How to easily restore an Rapid Recovery archive from AWS Glacier, we archive to AWS S3, then migrate S3 files to Glacier storage.

Question

Rapid Recovery working fine, archive to AWS S3 bucket which we have set up for glacier storage class. How do I restore from this? Rapid Recovery does not see the glacier storage.

Tudor.Popescu · Answer

Hi brian.lynch_69952: Glacier stored objects are not immediately available for restore. Waiting time wise, there are a few retrieval options, each with specific costs. Moreover, data cannot be accessed directly as it is provided as an S3 copy. You need to pay for the duration the archived data is available on S3. (I believe that there is a low cost option that makes the glacier data available on S3 for 24 hours only, but this may or may not suit your restore needs). Once the Rapid Recovery archive is made available as a S3 glacier data copy, you can attach it to a core as a read-only repository (if the archive was created via Rapid Recovery 6.x or you can import it if it is in the AppAssure format). More details re retrieving data from Amazon glacier is available here: docs.aws.amazon.com/.../restore-archived-objects.html Some FAQs may be consulted here: aws.amazon.com/.../ The document referenced above contains additional info re Archive Retrieval Options and various charges you may be subjected to. It may make sense to compare the ease of retrieval and costs with other Rapid Recovery supported Cloud providers such as Microsoft Azure, Rackspace or any other providers using the OpenStack technology. Hope that this helps.

Tudor.Popescu · Answer

Hi brian.lynch_69952: You need to make a difference between Amazon specific features and RapidRecovery specific ones. Technically speaking it is not possible to access AmazonAWS Glacier data directly. Amazon may come with this feature at some point but to my knowledge it is not available now and the general design of Glacier does not allow it. On the other side, Software Vendors/Integrators such as Rapid Recovery may include code to simplify the access process to different Amazon AWS containers and features. For instance I have written a script for Microsoft Azure which creates automatically the containers needed for archiving without requiring the user to log in the Azure console (which may be confusing to use for beginners), simplifying the overall process and extending Rapid Recovery usability. Returning to Amazon AWS and Rapid Recovery, such features that streamline the retrieval process may or may not be added in a future build (this depends on the decisions of the Product Manager), the underlying process will be the same: Glacier data will be temporarily copied to S3 for retrieval as this is how Glacier works.

Tim Seymour · Answer

Direct integration with Amazon Glacier is coming in the next release of Rapid Recovery. That should be out in the next couple of months. At that point you'll be able to essentially archive directly to Glacier and when doing a restore you'll be able to pull the data from Glacier and Rapid Recovery will ask you to choose the retrieval speed for the data and will show you what the costs associated with that retrieval speed will be.