Hi all,
I am running a nightly Cloud Sync Task on S3 Glacier.
The task, under Advanced Remote Options, has the Storage Class set to Glacier Deep Archive.
The task runs without errors since about 1 year.
Yet, it appears that S3 is storing everything under a Standard storage Class, not on Glacier.
If I try to move the bucket folders to Glacier, from the S3 UI, all object moves return the following error: CopyObjectOperationObject: storage class has unknown or unsupported value : DEEP_ARCHIVE
I have a very simple setup on AWS, this is my only bucket.
I haven’t had much success in using TrueNAS cloud sync tasks for syncing with my AWS S3 buckets. The data that I have is static and I needed to use the AWS Glacier Deep Archive storage tier. TrueNAS uses the open source Rclone backend for syncing S3 objects. Rclone does not offer the flexibility that I needed and kept trying to upload 50+ TB of data everytime I used TrueNAS cloud sync tasks with AWS Glacier. I suspect that you may be running into some compatibility issues as well, despite your sync being successful for the last year.
My solution was to use Developer Mode and then install the native AWS CLI. From there I wrote some scripts along with a systemd timer for my sync tasks.
This has worked really well for me. Maybe AWS only plays well with their own tools, especially when you start using anything other than a default storage tier? Just my opinion, try and run a sync using native AWS tools. I think with rclone something gets lost.
Another thing you could try is to create a VM or use a container with AWS CLI, give it access to your data and try and run a manual sync with AWS CLI and see what happens.