Clustering parameters

Whether to enable clustering on this application server.

Studio Read-only Mode

If you set the value of ClusteringEnabled to true in file config.xml on a particular application server and then restart the associated Studio, Studio becomes effectively read-only. In this context, read-only means:

• It is not possible to modify a Studio-managed resource. This applies, for example, to files that you open in the Gosu or Rules editor.
• If it is possible to modify a Studio-managed resource, it is not possible to save any modification you make to that resource. This applies, for example, to files that you open in the PCF editor.

To indicate the read-only status:

• Studio displays a padlock icon on the status bar that is visible only if Studio is in read-only mode. If you click the icon, Studio displays a modal message box indicating the reason why it is in read-only mode.
• Studio disables the Save button any time that Studio is in read-only mode.
• Studio changes the Save button tooltip in read-only mode to display the reason that save is not active in this mode. This is the same message that Studio shows if you click the padlock icon on the status bar.

Setting the value of configuration parameter ResourcesMutable to false provides the same Studio read-only behavior.

Default: false

Set for server: Yes

See also

• ResourcesMutable

The number of days to keep cluster member records before they can be deleted.

Default: 30

Minimum value: 1

Required: true

Sets the maximum number of retries for each cluster node to perform synchronized product model update operations. In a multi-node PolicyCenter cluster, you install product model changes on one server, then start that server to force the upgrade changes. After the first server completes the upgrade process, you can then start the other nodes in the server cluster simultaneously. Due to the number of concurrent server starts, it is possible for there to be contention for the shared update lock while updating the PolicyCenter product model on each node. This configuration parameter sets the maximum number of times that a server node attempts to acquire the update lock before failing.

This parameter works in conjunction with ClusteringProductModelUpdateRetryTimeout and ClusteringProductModelUpdateSleepTimeCeiling. See ClusteringProductModelUpdateSleepTimeCeiling for an explanation of how these parameters interact.

Default: 25 (retries)

Sets the time between successive attempts (retries) by a server node to perform synchronized product model update operations. In a multi-node PolicyCenter cluster, you install product model changes on one server, then start that server to force the upgrade changes. After the first server completes the upgrade process, you can then start the other nodes in the server cluster simultaneously. Due to the number of concurrent server starts, it is possible for there to be contention for the shared update lock while updating the PolicyCenter product model on each node. This configuration parameter sets the time to wait between successive attempts to acquire the update lock.

• (TIMEOUT) + (2 * TIMEOUT) + ... + ((n-1) * TIMEOUT) + (n * TIMEOUT)

This parameter works in conjunction with ClusteringProductModelUpdateMaxRetries and ClusteringProductModelUpdateSleepTimeCeiling. See ClusteringProductModelUpdateSleepTimeCeiling for an explanation of how these parameters interact.

Default: 5000 (milliseconds)

Sets the maximum time between successive attempts (retries) by a server node to perform synchronized product model update operations. In a multi-node PolicyCenter cluster, you install product model changes on one server, then start that server to force the upgrade changes. After the first server completes the upgrade process, you can then start the other nodes in the server cluster simultaneously. Due to the number of concurrent server starts, it is possible for there to be contention for the shared update lock while updating the PolicyCenter product model on each node. Each cluster server uses a retry/back-off approach in which each retry causes an increasing sleep time delay on that node. This configuration parameter sets the maximum allowable sleep time between successive attempts to acquire the update lock. PolicyCenter uses the minimum of the calculated sleep time and this parameter value to calculate the sleep time for each cluster node for each retry.

Default: 50000 (milliseconds)

Default: true

Set for server: Yes

Provides the BFO (Big Faceless Organization) license key needed for server-side PDF generation. If you do not provide a license key for a server, each generated PDF from that server contains a a large DEMO watermark on its face. The lack of a license key does not, however, prevent document creation entirely.

It is possible to set this value differently for each server node in the cluster.

Default: None

Whether PolicyCenter permits only those Java classes placed on a serialization whitelist to be deserialized. In some situations, it is possible for deserialized Java objects from untrusted sources to be used to perform remote code invocation attacks against a Guidewire application server.

For this configuration parameter:

• If false, PolicyCenter permits any Java class sent over a cluster channel to be deserialized, except for those classes specifically placed on a black (forbidden) list.
• If true, PolicyCenter permits only those Java classes placed on a white list to be deserialized if sent over a cluster channel. If enabled, it is not possible to place any class on the black list on the white list as well.

Define the permitted (whitelist) and forbidden (blacklist) Java classes in the following configuration files:

• serialization-whitelist.lst
• serialization-blacklist.lst

The Server Tools Serialization Info screen (under Info Pages) provides a means of monitoring the deserialization of Java classes.

Default: true

See also

• The Serialization Info screen