Chapter 4: Using the Replication Coordination Module

Recovering from switch active failover

When you use the SWITCH mode to fail over automatically, you must restart the active server and resume database connections, and so on, to recover the high-availability environment. This section describes the steps you must take to do this.

Table 4-6 uses the following acronyms:

ADB – the active database name
ADS – the active data server name
SDB – the standby database name
SDS – the standby data server name
LDS – the logical data server name
LDB – the logical database name

**Table 4-6: Steps to recover from automatic failover**
Step	Server	Example command line
1) Start the active Adaptive Server.	Active Adaptive Server	On Windows, `run [Active_ASE].bat` On UNIX, `dataserver -d [master_db_file] -R -c [config_file]`.
2) Resume the active Replication Server connections.	Replication Server	`resume connection to <ADS>.<ADB>`
3) Suspend connections to the standby Adaptive Server.	OpenSwitch	`rp_server_status <SDS>, 'LOCKED' rp_stop NULL, <SDS>, NULL, 1, 1`
4) Monitor the Replication Agent thread.	Standby Adaptive Server	`sp_help_rep_agent <SDB>, 'scan'` or watch for the last transaction in the active Adaptive Server
5 ) Stop the Replication Agent thread.	Standby Adaptive Server	`sp_stop_rep_agent <SDB>`
6) Switch the Replication Server connections.	Replication Server	`switch active connection for <LDS>.<LDB> to <ADS>.<ADB>`
7) Monitor the switching process.	Replication Server	`admin logical_status, <LDS>.<LDB>`
8) Start Active Adaptive Server Replication Agent.	Active Adaptive Server	`sp_start_rep_agent <ADB>`
9) Switch users back to active Adaptive Server.	OpenSwitch	`rp_server_status <ADS>, 'UP' rp_server_status <SDS>, 'UP' rp_switch NULL, <ADS>, NULL-1, <ADS> rp_start NULL, NULL, NULL`
10) Resume the standby Replication Server connections.	Replication Server	`Resume connection to <SDS>.<SDB>`

Unexpected failure of Replication Server

View this book as PDF