Examples

Use configure replication server with the set block_size to ‘block_size’ with shutdown clause to set the queue block size.

NoteYou must include the “with shutdown” clause after specifying the block size.

For convenience in the examples that follow, RSSD refers to both Adaptive Server as the Replication Server System Database (RSSD) and SQL Anywhere as the Embedded Replication Server System Database (ERSSD).

Example 1 This example sets the queue block size of the primary and replicate Replication Servers to 64KB in a simple replication system consisting of a primary database—pdb, a replicate database—rdb, the primary Replicate Server—PRS with its RSSD—pRSSD, and the replicate Replication Server—RRS with its RSSD—rRSSD:

        pRSSD      rRSSD
           |          |
pdb -----> PRS -----> RRS -----> rdb

To configure PRS:

  1. Suspend log transfer from all Replication Agents. At PRS, execute:

    suspend log transfer from all
    
  2. Quiesce PRS:

    admin quiesce_force_rsi
    
  3. Configure the block size at PRS:

    configure replication server
    set block_size to ‘64’ with shutdown
    

    When you execute the command, the command:

    1. Verifies that there is no subscription materialization in progress, otherwise an error message is returned.

    2. Verifies that all log transfer is suspended, otherwise an error message is returned.

    3. Verifies that all incoming routes are suspended, otherwise an error message is returned.

    4. Verifies that the Replication Server is quiesced, otherwise an error message is returned.

    5. Purges queues.

    6. Zeroes the values in the rs_locater RSSD system table to allow Replication Agents to resend transactions that may have not been applied to the replicate database when you started the queue block size change procedure.

    7. Sets the queue block size to 64KB.

    8. Shuts down Replication Server to clear any memory that may be allocated in the old queue block size.

  4. See the transaction log to verify if PRS is not materializing, if log transfer and routes are suspended, and if PRS is quiesced.

  5. Restart PRS. See “Starting Replication Server,” in Chapter 4, “Managing a Replication System” in the Replication Server Administration Guide Volume 1.

  6. See the PRS transaction log to verify that the block size is changed.

  7. Resume log transfer to allow Replication Agents to connect to PRS. At PRS execute:

    resume log transfer to all
    
  8. Check the RRS log file for information about data losses. Ignore data loss occurring from pRSSD to rRSSD by executing the ignore loss command on RRS.

    ignore loss from PRS.pRSSD to RRS.rRSSD
    

    See “Ignoring a loss”.

To configure RRS:

  1. Suspend log transfer from all Replication Agents. At PRS and RRS, execute:

    suspend log transfer from all
    
  2. Quiesce PRS:

    admin quiesce_force_rsi
    
  3. At all Replication Servers that originate routes to the RRS, suspend the routes:

    suspend route to destination_replication_server
    
  4. Quiesce RRS:

    admin quiesce_force_rsi
    
  5. Configure the block size at RRS:

    configure replication server
    set block_size to ‘64’ with shutdown
    
  6. See the transaction log to verify if RRS is not materializing, if log transfer and routes are suspended, and if RRS is quiesced.

  7. Restart RRS. See “Starting Replication Server,” in Chapter 4, “Managing a Replication System” in the Replication Server Administration Guide Volume 1.

  8. See the RRS transaction log to verify that the block size is changed.

  9. Resume log transfer to allow Replication Agents to connect to RRS, at RRS execute:

    resume log transfer to all
    
  10. Resume log transfer to allow Replication Agents to connect to PRS, at PRS execute:

    resume log transfer to all
    
  11. Resume the routes you suspended:

    resume route to destination_replication_server
    
  12. Check the PRS and RRS log files for information about data losses. Ignore data loss occurring between pRSSD and rRSSD if rRSSD is replicated to pRSSD by executing the ignore loss command on PRS.

    ignore loss from RRS.rRSSD to PRS.pRSSD
    

Example 2 This example sets the queue block size of the primary Replication Server to 64KB in a replication system with an intermediate route consisting of a primary database—pdb, a replicate database—rdb, the primary Replicate Server—PRS with its RSSD—pRSSD, the replicate Replication Server—RRS with its RSSD—rRSSD, and an intermediate Replication Server—IRS with its RSSD—iRSSD:

        pRSSD       iRSSD       rRSSD
           |           |           |
pdb -----> PRS ------> IRS ------> RRS -----> rdb

  1. Suspend log transfer from all Replication Agents. At PRS, execute:

    suspend log transfer from all
    
  2. Quiesce PRS:

    admin quiesce_force_rsi
    
  3. Configure the block size at PRS:

    configure replication server
    set block_size to ‘64’ with shutdown
    
  4. See the transaction log to verify if PRS is not materializing, if log transfer and routes are suspended, and if PRS is quiesced.

  5. Restart PRS. See “Starting Replication Server,” in Chapter 4, “Managing a Replication System” in the Replication Server Administration Guide Volume 1.

  6. See the PRS transaction log to verify that the block size is changed.

  7. Resume log transfer to allow Replication Agents to connect to PRS. At PRS execute:

    resume log transfer to all
    
  8. Check the IRS and RRS log files for information about data losses. Ignore data loss occurring from pRSSD to RRS and from pRSSD to iRSSD by executing the ignore loss command twice on IRS:

    ignore loss from PRS.pRSSD to RRS
    go
    ignore loss from PRS.pRSSD to IRS.iRSSD
    go
    
    

See the Replication Server Reference Manual for the full syntax, examples, and usage information for the commands in the examples.