Agent Incidents

Applies to: Tableau Advanced Management

Agents Unlicensed (critical)

Incompatible Agent Version (critical)

Agent Message Queue Credential Rotation Failure (critical)

Agent Down (warning)

In an effort to align with our company values of Equality, we have changed non-inclusive terminology where possible. Because changing terms in certain places can cause a breaking change, we maintain the existing terminology. So, you may continue to see the terms in CLI commands and options, installation folders, configuration files and other instances. For more information, see About Tableau Help(Link opens in a new window).

Agents Unlicensed

Agents Unlicensed incidents will be logged as critical when we detect that your Tableau Server has not been properly licensed to use Tableau Resource Monitoring Tool.

To monitor a Tableau Server using the Resource Monitoring Tool your Tableau Server must be licensed with the Tableau Advanced Management.

Agents will no longer collect data in the unlicensed state and send to the RMT Server. You can, however, continue to see the data that was previously collected when the environment was properly licensed.

To resolve this issue, license Tableau Server with Advanced Management. For more information, see About Tableau Advanced Management on Tableau Server.

Note: It may take up to an hour after adding a valid licence to Tableau Server, for the incident to be cleared.

There are other possible causes that can cause an Agents Unlicensed incident:

  • Tableau Server API credentials may not be correct. In the environment configuration page, verify that the username and password are correct and that the account used has the Server Administrator role on Tableau Server.

  • Unable to connect to Tableau Server REST API. Make sure that REST API is enabled on Tableau Server and that Tableau Server is running.

    To resolve the REST API connectivity issue, from the Admin menu, select Environments, click Edit Environment on the environment you want to modify, and in the Environment Details tab, click the Test Connection button in the Tableau Server REST API section on the right side of the page.

  • The RMT Server was unable to connect to Tableau Server, likely due to a network connectivity or similar issue.

Incompatible Agent Version

Agent-incompatible incidents will be logged as critical when one or more Agents in the environment are on a version that is not compatible with the RMT Server version.

To resolve the issue, you need to make sure that the Agent is on a version that is compatible with the current RMT Server version.

Use one of the following to resolve this issue:

  • We recommend that the RMT Server and all Agents be on the same version.
    • If the Agent is on a version that is earlier than the minimum compatible version, then you must upgrade the Agent to at least the minimum compatible version that is listed in the incident report.
    • If the Agent is on a version that is later than the RMT Server version, we recommend that you upgrade the RMT Server to that same later version as well. You may need to upgrade other Agents depending on the version that they are currently on.

    Upgrading Agents:

    Use the following steps to upgrade the Agent:

    1. Download the Agent setup program and copy it to the Tableau Server node where the Agent needs to be upgraded. The Agent should be on the same version as the RMT Server or use a version that is compatible with the RMT Server version. The incident details provide the minimum compatible version.

    2. Stop the Tableau Resource Monitoring Tool Agent service on all Tableau Server nodes using the following command:

      rmtadmin stop --agent

    3. Run the Agent setup program on the Agent that needs to be upgraded.

      Note: If you are using a non-default location, provide the custom install location by clicking Customize on the EULA page. Tableau recommends using \Tableau\Tableau Resource Monitoring Tool as the location for installing the Agent. Example non-default location: D:\Tableau\Tableau Resource Monitoring Tool.

    4. Confirm that the Agent is running and has been upgraded. You can view Agent registration status by navigating to Admin -> Environments -> Edit the Environment -> servers tab to see a full list of Tableau Server nodes and the status of the Agent.

    Upgrading RMT Server

    Upgrade the RMT Server if one or more Agents are on a later version.

    Use the following steps to upgrade RMT Server:

    1. Log in to the Resource Monitoring Tool in the browser and confirm it is working before starting the upgrade.

    2. Copy the RMT Server setup program to the machine where RMT Server is installed.

    3. Stop the Tableau Resource Monitoring Tool Agent Windows service on all Agents.

    4. Stop the Tableau Resource Monitoring Tool Windows service on the RMT server.

      Note: Once the services have been stopped, it is best practice to check for any Resource Monitoring Tool processes that are running after the services have been stopped: Any with tabrmt-agent or tabrmt-master. This does not include PostgreSQL or RabbitMQ.

    5. It can take a little time to fully stop all of the processes running on the RMT Server, so in Task Manager, wait until there are not any running processes with Resource Monitoring Tool in the name before the next step.

    6. Run the setup program on the RMT Server. This installs or upgrades any required prerequisites and then upgrades the RMT Server.

      Note: If you are using a non-default location, provide the custom install location by clicking Customize on the EULA page. Tableau recommends using \Tableau\Tableau Resource Monitoring Tool as the location for installing the RMT Server. Example non-default location: D:\Tableau\Tableau Resource Monitoring Tool.

    7. Confirm that the Agent and RMT Server are running. Start the Agent and RMT Server if they do not automatically restart after the upgrade is complete.

    Agent Message Queue Credential Rotation Failure

    Starting in version 2021.3, as a security best practice, the credentials for connections between Agent and Rabbit MQ must be unique to that Agent. The unique credentials are created during a new installation or updated when upgrading to 2021.3.

    During upgrade, if there are network issues, the credential update process can fail, resulting in a critical incident report. This, however, does not interrupt the upgrade process and the upgrade process will continue. After upgrade is complete, the Resource Monitoring tool will retry daily to create the unique credentials. A critical incident report is created for the initial failure and every subsequently failed retry. During this time, Agent will continue to work using the previous credentials until new credentials have been created.

    The issue might resolve by itself, but if it persists, do the following:

    • Verify that the Agents can connect to RMT Server.

    • Check for firewall issues between Agent and RMT Server.

    Agent Down

    Agent Down incidents will be logged as warning when the Tableau Resource Monitoring Tool is unable to communicate with the Agents on Tableau Server.

    Use the following steps to troubleshoot this issue:

    1. Make sure that the hardware that hosts the Agents is running and available to communicate with the RMT Server.
    2. Make sure that the Agent service is successfully registered and is running. You can view Agent registration status by navigating to Admin -> Environments -> Edit the Environment -> servers tab to see a full list of Tableau Server nodes and the status of the Agent.

    Agent polling and incident creation times

    • Agent sends a heartbeat message to the Resource Monitoring Tool server every 5 minutes.
    • The Resource Monitoring Tool server checks to see if the most recent heartbeat message from Agent is less than 15 minutes old.
      • If the latest heartbeat message received from the Agent is less than 15 minutes old, then the Agent is considered to be online and no incident is created, and any existing Agent Down incidents are cleared.
      • If the latest heartbeat message received from Agent is more than 15 minutes old, then it is considered to be the first failed attempt. Every minute thereafter, a check is done two more times. If after three attempts (first attempt and two retries) there is still no recent heartbeat message, then an Agent Down incident is created.

      Note: Based on the polling interval described above, it takes about 17 (15+1+1) minutes for an Agent down incident to be created after the Agent is offline.

    Who can do this

    Resource Monitoring Tool Administrator or a Resource Monitoring Tool user with Server/Environment Management role.

    Back to top
    Thanks for your feedback!Your feedback has been successfully submitted. Thank you!