Troubleshooting Replication Failures

Isolate replication failures, which may occur after subscriptions have successfully materialized.

Before you troubleshoot a replication failure, verify that data is not replicating. The troubleshooting procedures listed here can be executed separately and in any order; however, they are listed in the order in which they are most likely to solve the problem. These procedures assume that the replication system has:

Been installed and configured correctly,
Successfully completed subscription materialization, and
Correctly replicated data previousl.y

After executing the procedures, run the replication system to check if the problem has been solved. If replication is still not executing correctly:

Check the error logs for error messages.
Perform any of procedures that you skipped.

Verifying That Data is Not Replicating
Verify that data is not replicating before troubleshooting a replication failure.
Identifying the Data That is Failing to Replicate
Determine the specific subscriptions and columns that are failing to replicate.
Verifying That Replication Server Threads are Up
Use admin who_is_down to determine if the primary and destination Replication Server threads are up.
Replication Server Is Down
If the primary, destination, or both Replication Servers are down, analyze the error log for each server.
Checking for Queue Problems
Determine whether there is an increase in the number of duplicate transactions.
Verifying That All RepAgents are Up
Use sp_who to view the Adaptive Server RepAgent thread status.
Checking System Tables
Check the Replication Server System Database (RSSD) tables for error information.
Finding Failed Replication Component
To determine the component on which replication has stopped, turn on traces and examine the output of sysadmin dump_queue.
Checking for Route Problems
Use rs_helproute to check for route problems.

Parent topic: Troubleshooting Overview