Backup and restore

Before you back up the data, identify a new host on which you can restore the data from the current host.

• The new host must have the same operating system as the source host.
• The new host must meet the CPU and memory requirements.
  • For a single-node cluster, see CPU and memory requirements.
  • For a multi-node cluster, see CPU and memory requirements.

Complete the following steps on the new host:

1. Prepare your new host by completing the steps in the following sections:

   • Kernel parameters
     • For a single-node cluster, see Kernel parameters.
     • For a multi-node cluster, see Kernel parameters.
   • Packages and environment variables
     • For a single-node cluster, see Packages and environment variables.
     • For a multi-node cluster, see Packages and environment variables.
   • Firewall rules
     • For a single-node cluster, see Firewall rules.
     • For a multi-node cluster, see Firewall rules.

2. Add Instana repository.

3. Installing stanctl command-line tool.

4. Stop the current cluster by running the following command on the current host: stanctl down.

Use the following command to take a backup of the cluster. In a multi-node cluster, you must run this command on node0 (instana-0). The command creates a stanctl-backup-<timestamp>.tar.gz file in the current directory. To specify another location, use the --output-dir flag with the command.

stanctl cluster backup
 

Take snapshots of all the disks that you mounted on the current host.

For more information about how to take snapshots, refer to the documentation of your storage system or cloud provider.

Complete the following steps on your new host:

1. Create disks from the snapshots and mount them to the new host. Create one disk at a time from the snapshots to make sure to restore and mount the correct disk to the correct file system.

   • For a single-node cluster, complete the following steps:
     1. Create the directories.
     2. Add mount paths.
     3. Mount the file systems.
   • For a three-node cluster, complete the following steps:
     1. Create the directories.
     2. Add mount paths.
     3. Mount the file systems.
   • For a five-node cluster, complete the following steps:
     1. Create the directories.
     2. Add mount paths.
     3. Mount the file systems.

2. Copy the Stanctl backup archive (stanctl-backup-<timestamp>.tar.gz) file to the new host. In a multi-node cluster, you must copy the archive file to node0 (instana-0).

3. Restore data from the stanctl-backup-<timestamp>.tar.gz to the new host.

   • For a single-node cluster, run the following command:

     stanctl cluster restore -f <path/to/stanctl-backup-<timestamp>.tar.gz>
      

   • For a multi-node cluster, run the following command on node0 (instana-0):

     stanctl cluster restore --multi-node-enable --multi-node-ips=<node0IPaddress,node1IPaddress,node2IPaddress> -f <path/to/stanctl-backup-<timestamp>.tar.gz>
      

4. In the DNS settings configuration, be sure to replace the public IP address of your host with the public IP address of the new host.

   • For a single-node cluster, see DNS settings.
   • For a multi-node cluster, see DNS settings.

5. Install the Instana backend and data stores by using the following command. The command uses existing configurations that are located in the $HOME/.stanctl/instana.yaml file for the installation.

   • For a single-node cluster: stanctl up
   • For a multi-node cluster: stanctl up --multi-node-enable

You can use a configuration‑only backup and restore workflow when disk snapshots are not practical or when you need to preserve configuration before an upgrade. This workflow uses stanctl dump config-data and stanctl migrate to export and restore configuration data for Self-Hosted Standard Edition. It also fully supports air‑gapped environments.

Use this workflow in the following situations:

• You cannot create datastore or disk snapshots.
• You want to preserve configuration before an upgrade.
• You need to save tenant and application configuration without including analytics data.

Use the snapshot‑based backup and restore workflow when you need a complete system backup that includes metrics and analytics data.

The configuration backup includes all PostgreSQL-backed configuration data, including:

• Custom dashboards
• Alerts
• Users
• Application perspectives
• End‑user monitoring, which tracks websites (EUM)
• EUM mobile applications
• Events
• Tenant settings
• Unit settings
• Other configuration data stored in PostgreSQL