本節簡單介紹了PostgreSQL HA三種解決方案PAF（PostgreSQL Automatic Failover）、repmgr和Patroni的Pros（優勢）和Cons（劣勢）。

**PAF**
>PAF Pros
1.PAF provides the user a free hands-on configuration and setup of PostgreSQL.
2.PAF can handle node failures and trigger elections when the master goes down.
3.Quorum behavior can be enforced in PAF.
4.It will provide a complete high availability management solution for the resource, including start, stop, and monitor, and handle network isolation scenarios.
5.It’s a distributed solution, which enables the management of any node from another node.

PAF的優勢是PG的安裝配置相對自由、主節點當機的情況下可觸發選主、可實現強制仲裁、提供了一整套完整的HA和分散式管理方案。

>PAF Cons
1.PAF doesn’t detect if a standby is misconfigured with an unknown or non-existent node in recovery configuration. Node will be shown as slave, even if standby is running without connecting to the master/cascading standby node.
2.Requires an extra port (Default 5405) to be opened for the Pacemaker and Corosync components’ communication using UDP.
3.Does not support NAT-based configuration.
4.No pg_rewind support.

RAF的劣勢是不會檢測備機是否被錯誤配置為未知或不存在的節點、需要額外的埠用於Pacemaker和Corosync元件UDP通訊、不支援基於NAT配置以及不支援pg_rewind。

**repmgr**
>repmgr Pros
1.Repmgr provides utilities that help to setup primary and standby nodes and configure replication.
2.It doesn’t use any extra ports for communication. If you want to perform switchover, only then will it require passwordless SSH to be configured.
3.Provides notification by invoking the user scripts for the registered events.
4.Performs automatic failover in case of primary server failure.

repmgr的優勢是提供了工具配置主從節點和複製、不需要額外的通訊節點、可呼叫使用者指令碼提供通知以及可執行自動的failover。

>repmgr Cons
1.repmgr doesn’t detect if the standby is misconfigured with an unknown or non-existent node in recovery configuration. The node will be shown as standby even if it is running without connecting to the primary/cascading standby node.
2.Cannot retrieve the status of another node from a node where PostgreSQL service is down. Hence, it doesn’t provide a distributed control solution.
3.It doesn’t handle recovering the health of individual nodes.

repmgr的劣勢是不能檢測備機是否被錯誤配置為未知或不存在的節點、不能檢測遠端節點的狀態（不具備分散式解決方案）以及不能處理單個節點的恢復。

**Patroni**
>Patroni Pros
1.Patroni enables end-to-end setup up of the cluster.
2.Supports REST APIs and HAproxy integration.
3.Supports event notifications via callbacks scripts triggered by certain actions.
4.Leverages DCS for consensus.

Patroni的優勢是實現了叢集端對端的配置、支援REST APIs和HAProxy的整合、透過回撥指令碼支援事件通知、利用DCS達成共識。

>Patroni Cons
1.Patroni will not detect the misconfiguration of a standby with an unknown or non-existent node in recovery configuration. The node will be shown as a slave even if the standby is running without connecting to the master/cascading standby node.
2.User needs to handle setup, management, and upgrade of the DCS software.
3.Requires multiple ports to be open for components communication:
    REST API port for Patroni
    Minimum 2 ports for DCS

Patroni的劣勢是不能檢測備機是否被錯誤配置為未知或不存在的節點、使用者需要自行安裝配置DCS、需要多個埠用於元件之間的通訊。

**參考資料**
[Managing High Availability in PostgreSQL – Part I: PostgreSQL Automatic Failover](https://scalegrid.io/blog/managing-high-availability-in-postgresql-part-1/)
[Managing High Availability in PostgreSQL – Part II: Replication Manager](https://scalegrid.io/blog/managing-high-availability-in-postgresql-part-2/)
[Managing High Availability in PostgreSQL – Part III: Patroni](https://scalegrid.io/blog/managing-high-availability-in-postgresql-part-3/)

PostgreSQL DBA(121) - pgAdmin(HA：PAF vs repmgr vs Patroni)

相關文章