MariaDB Galera 集群未同步

MariaDB Galera Cluster is not syncing

我之前部署过MariaDB Cluster,最近才出现这个问题(之前没有出现过这个问题,不知道为什么)。

我有服务器1、2和3,我在服务器3上执行了INSERT命令,但是,服务器1和2的表保持不变。

3台服务器位于世界不同地区。 After the INSERT command, the state uuid remains the same.

这是服务器 1 的状态:
MariaDB [mysql]> show status like 'wsrep_%'; +------------------------------+----------------------------------------------------------+ | Variable_name | Value | +------------------------------+----------------------------------------------------------+ | wsrep_local_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_protocol_version | 7 | | wsrep_last_committed | 205 | | wsrep_replicated | 170 | | wsrep_replicated_bytes | 160481 | | wsrep_repl_keys | 664 | | wsrep_repl_keys_bytes | 9222 | | wsrep_repl_data_bytes | 140379 | | wsrep_repl_other_bytes | 0 | | wsrep_received | 46 | | wsrep_received_bytes | 26150 | | wsrep_local_commits | 170 | | wsrep_local_cert_failures | 0 | | wsrep_local_replays | 1 | | wsrep_local_send_queue | 0 | | wsrep_local_send_queue_max | 1 | | wsrep_local_send_queue_min | 0 | | wsrep_local_send_queue_avg | 0.000000 | | wsrep_local_recv_queue | 0 | | wsrep_local_recv_queue_max | 1 | | wsrep_local_recv_queue_min | 0 | | wsrep_local_recv_queue_avg | 0.000000 | | wsrep_local_cached_downto | 1 | | wsrep_flow_control_paused_ns | 0 | | wsrep_flow_control_paused | 0.000000 | | wsrep_flow_control_sent | 0 | | wsrep_flow_control_recv | 0 | | wsrep_cert_deps_distance | 7.482927 | | wsrep_apply_oooe | 0.009756 | | wsrep_apply_oool | 0.000000 | | wsrep_apply_window | 1.009756 | | wsrep_commit_oooe | 0.000000 | | wsrep_commit_oool | 0.000000 | | wsrep_commit_window | 1.000000 | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | wsrep_cert_index_size | 28 | | wsrep_causal_reads | 0 | | wsrep_cert_interval | 0.009756 | | wsrep_incoming_addresses | server1:3306,server2:3306,server3:3306 | | wsrep_evs_delayed | | | wsrep_evs_evict_list | | | wsrep_evs_repl_latency | 0.200155/0.201113/0.201752/0.000614937/4 | | wsrep_evs_state | OPERATIONAL | | wsrep_gcomm_uuid | c4f91b4f-fee1-11e5-8c4f-6e451c332f79 | | wsrep_cluster_conf_id | 3 | | wsrep_cluster_size | 3 | | wsrep_cluster_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | wsrep_local_bf_aborts | 6 | | wsrep_local_index | 0 | | wsrep_provider_name | Galera | | wsrep_provider_vendor | Codership Oy <info@codership.com> | | wsrep_provider_version | 25.3.14(r3560) | | wsrep_ready | ON | | wsrep_thread_count | 2 | +------------------------------+----------------------------------------------------------+

服务器 2 的状态:
MariaDB [(none)]> show status like 'wsrep_%'; +------------------------------+----------------------------------------------------------+ | Variable_name | Value | +------------------------------+----------------------------------------------------------+ | wsrep_local_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_protocol_version | 7 | | wsrep_last_committed | 225 | | wsrep_replicated | 35 | | wsrep_replicated_bytes | 25700 | | wsrep_repl_keys | 119 | | wsrep_repl_keys_bytes | 1757 | | wsrep_repl_data_bytes | 21703 | | wsrep_repl_other_bytes | 0 | | wsrep_received | 187 | | wsrep_received_bytes | 177793 | | wsrep_local_commits | 35 | | wsrep_local_cert_failures | 0 | | wsrep_local_replays | 1 | | wsrep_local_send_queue | 0 | | wsrep_local_send_queue_max | 1 | | wsrep_local_send_queue_min | 0 | | wsrep_local_send_queue_avg | 0.000000 | | wsrep_local_recv_queue | 0 | | wsrep_local_recv_queue_max | 4 | | wsrep_local_recv_queue_min | 0 | | wsrep_local_recv_queue_avg | 0.032086 | | wsrep_local_cached_downto | 9 | | wsrep_flow_control_paused_ns | 0 | | wsrep_flow_control_paused | 0.000000 | | wsrep_flow_control_sent | 0 | | wsrep_flow_control_recv | 0 | | wsrep_cert_deps_distance | 7.193548 | | wsrep_apply_oooe | 0.004630 | | wsrep_apply_oool | 0.000000 | | wsrep_apply_window | 1.004630 | | wsrep_commit_oooe | 0.000000 | | wsrep_commit_oool | 0.000000 | | wsrep_commit_window | 1.000000 | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | wsrep_cert_index_size | 28 | | wsrep_causal_reads | 0 | | wsrep_cert_interval | 0.009217 | | wsrep_incoming_addresses | server1:3306,server2:3306,server3:3306 | | wsrep_evs_delayed | | | wsrep_evs_evict_list | | | wsrep_evs_repl_latency | 0.200138/0.201917/0.203696/0.00177914/2 | | wsrep_evs_state | OPERATIONAL | | wsrep_gcomm_uuid | d562e272-fee1-11e5-b2a2-d3a6b5579aab | | wsrep_cluster_conf_id | 3 | | wsrep_cluster_size | 3 | | wsrep_cluster_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | wsrep_local_bf_aborts | 0 | | wsrep_local_index | 1 | | wsrep_provider_name | Galera | | wsrep_provider_vendor | Codership Oy <info@codership.com> | | wsrep_provider_version | 25.3.14(r3560) | | wsrep_ready | ON | | wsrep_thread_count | 2 | +------------------------------+----------------------------------------------------------+ 57 rows in set (0.01 sec)

server3的状态(如你所见,延迟显示全为0,但我不知道为什么)
MariaDB [(none)]> show status like 'wsrep_%'; +------------------------------+----------------------------------------------------------+ | Variable_name | Value | +------------------------------+----------------------------------------------------------+ | wsrep_local_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_protocol_version | 7 | | wsrep_last_committed | 245 | | wsrep_replicated | 5 | | wsrep_replicated_bytes | 4350 | | wsrep_repl_keys | 11 | | wsrep_repl_keys_bytes | 203 | | wsrep_repl_data_bytes | 3827 | | wsrep_repl_other_bytes | 0 | | wsrep_received | 226 | | wsrep_received_bytes | 208559 | | wsrep_local_commits | 1 | | wsrep_local_cert_failures | 0 | | wsrep_local_replays | 0 | | wsrep_local_send_queue | 0 | | wsrep_local_send_queue_max | 1 | | wsrep_local_send_queue_min | 0 | | wsrep_local_send_queue_avg | 0.000000 | | wsrep_local_recv_queue | 0 | | wsrep_local_recv_queue_max | 1 | | wsrep_local_recv_queue_min | 0 | | wsrep_local_recv_queue_avg | 0.000000 | | wsrep_local_cached_downto | 19 | | wsrep_flow_control_paused_ns | 0 | | wsrep_flow_control_paused | 0.000000 | | wsrep_flow_control_sent | 0 | | wsrep_flow_control_recv | 0 | | wsrep_cert_deps_distance | 7.022026 | | wsrep_apply_oooe | 0.000000 | | wsrep_apply_oool | 0.000000 | | wsrep_apply_window | 1.000000 | | wsrep_commit_oooe | 0.000000 | | wsrep_commit_oool | 0.000000 | | wsrep_commit_window | 1.000000 | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | wsrep_cert_index_size | 28 | | wsrep_causal_reads | 0 | | wsrep_cert_interval | 0.008811 | | wsrep_incoming_addresses | server1:3306,server2:3306,server3:3306 | | wsrep_evs_delayed | | | wsrep_evs_evict_list | | | wsrep_evs_repl_latency | 0/0/0/0/0 | | wsrep_evs_state | OPERATIONAL | | wsrep_gcomm_uuid | fd022144-fee1-11e5-a7a3-f23274fef9c3 | | wsrep_cluster_conf_id | 3 | | wsrep_cluster_size | 3 | | wsrep_cluster_state_uuid | c4f9e2e2-fee1-11e5-8648-a22b867b5a6e | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | wsrep_local_bf_aborts | 0 | | wsrep_local_index | 2 | | wsrep_provider_name | Galera | | wsrep_provider_vendor | Codership Oy <info@codership.com> | | wsrep_provider_version | 25.3.14(r3560) | | wsrep_ready | ON | | wsrep_thread_count | 2 | +------------------------------+----------------------------------------------------------+ 57 rows in set (0.00 sec)

所有三个服务器的 iptables 都设置为接受所有输入和输出流量。

日志显示所有服务器都已加入并与集群同步。

有谁知道为什么?谢谢

最后发现是应用程序使用MyISAM作为存储引擎的问题,导致了这个错误。改回InnoDB后没有报错

我建议你检查一下 table 引擎,因为 Galera 集群支持 InnoDB 引擎而不是 MyISAM。

所以这是一个简单的方法如何将 Mysql 数据库与 MyISAM tables 迁移到 Galera 和 InnoDB:

  1. 确保您的数据库模式不包含 InnoDB 引擎不支持的全文索引或任何其他结构
  2. 转储数据库架构
  3. 将转储字符串 "MYISAM" 替换为 "INNODB"
  4. 转储数据
  5. 在 Galera 集群中准备数据库用户(mysql.user table 未跨集群复制,因此您必须将数据库用户插入每个 mariadb 服务器)
  6. 导入模式(使用 innodb 引擎)
  7. 导入数据
  8. 清理转储文件

感谢https://support.qualityunit.com/718375-Migrate-MySQL-Database-with-Myisam-engine-to-MariaDB-Galera-Cluster