OCI GoldenGate Replication Lag guide with SQL Server Source - Troubleshooting Steps Needed
I'm dealing with I've encountered a strange issue with I'm stuck trying to I'm working with a important replication lag scenario with Oracle Cloud Infrastructure (OCI) GoldenGate while trying to set up a replication from an on-premises SQL Server to an OCI database. The configuration seems correct, but I have noticed that the lag time occasionally spikes to over 5 minutes, which is unacceptable for our use case. To troubleshoot, I've checked the following: - **GoldenGate Version**: Using GoldenGate 19c. - **Network Latency**: The network connection to OCI has been tested and shows a consistent latency of < 50ms. - **Batch Size**: Set the `FETCHSIZE` parameter to 1000 in the GoldenGate parameter file for the extract process. Here’s a snippet of my extract configuration: ``` EXTRACT ext1 USERIDALIAS ggs_user EXTTRAIL /u01/app/ogg/dirdat/lt TABLE source_db.dbo.*; FETCHSIZE 1000; ``` Additionally, I have checked the performance metrics in OCI and found that the target database is not under heavy load during these spikes. However, the GoldenGate process shows a steady increase in lag during peak application usage times. I've also tried increasing the `MAXSKIP` and `RBA` settings, but that didn’t seem to help. Furthermore, I examined the GoldenGate logs and encountered the following errors intermittently: ``` behavior: GG-100001 Fast-Extract process stopped due to a high-lag condition. ``` Could anyone suggest additional steps to diagnose or resolve this lag scenario? Any insights on tuning parameters for better performance or other best practices for GoldenGate replication with SQL Server would be greatly appreciated. Any help would be greatly appreciated! The stack includes Sql and several other technologies. Any ideas how to fix this? For reference, this is a production web app. Any suggestions would be helpful.