Shulou(Shulou.com)06/03 Report--

HACMP startup instructions:

In HACMP version 5.2 and before, you can see if the HACMP service is started by seeing if the clstrmgrES process is active. The specific methods are:

1. Clshowsrv-v command

2. C-SPOC tools in smit hacmp

3. The command c-g cluster

However, in HACMP versions 5.3 and 5.4, the clstrmgrES process is always active after the machine starts, so you cannot rely on the above method to determine whether the HACMP service is started or not. You can use the following methods to determine whether HACMP is started:

/ usr/es/sbin/cluster/utilities/clcheck_server grpsvcs | echo $?

If the output is 0, the HACMP service is not started, and if the output is 1, the HACMP service has been started. Full name is IBM High Availability Cluster Multiprocessing

HACMP interpretation

IBM high availability-multiprocessor-system cluster. HACMP is almost a cold backup solution for shared storage, provided that if one machine dies, the other machine will take over the application if the storage is not hung up, the application can be tolerated and the switching time is 10 minutes. If the business application cannot tolerate interruption, then Oracle's RAC is a better solution.

HACMP is not difficult, but since there are few opportunities to get in touch with it (since it is a cluster, of course it is an expensive system; since it is an expensive system, you can't start and stop at will), so I haven't had a chance to try. If you have a chance to try, write down the process notes and save them for future reference (screenshots are all AIX6.1).

[@ more@] part one: stop hacmp.

1. Start smitty Manager (execute under root user)

# smitty hacmp

2. Select the Manage HACMP Services interface to enter.

3. Select Stop Cluster Services on these nodes and press enter to stop hacmp.

The above steps can be directly entered into the later interface with # smitty clstop.

Part II: switching resource groups

If you need to replace the original A machine with the B machine in the hacmp, you need to switch the resource group, otherwise the disk array and the service IP cannot be used properly.

Still execute # smitty hacmp and select the HACMP Resource Group and Application Management option.

Select the Bring a Resource Group Online option and select the appropriate "node" (depending on whether you plan to start machine An or machine B)

Then, choose which resource group will be switched, which is usually your volume group.

Hit enter and see the interface below to indicate that the switch is successful.

Part 3: starting hacmp starting hacmp is similar to stopping hacmp. Use the command

# smitty clstart can do the operation.

Part IV: manually hanging the disk array to start the application.

When you want to use the two machines separately after stopping hacmp (regardless of the reason), because mounting the disk array is done in hacmp, if you do not start hacmp, the disk array cannot be mounted, so you need to mount the disk array manually. Execute the following two commands

# varyonvg rmvg-your vg name

# mount / home/oracle-your mount point

Similarly, the original application is placed in hacmp as a startup script, so if you don't start hacmp, you can't start the application automatically. If Oracle is running on the AIX, start Oracle and start the listener (you need to remind us that the listener is originally the service IP that listens to hacmp. To stop hacmp, you may need to modify the listening IP of the listener).

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.