Clustering: Troubleshooting

Clustering in CompleteFTP can provide significant advantages, such as load balancing and high availability, but like any distributed system, it may occasionally encounter issues. This section addresses common problems, their causes, and solutions to help you effectively maintain a stable and reliable CompleteFTP cluster.

1. Common Clustering Issues

1.1 Failure to Add a Secondary Server

Symptoms:

• Unable to add a secondary server to the cluster.
• Error messages related to connection failures or invalid credentials.

Potential Causes:

• IP Filtering Rules: The secondary server may not allow connections from the primary server’s IP address.
• Firewall Settings: The firewall on either the primary or secondary server may be blocking required ports (default: 14983 for SFTP-based communication).
• Incorrect Admin Credentials: The admin username or password entered when attempting to add the secondary server may be incorrect.
• Version Mismatch: The primary and secondary servers may be running different versions of CompleteFTP.

Solutions:

1. Verify IP Filtering Rules: Check that the secondary server allows connections from the primary server’s IP address. Adjust the IP filtering settings in the Admin Site on the secondary server to permit connections from the primary server.
2. Check Firewall Settings: Ensure the firewalls on both the primary and secondary servers are configured to allow communication on port 14983 for SFTP-based server-to-server communication.
3. Double-check Admin Credentials: Ensure the admin account exists on the secondary server and the correct username and password are being used. Verify that the account has the necessary permissions to join the cluster.
4. Confirm Version Compatibility: Verify that both the primary and secondary servers are running the same version of CompleteFTP. If not, upgrade the servers to the same version before proceeding.

1.2 Synchronization Failures Between Primary and Secondary Servers

Symptoms:

• Secondary servers fail to synchronize with the primary server.
• Error messages such as "Synchronization failed" or "Unable to connect to primary server."

Potential Causes:

• Network configuration: Ensure that both the primary and secondary servers are allowed to communicate over the network. Common causes include closed firewall ports or incorrect IP filtering.
• Version mismatch: All servers in the cluster must be running the same version of CompleteFTP. Version differences can cause synchronization to fail.
• IP Filtering: The IP filtering settings on the secondary servers may not allow incoming connections from the primary server.

Solutions:

1. Verify firewall and network settings: Ensure the necessary ports (default: 14983 for SFTP) are open between the primary and secondary servers. Also, ensure the network allows bidirectional communication.
2. Check IP filtering rules: On each secondary server, allow the primary server’s IP address through the IP filtering settings. To modify these settings, connect to each secondary server, go to the Admin Site settings in the Sites panel, and ensure the primary’s IP is whitelisted.
3. Confirm version compatibility: Check that all servers in the cluster are running the same version of CompleteFTP. If needed, upgrade all servers to the same version before attempting to synchronize again.
4. Force synchronization: In the CompleteFTP Manager, go to the Servers panel and select "Force configuration update" on the problematic secondary server.

1.3 File Synchronization Issues

Symptoms:

• Files are not synchronized between servers in the cluster.
• Inconsistent file availability across different servers.

Potential Cause:

• Misunderstanding of the clustering feature: CompleteFTP clustering does not synchronize files between servers—it only synchronizes server configurations.

Solutions:

• External file synchronization: Use an external method such as DFS (Distributed File System) or shared network storage (NAS/SAN) to manage file synchronization across servers. Ensure that your file synchronization solution is properly configured and suited to your network architecture.

1.4 Primary Server Fails or Becomes Unavailable

Symptoms:

• The primary server crashes or becomes unresponsive.
• Secondary servers can no longer receive configuration updates.

Potential Causes:

• Hardware failure or software crash on the primary server.
• Network issues causing loss of connectivity between the primary and secondary servers.

Solutions:

1. Promote a secondary server to primary: In the CompleteFTP Manager, right-click the desired secondary server and select "Select as primary". This promotes the secondary to act as the new primary server, restoring configuration synchronization across the cluster.
2. Fix or replace the old primary server: Once the new primary is functioning, troubleshoot the failed server or decommission it. If the original primary server is restored, you may want to re-add it as a secondary.

1.5 Licensing Issues During Clustering Setup

Symptoms:

• License activation errors during the addition of secondary servers.
• Secondary servers fail to synchronize licensing from the primary.

Potential Causes:

• Missing licenses for secondary servers.
• Network connectivity issues preventing license validation.

Solutions:

1. Check licensing requirements: Each server in a cluster requires its own Enterprise edition license. Verify that licenses are valid and activated on both primary and secondary servers.
2. License troubleshooting: If licensing issues persist, contact support for assistance with reactivating or validating your licenses.

1.6 Unbalanced Server Load

Symptoms:

• Client connections are not properly distributed across servers in the cluster.
• Some servers experience heavy loads while others remain underutilized.

Potential Causes:

• Load balancer configuration: An external load balancer may not be properly routing traffic across all secondary servers, resulting in uneven load distribution.

Solutions:

1. Review the external load balancer configuration: Ensure that the load balancer is configured to distribute traffic evenly across all servers in the cluster. This includes setting up appropriate rules and ensuring health checks are working correctly.
2. Sticky sessions for HTTP/S: If you are using HTTP or HTTPS, ensure that sticky sessions (session persistence) are enabled to route clients to the same server for the duration of their session.
3. Test and adjust load distribution: Regularly test how traffic is being distributed by the load balancer to ensure that no server is being overloaded while others are idle. Adjust load balancer rules as necessary.

Note: CompleteFTP itself is not responsible for managing the distribution of client connections. Proper traffic balancing must be handled by the external load balancer used in your environment.

2 Diagnostic Tools and Logs

2.1 Reviewing Logs

CompleteFTP provides detailed logs that can help diagnose issues in the cluster. Logs can be accessed directly from the server’s file system.

Log Location: Logs for each server are located in C:\ProgramData\Enterprise Distributed Technologies\Complete FTP\Logs on each server. These logs can provide insights into any network-related or configuration errors affecting the clustering setup.

It is recommended that users access secondary server logs directly from the file system, as the CompleteFTP Manager typically only shows logs for the server to which it is connected (usually the primary server).

2.2 Enabling Debug Logging

In the event of recurring issues, you can increase the level of detail captured in the logs by enabling debug logging. This can help capture more specific information about the server's operations and errors.

How to Enable Debug Logging: In the CompleteFTP Manager, go to Settings > Logging and set the logging level to Debug for more detailed diagnostic information.

3. Preventative Maintenance and Best Practices

• Regular Backups: Regularly back up the primary server’s configuration to ensure you can quickly restore your cluster in the event of a failure.
• Monitor Server Health: Use monitoring tools to keep track of server performance, network connectivity, and storage usage to prevent resource exhaustion.
• Test Failover Procedures: Periodically test your failover procedures by promoting a secondary server to primary and verifying that synchronization and configuration management work as expected.
• Keep Servers Up-to-Date: Ensure all servers are running the latest version of CompleteFTP to benefit from bug fixes and performance improvements.

By following these troubleshooting steps and best practices, you can maintain a robust and resilient CompleteFTP cluster that meets your high availability and performance needs.