Major Upgrades¶

Overview¶

This guide provides step-by-step instructions on how to upgrade your MicroCeph cluster to a new major release.

Follow these steps carefully to prevent to ensure a smooth transition.

In the code examples below an upgrade to the Squid stable release is shown. The procedure should apply to any major release upgrade in a similar way however.

Procedure¶

Prerequisites¶

Firstly, before initiating the upgrade, ensure that the cluster is healthy. Use the below command to check the cluster health:

sudo ceph -s

Note: Do not start the upgrade if the cluster is unhealthy.

Secondly, review the release notes to check for any version-specific information.

Optional but Recommended: Preparation Steps¶

Carry out these precautionary steps before initiating the upgrade:

Back up your data: as a general precaution, it is recommended to take a backup of your data (such as stored S3 objects, RBD volumes, or cephfs filesystems).
Prevent OSDs from dropping out of the cluster: Run the following command to avoid OSDs from unintentionally dropping out of the cluster during the upgrade process:

sudo ceph osd set noout

Upgrading Each Cluster Node¶

If your cluster is healthy, proceed with the upgrade by refreshing the snap on each node using the following command:

sudo snap refresh microceph --channel squid/stable

Be sure to perform the refresh on every node in the cluster.

Verifying the Upgrade¶

Once the upgrade process is done, verify that all components have been upgraded correctly. Use the following command to check:

sudo ceph versions

Unsetting Noout¶

If you had previously set noout, unset it with this command:

sudo ceph osd unset noout

You have now successfully upgraded your Ceph cluster.