png-wh_500x0-wm_3 -wmp_4-s_1030923486.png "style=" Float:none; "title=" failed to recover the node "alt=" Wkiom1gkjciymp3kaaumxot9nko812.png-wh_50 "/ >

3. One of the primary nodes is down, this is the normal service for the fault primary node, when the key value is in the slot of the failed node, the normal operation cannot be performed, and the primary node operation of the normal service is required after exiting the heavy connection.

650) this.width=650; "Src=" Http://s2.51cto.com/wyfs02/M00/8A/08/wKiom1gkJcrxl5KqAAYku6r-PpE919.png-wh_500x0-wm_3 -wmp_4-s_1982909918.png "style=" Float:none "title=" a Main library outage "alt=" wkiom1gkjcrxl5kqaayku6r-ppe919.png-wh_50 "/>

4. When two primary nodes are down, the cluster stops the service.

650) this.width=650; "Src=" Http://s5.51cto.com/wyfs02/M01/8A/04/wKioL1gkJc2zDmTkAAUtcJAhfdU950.png-wh_500x0-wm_3 -wmp_4-s_841481751.png "style=" Float:none; "title=" Another main library down "alt=" wkiol1gkjc2zdmtkaautcjahfdu950.png-wh_50 "/>

When Cluster-require-full-coverage is yes, the cluster is not available when the main library that is responsible for a slot is offline and there is no corresponding recovery from the library, the case is demonstrated below.

1. Check the configuration for Yes.

650) this.width=650; "Src=" Http://s4.51cto.com/wyfs02/M01/8A/08/wKiom1gkK66BW1ImAAB5vivVus0510.png-wh_500x0-wm_3 -wmp_4-s_788112457.png "style=" Float:none; "title=" 5.png "alt=" Wkiom1gkk66bw1imaab5vivvus0510.png-wh_50 "/>

2. There is a primary node outage is the failure of the node to recover, the entire cluster is not available at this time.

650) this.width=650; "Src=" Http://s1.51cto.com/wyfs02/M01/8A/04/wKioL1gkK83xOjMaAAST6H4z6Ek657.png-wh_500x0-wm_3 -wmp_4-s_3691989319.png "title=" 6.png "alt=" Wkiol1gkk83xojmaaast6h4z6ek657.png-wh_50 "/>

This article is from the "Floating Phoenix Year" blog, please be sure to keep this source http://liuzhanbin.blog.51cto.com/10060150/1871514

The Cluster-require-full-coverage of Redis-cluster