Upgrading Transfer Nodes Using the First-Boot Script

If you need to upgrade or update the default cluster node images provided by Aspera, you can modify the first-boot script executed during the initialization process of each cluster node. This script, located at /opt/aspera/atcm-repo/atcm-upgrade.sh on the Cluster Manager, is executed by every new node of all clusters tied to that Cluster Manager. Everything in the script is executed during the first bootup of a node before it becomes operational, so prevent running any lengthy procedures such as installing large software packages (to do so, see Customizing the Cluster Manager or Cluster Node Images). Instead, use the upgrade script to make simple configuration changes to the operating system such as creating additional system users or configuring the firewall.

Important: This upgrade script is lost if the instance is terminated or restored with the restore procedure (Recover a Failed ATCM Instance).

By default, the script provides two basic functions available for use:

createuser() - Create a system user on the cluster node.
download() - Download a file from the cluster manager to the cluster node.

At the bottom of the script, you can see an example for both functions:

log "========== $0 starts. Repository host [$HOST]  =========="

# createuser myuser mygroup mypassword

# download /opt/aspera/atcm/lib/atcm-main.jar /opt/aspera/atcm/lib/.

log "========== $0 done. [$RET] =========="

Un-comment the createuser script and specify your own user, group, and password to automatically create those users with their associated passwords upon initialization of each cluster node.

Important: We recommend using the new Access Keys for initiating transfers to the cluster nodes. This example is provided in the event that you require new system users.