ITPUX技术网

交流 . 资讯 . 分享
Make progress together!
Linux操作系统下Oracle11g R2 RAC 安装配置教程
Oracle数据库培训-备份恢复-性能优化-集群容灾
Oracle数据库DBA高级工程师培训视频

配置redis主从复制和sentinel模式

内容发布:luashin| 发布时间:2017-6-8 10:11:42
配置redis主从复制和sentinel模式

测试环境:
机器:192.168.110.132
redis主端口:6379
redis从端口:6380
redis从端口:6381
sentinel端口:26379
操作系统版本:CentOS release 6.5 (Final)
redis版本:3.2.6

linux系统安装redis:
1、下载redis:
登陆redis官网https://redis.io/download,下载最新稳定版源码包redis-3.2.6.tar.gz。解压后进行编译即可。
tar -zxvf redis-3.2.6.tar.gz -C /usr/local/redis && cd /usr/local/redis && make

2、编译配置文件redis.conf
redis.conf配置文件:
bind 0.0.0.0
protected-mode yes
port 6379
tcp-backlog 511
timeout 0
tcp-keepalive 300
daemonize yes
supervised no
pidfile /var/run/redis_6379.pid
loglevel notice
logfile "/tmp/redis6379.log"
databases 16
save 900 1
save 300 10
save 60 10000
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename dump6379.rdb
dir /data/redis/
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay 5
repl-disable-tcp-nodelay no
slave-priority 100
appendonly yes
appendfilename "appendonly6379.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage 100
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit 5000
slowlog-log-slower-than 10000
slowlog-max-len 128
latency-monitor-threshold 0
notify-keyspace-events ""
hash-max-ziplist-entries 512
hash-max-ziplist-value 64
list-max-ziplist-size -2
list-compress-depth 0
set-max-intset-entries 512
zset-max-ziplist-entries 128
zset-max-ziplist-value 64
hll-sparse-max-bytes 3000
activerehashing yes
client-output-buffer-limit normal 0 0 0
client-output-buffer-limit slave 256mb 64mb 60
client-output-buffer-limit pubsub 32mb 8mb 60
hz 10
aof-rewrite-incremental-fsync yes

redis6380.cnf配置文件:
bind 0.0.0.0
protected-mode yes
port 6380
tcp-backlog 511
timeout 0
tcp-keepalive 300
daemonize yes
supervised no
pidfile "/var/run/redis_6380.pid"
loglevel notice
logfile "/tmp/redis6380.log"
databases 16
save 900 1
save 300 10
save 60 10000
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename "dump6380.rdb"
dir "/data/redis"
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay 5
repl-disable-tcp-nodelay no
slave-priority 10
appendonly yes
appendfilename "appendonly6380.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage 100
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit 5000
slowlog-log-slower-than 10000
slowlog-max-len 128
latency-monitor-threshold 0
notify-keyspace-events ""
hash-max-ziplist-entries 512
hash-max-ziplist-value 64
list-max-ziplist-size -2
list-compress-depth 0
set-max-intset-entries 512
zset-max-ziplist-entries 128
zset-max-ziplist-value 64
hll-sparse-max-bytes 3000
activerehashing yes
client-output-buffer-limit normal 0 0 0
client-output-buffer-limit slave 256mb 64mb 60
client-output-buffer-limit pubsub 32mb 8mb 60
hz 10
aof-rewrite-incremental-fsync yes

redis6381.conf配置文件:
bind 0.0.0.0
protected-mode yes
port 6381
tcp-backlog 511
timeout 0
tcp-keepalive 300
daemonize yes
supervised no
pidfile "/var/run/redis_6381.pid"
loglevel notice
logfile "/tmp/redis6381.log"
databases 16
save 900 1
save 300 10
save 60 10000
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename "dump6381.rdb"
dir "/data/redis"
slaveof 192.168.110.132 6380
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay 5
repl-disable-tcp-nodelay no
slave-priority 100
appendonly yes
appendfilename "appendonly6381.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage 100
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit 5000
slowlog-log-slower-than 10000
slowlog-max-len 128
latency-monitor-threshold 0
notify-keyspace-events ""
hash-max-ziplist-entries 512
hash-max-ziplist-value 64
list-max-ziplist-size -2
list-compress-depth 0
set-max-intset-entries 512
zset-max-ziplist-entries 128
zset-max-ziplist-value 64
hll-sparse-max-bytes 3000
activerehashing yes
client-output-buffer-limit normal 0 0 0
client-output-buffer-limit slave 256mb 64mb 60
client-output-buffer-limit pubsub 32mb 8mb 60
hz 10
aof-rewrite-incremental-fsync yes

sentinel配置文件:
bind 0.0.0.0
protected-mode yes
port 26379
dir /tmp
sentinel monitor mymaster 192.168.110.132 6379 1
sentinel down-after-milliseconds mymaster 10000
sentinel parallel-syncs mymaster 1
sentinel failover-timeout mymaster 18000

3、启动redis。
    启动redis十分简单,直接redis-server /path/redis.conf即可。
    redis-server /usr/local/redis/redis.conf
    redis-server /usr/local/redis/redis6380.conf
    redis-server /usr/local/redis/redis6381.conf

4、检查redis主从状态:
6379端口:
127.0.0.1:6379> set title helloOK
127.0.0.1:6379> get title"hello"
127.0.0.1:6379> info Replication# Replicationrole:masterconnected_slaves:1slave0:ip=192.168.110.132,port=6380,state=online,offset=1,lag=0master_repl_offset:1repl_backlog_active:1repl_backlog_size:1048576repl_backlog_first_byte_offset:2repl_backlog_histlen:0

6380端口:
127.0.0.1:6380> get title"hello"
127.0.0.1:6380> info Replication# Replicationrole:slavemaster_host:192.168.110.132master_port:6379master_link_status:upmaster_last_io_seconds_ago:0master_sync_in_progress:0slave_repl_offset:2115slave_priority:10slave_read_only:1connected_slaves:0master_repl_offset:0repl_backlog_active:0repl_backlog_size:1048576repl_backlog_first_byte_offset:0repl_backlog_histlen:0

6381端口:
127.0.0.1:6381> get title"hello"
127.0.0.1:6381> info Replication# Replicationrole:slavemaster_host:192.168.110.132master_port:6379master_link_status:upmaster_last_io_seconds_ago:2master_sync_in_progress:0slave_repl_offset:5970slave_priority:100slave_read_only:1connected_slaves:0master_repl_offset:0repl_backlog_active:0repl_backlog_size:1048576repl_backlog_first_byte_offset:0repl_backlog_histlen:0

5、启动sentinel:
    redis-server /usr/local/redis/sentinel.conf --sentinel

6、杀掉redis的6379端口进程,观察failover
    46030:X 13 Apr 10:15:46.085 * Increased maximum number of open files to 10032 (it was originally set to 1024).
                _._                                                
           _.-``__ ''-._                                            
      _.-``    `.  `_.  ''-._           Redis 3.2.6 (00000000/0) 64 bit
  .-`` .-```.  ```\/    _.,_ ''-._                                 
(    '      ,       .-`  | `,    )     Running in sentinel mode
|`-._`-...-` __...-.``-._|'` _.-'|     Port: 26379
|    `-._   `._    /     _.-'    |     PID: 46030
  `-._    `-._  `-./  _.-'    _.-'                                 
|`-._`-._    `-.__.-'    _.-'_.-'|                                 
|    `-._`-._        _.-'_.-'    |           http://redis.io      
  `-._    `-._`-.__.-'_.-'    _.-'                                 
|`-._`-._    `-.__.-'    _.-'_.-'|                                 
|    `-._`-._        _.-'_.-'    |                                 
  `-._    `-._`-.__.-'_.-'    _.-'                                 
      `-._    `-.__.-'    _.-'                                      
          `-._        _.-'                                          
              `-.__.-'                                             
46030:X 13 Apr 10:15:46.086 # WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.
46030:X 13 Apr 10:15:46.097 # Sentinel ID is 89f3e9eeacc7b99b9c95d7c20d25a241d6c1ccd9
46030:X 13 Apr 10:15:46.097 # +monitor master mymaster 192.168.110.132 6379 quorum 1
46030:X 13 Apr 10:15:46.098 * +slave slave 192.168.110.132:6380 192.168.110.132 6380 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:16:36.359 * +slave slave 192.168.110.132:6381 192.168.110.132 6381 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.344 # +sdown master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.344 # +odown master mymaster 192.168.110.132 6379 #quorum 1/1
46030:X 13 Apr 10:19:37.344 # +new-epoch 1
46030:X 13 Apr 10:19:37.344 # +try-failover master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.348 # +vote-for-leader 89f3e9eeacc7b99b9c95d7c20d25a241d6c1ccd9 1
46030:X 13 Apr 10:19:37.348 # +elected-leader master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.348 # +failover-state-select-slave master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.415 # +selected-slave slave 192.168.110.132:6380 192.168.110.132 6380 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.415 * +failover-state-send-slaveof-noone slave 192.168.110.132:6380 192.168.110.132 6380 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.477 * +failover-state-wait-promotion slave 192.168.110.132:6380 192.168.110.132 6380 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.983 # +promoted-slave slave 192.168.110.132:6380 192.168.110.132 6380 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:37.983 # +failover-state-reconf-slaves master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:38.051 * +slave-reconf-sent slave 192.168.110.132:6381 192.168.110.132 6381 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:38.194 * +slave-reconf-inprog slave 192.168.110.132:6381 192.168.110.132 6381 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:39.217 * +slave-reconf-done slave 192.168.110.132:6381 192.168.110.132 6381 @ mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:39.299 # +failover-end master mymaster 192.168.110.132 6379
46030:X 13 Apr 10:19:39.299 # +switch-master mymaster 192.168.110.132 6379 192.168.110.132 6380
46030:X 13 Apr 10:19:39.299 * +slave slave 192.168.110.132:6381 192.168.110.132 6381 @ mymaster 192.168.110.132 6380
46030:X 13 Apr 10:19:39.299 * +slave slave 192.168.110.132:6379 192.168.110.132 6379 @ mymaster 192.168.110.132 6380
46030:X 13 Apr 10:19:49.319 # +sdown slave 192.168.110.132:6379 192.168.110.132 6379 @ mymaster 192.168.110.132 6380
由于6380的配置文件中slave-priority配置为10,比6381的slave-priority的100要小,故优先选择6380为master。因为sentinel配置文件中down-after-milliseconds设置为10s,所以failover在10s后确认master down后自动选举6380为新的master。

7、查看日志:
当选举6380为新的master的时候,原先6381会重新指向6380,并清理自身旧的数据,从6380重新sync全量的数据过来。
6380的日志:
45995:S 13 Apr 10:19:37.342 # Error condition on socket for SYNC: Connection refused
45995:M 13 Apr 10:19:37.477 * Discarding previously cached master state.
45995:M 13 Apr 10:19:37.478 * MASTER MODE enabled (user request from 'id=4 addr=192.168.110.132:46885 fd=7 name=sentinel-89f3e9ee-cmd a
ge=231 idle=0 flags=x db=0 sub=0 psub=0 multi=3 qbuf=0 qbuf-free=32768 obl=36 oll=0 omem=0 events=r cmd=exec')
45995:M 13 Apr 10:19:37.478 # CONFIG REWRITE executed with success.
45995:M 13 Apr 10:19:38.755 * Slave 192.168.110.132:6381 asks for synchronization
45995:M 13 Apr 10:19:38.755 * Full resync requested by slave 192.168.110.132:6381
45995:M 13 Apr 10:19:38.755 * Starting BGSAVE for SYNC with target: disk
45995:M 13 Apr 10:19:38.756 * Background saving started by pid 46050
46050:C 13 Apr 10:19:38.766 * DB saved on disk
46050:C 13 Apr 10:19:38.767 * RDB: 0 MB of memory used by copy-on-write
45995:M 13 Apr 10:19:38.855 * Background saving terminated with success
45995:M 13 Apr 10:19:38.855 * Synchronization with slave 192.168.110.132:6381 succeeded

6381的日志:
46039:S 13 Apr 10:19:37.746 # Error condition on socket for SYNC: Connection refused
46039:S 13 Apr 10:19:38.051 * Discarding previously cached master state.
46039:S 13 Apr 10:19:38.051 * SLAVE OF 192.168.110.132:6380 enabled (user request from 'id=4 addr=192.168.110.132:41739 fd=7 name=senti
nel-89f3e9ee-cmd age=182 idle=0 flags=x db=0 sub=0 psub=0 multi=3 qbuf=145 qbuf-free=32623 obl=36 oll=0 omem=0 events=r cmd=exec')
46039:S 13 Apr 10:19:38.052 # CONFIG REWRITE executed with success.
46039:S 13 Apr 10:19:38.755 * Connecting to MASTER 192.168.110.132:6380
46039:S 13 Apr 10:19:38.755 * MASTER <-> SLAVE sync started
46039:S 13 Apr 10:19:38.755 * Non blocking connect for SYNC fired the event.
46039:S 13 Apr 10:19:38.755 * Master replied to PING, replication can continue...
46039:S 13 Apr 10:19:38.755 * Partial resynchronization not possible (no cached master)
46039:S 13 Apr 10:19:38.756 * Full resync from master: 065cade6f55c04c839436d5d0d2c6abc56a5d862:1
46039:S 13 Apr 10:19:38.855 * MASTER <-> SLAVE sync: receiving 102 bytes from master
46039:S 13 Apr 10:19:38.855 * MASTER <-> SLAVE sync: Flushing old data
46039:S 13 Apr 10:19:38.855 * MASTER <-> SLAVE sync: Loading DB in memory
46039:S 13 Apr 10:19:38.855 * MASTER <-> SLAVE sync: Finished with success
46039:S 13 Apr 10:19:38.856 * Background append only file rewriting started by pid 46051
46039:S 13 Apr 10:19:38.897 * AOF rewrite child asks to stop sending diffs.
46051:C 13 Apr 10:19:38.897 * Parent agreed to stop sending diffs. Finalizing AOF...
46051:C 13 Apr 10:19:38.897 * Concatenating 0.00 MB of AOF diff received from parent.
46051:C 13 Apr 10:19:38.897 * SYNC append only file rewrite performed
46051:C 13 Apr 10:19:38.897 * AOF rewrite: 0 MB of memory used by copy-on-write
46039:S 13 Apr 10:19:38.956 * Background AOF rewrite terminated with success
46039:S 13 Apr 10:19:38.956 * Residual parent diff successfully flushed to the rewritten AOF (0.00 MB)
46039:S 13 Apr 10:19:38.956 * Background AOF rewrite finished successfully
46039:S 13 Apr 10:31:31.097 * 1 changes in 900 seconds. Saving...
46039:S 13 Apr 10:31:31.097 * Background saving started by pid 46071
46071:C 13 Apr 10:31:31.109 * DB saved on disk
46071:C 13 Apr 10:31:31.109 * RDB: 0 MB of memory used by copy-on-write
46039:S 13 Apr 10:31:31.198 * Background saving terminated with success



上一篇:Linux环境下redis安装与配置
下一篇:Redis高可用方案(redis主从+keepalived+sentinel)
回复

使用道具 举报

1框架
您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

快速回复 返回顶部 返回列表