The Synchronize Object (SYNCOBJ), Synchronize IFS (SYNCIFS), and Synchronize DLO (SYNCDLO) commands provide versatility for synchronizing objects and their authority attributes.
Where to run: The synchronize commands can be run from either system. However, if you run these commands from a target system, you must specify the name of a data group to avoid overwriting the objects on the source system.
Identifying what to synchronize: On each command, you can identify objects to synchronize by specifying a data group, a subset of a data group, or by specifying objects independently of a data group.
-
When you specify a data group, its source system determines the objects to synchronize. The objects to be synchronized by the command are the same as those identified for replication by the data group. For example, specifying a data group on the SYNCOBJ command, will synchronize the same library-based objects as those configured for replication by the data group.
-
If you specify a data group as well as specify additional object information in command parameters, the additional parameter information is used to filter the list of objects identified for the data group.
-
When no data group is specified, the local system becomes the source system and a target system must be identified. The list of objects to synchronize is generated on the local system. For more information about the object selection criteria used when no data group is specified on these commands, see Object selection for Compare and Synchronize commands.
Each command has a Synchronize authorities parameter to indicate whether authority attributes are synchronized. By default, the object and all authority-related attributes are synchronized. You can also synchronize only the object or only the authority attributes of an object. Authority attributes include ownership, authorization list, primary group, public and private authorities.
When you use the SYNCOBJ command to synchronize only the authorities for an object and a data group name is not specified, if any files processed by the command are cooperatively processed and a data group that contains these files is active, the command could fail if the database apply job has a lock on these files.
When to run: Each command can be run when the data group is in an active or an inactive state. You can synchronize objects whether or not the data group is active.
Using the SYNCOBJ, SYNCIFS, and SYNCDLO commands during off-peak usage or when the objects being synchronized are in a quiesced state reduces contention for object locks.
When using the SYNCIFS command for a data group configured for user journal replication, the data group can be active but it should not have a backlog of unprocessed entries.
Additional parameters: On each command, the following parameters provide additional control of the synchronization process.
-
The Save active parameter provides the ability to save the object in an active environment using IBM's save while active support. Values supported are the same as those used in related IBM commands. When you use this capability, the following parameters further qualify save while active operations:
-
On the SYNCOBJ and SYNCDLO commands, the Save active wait time parameter specifies the amount of time to wait for a commit boundary or for a lock on an object. If a lock is not obtained in the specified time, the object is not saved. If a commit boundary is not reached in the specified time, the save operation ends and the synchronization attempt fails.
-
On the SYNCIFS command, the Save active option parameter defaults to *NONE, which is appropriate for most users. Optionally, you can specify that the IBM command (SAV) should use the value *ALWCKPWRT for its SAVACTOPT parameter. This allows the object being saved to be opened for write when the checkpoint is achieved.
-
The Maximum sending size (MB) parameter specifies the maximum size that an object can be in order to be synchronized. For more information, see Limiting the maximum sending size .
-
The Ignored attribute processing parameter specifies how to process any attributes of the object being synchronized that are configured to be ignored by audits. Attributes that are configured to be ignored streamline distributing data in non-switchable data groups and in data groups whose target node is a replicate node of an application group. The default value, *KEEP, does not synchronize the ignored attributes and keeps the current values on each system in the data group. The value *REPLACE allows you to synchronize the ignored attributes using values from the source system objects. For more information, see Ignoring attribute differences in distributed environ-ments.