Server Status Tab

The Controller is responsible for accepting work messages from the Administrator and distributing those work orders to the Event Producer and Event Consumer. The Controller manages error messages, as well as work messages and is responsible for starting and stopping the Event Consumer. It resides on the same system as the Event Consumer but does not need to be on the same system as the Event Producer or the target database.

The Event Producer resides on the same system as Adabas and is loaded into the process space of the Adabas nucleus. When changes are made in Adabas files that are identified for replication, the Event Producer stores those changes in a persistent store on the hard drive. When it receives an ET for a transaction, the Event Producer assembles all the elements of the transaction into a message and puts it on the message queue.

The Event Consumer is responsible for taking transactions off the message queue and executing them on the target database. Once a transaction has been successfully committed, it will be removed from the message queue. There is one Event Consumer for each source DBID/target database pair. If two Adabas DBID's are being replicated to a single target database, there will be two Event Consumers; one for each source/target combination. In addition to transaction processing, the Event Consumer is also responsible for Initial State processing. The number of initial states or transactions it can process in parallel is determined by the Parallel transaction count setting on the Config Servers dialog. This dialog can be accessed from the Servers menu on the Replication Design Tab. Under each Consumer entry on the grid is a list of replications currently being processed for that Event Consumer. Each replication entry has the following columns:

Column name	Description
Replication	This is the name of the source table followed by the name of the target table.
State Description	This tells the current state of the replication. See below for a complete list of State Descriptions.
Initial State Start	This is the date and time the last initial state was started for this replication.
Initial State End	This is the date and time the last initial state ended.
Total (HH:MM:SS)	The total time the last initial state took for this replication.
Messages	If there are any error messages associated with this replication, they will be displayed here. You can double click on this field to display the entire message in an edit box for easier reading.

The Event Consumer can show performance statistics about the replications if the controller has the CTRL_PERF_ENABLED registry setting set to 1. The amount of time each statistics period is collected before it is recorded is 30 seconds by default and can be set with the CTRL_PERF_QUERY_INTERVAL registry setting. Once the controller is recording and sending back statistics, they can be hidden on the Replication Administrator for clarity by unchecking View -> Show Statistics. The following statistics are recorded:

Statistic	Description
Consumer start date	The day and time the consumer first started replicating after a deploy.
Status Interval length	The length of time in seconds for which statistics are gathered before recording them. Set in the registry on the controller.
Statistics Totals	Consumer Since Start - These are the total values collected since the engine was started with the first deploy. Last Interval - These are the total values collected during the last statistics recording interval (as set by the CTRL_PERF_QUERY_INTERVAL registry setting).
Statistics Rates	Consumer Total - These are the rates for this consumer. Rates are determined as amount per second. Record Loader (i) - These are the rates for each Record loader that the consumer is running. The number of Record loaders is set in the Replication administrator Config Servers screen with the "Parallel transaction count".
Initial States	The number of initial states processed by this consumer.
Transactions	Number of transactions the consumer has processed and committed to the target.
Rows Processed	Number of individual rows processed by the consumer and sent to the target.
Events	Number of messages from the source processed and sent to the target by the consumer.
Inserts	Number of rows inserted into the target by the consumer.
Updates	Number of rows updated in the target by the consumer.
Deletes	Number of rows deleted in the target by the consumer.
Errors	Number of errors recorded during replication by the consumer.

State	Description
Replicating	Active and replicating.
Not replicating - offline	This indicates that either the target database has gone down or replication has been stopped by pressing the Stop Targets button (see below). This is a Consumer status message.
Replication offline	This indicates that an individual replication is offline either due to an unrecoverable error on the target or due to pressing the Stop Targets button (see below). This is a Replication status message.
Processing Initial State...	On the Consumer, this indicates that it is processing initial states. Each replication that is actively being processed will also indicate this status.
Initial State Pending	This indicates that an initial state has been requested for a replication, but it has not started yet.
Initial State Complete	This indicates that an initial state has finished for this replication but the overall process has not yet been completed by the Event Consumer. Once all pending initial states have completed, the consumer and all replications with a successful initial state will go to a Replicating status.
Initial State Failed	This indicates that an error occurred during initial state processing. Check the Messages field for more information about the error.
Deploy Failed	This indicates that an error occurred during the deploy process. Check the Messages field for all components for more information about the error.
No Active Configuration	This indicates that nothing has been deployed to the controller.
Status of the System Unknown/Unable to contact controller	This indicates that the Replication Administrator was not able to contact the controller to get status. Common causes of this message are the controller is not running or the message queues on either the Administrator machine, the EP machine or the controller machine are not running.