Troubleshooting secondary TeamCity nodes failing to start in a multi-node setup

Symptoms

In a multi-node TeamCity setup, nodes fail to start when initialized in parallel. The presented error states, "It seems that there is another running TeamCity installation that is connected to the same database."

Cause

The problem occurs due to incorrect configurations related to node IDs, node-specific data directories, URLs, JVM options, and a potential misstep in expecting configurations in nodes-config.xml instead of the required JVM options.

Resolution

To resolve the issue, follow these steps:

1. Update JVM Options (Do not use nodes-config.xml for HA setup):

   • Set up the required JVM options for each node. These options include:
     • teamcity.server.nodeId=<node_ID> (Unique identifier for each node.)
     • teamcity.server.rootURL=<root_URL> (Root URL of the node for inter-node communication.)
     • teamcity.node.data.path=<Node-specific Data Directory> (Location of the node’s data.)
     • teamcity.data.path=<TeamCity Data Directory> (Optional - Path to the mount point of a shared TeamCity Data Directory)
     • teamcity.server.responsibilities=<responsibilities_list> (Optional - Comma-separated list of responsibilities to assign to this node)
   • Example configuration for a node:

   TEAMCITY_SERVER_OPTS=-Dteamcity.server.nodeId=<node_ID> -Dteamcity.server.rootURL=<node_root_URL> -Dteamcity.data.path=<TeamCity Data Directory> -Dteamcity.node.data.path=<Node-specific Data Directory>
   

2. Check for Application Logs:

   • Additional logs like teamcity-server.log, teamcity-startup.log, and teamcity-sql.log may provide further insights into anomalies during startup.

3. General Recommendations:

   • Review and apply configurations described in the official TeamCity Multinode Setup documentation: Multinode Setup.

By following the above steps, you should be able to resolve the initialization issue for secondary TeamCity nodes. If further problems persist, consider reaching out to TeamCity Support with updated logs and configurations or requesting a troubleshooting call for complex scenarios.