ansible roles 案例2 二进制部署redis cluster
集群规划
此role 模拟3台节点，每个节点运行两个redis，共计6节点。
仓库地址
github:
目录结构
创建redis_cluster roles
ansible-galaxy init redis_cluster
roles/redis_cluster/ ├── defaults │ └── main.yml ├── files │ ├── redis-6.2.11 # 编译后的文件，此role需要提前编译 │ ├── redis-6.2.11.tar.gz │ ├── redis-benchmark # 下面是编译好的src目录下的二进制执行命令文件 │ ├── redis-check-aof │ ├── redis-check-rdb │ ├── redis-cli │ ├── redis-server │ └── redis-trib.rb ├── handlers │ └── main.yml ├── hosts │ └── inventory.yaml ├── meta │ └── main.yml ├── README.md ├── tasks │ ├── add_user.yml │ ├── check_cluster.yml │ ├── check_reids.yml │ ├── chown_dir.yml │ ├── copy_bin.yml │ ├── create_cluster.yml │ ├── create_redis_dir.yml │ ├── main.yml │ ├── rsync_dir.yml │ ├── set_sysctl.yml │ ├── start_redis.yml │ ├── template_redis_conf.yml │ └── template_redis_service.yml ├── templates │ ├── redis.conf.j2 # redis配置文件 │ └── redis.service.j2 # redis启动文件 ├── tests │ ├── inventory │ └── test.yml └── vars └── main.yml
静态文件
编译redis：略
主机清单
[root@master-1 roles]# cat roles/redis_cluster/hosts/inventory.yaml all: hosts: {} children: redis_cluster: vars: base_port: 6379 redis_data_base_dir: /opt/redis_cluster cluster_node: "192.168.43.129:6379 192.168.43.129:6380 192.168.43.130:6381 192.168.43.130:6382 192.168.43.131:6383 192.168.43.131:6384" hosts: node1: ansible_host: 192.168.43.129 index: 0 # 索引，后面可以用base_port变量与index来动态生成一些配置 create_node: true # 创建集群与检查集群的判断配置 node2: ansible_host: 192.168.43.129 index: 1 node3: ansible_host: 192.168.43.130 index: 2 node4: ansible_host: 192.168.43.130 index: 3 node5: ansible_host: 192.168.43.131 index: 4 node6: ansible_host: 192.168.43.131 index: 5
变量设置
[root@master-1 roles]# cat roles/redis_cluster/vars/main.yml --- # vars file for redis-_cluster password: 123456 rewrite_percentage: 25 # aof文件压缩比 cluster_enabled: 'yes' listen_port: "{{ base_port + index }}" # 动态生成redis listen port data_dir: "{{ redis_data_base_dir }}/{{ base_port + index }}" # 动态生成redis家目录 maxmemory_policy: volatile-lru # 淘汰策略
渲染模板
redis配置文件
[root@master-1 roles]# cat roles/redis_cluster/templates/redis.conf.j2 bind 0.0.0.0 protected-mode yes port {{ listen_port }} # 动态生成，因为每个节点的index不同，当执行到本节点，取index变量，就可以动态生成port tcp-backlog 511 timeout 0 tcp-keepalive 300 daemonize yes pidfile {{ data_dir }}/redis_{{ listen_port }}.pid loglevel notice logfile {{ data_dir }}/logs # 这个也是一样 databases 16 always-show-logo no set-proc-title yes proc-title-template "{title} {listen-addr} {server-mode}" save 3600 1 save 300 100 save 60 10000 stop-writes-on-bgsave-error no rdbcompression yes rdbchecksum yes dbfilename dump.rdb rdb-del-sync-files no dir {{ data_dir }}/data masterauth {{ password }} replica-serve-stale-data yes replica-read-only yes repl-diskless-sync no repl-diskless-sync-delay 30 repl-diskless-load disabled repl-ping-replica-period 10 repl-timeout 60 repl-disable-tcp-nodelay no repl-backlog-size 512mb repl-backlog-ttl 3600 replica-priority 100 acllog-max-len 128 requirepass {{ password }} maxclients 10000 maxmemory {{ ansible_memtotal_mb // 2 }}mb # 获取主机信息后，动态设置redis maxmemory maxmemory-policy {{ maxmemory_policy }} lazyfree-lazy-eviction no lazyfree-lazy-expire no lazyfree-lazy-server-del no replica-lazy-flush no lazyfree-lazy-user-del no lazyfree-lazy-user-flush no oom-score-adj no oom-score-adj-values 0 200 800 disable-thp yes appendonly yes appendfilename "appendonly.aof" appendfsync everysec no-appendfsync-on-rewrite no auto-aof-rewrite-percentage {{ rewrite_percentage }} auto-aof-rewrite-min-size 64mb aof-load-truncated yes aof-use-rdb-preamble yes lua-time-limit 5000 cluster-enabled {{cluster_enabled }} cluster-config-file nodes-{{ listen_port }}.conf # 集群配置文件 cluster-node-timeout 15000 cluster-replica-validity-factor 10 cluster-migration-barrier 1 cluster-require-full-coverage no 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 stream-node-max-bytes 4096 stream-node-max-entries 100 activerehashing yes client-output-buffer-limit normal 0 0 0 client-output-buffer-limit replica 256mb 64mb 60 client-output-buffer-limit pubsub 32mb 8mb 60 hz 10 dynamic-hz yes aof-rewrite-incremental-fsync yes rdb-save-incremental-fsync yes jemalloc-bg-thread yes
redis启动文件
[root@master-1 roles]# cat roles/redis_cluster/templates/redis.service.j2 [Unit] Description=Redis In-Memory Data Store After=network.target [Service] Type=forking ExecStart=/usr/local/sbin/redis-server {{ data_dir }}/redis.conf --supervised systemd # 根据redis 不同家目录的配置文件启动redis ExecStop=/usr/local/sbin/redis-cli -p {{ listen_port }} shutdown # 根据端口生成配置文件 Restart=always User=redis # redis 启动用户 Group=redis LimitNOFILE=65535 LimitNPROC=65535 # 这指定了一个运行时目录，用于存储服务运行时的临时文件。这个目录在服务停止时会被清理。 # 这设置了运行时目录的权限。0755意味着目录的所有者可以读、写、执行，而组用户和其他用户只能读和执行。 # RuntimeDirectory=redis # RuntimeDirectoryMode=0755 [Install] WantedBy=multi-user.target
task任务
main文件
[root@master-1 tasks]# cat main.yml --- # tasks file for redis-_cluster - include: set_sysctl.yml - include: add_user.yml - include: create_redis_dir.yml - include: rsync_dir.yml - include: template_redis_conf.yml - include: template_redis_service.yml - include: copy_bin.yml - include: chown_dir.yml - include: start_redis.yml - include: check_reids.yml - include: create_cluster.yml - include: check_cluster.yml
创建redis家目录
[root@master-1 tasks]# cat create_redis_dir.yml - name: create redis dir ansible.builtin.file: path: "{{ data_dir }}" state: directory
传输redis文件到redis 不同家目录
[root@master-1 tasks]# cat rsync_dir.yml - name: copy redis dir #ansible.builtin.copy: # 如果 path 是目录，则递归复制。在这种情况下，如果 path 以“/”结尾，则只有该目录的内部内容才会复制到目标。否则，如果它不以“/”结尾，则复制目录本身及其所有内容。 # src: redis-6.2.11/ # 如果 `dest' 是不存在的路径，并且 `dest' 以“/”结尾或 `src' 是目录，则创建 `dest'。 # 如果 `src' 和 `dest' 都是文件，则不会创建 `dest' 的父目录，如果该目录尚不存在，则任务失败。 # dest: "{{ data_dir }}/" ansible.builtin.synchronize:
# /表示文件目录下所有文件，不加/表示目录本身 src: redis-6.2.11/ dest: "{{ data_dir }}/"
# 报错时间，权限，符号，链接不变 archive: yes
# 压缩，加速传输 compress: yes owner: yes group: yes
# 保留权限不变 perms: yes
# 是否删除源文件 delete: no
# 这个配置是让两台主机之间同步数据，而不是在ansible本机 #delegate_to: "{{ inventory_hostname }}"
拷贝redis bin文件
[root@master-1 tasks]# cat copy_bin.yml - name: copy redis bin ansible.builtin.copy: src: "{{ item }}" dest: "/usr/local/sbin/{{ item }}" mode: 0551 force: true loop: - redis-cli - redis-check-aof - redis-check-rdb - redis-server - redis-trib.rb
启动redis
[root@master-1 tasks]# cat start_redis.yml - name: start redis ansible.builtin.systemd:
# 在执行到node时，获取到当前node节点的port，并启动 name: redis-{{listen_port}} daemon_reload: yes state: started
检查redis进程及端口
[root@master-1 tasks]# cat check_reids.yml - name: check redis port & process ansible.builtin.shell: | ps -C redis-server --no-headers |wc -l # 统计进程数量 register: redis_process_count failed_when: redis_process_count.stdout | int != 2 # 转为数字当不等于2 changed_when: false # 此操作不认为发生改变
创建集群
[root@master-1 tasks]# cat create_cluster.yml - name: create redis cluster ansible.builtin.shell: | echo "yes" |redis-cli -a {{ password }} --cluster create {{ cluster_node }} --cluster-replicas 1
sleep 5 # 这个可以避免集群刚创建成功就去检测，就会检测不通过，等初始化完成 when: create_node is defined and create_node # 只在打了该标签的节点执行
检查集群状态
[root@master-1 tasks]# cat check_cluster.yml - name: check cluster status ansible.builtin.shell: | redis-cli -a {{ password }} -h {{ ansible_host }} -p {{ listen_port }} cluster info # 输出集群信息 register: cluster_status when: create_node is defined and create_node # 只在打了该标签的节点运行 changed_when: false failed_when: cluster_status.rc != 0 and 'cluster_state:ok' not in cluster_status.stdout # 当cluster info执行结果返回不为0 和该字段不在输出结果
- name: output resister
ansible.builtin.debug:
var: cluster_status # 输出变量所有结果，也可以用cluster_status.stdout,cluster_status.stderr等字段。
- name: output msg ansible.builtin.debug: msg: "集群状态检查成功..." when: create_node is defined and create_node and cluster_status.rc == 0
执行role
[root@master-1 roles]# ansible-playbook -i roles/redis_cluster/hosts/inventory.yaml main_redis_all.yml --list-hosts playbook: main_redis_all.yml play #1 (redis_cluster): redis_cluster TAGS: [] pattern: [u'redis_cluster'] hosts (6): node1 node3 node2 node5 node4 node6 [root@master-1 roles]# ansible-playbook -i roles/redis_cluster/hosts/inventory.yaml main_redis_all.yml PLAY [redis_cluster] ********************************************************************************************************************************************************** TASK [Gathering Facts] ******************************************************************************************************************************************************** ok: [node1] ok: [node2] ok: [node5] ok: [node4] ok: [node3] ok: [node6] TASK [redis_cluster : set vm.overcommit_memory] ******************************************************************************************************************************* ok: [node1] ok: [node4] ok: [node3] ok: [node2] ok: [node5] ok: [node6] TASK [redis_cluster : add redis user] ***************************************************************************************************************************************** ok: [node3] ok: [node4] ok: [node5] ok: [node1] ok: [node2] ok: [node6] TASK [redis_cluster : create redis dir] *************************************************************************************************************************************** ok: [node5] ok: [node3] ok: [node4] ok: [node1] ok: [node2] ok: [node6] TASK [redis_cluster : copy redis dir] ***************************************************************************************************************************************** changed: [node5] changed: [node2] changed: [node1] changed: [node4] changed: [node3] changed: [node6] TASK [redis_cluster : 设置 Redis 配置] ******************************************************************************************************************************************** changed: [node5] changed: [node4] changed: [node2] changed: [node1] changed: [node3] changed: [node6] TASK [redis_cluster : 设置 Redis 启动配置] ****************************************************************************************************************************************** ok: [node1] ok: [node5] ok: [node2] ok: [node4] ok: [node3] ok: [node6] TASK [redis_cluster : chown redis_dir] **************************************************************************************************************************************** changed: [node3] changed: [node1] ok: [node2] ok: [node4] changed: [node5] ok: [node6] TASK [redis_cluster : start redis] ******************************************************************************************************************************************** changed: [node5] changed: [node3] changed: [node4] changed: [node2] changed: [node1] changed: [node6] TASK [redis_cluster : check redis port & process] ***************************************************************************************************************************** ok: [node5] ok: [node4] ok: [node3] ok: [node1] ok: [node2] ok: [node6] TASK [redis_cluster : create redis cluster] *********************************************************************************************************************************** skipping: [node3] skipping: [node2] skipping: [node5] skipping: [node4] skipping: [node6] changed: [node1] TASK [redis_cluster : output cluster info] ************************************************************************************************************************************ skipping: [node3] skipping: [node2] skipping: [node5] skipping: [node4] skipping: [node6] ok: [node1] # 可以看到判断标签条件已生效，其他节点被跳过了 TASK [redis_cluster : output resister] **************************************************************************************************************************************** ok: [node1] => { "cluster_status": { "changed": false, "cmd": "redis-cli -a 123456 -h 192.168.43.129 -p 6379 cluster info\n", "delta": "0:00:00.063657", "end": "2025-02-25 14:30:36.592194", "failed": false, "failed_when_result": false, "rc": 0, "start": "2025-02-25 14:30:36.528537", "stderr": "Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.", "stderr_lines": [ "Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe." ], "stdout": "cluster_state:ok\r\ncluster_slots_assigned:16384\r\ncluster_slots_ok:16384\r\ncluster_slots_pfail:0\r\ncluster_slots_fail:0\r\ncluster_known_nodes:6\r\ncluster_size:3\r\ncluster_current_epoch:6\r\ncluster_my_epoch:1\r\ncluster_stats_messages_ping_sent:14\r\ncluster_stats_messages_pong_sent:15\r\ncluster_stats_messages_sent:29\r\ncluster_stats_messages_ping_received:10\r\ncluster_stats_messages_pong_received:14\r\ncluster_stats_messages_meet_received:5\r\ncluster_stats_messages_received:29", "stdout_lines": [ "cluster_state:ok", "cluster_slots_assigned:16384", "cluster_slots_ok:16384", "cluster_slots_pfail:0", "cluster_slots_fail:0", "cluster_known_nodes:6", "cluster_size:3", "cluster_current_epoch:6", "cluster_my_epoch:1", "cluster_stats_messages_ping_sent:14", "cluster_stats_messages_pong_sent:15", "cluster_stats_messages_sent:29", "cluster_stats_messages_ping_received:10", "cluster_stats_messages_pong_received:14", "cluster_stats_messages_meet_received:5", "cluster_stats_messages_received:29" ] } } ok: [node3] => { "cluster_status": { "changed": false, "skip_reason": "Conditional result was False", "skipped": true } } ok: [node2] => { "cluster_status": { "changed": false, "skip_reason": "Conditional result was False", "skipped": true } } ok: [node5] => { "cluster_status": { "changed": false, "skip_reason": "Conditional result was False", "skipped": true } } ok: [node4] => { "cluster_status": { "changed": false, "skip_reason": "Conditional result was False", "skipped": true } } ok: [node6] => { "cluster_status": { "changed": false, "skip_reason": "Conditional result was False", "skipped": true } } TASK [redis_cluster : output msg] ********************************************************************************************************************************************* ok: [node1] => { "msg": "集群状态检查成功..." } skipping: [node3] skipping: [node2] skipping: [node5] skipping: [node4] skipping: [node6] PLAY RECAP ******************************************************************************************************************************************************************** node1 : ok=14 changed=5 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0 node2 : ok=11 changed=3 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0 node3 : ok=11 changed=4 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0 node4 : ok=11 changed=3 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0 node5 : ok=11 changed=4 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0 node6 : ok=11 changed=3 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0
遇到问题
1. copy权限发生变化，并且设置后权限没达到预期
错误示例
- name: copy redis bin ansible.builtin.copy: src: "{{ item }}" dest: "/usr/local/sbin/{{ item }}" mode: 551 force: true loop: - redis-cli - redis-check-aof - redis-check-rdb - redis-server - redis-trib.rb
结果
----r--rwt 1 root root 9548592 2月 24 14:23 redis-check-aof ----r--rwt 1 root root 9548592 2月 24 14:23 redis-check-rdb ----r--rwt 1 root root 5004456 2月 24 14:23 redis-cli ----r--rwt 1 root root 9548592 2月 24 14:23 redis-server ----r--rwt 1 root root 3600 2月 24 14:23 redis-trib.rb
分析：
mode需要写八进制
- name: copy redis bin ansible.builtin.copy: src: "{{ item }}" dest: "/usr/local/sbin/{{ item }}" mode: 0551 force: true loop: - redis-cli - redis-check-aof - redis-check-rdb - redis-server - redis-trib.rb
2. 解决只在其中一台创建集群，而不是定义的所有主机
hosts: node1: ansible_host: 192.168.43.129 index: 0 create_node: true - name: check cluster status ansible.builtin.shell: | redis-cli -a {{ password }} -h {{ ansible_host }} -p {{ listen_port }} cluster info |grep 'cluster_state:ok' register: cluster_status when: create_node is defined and create_node # 加判断 changed_when: false - name: output msg ansible.builtin.debug: msg: "集群状态检查成功..." when: create_node is defined and create_node and cluster_status.rc == 0 加判断
3. 在启动redis时错误使用了loop
[root@master-1 tasks]# cat start_redis.yml - name: start redis ansible.builtin.systemd: name: redis-{{listen_port}} daemon_reload: yes state: started loop: "{{ groups['redis_cluster'] }}" # 不需要
当使用loop后，等于是在一台节点上重复启动6次，所以不需要使用loop。
4. 集群状态检测when条件没生效
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - name: check cluster status ansible.builtin.shell: | redis-cli -a {{ password }} -h {{ ansible_host }} -p {{ listen_port }} cluster info register: cluster_status when: create_node is defined and create_node changed_when: false failed_when: cluster_status.rc != 0 and 'cluster_state:ok' not in cluster_status.stdout - name: output resister ansible.builtin.debug: var: cluster_status - name: output msg ansible.builtin.debug: msg: "集群状态检查成功..." when: create_node is defined and create_node and cluster_status.rc == 0 TASK [redis_cluster : output resister] **************************************************************************************************************************************** ok: [node1] => { "cluster_status": { "changed": false, "cmd": "redis-cli -a 123456 -h 192.168.43.129 -p 6379 cluster info\n", "delta": "0:00:00.035331", "end": "2025-02-25 13:29:46.126410", "failed": false, "failed_when_result": false, "rc": 0, "start": "2025-02-25 13:29:46.091079", "stderr": "Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.", "stderr_lines": [ "Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe." ], "stdout": "cluster_state:fail\r\ncluster_slots_assigned:16384\r\ncluster_slots_ok:16384\r\ncluster_slots_pfail:0\r\ncluster_slots_fail:0\r\ncluster_known_nodes:6\r\ncluster_size:3\r\ncluster_current_epoch:6\r\ncluster_my_epoch:1\r\ncluster_stats_messages_ping_sent:8\r\ncluster_stats_messages_pong_sent:12\r\ncluster_stats_messages_sent:20\r\ncluster_stats_messages_ping_received:7\r\ncluster_stats_messages_pong_received:8\r\ncluster_stats_messages_meet_received:5\r\ncluster_stats_messages_received:20", "stdout_lines": [ "cluster_state:fail", "cluster_slots_assigned:16384", "cluster_slots_ok:16384", "cluster_slots_pfail:0", "cluster_slots_fail:0", "cluster_known_nodes:6", "cluster_size:3", "cluster_current_epoch:6", "cluster_my_epoch:1", "cluster_stats_messages_ping_sent:8", "cluster_stats_messages_pong_sent:12", "cluster_stats_messages_sent:20", "cluster_stats_messages_ping_received:7", "cluster_stats_messages_pong_received:8", "cluster_stats_messages_meet_received:5", "cluster_stats_messages_received:20" ] } }
但是从输出来看，"cluster_state:fail，我在想是不是failed_when条件没执行吗。failed_when: cluster_status.rc != 0 and 'cluster_state:ok' not in cluster_status.stdout
分析：
failed_when: cluster_status.rc != 0 and 'cluster_state:ok' not in cluster_status.stdout
这个判断条件从结果来看cluster_status.rc != 0 为false，'cluster_state:ok' not in cluster_status.stdout为true，and条件判断来说，只有两个都为true，结果为true，所以说结果为false，条件不成立，failed_when为false。
整改
将条件改成false，只要有一个条件为false，条件为false，failed_when为false，结果成了，task失败。
failed_when: cluster_status.rc != 0 or 'cluster_state:ok' not in cluster_status.stdout
5. 集群状态检测不通过
执行集群状态检测
TASK [redis_cluster : check cluster status] *********************************************************************************************************************************** skipping: [node3] skipping: [node2] skipping: [node5] skipping: [node4] skipping: [node6] fatal: [node1]: FAILED! => {"changed": false, "cmd": "redis-cli -a 123456 -h 192.168.43.129 -p 6379 cluster info\n", "delta": "0:00:00.037506", "end": "2025-02-25 13:53:21.595669", "failed_when_result": true, "rc": 0, "start": "2025-02-25 13:53:21.558163", "stderr": "Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.", "stderr_lines": ["Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe."], "stdout": "cluster_state:fail\r\ncluster_slots_assigned:16384\r\ncluster_slots_ok:16384\r\ncluster_slots_pfail:0\r\ncluster_slots_fail:0\r\ncluster_known_nodes:6\r\ncluster_size:3\r\ncluster_current_epoch:6\r\ncluster_my_epoch:1\r\ncluster_stats_messages_ping_sent:8\r\ncluster_stats_messages_pong_sent:13\r\ncluster_stats_messages_sent:21\r\ncluster_stats_messages_ping_received:8\r\ncluster_stats_messages_pong_received:8\r\ncluster_stats_messages_meet_received:5\r\ncluster_stats_messages_received:21", "stdout_lines": ["cluster_state:fail", "cluster_slots_assigned:16384", "cluster_slots_ok:16384", "cluster_slots_pfail:0", "cluster_slots_fail:0", "cluster_known_nodes:6", "cluster_size:3", "cluster_current_epoch:6", "cluster_my_epoch:1", "cluster_stats_messages_ping_sent:8", "cluster_stats_messages_pong_sent:13", "cluster_stats_messages_sent:21", "cluster_stats_messages_ping_received:8", "cluster_stats_messages_pong_received:8", "cluster_stats_messages_meet_received:5", "cluster_stats_messages_received:21"]} - name: check cluster status ansible.builtin.shell: | redis-cli -a {{ password }} -h {{ ansible_host }} -p {{ listen_port }} cluster info register: cluster_status when: create_node is defined and create_node changed_when: false failed_when: cluster_status.rc != 0 or 'cluster_state:ok' not in cluster_status.stdout - name: output resister ansible.builtin.debug: var: cluster_status - name: output msg ansible.builtin.debug: msg: "集群状态检查成功..." when: create_node is defined and create_node and cluster_status.rc == 0
问题：当我手动执行，发现状态是cluster_state:ok的，我不明白为什么ansible执行这个命令为什么cluster_state:fail
[root@master-1 roles]# redis-cli -a 123456 -h 192.168.43.129 -p 6379 cluster info Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe. cluster_state:ok cluster_slots_assigned:16384 cluster_slots_ok:16384 cluster_slots_pfail:0 cluster_slots_fail:0 cluster_known_nodes:6 cluster_size:3 cluster_current_epoch:6 cluster_my_epoch:1 cluster_stats_messages_ping_sent:33 cluster_stats_messages_pong_sent:37 cluster_stats_messages_sent:70 cluster_stats_messages_ping_received:32 cluster_stats_messages_pong_received:33 cluster_stats_messages_meet_received:5 cluster_stats_messages_received:70
分析：
从redis日志来看，redis集群需要链接其他节点做初始化操作，如果集群刚创建好就去检测，可能会失败。
解决
加上睡眠5s，等集群初始化完成后再去检测
- name: create redis cluster ansible.builtin.shell: | echo "yes" |redis-cli -a {{ password }} --cluster create {{ cluster_node }} --cluster-replicas 1 sleep 5 when: create_node is defined and create_node
本文来自博客园，作者：不会跳舞的胖子，转载请注明原文链接：https://www.cnblogs.com/rtnb/p/18735680
浙公网安备 33010602011771号