If a shutdown was unintentional or you need to resume operations, reattach the client by starting the Extract:
SQL> SELECT CAPTURE_NAME, STATUS, CLIENT_STATUS, ERROR_NUMBER, ERROR_MESSAGE FROM DBA_CAPTURE WHERE CAPTURE_NAME LIKE 'OGG$CAP_%';
When you see the message "OGG capture client successfully detached from goldengate capture," it means the GoldenGate Extract process (the client) has cleanly terminated its connection to the database-side logmining server. Common Causes of Detachment
To determine if this message requires action, you must look at the of the logs. Situation A: Normal (No Action Required)
The ggserr.log shows the client attaching and successfully detaching every few minutes, resulting in high replication lag. If a shutdown was unintentional or you need
First, check the overall health of your processes:
If the Extract encounters a critical error—such as hitting an unhandled data type, running out of memory, or losing network connectivity to a remote deployment—it will attempt to shut down safely. Part of this safe shutdown sequence involves detaching from the database server to prevent orphaned sessions.
The word "successfully" is critical. It implies that the detachment was planned or handled gracefully by the software architecture, rather than resulting from a crash, network abort, or sudden process termination.
The GoldenGate instance or the source database was being restarted. First, check the overall health of your processes:
Look for surrounding messages. If you see STOP requested by user , the detachment is normal. If you see ORA- errors immediately preceding the detachment, the client detached because the database encountered an issue. Step 3: Check Database-Side Capture Status
Look at the bottom of the file for the specific Oracle error (ORA-XXXXX) or OGG error that preceded the detachment. Step 2: Check Database Alert Logs and Trace Files
If you manually stop the Extract ( STOP EXTRACT ), this message confirms that the client has disconnected cleanly from the database resources.
In environments with low transactional volume, an Integrated Extract might hit an idle timeout threshold. If no new redo data is generated for an extended period, or if network instability disrupts the IPC (Inter-Process Communication) between GoldenGate and the database instance, the client may detach to prevent resource locking. 4. Critical Database Errors (ORA- Errors) It implies that the detachment was planned or
Configure the GoldenGate Manager process with AUTORESTART parameters to automatically revive the Extract if it detaches due to transient network or database blips.
In Oracle GoldenGate (OGG), the message "OGG Capture Client successfully detached from GoldenGate Capture"
The "detached" message signifies that the communication link between the GoldenGate Extract process and the underlying database capture engine has been cleanly severed. In simpler terms, the client (GoldenGate) told the server (the Database) it was finished, and the database acknowledged the disconnection. Common Scenarios for This Message 1. Graceful Shutdown
The Extract is waiting for the database to complete processing a massive, uncommitted long-running transaction.
A: First, stop the Extract with STOP EXTRACT <group> . Then, unregister it from the database with UNREGISTER EXTRACT <group> DATABASE before deleting it with DELETE EXTRACT <group> .
If the STREAMS_POOL_SIZE is too small, the logminer may force a detachment. Logminer Errors: Look for ORA-01280 or similar errors. 3. Verify Extract Checkpoints