You can remove a Cluster node using the Dynatrace Managed Cluster Management Console. However, as removing a Cluster node means uninstalling Dynatrace Managed, you can alternatively use the command prompt to remove a Cluster node. Both approaches are described below.
Recommended
Before you can remove a node from a Managed Cluster, you must first disable OneAgent traffic to prevent data flow to the node during its removal.
To disable OneAgent traffic on a Cluster node, follow these steps.
Before proceeding with node removal, it's recommended that you wait for at least the duration of the transaction storage data retention period (up to 35 days). This is because although all metrics are replicated in the Managed Cluster, the raw transaction data is not stored redundantly. Removing a node before its transaction storage data retention period expires might impact code-level and user-session data analysis.
Remove no more than one node at a time. To avoid data loss, allow 24 hours before removing any subsequent nodes. This is because it takes up to 24 hours for the long-term metrics replicas to be automatically redistributed on the remaining nodes.
If you have Managed Cluster backup enabled and you remove a node, the backup files for that node remain in your backup space. If you want to delete the backup for the removed node, delete the <path-to-backup>/<UUID>/node_<node_id> folder.
In this example path:
rmdir -r /mnt/backup/bckp/c9dd47f0-87d7-445e-bbeb-26429fac06c6/node_3/
the parts of the path are as follows:
/mnt/backup/bckp/ = <path-to-backup>c9dd47f0-87d7-445e-bbeb-26429fac06c6 = <UUID>3 = <node_id>After you remove an existing Cluster node, you must wait 72 hours before installing a new node on a host that shares the same IP address (for example, when you reinstall on the same host).
To remove a disabled Cluster node, follow these steps.
Dynatrace Managed shows dead and removed nodes for seven days, so you do not need the cleanup API during regular node removal. Use it only to remove a dead node that's no longer displayed on the Cluster nodes page but is still registered in the Managed Cluster.
To clean up the dead node registration, run the Cluster Management API call to the following endpoint: onpremise/nodeManagement/deadNodeCleaning.
See the Dynatrace API explorer in the Cluster Management Console for details.
/nodeManagement/deadNodeCleaning section to try it out.Recommended
Removing a Cluster node using the Cluster Management Console is recommended. However, there's also an alternative option, via public API: /api/v1.0/onpremise/nodeManagement/triggerRemoveNode.
If the recommended methods for removing Cluster nodes (Cluster Management Console and API call) don't work and you're still facing issues, use the following command prompt.
Running the script deletes all Dynatrace Managed data from the machine.
Log in to the Linux machine where Dynatrace Managed is installed.
Go to the PRODUCT_PATH main directory for Dynatrace Managed binaries (default: /opt/dynatrace-managed).
Using root rights, run the following command:
/bin/sh uninstall-dynatrace.sh