• 集群節(jié)點(diǎn)信息
• 這時(shí)候停掉一個(gè)master 172.30.60.31
  • 從集群信息集中我們可以看到172.30.60.31的slave是172.30.60.41，查看41的日志，發(fā)現(xiàn)他成為了新的master
• 這時(shí)候我們?cè)趯?72.30.60.41也殺死，會(huì)發(fā)現(xiàn)集群異常了
  • 嘗試把172.30.60.31啟動(dòng)，發(fā)現(xiàn)集群還是異常的，因?yàn)槁?lián)系不到他的master，查看31的日志
  • 我們嘗試手動(dòng)把172.30.60.31升級(jí)為master
  • 因?yàn)槁?lián)系不到master，直接強(qiáng)制升級(jí)
  • 查看集群狀態(tài)，發(fā)現(xiàn)恢復(fù)正常了
• 當(dāng)集群中一個(gè)master和他的slave都掛了的情況下(因測(cè)試主從發(fā)生了變化，不用在意)，集群異常
  • 假如這兩個(gè)節(jié)點(diǎn)都無(wú)法啟動(dòng)了，這時(shí)候應(yīng)該怎么辦？嘗試把172.30.60.41從節(jié)點(diǎn)升級(jí)為主
  • 首先把172.30.60.41設(shè)置為172.30.60.31的從節(jié)點(diǎn):?

    redis-cli -p 7000 -h 172.30.60.41 cluster replicate 714bae586a45160a4d6869e4547b4aca0361fd0e

  • 
  • 然后升級(jí)172.30.60.41升級(jí)為主節(jié)點(diǎn)，然后查看集群狀態(tài),恢復(fù)正常了
• 也可以新增一個(gè)節(jié)點(diǎn)到集群中 172.30.100.114 7777

redis-cli -p 7000 cluster meet 172.30.100.114 7000

# 因?yàn)閽斓舻膍aster slots是5461-10922，所以這里指定5461

redis-cli -p 7777 -h 172.30.100.114?CLUSTER ADDSLOTS 5461

redis-cli -p 7777 -h 172.30.100.114

# 強(qiáng)制升級(jí)為主節(jié)點(diǎn)

cluster failover force

# 查看集群狀態(tài)

172.30.100.114:7777> cluster nodes

8104a927a687e932b906855a83c8ffee7d746798 172.30.100.114:7777@17777 myself,master - 0 1739526224000 33 connected 5461-10922

8a35a0bdfa37946b2cf9e5b1813360451f004b81 172.30.60.61:7000@17000 slave,fail - 1739525460865 0 0 disconnected

3a84c93ed5bcbf67a1ec746501e6574c736e525d 172.30.60.21:7000@17000 master - 0 1739526225404 17 connected 10923-16383

7fa07aae1f0ecaeee805a578cadd52184f6214a3 172.30.60.11:7000@17000 master - 0 1739526227000 13 connected 0-5460

f198e3277009f6dbbf0899bd26c9954cc15b977c 172.30.60.51:7000@17000 slave 3a84c93ed5bcbf67a1ec746501e6574c736e525d 0 1739526227416 17 connected

714bae586a45160a4d6869e4547b4aca0361fd0e 172.30.60.31:7000@17000 master,fail - 1739525459356 0 32 disconnected

e509a47a20eedc0e9a9a417d203dbf3161ce8041 172.30.60.41:7000@17000 slave 7fa07aae1f0ecaeee805a578cadd52184f6214a3 0 1739526225000 13 connected

172.30.100.114:7777> cluster info

cluster_state:ok

cluster_slots_assigned:16384

cluster_slots_ok:16384

cluster_slots_pfail:0

cluster_slots_fail:0

cluster_known_nodes:7

cluster_size:3

cluster_current_epoch:33

cluster_my_epoch:33

cluster_stats_messages_ping_sent:771

cluster_stats_messages_pong_sent:764

cluster_stats_messages_auth-req_sent:6

cluster_stats_messages_sent:1541

cluster_stats_messages_ping_received:759

cluster_stats_messages_pong_received:769

cluster_stats_messages_meet_received:1

cluster_stats_messages_auth-ack_received:2

cluster_stats_messages_update_received:1

cluster_stats_messages_received:1532