- 최신
- 최다 투표
- 가장 많은 댓글
To change your S3 objects from Glacier Instant Retrieval to Deep Archive, you have a few options:
For objects already in Glacier Instant Retrieval:
- You can use the AWS CLI to copy the objects in place while changing their storage class. This approach doesn't duplicate your data but updates the storage class:
aws s3 cp s3://your-bucket/ s3://your-bucket/ --recursive --storage-class DEEP_ARCHIVE
- You can create a new lifecycle policy to transition objects from Glacier Instant Retrieval to Deep Archive. However, there's an important limitation: objects in Glacier Instant Retrieval cannot be directly transitioned to most other storage classes except for Glacier Deep Archive.
For future objects:
- Update your existing lifecycle configuration to transition objects directly to Deep Archive instead of Glacier Instant Retrieval.
Important considerations:
- Simply disabling the current lifecycle rule won't change the storage class of objects already in Glacier Instant Retrieval
- Using the CLI copy command doesn't create duplicate objects but does count as a new PUT operation, which may incur some fees
- When transitioning between storage classes, be aware of minimum storage duration requirements to avoid additional charges
- The storage class change isn't immediate and may take some time to complete
If you're managing a large number of objects, the lifecycle approach is more automated, while the CLI approach gives you more immediate control over the transition.
Sources
Setting the storage class of an object - Amazon Simple Storage Service
S3 bucket lifecycle policy through CLI | AWS re:Post
Setting an S3 Lifecycle configuration on a bucket - Amazon Simple Storage Service
Transitioning objects using Amazon S3 Lifecycle - Amazon Simple Storage Service
You need to do a couple of things -
- Update your existing lifecycle rule to move objects from Glacier to Deep Archive. See screenshot.
- Update your current process for uploading objects to S3 and make sure to specify the storage class as suggested by repost agent.
Note, S3 bucket don't have a default setting that you can override for storage class. The default is Standard.
Thank you very much for the fast response :)
What I attempt is
- Run the aws s3 cp s3://your-bucket/ s3://your-bucket/ --recursive --storage-class DEEP_ARCHIVE command
- Changed my Lifecycle jobs to send to Deep Archive
