Upcoming changes to cloud object storage

by Bruno Lago

The Catalyst Cloud object storage service is about to get even better. It will replicate your objects across regions and introduce features that are very useful for web applications, such as object versioning.

While most customers will not be affected by this change, we want to ensure everyone receives adequate time to prepare for it, hence this notification.

What will change?

The change is very simple. Currently you call our API using the following format:


Going forward we will prefix containers with your tenant ID:


Will this affect my code?

If you have been using the S3 API using the Boto library or the Swift API using the Python Swift library with Keystone V2 or V3 for authentication, this change will not affect you and there is nothing you need to do.

If you have been using the APIs via cURL or libraries that do not prefix the object path with the tenant_id, then you will need to make a small change to your code to introduce this prefix.

Will this affect my container names?

Currently you can have two different containers with the same name in both regions. Going forward, container (and its objects) will be visible in all regions. As a result, if you have a tenant with the same container name in both regions (eg: /backup), we would like to ask you to rename the containers so they become unique (eg: /backup-por and /backup-wlg).

If we identify that you have containers with clashing names, prior to the migration, we may give you a call and help you rename them to prepare for this change.

If you have any questions in relation to this change and need help from Catalyst, feel free to raise a support ticket via our dashboard or call your account manager.

When should I make these changes?

The transition to the new object storage service will happen on Sunday 13 November 2016.


Kind regards,
The Catalyst Cloud team