Qumulo in AWS: Upgrade an Unconfigured Node

IN THIS ARTICLE

Outlines how to start an unconfigured node and upgrade it to a newer version of Qumulo Core

REQUIREMENTS

• Cloud Cluster with Qumulo Core 2.12.4 and above 
• AWS Console access
• Command line (qq CLI) tools installed via API & Tools in the Web UI

DETAILS

One of the perks of hosting your storage in the cloud is the ability to dynamically add more as your storage needs increase. Starting up a new node is a simple process; if the new node uses an older version of Qumulo Core than your cluster, it must be upgraded before you can add it. The process of upgrading a new node is simply a matter of starting the node up and then upgrading it via the QQ CLI. Once the node's software version matches the one your deployment uses, you can add it to your cluster.

For details on how to launch a new node, check out the Qumulo in AWS: Launch a New Node for an Existing Cloud Cluster article.

Upgrade an Unconfigured Node

To upgrade the node, you'll need to upload the upgrade .qimg to a node-accessible location. For AWS, we recommend uploading the .qimg file to S3 and then generating a pre-signed object URL.

1. Obtain a .qimg file for the version of software that your new node can be upgraded to via our Product Releases page.

2. Upload the upgrade file by running the following command with the aws cli utility configured and authenticated:
   

   aws s3 cp <local qimg local file path> <object path>

3. Generate a pre-signed URL to grant the new node permission to read the upgrade file. Note that this pre-signed URL is used in the steps below. Keep in mind that this step is unnecessary if the node has permission to access the object directly (e.g. permission inherited through the VPC).
   

   aws s3 presign <object path>

4. Connect to the new node via ssh using its IP address.
   

   ssh admin@node_IP_address

5. Prepare the upgrade by running the following command using the path to the pre-signed URL generated for the upgrade image you uploaded to S3.
   

   qq upgrade_config_set --path <pre-signed URL to qimg> --target "prepare"

6. Issue the following command to monitor the ‘prepare’ status and wait for it to return.
   

   qq upgrade_status --monitor

7. Once you see UPGRADE_PREPARED, run the following command to start the upgrade.
   

   qq upgrade_config_set --path <pre-signed URL to qimg> --target "arm"

8. After a few minutes, the node will restart. Check that the upgrade was successful by running the following command and verifying the new version number:
   

   qq version

Your node has been upgraded! Once the software versions match between the cluster and the new node, you can successfully add it to your cluster using the Qumulo in AWS: Add a Node to an Existing Cloud Cluster article as a guide. If the new node is still running an older version, repeat this process with subsequent upgrades until the versions match.

Troubleshooting

If you are unable to successfully upgrade your unconfigured node, test that the node can access the qimg file.

1. Connect to the new node via ssh using its IP address.
   

   ssh admin@node_IP_address

2. Run the following command to verify access:
   

   curl "<object url>" --range 0-10

If the curl command does not work and you receive an error, attempt the following:

• Use a pre-signed URL (if not already)
• Check that the object is accessible from another computer in the same VPC via the same URL
• Troubleshoot your s3 VPC endpoint connection

Once you no longer receive an error using the curl command above, re-attempt to upgrade the node. If the upgrade is unsuccessful again, contact Qumulo Care.

RESOLUTION

You should now be able to successfully upgrade an unconfigured node for an existing cloud cluster in AWS

ADDITIONAL RESOURCES

Qumulo in AWS: Launch a New Node for an Existing Cloud Cluster

Qumulo in AWS: Add a Node to an Existing Cloud Cluster

QQ CLI: Cluster Configuration

Like what you see? Share this article with your network!