8.7 Redis集群之故障自动转移

已于 2023-02-01 23:20:58 修改
阅读量283 收藏
点赞数
分类专栏： 从头开始学Redis 文章标签： redis java 数据库
于 2023-02-01 23:15:44 首次发布
未经博主同意，禁止转载。
本文链接：https://blog.csdn.net/Jgx1214/article/details/128841961
版权

1.集群节点信息
在这里插入图片描述
6372是6375的从节点。

2.挂掉6375，观察6372是否会晋升为主节点继续提供服务

ps -ef | grep redis
kill -9 5928

在这里插入图片描述
在这里插入图片描述
在这里插入图片描述
从上图中可以看到，6372晋升为主节点，并正常提供服务。

3.6372节点日志变化

#主节点下线
22398:S 01 Feb 2023 22:50:04.149 # Connection with master lost.
22398:S 01 Feb 2023 22:50:04.149 * Caching the disconnected master state.
#重连主节点start(超时时间在配置文件中)
22398:S 01 Feb 2023 22:50:04.149 * Reconnecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:04.149 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:04.149 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:04.779 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:04.779 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:04.779 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:05.784 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:05.784 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:05.784 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:06.789 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:06.789 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:06.789 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:07.794 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:07.794 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:07.794 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:08.799 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:08.799 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:08.799 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:09.804 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:09.804 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:09.805 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:10.809 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:10.809 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:10.810 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:11.814 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:11.814 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:11.815 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:12.820 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:12.820 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:12.820 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:13.825 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:13.825 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:13.825 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:14.830 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:14.830 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:14.831 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:15.836 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:15.836 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:15.838 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:16.842 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:16.842 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:16.842 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:17.847 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:17.847 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:17.847 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:18.852 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:18.852 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:18.852 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:19.857 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:19.858 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:19.858 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:20.863 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:20.863 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:20.865 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:21.871 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:21.871 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:21.871 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:22.876 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:22.876 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:22.876 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:23.880 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:23.880 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:23.881 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:24.885 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:24.885 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:24.886 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:25.890 * FAIL message received from b8010c02266eb3a4fb23e2724b2a334f34722e30 about e99c4338f171bd3ee45d7597aa4336ff2a5dc41f
22398:S 01 Feb 2023 22:50:25.890 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:25.891 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:25.891 # Start of election delayed for 985 milliseconds (rank #0, offset 49342).
22398:S 01 Feb 2023 22:50:25.891 # Cluster state changed: fail
22398:S 01 Feb 2023 22:50:25.891 # Error condition on socket for SYNC: Connection refused
22398:S 01 Feb 2023 22:50:26.896 * Connecting to MASTER 121.4.191.137:6375
22398:S 01 Feb 2023 22:50:26.896 * MASTER <-> REPLICA sync started
22398:S 01 Feb 2023 22:50:26.896 # Starting a failover election for epoch 9.
22398:S 01 Feb 2023 22:50:26.900 # Error condition on socket for SYNC: Connection refused
#重连主节点end
#从节点成为新的master
22398:S 01 Feb 2023 22:50:26.904 # Failover election won: I'm the new master.
22398:S 01 Feb 2023 22:50:26.904 # configEpoch set to 9 after successful failover
22398:M 01 Feb 2023 22:50:26.904 * Discarding previously cached master state.
22398:M 01 Feb 2023 22:50:26.904 # Setting secondary replication ID to db262f1317558dc051af4f52f0e50faa73bff022, valid up to offset: 49343. New replication ID is d24235ee6476b3ef0d675f45a63656cd1a1155a7
22398:M 01 Feb 2023 22:50:26.904 # Cluster state changed: ok

4.重启6375节点，观察集群节点信息

cd /usr/local/redis/src
redis-server  /usr/local/redis/cluster/conf/redis-6375.conf
ps -ef | grep redis

在这里插入图片描述
在这里插入图片描述
可以看到，6375成为6372的从节点。

5.6375节点日志变化

30013:C 01 Feb 2023 23:09:45.332 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo
30013:C 01 Feb 2023 23:09:45.332 # Redis version=6.2.6, bits=64, commit=00000000, modified=0, pid=30013, just started
30013:C 01 Feb 2023 23:09:45.332 # Configuration loaded
30013:M 01 Feb 2023 23:09:45.333 * monotonic clock: POSIX clock_gettime
30013:M 01 Feb 2023 23:09:45.334 * Node configuration loaded, I'm e99c4338f171bd3ee45d7597aa4336ff2a5dc41f
30013:M 01 Feb 2023 23:09:45.334 * Running mode=cluster, port=6375.
30013:M 01 Feb 2023 23:09:45.334 # WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.
30013:M 01 Feb 2023 23:09:45.334 # Server initialized
30013:M 01 Feb 2023 23:09:45.334 # WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.
30013:M 01 Feb 2023 23:09:45.335 * DB loaded from append only file: 0.000 seconds
30013:M 01 Feb 2023 23:09:45.335 * Ready to accept connections
30013:M 01 Feb 2023 23:09:45.336 # Configuration change detected. Reconfiguring myself as a replica of 58ba15be743b298766cafea04646ea9bfbfe3389
30013:S 01 Feb 2023 23:09:45.336 * Before turning into a replica, using my own master parameters to synthesize a cached master: I may be able to synchronize with the new master with just a partial transfer.
#连接到主节点6372
30013:S 01 Feb 2023 23:09:45.336 * Connecting to MASTER 121.4.211.73:6372
30013:S 01 Feb 2023 23:09:45.336 * MASTER <-> REPLICA sync started
30013:S 01 Feb 2023 23:09:45.336 # Cluster state changed: ok
30013:S 01 Feb 2023 23:09:45.340 * Non blocking connect for SYNC fired the event.
30013:S 01 Feb 2023 23:09:45.340 * Master replied to PING, replication can continue...
30013:S 01 Feb 2023 23:09:45.341 * Trying a partial resynchronization (request ffb52f67e20e4d272cc09de8318d5d834a93293c:1).
30013:S 01 Feb 2023 23:09:45.341 * Full resync from master: d24235ee6476b3ef0d675f45a63656cd1a1155a7:49342
30013:S 01 Feb 2023 23:09:45.341 * Discarding previously cached master state.
30013:S 01 Feb 2023 23:09:45.365 * MASTER <-> REPLICA sync: receiving 200 bytes from master to disk
30013:S 01 Feb 2023 23:09:45.365 * MASTER <-> REPLICA sync: Flushing old data
30013:S 01 Feb 2023 23:09:45.365 * MASTER <-> REPLICA sync: Loading DB in memory
30013:S 01 Feb 2023 23:09:45.371 * Loading RDB produced by version 6.2.6
30013:S 01 Feb 2023 23:09:45.371 * RDB age 0 seconds
30013:S 01 Feb 2023 23:09:45.371 * RDB memory usage when created 2.57 Mb
30013:S 01 Feb 2023 23:09:45.372 # Done loading RDB, keys loaded: 1, keys expired: 0.
#完成复制
30013:S 01 Feb 2023 23:09:45.372 * MASTER <-> REPLICA sync: Finished with success
30013:S 01 Feb 2023 23:09:45.372 * Background append only file rewriting started by pid 30019
30013:S 01 Feb 2023 23:09:45.397 * AOF rewrite child asks to stop sending diffs.
30019:C 01 Feb 2023 23:09:45.397 * Parent agreed to stop sending diffs. Finalizing AOF...
30019:C 01 Feb 2023 23:09:45.397 * Concatenating 0.00 MB of AOF diff received from parent.
30019:C 01 Feb 2023 23:09:45.398 * SYNC append only file rewrite performed
30019:C 01 Feb 2023 23:09:45.398 * AOF rewrite: 0 MB of memory used by copy-on-write
30013:S 01 Feb 2023 23:09:45.436 * Background AOF rewrite terminated with success
30013:S 01 Feb 2023 23:09:45.436 * Residual parent diff successfully flushed to the rewritten AOF (0.00 MB)
30013:S 01 Feb 2023 23:09:45.436 * Background AOF rewrite finished successfully
lwen.steven
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
Redis集群节点操作方法，及如何测试自动故障转移
01-07
Redis集群以及自动故障转移测试 https://www.cnblogs.com/wy123/p/10325904.html 执行命令： 1、查询出要宕机的redis集群主节点信息 redis-cli -h 30.74.3.50 -p 7224 -c –raw 进入redis服务命令行客户端窗口 ...
redis cluster master failover问题
阿拉斯加大闸蟹的博客
02-24 5754
背景： 在进行主从切换演练的过程中，发现client应用端会在master下线后，slave选举成为主且节点topology拓扑更新完毕的过程中出现报错信息 默认的时间是60s，我这里配置的是300ms io.lettuce.core.RedisCommandTimeoutException: Command timed out after 300 millisecond(s) at io.lettuce.core.ExceptionFactory.createTimeoutExc
Redis 学习笔记（十二）Redis 复制功能详解 ----- (error) READONLY You can't write against a read only slave...
daxingsuo3996的博客
04-18 3924
Redis 复制(Replication)1. 复制介绍分布式数据库为了获取更大的存储容量和更高的并发访问量，会将原来集中式数据库中的数据分散存储到多个通过网络连接的数据存储节点上。Redis为了解决单点数据库问题，会把数据复制多个副本部署到其他节点上，通过复制，实现Redis的高可用性，实现对数据的冗余备份，保证数据和服务的高度可靠性。 2. 复制的建立建立复制的配置方式有三种。 ...
解决Redis/Codis Connection with master lost(复制超时)问题
weixin_45866907的博客
09-29 210
解决Redis/Codis Connection with master lost(复制超时)问题
解决Redis Connection with master lost(复制超时)问题
花飘万家雪
01-11 1732
今天在线上环境中遇到了codis-server报警，按照常规处理流程进行处理，报错步骤如下： 首先将codis-slave的rdb文件移除，并重启codis-slave 在codis-dashbord中将codis-slave移除问题codis group 将codis-slave重新加入codis group，并测试在codis-master中写入数据，查看codis-slave中能否正常读取数据 没想到在新加入codis group同步数据时发生以下报错： [13029] 15.
故障分析 | redis cluster 从库无法自动恢复同步案例一则
ActionTech的博客
02-16 873
作者：任坤现居珠海，先后担任专职 Oracle 和 MySQL DBA，现在主要负责 MySQL、mongoDB 和 Redis 维护工作。本文来源：原创投稿*爱可生开源社区出品，原创内容未经授权不得随意使用，转载请联系小编并注明来源。
# Error condition on socket for SYNC: Connection refused
最新发布
weixin_68919548的博客
09-19 412
Error condition on socket for SYNC: Connection refused
启动Redis警告:WARNING overcommit_memory is set to 0 Background save may fail under low memory condition
我的开发博客(公众号:Web后端技术)
07-22 1万+
启动redis里有一个警告： $ src/redis-server # Server initialized # WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command
【Redis】解决WARNING overcommit_memory is set to 0 Background save may fail under low memory condition.
源林氏的笔记本
09-02 4412
在内存不足的情况下，后台保存可能会失败。若要解决此问题，请将“vm.overcommit_memory = 1”添加到/etc/sysctl.conf，然后重新启动或运行命令“sysctl vm.overcommit_memory=1”使其生效。不管是linux直装，还是在docker环境中，启动redis时，报如下错误。编辑主机的/etc/sysctl.conf文件，添加或者配置，保存。重启Redis服务即可。
redis集群一键自动部署脚本.rar
02-26
redis集群自动搭建脚本,按照脚本中说明,简单修改ip 端口等参数 便可执行脚本自动搭建redis集群,适用centos7.X版本
Redis源码解析:集群手动故障转移、从节点迁移详解
09-09
主要介绍了Redis源码解析:集群手动故障转移、从节点迁移的相关内容，涉及通过集群定时器函数clusterCron实现从节点迁移等知识，具有一定参考价值，需要的朋友可以了解。
详细分析Redis集群故障
09-09
主要介绍了详细分析Redis集群故障的相关内容，具有一定的参考价值，这里分享给大家，供需要的朋友参考。
Redis集群系列十二 —— 故障转移二
旷野历程
01-03 752
故障转移二
Redis在Linux系统的配置优化
m0_67900727的博客
03-15 1575
一. 内存分配控制 1. vm.overcommit_memory Redis在启动时可能会出现这样的日志： # WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the comman
redis集群故障的发现与转移
keehom的博客
07-13 649
Sentinel 系统通过向集群发送心跳包来检测节点状态，并在检测到故障时与其他 Sentinel 节点协同工作，以确定新的主节点并执行故障转移。b. 集群会选举一个新的主节点来接管故障主节点的槽（slot）。3. 故障转移：当一个主节点发生故障时，集群会执行故障转移以维护系统的高可用性。c. 新的主节点开始复制故障主节点的数据，以便在故障转移完成后提供完整的数据服务。d. 集群中的其他节点更新它们的配置，以将故障主节点的槽重新映射到新的主节点。a. 集群中的其他节点会检测到主节点已下线。
【redis】cluster配置和故障转移
alwayszmj
04-15 1121
redis cluster配置和故障转移
Redis启动出现WARNING overcommit_memory is set to 0...解决
热门推荐
yukuleshui的博客
02-22 1万+
1.启动redis。 redis-server redis.conf路径+文件 出现警告 WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add ‘vm.overcommit_memory = 1’ to > > > /etc/sysctl.conf and then reboot or run the command
Redis 部分复制详解
小楼一夜听春雨，深巷明朝卖杏花
03-06 3477
 对数据库进行备份 1．对数据库进行全备 1．使用backup database 命令执行备份 RMAN> BACKUP DATABASE; 执行上述命令后将对目标数据库中的所有数据文件进行备份，由于没有显式指定 FORMAT 参数，rman 会自动为每个备份片段命名，并保存在ORACLE_HOME/database目录下。 当然，也可以通过显式指定format
6.3 Redis主从复制之runid和偏移量
Steven的博客
03-21 2981
1.数据同步 Redis在2.8及以上版本使用psync命令完成主从数据同步，同步过程分为全量复制和部分复制。 (1).全量复制 一般用于初次复制场景，Redis早期支持的复制功能只有全量复制，它会把主节点全部数据一次性发送给从节点，当数据量较大时，会对主从节点和网络造成很大的开销。 (2).部分复制 用于处理在主从复制中因网络闪断等原因造成的数据丢失场景，当从节点再次连上主节点后，如果条件允许，主节点会补发丢失数据给从节点。因为补发的数据远远小于全量数据，可以有效避免全量复制的过高开销。 部分复制是对老版
redis集群节点故障恢复
06-09
当 Redis 集群中的某个节点出现故障时，我们需要进行故障恢复操作以保证集群的正常工作。以下是一个简单的 Redis 集群节点故障恢复步骤： 1. 检查故障节点：首先需要检查出现故障的 Redis 节点，可以通过查看 Redis...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值