Stage to production

1. Accounts

   One profile per AWS account, signed in with access keys or an IAM Identity Center session. While the console is open, Douklar checks each account at the interval you choose.

   Accounts and credentialsSee the Accounts screen

   

   Accounts

   Captured from the console with sample data.

   Close

2. Discovery

   Douklar keeps a catalog of buckets and repositories for each account, and refreshes it when you ask.

   Discovery and mappings

3. Mappings

   Pair each source with a destination, per service. Renames are allowed, and mappings are grouped by app.

   Discovery and mappingsSee the Mappings screen

   

   Mappings

   • Suggested destinations

     Destinations are suggested from your cross-account defaults.

   • One table per service

     ECR, S3, CodeCommit and DynamoDB each have a table, with a count on the tab.

   • YAML or one at a time

     Edit a whole table as YAML, or add mappings one by one.

   Captured from the console with sample data.

   Close

4. Preflight

   Checks before the run starts: does the destination exist, how do the encryption keys compare, does a TTL need a decision, will the credentials last. Problems show up here, not halfway through a cutover.

   Preflight and Review

5. Run

   Review the plan for each service, then start one queued run. A worker picks it up, and you can watch its logs live.

   Running a syncSee the Sync screen

   

   Sync

   • Scope, review, run

     Every run goes through three steps: choose the scope, review the plan, then run it.

   • Same account or cross account

     Pick the mode. Regions for ECR, CodeCommit and DynamoDB are set here; S3 bucket Regions are found for you.

   • Several targets

     One run can copy to more than one target account.

   Captured from the console with sample data.

   Close

6. Verification

   After the run, Douklar checks what arrived: image digests and referrers, refs, items, and objects copied server-side.

   Running a sync

7. Activity

   Every run is kept: counts per service, logs and the outcome of each resource. Re-run a move, or retry only what failed.

   Activity and evidenceSee the Activity screen

   

   Activity

   • Filter by state

     Running, completed, warning, failed or cancelled.

   • Failed runs stay

     A failed run keeps its place beside the completed ones, newest first.

   • Route of every run

     Source to destination; "+1" marks a run that went to one more target.

   Captured from the console with sample data.

   Close

Print the cutover checklist: before, during and after a move, built from these steps.