
AIOZ's documentation covers a Backblaze B2 migration path separately from its Amazon S3 one, and the difference is not just branding. B2 doesn't run through rclone's s3 remote type at all, it uses type = b2, which means this migration is a source-and-destination remote type change, not the same-type credential swap the S3 migration is. This guide covers the actual b2-to-s3 config, the sync command, and the one gotcha AIOZ's docs name for this path specifically.
TL;DR:
b2 remote (account and key fields), while AIOZ Storage stays on the s3 remote with provider = Other, two different remote types in one sync command.b2: source instead of s3:.--ignore-size fix apply here too.Yes, and AIOZ publishes a dedicated guide for it, separate from the Amazon S3 migration guide. The mechanism is still rclone, but the setup underneath is different: Backblaze B2 has its own native rclone backend (type = b2), while AIOZ Storage is reached through rclone's generic S3-compatible backend (type = s3, provider = Other). One sync command bridges both, but you're configuring two different kinds of remote, not two flavors of the same one.
Migrating from Amazon S3 is a same-remote-type change: both sides use type = s3, and only the provider field and credentials flip. B2 breaks that pattern. Since Backblaze's own B2-vs-S3-compatible-API comparison confirms the Native API predates its later S3-compatible one and rclone built B2 support around that native API first, your source remote for this migration uses entirely different config fields, account and key, not access_key_id and secret_access_key. The destination side is identical to the S3 migration though, since AIOZ Storage itself is S3-compatible regardless of what you're migrating from.
Two credentials, from two different places. From Backblaze, you need either your Account ID or an Application Key ID, plus the matching Application Key, generated from the B2 dashboard. From AIOZ Storage, you need an access grant created with the "S3 Credentials" type, which returns an Access Key, Secret Key, and endpoint. How access grants work covers what that credential actually is, but this migration only needs the three values it hands back.
Add a source remote for Backblaze B2:
[b2]
type = b2
account = <your-b2-account-or-keyid>
key = <your-b2-application-key>Then add a destination remote for AIOZ Storage, identical to what the S3 migration guide uses:
[aioz-storage]
type = s3
provider = Other
access_key_id = <your-aioz-storage-access-key-id>
secret_access_key = <your-aioz-storage-secret-access-key>
endpoint = <your-aioz-storage-endpoint>Notice the AIOZ Storage block never changes based on where you're migrating from. It's the source block that shape-shifts to match whatever backend you're leaving, B2's native fields here, S3's access_key_id/secret_access_key/region fields if you were coming from Amazon instead. AIOZ's migration guide leaves the destination endpoint as a placeholder, but its SDK documentation confirms the real value: https://s3.aiozstorage.network.
Possibly, and it's worth checking before you kick off a large sync. Backblaze B2's own pricing gives free egress up to 3x your average monthly stored data, then charges $0.01/GB beyond that. A one-time full-bucket migration reads your entire stored volume out through B2's egress path, so if you're already running close to that 3x ceiling from normal application traffic, the migration read can push you over it. For most accounts, a single full sync stays comfortably inside the free allowance, but it's a real cost some migration guides skip mentioning entirely. Check your current egress usage against that 3x threshold before syncing a bucket in the multi-terabyte range.
With both remotes configured, sync your bucket over:
rclone sync --progress b2:your-bucket aioz-storage:your-bucketThen check that everything landed intact before you touch your application's configuration:
rclone check --progress b2:your-bucket aioz-storage:your-bucketOnly cut your application over to AIOZ Storage's endpoint and credentials once that check comes back clean.
AIOZ's B2 migration guide flags the same error as the S3 guide: corrupted on transfer: sizes differ. The fix is identical too, rerun with --ignore-size appended:
rclone sync --progress b2:your-bucket aioz-storage:your-bucket --ignore-sizeAIOZ doesn't explain why this happens on either migration path, so treat it as a documented workaround rather than a diagnosed cause. And as with the S3 migration, this guide covers object data only. B2's own bucket lifecycle rules, file versioning, and application key permission scopes have no stated equivalent in AIOZ's documentation, verify those separately if your B2 setup depends on them before you cut over.
Does migrating from Backblaze B2 to AIOZ Storage use the same rclone remote type as migrating from S3?
No. B2 uses rclone's native b2 remote type with account and key fields. AIOZ Storage's own side always uses the s3 remote type with provider = Other, regardless of what you're migrating from.
What rclone command migrates data from Backblaze B2 to AIOZ Storage?rclone sync --progress b2:your-bucket aioz-storage:your-bucket, with a B2 source remote and an AIOZ Storage destination remote already configured.
What credentials does Backblaze B2 require for this migration?
Your Account ID or Application Key ID, plus the matching Application Key, both generated from the Backblaze B2 dashboard.
Does the "corrupted on transfer: sizes differ" error happen on this migration too?
Yes. It's the same error AIOZ's S3 migration guide names, and the same fix applies: rerun the sync with --ignore-size appended.
Do I need a different AIOZ Storage endpoint for a B2 migration versus an S3 migration?
No. The AIOZ Storage destination remote (https://s3.aiozstorage.network) is identical regardless of which service you're migrating from.
Does this migration guide cover B2's file versioning or lifecycle rules?
No. AIOZ's documentation covers object data transfer only. Verify B2 versioning, lifecycle rules, or application key scopes separately if your setup depends on them.
Does Backblaze B2 charge for the data transferred out during a migration?
Possibly. B2 gives free egress up to 3x your average monthly stored data, then $0.01/GB beyond that. A large one-time migration sync counts against that allowance, so check your current usage against the threshold first if your bucket is in the multi-terabyte range.

AI agent sandboxes need external storage for files that outlive the sandbox. Real providers like Modal and E2B mount S3-compatible buckets, AIOZ Storage included.

AIOZ storage can back LangGraph's S3 checkpoint offload tier, but not the whole backend. Here is the real DynamoDB-plus-S3 setup and its credential gap.

AIOZ storage for AI datasets means S3-compatible buckets for training data and model outputs, no native versioning or lifecycle policies. Here is the honest scope.

AIOZ storage for AI workloads means S3-compatible object storage for datasets, checkpoints, and model outputs. No vector database. Here is what is real.

Add AIOZ storage team members through the dashboard's 3-step wizard: name and password, per-bucket permissions, and a one-time credential download.

No Postman collection to import. AIOZ Storage docs show building raw S3 requests by hand, authenticated with AWS Signature and your access grant keys.