Oracle GoldenGate Network Troubleshooting: Practical Tools and Tips

Overview

When Oracle GoldenGate replication begins to lag or behave unpredictably, one of the most critical areas to investigate is the network. GoldenGate depends on a consistent and reliable transport layer to move transactional data in real time, which means any issues—whether related to latency, packet loss, jitter, firewall restrictions, or routing inconsistencies—can disrupt replication flow or lead to performance degradation.

Although certain tuning parameters exist at both the operating system (e.g., TCP socket buffers, kernel settings) and GoldenGate level (e.g., BATCHSQL, GROUPTRANSOPS, TCPBUFSIZE, COMPRESS, Parallel Replicat), these can only optimize replication performance within the boundaries of the underlying network conditions. Ultimately, the effectiveness of GoldenGate in high-volume or distributed environments is still highly dependent on the health, stability, and configuration of the network path between source and target.

This blog introduces a set of practical tools and techniques to help you troubleshoot GoldenGate replication issues by examining network-level behavior.

Key Tools for Network Troubleshooting

1. ping – quick health check that verifies reachability and measures latency.

• A TCP/IP utility that transmits a datagram to another host, specified in the command.
• If the network is functioning properly, the receiving host returns the datagram.

What to Look For:

• RTT (min/avg/max/mdev) – latency profile
• Packet Loss – must be 0% for GG health
• High mdev – inconsistent timing (jitter), often missed but critical

 Basic Usage:

 

Sample Output:

Simulate GoldenGate Load:

 

Sample Output:

 

Quiet Summary Only:

 

Quiet Output Sample:

 

GoldenGate Tip: Use -s to simulate real traffic size. Watch for high mdev, even with 0% loss—it can cause Extract/Replicat lag.

2. traceroute / tracert – trace the path and used to identify latency hops, misroutes, and intermittent path flaps.

• A TCP/IP utility that determines the route data takes to get to a particular destination.
  • Shows the exact path data takes to reach the server, listing each stop (router) along the way and how long each stop takes.
• Can help determine a point of latency or where packets are being lost in the network.

Example:

 

Output:

The example output above shows:

• The intermediate hops (routers) between the source and target network destinations.
• The amount of time it took for the data packet to be received and acknowledged along the path.
• Each hop shows the packet was received by different routers.
  • By default, traceroute send 3 packets to each hop.
  • There is no guarantee that each packet will be routed in the same way.
  • Load balancing, network congestion, or a host of other factors determine how a packet may be routed.

 

OCI Note: In OCI and cloud-based environments, ping is often blocked, especially when crossing VCNs or between on-prem and cloud. In such cases, using traceroute with TCP-based packets is more effective.

TCP-Based Traceroute Example:

Example:

-T: Sends TCP SYN packets instead of UDP

-p: Target port (e.g., 443 for HTTPS or NGINX-based GoldenGate Microservices)

Output:

3. ifconfig / ipconfig – Network Interface Status

• A Linux (ifconfig) / Windows (ipconfig) TCP/IP utility that verifies network settings and connections.
• It returns a host’s IP address, subnet mask and default gateway, alongside other important network information.

Linux:

 

Output:

 

Windows:

 

Output:

4. netstat – Port Visibility

• A utility that shows the status of each active network connection.
• This tool is useful for finding out what services are running on a particular system.

Example:

Output:

5. tcpdump – Packet Capture

• A utility that is used to obtain packet information from a query string sent to the network interface.

Example:

 

Output:

6. iptables – Linux Firewall Validation

• A Linux firewall program that protects a network.
• You can use this tool if you suspect that your firewall may be too restrictive or too lenient.

Example:

 

Output:

7. route – Routing Table Review

• A command that enables manual updating of the routing table.
• It can be used to troubleshoot static routing problems in a network.

Example:

 

Output:

8. ss – Socket Statistics

• A modern replacement for netstat, used for displaying active socket connections, including ports used by GoldenGate.

Example:

 

Output:

9. nc – Netcat for Port Connectivity Tester

• nc (Netcat) is a powerful utility for testing TCP port connectivity and can be an effective substitute for ping, particularly when ICMP is blocked or disabled. It’s especially useful for verifying GoldenGate Service manager, Receiver, Admin Server ports are reachable and listening.

Example:

Output:

10. curl -v telnet://<ip>:<port> – TCP Client simulation

• curl is traditionally used for HTTP calls, but when used with the telnet:// protocol, it can serve as a substitute for ping or nc when testing raw TCP connectivity—particularly useful for verifying GoldenGate Service Manager, Receiver, or Admin Server ports through firewalls, proxies, or constrained environments like OCI.

Example:

Output:

A successful response confirms that the port is reachable and listening. This method works even when ping is blocked by firewalls or ICMP is disabled—making it a powerful lightweight alternative to nc or telnet for quick connectivity checks.

Diagnosis Example: High mdev with No Loss

 

• 0% packet loss
• mdev = 68.9ms → Indicates severe jitter
• Replication may lag, Extract may appear stuck or in recovery

Additional References

To further expand on the topics covered in this blog, consider reviewing these Oracle GoldenGate A-Team and product blog articles:

• Optimizing Network Performance in Hub-Based Oracle GoldenGate Extract Configurations

• How to Use Network Path Analyzer to Troubleshoot OCI GoldenGate Connectivity Issues

These resources provide in-depth insights into optimizing performance and diagnosing advanced connectivity problems in OCI and distributed OGG deployments.

Summary Table

Tool	Purpose
ping	Check latency and packet loss
traceroute	Identify routing problems and delays
ifconfig / ipconfig	Interface and IP validation
netstat	Port state and session activity
tcpdump	Packet-level inspection of replication
iptables	Ensure firewall allows GG traffic
route	Confirm gateway and route correctness
ss	Modern socket viewer showing port bindings and process associations
nc (Netcat)	Substitute for ping; validate port-lvel connectivity for GoldenGate components
curl -v telnet://<ip>:<port>	Lightweight TCP port test when ICMP or Netcat is blocked; ideal for firewall checks

Conclusion

Network stability is a critical foundation for Oracle GoldenGate replication. Issues like high latency, jitter (mdev), or intermittent connectivity often manifest as lagging Extracts, stuck Replicats, or recovery loops—even when database and GoldenGate configurations are sound.

By using the diagnostic tools outlined in this blog (ping, traceroute, tcpdump, netstat, nc, ss, etc.), you can isolate and validate network issues before escalating. These tools also improve collaboration with DevOps and Network teams by providing clear, actionable evidence.

Leveraging these tools and insights enables faster diagnosis and resolution of network-related issues, helping maintain stable and efficient GoldenGate replication across diverse environments.