Restore Without Reinstall (Factory-Installed)¶

Overview

This procedure performs an in-place restore on a factory-installed AIO-SX subcloud without reinstalling the operating system.

Unlike the standard restore procedure described in Run Restore Playbook Locally on the Controller, which requires a full OS reinstall from the ISO before running the restore playbook, this procedure restores platform configuration and state directly on an intact factory installation.

The restore playbook (restore_platform.yml) is the same in both cases and uses restore_mode=optimized. However, in this procedure, the playbook auto-detects the factory-installed state by checking for the /var/lib/factory-install/complete file and skips the OS-level recovery steps that are not required on an intact factory installation.

Note

This is supported only for AIO-SX subclouds.
The factory system must be configured with Rook-ceph or LVM CSI (baremetal ceph is not supported).
The backup and the factory system must use the same persistent volume storage solution. You cannot restore a Rook-ceph backup onto an LVM CSI factory system or vice versa.

Prerequisites

The system is factory-installed and the file /var/lib/factory-install/complete, created by the factory install, exists on the system.

Procedure

Local Restore (Manual Procedure)¶

This manual procedure is for standalone systems. For subcloud restore in a Distributed Cloud environment, see the Subcloud Restore documentation.

Note

The backup tarball must be available locally on the factory-installed system (for example, at /opt/platform-backup/localhost).

Run the restore playbook, specifying the location of the backup tarball on the node:

$ ansible-playbook /usr/share/ansible/stx-ansible/playbooks/restore_platform.yml -v \
  -e "ansible_become_pass=<become_password>" \
  -e "admin_password=<admin_password>" \
  -e "initial_backup_dir=/opt/platform-backup/localhost" \
  -e "backup_filename=$(ls /opt/platform-backup/localhost/platform_backup.tgz | tail -n -1 | xargs basename)" \
  -e "restore_mode=optimized"

After the playbook completes, re-source openrc to pick up the restored admin credentials from keyring, then unlock the controller:

Note

This step is specific to the factory-installed restore flow. Unlike the standard optimized restore, re-sourcing openrc is required to pick up the restored admin credentials before proceeding.
```
$ source /etc/platform/openrc
~(keystone_admin)$ system host-unlock controller-0
```

Postrequisites

After the host is unlocked, complete the restore using the following command:

~(keystone_admin)$ system restore-complete

Remote Restore (Subcloud)¶

For subcloud restore in a Distributed Cloud environment, use the ISO-based restore workflow. This workflow automates the complete restore process, including updating the subcloud deployment state on the System Controller. For more information, see the Subcloud Restore documentation.