Amazon Glacier considerations and retrieval options
Adding an Amazon Glacier type tier to a file system creates a vault in Amazon Glacier.
The Amazon Glacier tier is an offline tier. Read, write, and truncation file operations fail with an EIO error for files moved to the Amazon Glacier tier.
When files are moved to an Amazon Glacier tier, an archive is created per file.
If you want to read or modify the data that is moved to Amazon Glacier, move the data back to on-premises using policies.
See the Amazon Glacier website for storage and retrieval costs.
The maximum file size for moving files to Amazon Glacier is 4 GB.
The Amazon Glacier cloud tier usage statistics are not immediately reflected.
Amazon Glacier archive retrieval options:
Expedited - Retrievals typically complete within 1-5 minutes.
The expedited option is expensive and you should use it conservatively.
Files moved from the Amazon Glacier tier with the expedited option might return the following error:
InsufficientCapacityException (503 service unavailable)
This error occurs if there is insufficient capacity to process the expedited request. This error only applies to expedited retrievals and not to standard or bulk retrievals.
Standard - Retrievals typically complete within 3-5 hours.
Bulk - Retrievals typically complete within 5-12 hours.
Note:
The retrieval operation is asynchronous whereas the upload operation is synchronous for the Amazon Glacier cloud. When a file is moved to the Amazon Glacier cloud, all the data in the file is written as one archive in a glacier vault. When a file is retrieved from the Amazon Glacier cloud, a job gets initiated and the output gets downloaded after the job is completed.
Note:
If you execute a policy to move files from Amazon Glacier to the primary file system, you should check if all the files are moved back using the storage tier listfiles command. If any of the files do not get listed, Veritas recommends that you re-run the policy.