【我和 openGauss 的故事】为集群增加 VIP
作者：daydayup
- 2023-08-07 北京
本文字数：5762 字
阅读完需：约 19 分钟
【我和 openGauss 的故事】为集群增加 VIP
lqkitten [openGauss](javascript:void(0);) 2023-08-04 18:01 发表于四川
openGauss 发布以来，原生支持一主多备，RTO<10S,高可用性能大大增强。自 openGauss3.0 开始，更新了集群管理套件 CM，易用性也得到了提高。但对于客户端来说，数据库端的切换，需要手工完成。
openGauss 增加 VIP 后，客户端的连接就如连接 ORACLE RAC 的 scan VIP 一样，对于服务端的切换无感知。
要使用 VIP，可以在安装前规划，在配置文件中指定，也可以对已安装的集群进行手工增加。下面就测试手工增加方法。
1.已安装集群的相关信息
数据库版本 gsql -Vgsql (openGauss 5.0.0 build a07d57c3) compiled at 2023-03-29 03:37:13 commit 0 last mr
复制代码
集群状态
[omm@db1 srv]$ cm_ctl query -Cv[ CMServer State ]
node instance state-----------------------1 db1 1 Primary2 db2 2 Standby3 db3 3 Standby
[ Cluster State ]
cluster_state : Normalredistributing : Nobalanced : Yescurrent_az : AZ_ALL
[ Datanode State ]
node instance state | node instance state | node instance state---------------------------------------------------------------------------------------------------------1 db1 6001 P Primary Normal | 2 db2 6002 S Standby Normal | 3 db3 6003 S Standby Normal
[omm@db1 srv]$ gs_om -t status --detail[ CMServer State ]
node node_ip instance state-----------------------------------------------------------------------1 db1 192.168.56.11 1 /opt/huawei/data/cmserver/cm_server Primary2 db2 192.168.56.12 2 /opt/huawei/data/cmserver/cm_server Standby3 db3 192.168.56.13 3 /opt/huawei/data/cmserver/cm_server Standby
[ Cluster State ]
cluster_state : Normalredistributing : Nobalanced : Yescurrent_az : AZ_ALL
[ Datanode State ]
node node_ip instance state-------------------------------------------------------------------------1 db1 192.168.56.11 6001 /opt/huawei/install/data/dn P Primary Normal2 db2 192.168.56.12 6002 /opt/huawei/install/data/dn S Standby Normal3 db3 192.168.56.13 6003 /opt/huawei/install/data/dn S Standby Normal
复制代码
2.给 omm 用户增加 sudo 权限，三台机器都执行
echo “omm ALL=(ALL) NOPASSWD:ALL”>>/etc/sudoers
3. 在主库上添加 VIP
添加前
[omm@db1 cm_agent]$ ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:04:f9:a3 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global dynamic noprefixroute enp0s3 valid_lft 74572sec preferred_lft 74572sec inet6 fe80::c8c2:7f4c:914f:a32d/64 scope link noprefixroute valid_lft forever preferred_lft forever3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:50:47:12 brd ff:ff:ff:ff:ff:ff inet 192.168.56.11/24 brd 192.168.56.255 scope global noprefixroute enp0s8 valid_lft forever preferred_lft forever inet6 fe80::890a:d968:b65d:f59a/64 scope link noprefixroute valid_lft forever preferred_lft forever ifconfig enp0s8:15400 192.168.56010 netmask 255.255.255.0 up
复制代码
添加后
[omm@db1 cm_agent]$ ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:04:f9:a3 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global dynamic noprefixroute enp0s3 valid_lft 74572sec preferred_lft 74572sec inet6 fe80::c8c2:7f4c:914f:a32d/64 scope link noprefixroute valid_lft forever preferred_lft forever3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:50:47:12 brd ff:ff:ff:ff:ff:ff inet 192.168.56.11/24 brd 192.168.56.255 scope global noprefixroute enp0s8 valid_lft forever preferred_lft forever inet 192.168.56.10/24 brd 192.168.56.255 scope global secondary enp0s8:15400 inet6 fe80::890a:d968:b65d:f59a/64 scope link noprefixroute valid_lft forever preferred_lft forever
复制代码
4.给集群添加 VIP 资源 VIP 作为 openGauss 的资源管理
[omm@db2 cm_agent]$cm_ctl res --add --res_name="VIP_az1" --res_attr="resources_type=VIP,float_ip=192.168.56.10"cm_ctl: add res(VIP_az1) success.
复制代码
将每个实例加到资源中
[omm@db2 cm_agent]$ cm_ctl res --edit --res_name="VIP_az1" --add_inst="node_id=1,res_instance_id=6001" --inst_attr=base_ip=192.168.56.11cm_ctl: edit res(VIP_az1) success. [omm@db2 cm_agent]$ cm_ctl res --edit --res_name="VIP_az1" --add_inst="node_id=2,res_instance_id=6002" --inst_attr=base_ip=192.168.56.12cm_ctl: edit res(VIP_az1) success.
[omm@db2 cm_agent]$ cm_ctl res --edit --res_name="VIP_az1" --add_inst="node_id=2,res_instance_id=6003" --inst_attr=base_ip=192.168.56.13cm_ctl: edit res(VIP_az1) success.
复制代码
查询 VIP 在哪个节点
[omm@db3 ~]$ cm_ctl show
[ Network Connect State ]
Network timeout: 6sCurrent CMServer time: 2023-08-03 06:18:42Network stat('Y' means connected, otherwise 'N'):| \ | Y | Y || Y | \ | Y || Y | Y | \ |
[ Node Disk HB State ]
Node disk hb timeout: 200sCurrent CMServer time: 2023-08-03 06:18:43Node disk hb stat('Y' means connected, otherwise 'N'):| N | N | N |
[ FloatIp Network State ]
node instance base_ip float_ip_name float_ip----------------------------------------------------------1 db1 6001 192.168.56.11 VIP_az1 192.168.56.10
复制代码
模拟主节点故障
[omm@db3 ~]$ cm_ctl stop -n 1cm_ctl: stop the node: 1.cm_ctl: stop node, nodeid: 1...........cm_ctl: stop node successfully.
复制代码
主节点切换到节点 2，VIP 也到了节点 2
[omm@db1 cm_agent]$ cm_ctl show
[ Network Connect State ]
Network timeout: 6sCurrent CMServer time: 2023-08-03 06:19:40Network stat('Y' means connected, otherwise 'N'):| \ | N | N || N | \ | Y || N | Y | \ |
[ Node Disk HB State ]
Node disk hb timeout: 200sCurrent CMServer time: 2023-08-03 06:19:41Node disk hb stat('Y' means connected, otherwise 'N'):| N | N | N |
[ FloatIp Network State ]
node instance base_ip float_ip_name float_ip----------------------------------------------------------2 db2 6002 192.168.56.12 VIP_az1 192.168.56.10
复制代码
节点 1 的 IP，已没有 192.168.56.10
[omm@db1 cm_agent]$ ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:04:f9:a3 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global dynamic noprefixroute enp0s3 valid_lft 74572sec preferred_lft 74572sec inet6 fe80::c8c2:7f4c:914f:a32d/64 scope link noprefixroute valid_lft forever preferred_lft forever3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:50:47:12 brd ff:ff:ff:ff:ff:ff inet 192.168.56.11/24 brd 192.168.56.255 scope global noprefixroute enp0s8 valid_lft forever preferred_lft forever inet6 fe80::890a:d968:b65d:f59a/64 scope link noprefixroute valid_lft forever preferred_lft forever
复制代码
节点 2 的 IP，已增加 192.168.56.10
[omm@db2 ~]$ ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:04:f9:a3 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global dynamic noprefixroute enp0s3 valid_lft 74550sec preferred_lft 74550sec inet6 fe80::c8c2:7f4c:914f:a32d/64 scope link noprefixroute valid_lft forever preferred_lft forever3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 08:00:27:41:73:29 brd ff:ff:ff:ff:ff:ff inet 192.168.56.12/24 brd 192.168.56.255 scope global noprefixroute enp0s8 valid_lft forever preferred_lft forever inet 192.168.56.10/24 brd 192.168.56.255 scope global secondary enp0s8:15400 valid_lft forever preferred_lft forever inet6 fe80::5373:d66d:7a39:ddc2/64 scope link noprefixroute valid_lft forever preferred_lft forever
复制代码
资源配置文件
[omm@db1 cm_agent]$ cat cm_resource.json{ "resources": [{ "name": "VIP_az1", "resources_type": "VIP", "instances": [{ "node_id": 1, "res_instance_id": 6001, "inst_attr": "base_ip=192.168.56.11" }, { "node_id": 2, "res_instance_id": 6002, "inst_attr": "base_ip=192.168.56.12" }, { "node_id": 3, "res_instance_id": 6003, "inst_attr": "base_ip=192.168.56.13" }], "float_ip": "192.168.56.10" }]
复制代码
同步配置文件到其余节点
scp cm_resource.json db2:/opt/huawei/data/cmserver/cm_agentscp cm_resource.json db3:/opt/huawei/data/cmserver/cm_agent
复制代码
启动节点 1
[omm@db3 ~]$ cm_ctl start -n 1cm_ctl: start the node: 1.cm_ctl: start node, nodeid: 1...........cm_ctl: start node successfully. [omm@db1 cm_agent]$ gs_om -t status --detail[ CMServer State ]
node node_ip instance state-----------------------------------------------------------------------1 db1 192.168.56.11 1 /opt/huawei/data/cmserver/cm_server Standby2 db2 192.168.56.12 2 /opt/huawei/data/cmserver/cm_server Primary3 db3 192.168.56.13 3 /opt/huawei/data/cmserver/cm_server Standby
[ Cluster State ]
cluster_state : Normalredistributing : Nobalanced : Nocurrent_az : AZ_ALL
[ Datanode State ]
node node_ip instance state-------------------------------------------------------------------------1 db1 192.168.56.11 6001 /opt/huawei/install/data/dn P Standby Normal2 db2 192.168.56.12 6002 /opt/huawei/install/data/dn S Primary Normal3 db3 192.168.56.13 6003 /opt/huawei/install/data/dn S Standby Normal
复制代码
现在 CM 的主节点和数据库的主节点在同一机器上了。
划线
评论
复制
发布于: 刚刚阅读数: 5
daydayup
关注
还未添加个人签名 2023-07-18 加入
还未添加个人简介
评论