6.5. 使用 ASCS 实例崩溃节点

验证

1. 检查其他节点中的集群是否隔离崩溃的节点：

   [root]# pcs stonith history
   reboot of node1 successful: delegate=node2, client=pacemaker-controld.1468, origin=node2, completed=...
   1 event found

2. 检查集群是否在剩余的节点上启动 ASCS 资源。在 2 节点集群中，这会导致同一节点上运行的 ASCS 和 ERS ：

   [root]# pcs resource
     * Resource Group: grp_S4H_ASCS20:
       * rsc_vip_S4H_ASCS20        (ocf:heartbeat:IPAddr2):         Started node2
       * rsc_SAPStartSrv_S4H_ASCS20        (ocf:heartbeat:SAPStartSrv):     Started node2
       * rsc_SAPInstance_S4H_ASCS20        (ocf:heartbeat:SAPInstance):     Started node2
     * Resource Group: grp_S4H_ERS29:
       * rsc_vip_S4H_ERS29 (ocf:heartbeat:IPAddr2):         Started node2
       * rsc_SAPStartSrv_S4H_ERS29 (ocf:heartbeat:SAPStartSrv):     Started node2
       * rsc_SAPInstance_S4H_ERS29 (ocf:heartbeat:SAPInstance):     Started node2

3. 在隔离的节点再次运行后，检查集群是否自动将 ERS 实例移到之前失败的节点：

   [root]# pcs resource
     * Resource Group: grp_S4H_ASCS20:
       * rsc_vip_S4H_ASCS20        (ocf:heartbeat:IPAddr2):         Started node2
       * rsc_SAPStartSrv_S4H_ASCS20        (ocf:heartbeat:SAPStartSrv):     Started node2
       * rsc_SAPInstance_S4H_ASCS20        (ocf:heartbeat:SAPInstance):     Started node2
     * Resource Group: grp_S4H_ERS29:
       * rsc_vip_S4H_ERS29 (ocf:heartbeat:IPAddr2):         Started node1
       * rsc_SAPStartSrv_S4H_ERS29 (ocf:heartbeat:SAPStartSrv):     Started node1
       * rsc_SAPInstance_S4H_ERS29 (ocf:heartbeat:SAPInstance):     Started node1

   较早配置的 ASCS 资源粘性和您先前配置的约束可确保 ASCS 实例保持原位，以避免再次中断服务。