Manual peer recovery processing is not the default setting; you
must enable it through configuration. Administrator intervention is then required
to trigger any peer recovery processing.
About this task
Perform this task if you want to use manual peer
recovery for the transaction service. A scenario where you might want to use
manual peer recovery is when your file system does not provide the required
level of file locking support, and no constraints are in place to ensure that
overloading or network partitioning does not occur. For more information,
see How to choose between automated and manual transaction peer recovery
.
After
you configure manual peer recovery, your intervention is required only when
a server fails and cannot be restarted; in this case, use the administrative
console to specify which peer server performs recovery processing for the
failed server.
Manual peer recovery configuration is provided by a group
of policies known as static policies, where one policy definition
is provided per application server. Individual definitions are required to
define server-specific configuration within the policy which, in this case,
is the identity of the server that initiates a peer recovery process.
Procedure
- Create the required static policy definitions.
- In the administrative console click . All the members of a cluster reside in a single core group,
which by default is the DefaultCoreGroup core group. Click the core group
that contains your cluster. The Core groups configuration
panel is displayed.
- Under Additional Properties, click Policies.
The current set of policies is displayed. You can identify the
current policy that applies to transactional high availability (HA) by the
match criteria type = WAS_TRANSACTIONS. The default policy has a
policy type of One of N policy. If this policy exists,
you do not have to remove it because it is overridden by the static policy
that you are creating.
- Click New to create a new policy.
- Select Static policy and click Next.
- Provide a name for the policy. This value is a free-form text
string that is used only as a label in the administrative console. However,
to assist readability in the administrative console, use a name that associates
the policy with a particular server, for example, TM-SERVER1 policy.
- Enter a description if required, and click OK.
- Set values for the Is alive timer and
the Quorum options, if required.
- Under Additional Properties, click Match
criteria to open an empty match criteria panel.
- Click New to create a new match criterion.
Match criteria are key-value pairs that are used to define the scope
of a policy.
- Enter type in the Name field
and WAS_TRANSACTIONS in the Value field. Click OK.
This match criterion is used to associate the policy with the transaction
service.
- Click New to create a second new match
criterion.
- Enter GN_PS in the Name field
and cell_name\node_name\server_name in
the Value field, where cell_name, node_name and server_name are
the names of your cell, node and server. For example, dmgrCell\appnode1\server1.
Click OK. This match criterion is used
to associate the policy with a particular server.
- Return to the configuration page for the new policy, and under Additional
Properties, click Static group servers.
The Static group servers panel is displayed. This panel lists
all the servers in the core group and classifies them as either core group
servers or static group servers.
- From the Core group servers list, select
the server that is associated with the policy and click Add >> to
move it to the Static group servers list. The
static group servers list is the set of servers that attempts to own the recovery
logs at the same time. Adding an incorrect server to this list can compromise
data integrity. For normal operation, follow these rules to guarantee data
integrity:
- Add only one server to the list of static group servers. Adding two servers
causes recovery log contention, because both servers attempt to own the associated
recovery logs. The exception to this rule is adding a second server as part
of manual peer recovery initiation. For more information, see Managing manual peer recovery of the transaction service
.
- Add only the server that is associated with the policy to the list. Adding
a different server prevents the home server from owning its recovery logs,
and stops the home server from starting correctly.
- Repeat the procedure in step 1 for each server in the cluster.
- Save your changes to the master configuration, ensuring that you
select the Synchronize changes with Nodes check box.
- Stop and restart the cluster members for the changes to take effect.
Results
Manual transaction peer recovery is configured for all the servers
in the cluster. Automatic transaction peer recovery can no longer occur.