AIX双机热备实战操作

1. 环境准备 1.1 创建Concurrent VG 1.1.1 主机A创建Concurrent VG

# smit vg

Add a Volume Group with Data Path Devices

 VOLUME GROUP name                                       [datavg] +

Physical partition SIZE in megabytes                              256

* PHYSICAL VOLUME names        [vpath0 vpath1 vpath2 vpath3 vpath4 vpath5 vpath6 vpath7 vpath8 vpath9 vpath10 vpath11 vpath12 vpath13 vpath14 vpath15 vpath16 vpath17 vpath18 vpath19 vpath20 vpath21 vpath22 vpath23 vpath24 vpath25 vpath26 vpath27 vpath28 vpath29 vpath30]

Force the creation of a volume group                              no +

Activate volume group AUTOMATICALLY                  no +

at system restart?

Volume Group MAJOR NUMBER                                 [61]

Create VG Concurrent Capable?                                     yes

Auto-varyon in Concurrent Mode? no                                                                                                 

Create a big VG format Volume Group?           yes                                                                                                 

Create a scalable VG format Volume Group?       no 

*big vg是IBM为Oracle等数据库特定的一种VG。

1.1.2 主机A导出datavg信息给主机B

#smit vg

Export a Volume Group

   * VOLUME GROUP name                          [datavg]

导出后本节点看不到datavg。

1.1.3 主机A导入datavg信息

#smit vg

Import a Volume Group

   VOLUME GROUP name                       [datavg]

*PHYSICAL VOLUME name                    [vpath0]

       Volume Group MAJOR NUMBER               [61]

 

 

1.1.4 主机B导入datavg信息

# smitty vg

    Import a Volume Group

Type or select values in entry fields.

Press Enter AFTER making all desired changes.    [Entry Fields]

 

VOLUME GROUP name                       [datavg]

* PHYSICAL VOLUME name                  [vpath0] +

Volume Group MAJOR NUMBER                [61] +#

两台主机的VG major number要保持一致！

1.1.5 查看VG状态

创建成功后，查看主机A和主机B的pv：

# lspv

hdisk0  00f64acac2b7e530  rootvg  active

hdisk1  00f64aca6b8e4066  rootvg  active

vpath0  00f6391f42b893a7  datavg 

vpath1  00f6391f42b89562  datavg 

vpath2  00f64aca59fb8bd1  datavg 

vpath3  00f64aca5f5354fb  datavg 

vpath4  00f64aca5f53567c  datavg 

vpath5  00f64aca5f5357db  datavg 

vpath6  00f64aca5f535988  datavg 

vpath7  00f64aca5f535afb  datavg 

vpath8  00f64aca5f535c50  datavg 

vpath9  00f64aca5f535da0  datavg 

vpath10  00f64aca5f535ef7   datavg 

vpath11  00f64aca5f53606f   datavg 

vpath12  00f64aca5f5361d6   datavg 

vpath13  00f64aca5f53632a   datavg 

vpath14  00f64aca5f536484   datavg 

vpath15  00f64aca5f5365d9   datavg 

vpath16  00f64aca5f53672d   datavg 

vpath17  00f64aca5f536919   datavg 

vpath18  00f64aca5f536a86   datavg 

vpath19  00f64aca5f536c28   datavg 

vpath20  00f64aca5f536dad   datavg 

vpath21  00f64aca5f536f08   datavg 

vpath22  00f64aca5f5370cf   datavg 

vpath23  00f64aca5f537253   datavg 

vpath24  00f64aca5f5373d9   datavg 

vpath25  00f64aca5f537542   datavg 

vpath26  00f64aca5f5376da   datavg 

vpath27  00f64aca5f537873   datavg 

vpath28  00f64aca5f5379d3   datavg  

vpath29  00f64aca5f537b34   datavg  

vpath30  00f64aca5f537cd0   datavg  

vpath31  00f64aca5f59c330   None

vpath32  00f64aca5f59c4d4   None

vpath33  00f64aca792ecee2   None

vpath34  00f64aca792ed073   None

vpath35  00f6391f5afa6ccf    None    

vpath36  none           None    

vpath37  none           None    

vpath38  none           None    

vpath39  00f6391f76699d1f      None  

1.2 心跳线配置及测试

在选择哪个串口作为心跳线的端口时，原则是这样：

在串口富裕的情况下，尽量不要选择第1个和第2个。（因为第一个串口一般做console，第二个常用作远程维护程序使用）。

如果不这样，虽然同步和简单的测试没有问题，但在实际应用中回出现HACMP经常自动切换，所以应该不使用第1和第2个串口。

1.2.1 心跳线配置

#smitty tty

Add a TTY

  TTY type                                            tty

  TTY interface                                       rs232

  Description                                     Asynchronous Terminal

  Parent adapter                                      sa0

* PORT number                                        [0]     +

  Enable LOGIN                                        disable +

  BAUD rate                                           [9600] 

  PARITY                                             [none] +

  BITS per character                                     [8]    +

  Number of STOP BITS                                 [1]    +

  TIME before advancing to next port setting                 [0]    +

  TERMINAL type                                      [dumb]

  FLOW CONTROL to be used                            [xon]

 

主机2步骤与主机1相同。

1.2.2 心跳线测试

在两台主机上，通过lsdev –Cc tty查看新增的串行口，会有类似下面的内容：

tty0 Available 00-00-S3-00 Asynchronous Terminal

 

序号	主机	内容
1.	主机1	stty
2.	主机2	stty 这时在两台主机的命令行下会有内容的显示，否则，tty配置失败。示例如下： speed 9600 baud; -parity hupcl eol2 = ^? brkint -inpck -istrip icrnl -ixany ixoff onlcr tab3 echo echoe echok
3.	主机1	cat /etc/hosts > /dev/tty0
4.	主机2	cat < /dev/tty0 这时在主机2的命令行下有主机1的/etc/hosts文件的内容，否则，tty配置失败。

 

2. 安装HACMP 2.1 安装HACMP5.4

将HACMP V5.4 软件CD放入CD-ROM，在主控台上执行以下操作：

#smitty installp

Install Software

INPUT device / directory for software              [/dev/cd0]

SOFTWARE to install                [不选择cluster.haview,

cluster.hativoli]

        ACCEPT new license agreements?                 yes

2.2 升级到HACMP 5.4.1

由于Oracle 10g RAC需要HACMP版本为5.4.1，可以从IBM官方网站下载5.4.1的最新补丁。

#smitty installp

Update Installed Software to Latest Level (Update All)

    ACCEPT new license agreements?                 yes

2.3 查看HACMP版本

#lslpp -l |grep cluster

cluster.adt.es.client.include

cluster.adt.es.client.samples.clinfo

cluster.adt.es.client.samples.clstat

cluster.adt.es.client.samples.libcl

cluster.adt.es.java.demo.monitor

cluster.doc.en_US.es.html  5.4.1.0  COMMITTED  HAES Web-based HTML

cluster.doc.en_US.es.pdf   5.4.1.0  COMMITTED  HAES PDF Documentation - U.S.

cluster.es.cfs.rte         5.4.1.6  COMMITTED  ES Cluster File System Support

cluster.es.client.lib      5.4.1.7  COMMITTED  ES Client Libraries

cluster.es.client.rte     5.4.1.10  COMMITTED  ES Client Runtime

cluster.es.client.utils    5.4.1.9  COMMITTED  ES Client Utilities

cluster.es.client.wsm      5.4.1.7  COMMITTED  Web based Smit

cluster.es.cspoc.cmds     5.4.1.11  COMMITTED  ES CSPOC Commands

cluster.es.cspoc.dsh       5.4.1.0  COMMITTED  ES CSPOC dsh

cluster.es.cspoc.rte       5.4.1.6  COMMITTED  ES CSPOC Runtime Commands

cluster.es.plugins.dhcp    5.4.1.0  COMMITTED  ES Plugins - dhcp

cluster.es.plugins.dns     5.4.1.0  COMMITTED  ES Plugins - Name Server

cluster.es.plugins.printserver

cluster.es.server.cfgast   5.4.1.0  COMMITTED  ES Two-Node Configuration

cluster.es.server.diag    5.4.1.11  COMMITTED  ES Server Diags

cluster.es.server.events  5.4.1.11  COMMITTED  ES Server Events

cluster.es.server.rte     5.4.1.11  COMMITTED  ES Base Server Runtime

cluster.es.server.testtool

cluster.es.server.utils   5.4.1.11  COMMITTED  ES Server Utilities

cluster.es.worksheets      5.4.1.5  COMMITTED  Online Planning Worksheets

cluster.license            5.4.1.1  COMMITTED  HACMP Electronic License

cluster.msg.en_US.cspoc    5.4.1.0  COMMITTED  HACMP CSPOC Messages - U.S.

cluster.msg.en_US.es.client

cluster.msg.en_US.es.server

cluster.es.client.lib      5.4.1.7  COMMITTED  ES Client Libraries

cluster.es.client.rte     5.4.1.10  COMMITTED  ES Client Runtime

cluster.es.cspoc.rte       5.4.0.0  COMMITTED  ES CSPOC Runtime Commands

cluster.es.server.diag     5.4.0.0  COMMITTED  ES Server Diags

cluster.es.server.events   5.4.0.0  COMMITTED  ES Server Events

cluster.es.server.rte     5.4.1.11  COMMITTED  ES Base Server Runtime

cluster.es.server.utils   5.4.1.11  COMMITTED  ES Server Utilities

cluster.man.en_US.es.data  5.4.1.6  COMMITTED  ES Man Pages - U.S. English

 

 

3. OA数据库HACMP配置

 以下操作只需在一个节点上操作即可。

# smitty hacmp                                      

   Extended Configuration                    

     Extended Topology Configuration   

       Configure an HACMP Cluster

         Add/Change/Show an HACMP Cluster 

* Cluster Name              [oadb_cl]                        

节点

# smitty hacmp                                      

   Extended Configuration                    

        Extended Topology Configuration   

           Configure HACMP Nodes  

                 Add a Node to the HACMP Cluste

                    *NodeName                         [主机A]                 

                     Communication Path to Node          [主机A] +

以同样方法将主机B添加到Cluster中

                     *NodeName                        [主机B]                 

                     Communication Path to Node          [主机B] +

网络 网络

# smitty hacmp                                      

  Extended Configuration                    

      Extended Topology Configuration

        Configure HACMP Networks

             Add a Network to the HACMP Cluster

              * NetworkName                       [net_ether_01]          

* NetworkType                         ether                  

* Netmask                            [255.255.255.0] +

* Enable IP Address Takeover via IP Aliases  [No]        +

               IP Address Offset for Heartbeating over IP Aliases []                      

     

* Enable IP Address Takeover via IP Aliases  [Yes]  此选项决定了HACMP的IP切换方式，但值得一提的是只有“boot”、“standby”、“service”三个IP分别为三个不同网段时必须选用IP Aliases方式。如果“boot”、“standby”其中一个与“service”为同一个网段时必须选用IP Replace，则此选项应选“NO”。

# smitty hacmp                                      

  Extended Configuration                    

        Extended Topology Configuration

             Configure HACMP Networks

                Add a Network to the HACMP Cluster

                                     * NetworkName                [net_rs232_01]

* Network Type                 rs232 

 

通信接口

# smitty hacmp                                      

Extended Configuration                    

    Extended Topology Configuration

       Configure HACMP Networks

          Configure HACMP Communication Interfaces/Devices

             Add Communication Interfaces/Device

                Add Pre-defined Communication Interfaces and Devices

                    Communication Interfaces

                           * IP Label/Address       [主机A] +

* Network Type          ether                   

* Network Name         net_ether_01            

* Node Name           [主机A]    +

                         Network Interface      []

      同样方法添加另一个通信接口到”net_ether_01”

                          * IP Label/Address       [主机B] +

* Network Type          ether                   

* Network Name         net_ether_01            

* Node Name           [主机B]    +

                         Network Interface      []

      同样方法创建另一个通信接口

                        * IP Label/Address       [主机A_priv] +

* Network Type          ether                   

* Network Name         net_ether_02            

* Node Name           [主机A]    +

                        Network Interface      []

                        * IP Label/Address       [主机B_ priv] +

* Network Type          ether                   

* Network Name         net_ether_02            

* Node Name           [主机B]    +

                        Network Interface      []

# smitty hacmp                                      

 Extended Configuration                    

    Extended Topology Configuration

      Configure HACMP Networks

          Configure HACMP Communication Interfaces/Devices

             Add Communication Interfaces/Device

                 Add Pre-defined Communication Interfaces and Devices

                     Communication Devices

                         * Device Name           [osdb1_tty0]          

* Network Type           rs232                  

* Network Name          net_rs232_01           

* Device Path            [/dev/tty0]             

* Node Name            [主机A]    +

      以同样方法创建另一个串口通信设备

                         * Device Name           [osdb2_tty0]          

* Network Type           rs232                  

* Network Name          net_rs232_01           

* Device Path            [/dev/tty0]             

* Node Name            [主机B]    +

资源 Service IP

# smitty hacmp

Extended Configuration

Extended Resource Configuration

HACMP Extended Resources Configuration

Configure HACMP Service IP Labels/Addresses

Add a Service IP Label/Address

Configurable on Multiple Nodes

* IP Label/Address  主机A_svc

* Network Name    net_ether_01

同样方法配置第二个Service IP

* IP Label/Address  主机B_svc

* Network Name    net_ether_01

 

此处配置Service IP仅仅为了让HACMP区分Boot IP和Service IP，不配置Verify时会报错

3.6.2 创建资源组

# smitty hacmp                                      

 Extended Configuration                    

    Extended Resource Configuration

HACMP Extended Resource Group Configuration

           Add a Resource Group

* Resource Group Name                    [xckyres]              

                          Participating Nodes (Default Node Priority)[主机A 主机B]                     

                Startup Policy          Online On All Available Nodes > +

                  Fallover Policy               Bring Offline > +

                        Fallback Policy                Never Fallback > +

        

“Participating Nodes”选项决定了资源组中节点的优先级，写在前面的节点其优先级高于后面的节点。

 根据实际需求决定资源组的：“Startup Policy”、“Fallover Policy”、“Fallback Policy”。

Startup Policy：

n  Online On Home Node Only： 只在主节点启动。在 “Participating Nodes”中写在第一位的节点启动。

n  Online On First Available Node：在第一个启动的节点启动。在“Participating Nodes”中所选择的所有节点中最先启动的节点上启动。

n  Online Using Distribution Policy：按照分布策略启动

n  Online On All Available Nodes：在所有启动的节点中启动。做并发群集时选择此项。如：Oracle RAC

Fallover Policy：

n  Fallover To Next Priority Node In The List：节点失败时资源组迁移到下一优先级节点

n  Fallover Using Dynamic Node Priority：节点失败时动态选择迁移节点

n  Bring Offline (On Error Node Only)：将资源组下线

Fallback Policy：

n  Fallback To Higher Priority Node In The List：节点恢复时资源组返回优先级高的节点

n  Never Fallback：不进行资源组回迁

# smitty hacmp                                      

Extended Configuration                     

    Extended Resource Configuration

        Change/Show Resources and Attributes for a Resource Group

Resource Group Name             oadbcon_rg

Participating Nodes (Default Node Priority)     主机A 主机B

Startup Policy                  Online On All Available Nodes

Fallover Policy             Bring Offline (On Error Node Only)

Fallback Policy                             Never Fallback

Concurrent Volume Groups                    [datavg]   +

Use forced varyon of volume groups, if necessary    false    +

Automatically Import Volume Groups             false   +

Application Servers                           []     +

Tape Resources                              []      +

Raw Disk PVIDs                             []     +

Disk Fencing Activated                       false    +

Fast Connect Services                         []     +

Communication Links                         []     +

Workload Manager Class                       []    +

Miscellaneous Data                           []

 

配置

# smitty hacmp                                      

Extended Configuration

   Extended Verification and Synchronization

     * Verify, Synchronize or Both                            [Both] +

* Automatically correct errors found during            [Interactively] +

          verification?                                                              

* Force synchronization if verification fails?                [No] +

* Verify changes only?                                  [No] +

* Logging                                         [Standard] +

停止HACMP （快速路径：#smitty clstart）

# smitty hacmp

System Management (C-SPOC)

    Manage HACMP Services

        Start Cluster Services

          * Start now, on system restart or both                now  

            Start Cluster Services on these nodes       [主机A 主机B]                 

* Manage Resource Groups                  Automatically

  BROADCAST message at startup?                true

  Startup Cluster Information Daemon?              false 

            Ignore verification errors?                       false

            Automatically correct errors found during         Interactively         

            cluster start?    

 

（快速路径：#smitty clstop）

# smitty hacmp

System Management (C-SPOC)

   Manage HACMP Services

      Stop Cluster Services

          * Stop now, on system restart or both             now+

  Stop Cluster Services on these nodes       [主机A 主机B]

  BROADCAST cluster shutdown?               true+

4.1 HACMP同步失败

报错信息如下：

ERROR: Verification of Cluster Topology for RSCT failed.

See “/var/ha/log/topsvcs.default” for detailed information.

WARNING: File ‘netmon.cf’ is missing or empty on node gdapp1. This file is needed for a cluster with the single-adapter network net_rs232_01. Please create ‘netmon.cf’ file on node gdapp1 as described in ‘HACMP Planning and Installation Guide’.

WARNING: File ‘netmon.cf’ is missing or empty on node gdapp2. This file is needed for a cluster with the single-adapter network net_rs232_01. Please create ‘netmon.cf’ file on node gdapp2 as described in ‘HACMP Planning and Installation Guide’.

 

解决方法：

到两个节点的/usr/sbin/cluster目录下检查是否有netmon.cf文件，如果没有，请增加该文件，并增加内容；如果有，请检查内容，内容如下：

10.19.98.18

10.19.98.22

10.19.98.17

10.19.98.19

192.168.254.3

192.168.254.4

4.2 不能正常停止HACMP

 

以root用户执行

#ps -ef|grep ora

保证没有oracle进程运行

 

查看 hacmp进程

lssrc -g cluster 

 

可用 stopsrc -s service_name来停HACMP的子系统，例如：

#stopsrc -s clinfo