What changes for you
Recovery can reconstruct supported failed direct loads, staged streams, comparisons, indexes, and foreign keys, then rerun configured validation.
Operations · Recovery
Keep failure visible. Replay the work Omni Loader can recover safely.
Omni Loader records runs, stages, tables, task status, warnings, errors, and failed work payloads. A reported failed task can be cleaned up and replayed from the recovery workflow.
Failed-work recovery Inspect · clean · replay
Retry isolated
Inputs
Completed workREADY
Reported failureREADY
Saved task payloadREADY
Run history Inspect · clean · replay
Outputs
Keep completedREADY
Clean partial stateREADY
Replay failed unitREADY
The job to be done
Move the full dataset without turning failure into a restart.
A large migration is not finished when the first rows arrive. Omni Loader keeps parallel work productive, adapts the loading path to the target, and records enough progress to recover the part that failed.
The local state store tracks projects, runs, stages, tables, and work status while logs capture summaries, warnings, errors, SQL, and troubleshooting detail.
Why this path matters
Keep each migration run inspectable.
The local state store tracks projects, runs, stages, tables, and work status while logs capture summaries, warnings, errors, SQL, and troubleshooting detail.
The outcome
We will identify retryable work, cleanup behavior, validation, and the operational response to a process-level interruption.
The payoff
See what the capability changes.
Run history
Reviewable execution
Inspect stages, tables, progress, warnings, and errors.
Failed tasks
Serialized replay
Recover supported work after a task reports failure.
Scoped
Target cleanup
Clean the failed table or filter range before selected retries.
Where the work happens
The details that make the result usable.
Read each capability as part of the operating path. The important question is not whether a feature exists, but what work it removes and what evidence it leaves behind.
At a glance
A visual summary of the work.
A host or power loss before a task records its failure may not leave the serialized payload required for replay. Plan process-level restarts and validation accordingly.
Operational record
Keep each migration run inspectable.
The local state store tracks projects, runs, stages, tables, and work status while logs capture summaries, warnings, errors, SQL, and troubleshooting detail.
Supported replay
Retry work that finished with an explicit failure.
Recovery can reconstruct supported failed direct loads, staged streams, comparisons, indexes, and foreign keys, then rerun configured validation.
- Direct load tasks
- Remote-storage stream tasks
- Comparison tasks
- Index and foreign-key creation
Clear boundary
Failed-task recovery is not a distributed durable queue.
A host or power loss before a task records its failure may not leave the serialized payload required for replay. Plan process-level restarts and validation accordingly.
Before you choose
Check the path against your environment.
Use the page-specific details below as a short discovery checklist. They are the conditions and work areas that shape the capability, not generic product promises.
- Direct load tasks
- Remote-storage stream tasks
- Comparison tasks
- Index and foreign-key creation
The path to confidence
See the work. Make the call. Prove the result.
The page keeps the product detail close to the decision it supports. Your team can see what the software handles automatically, what it changes for the target, and what the result leaves ready to operate.
Keep each migration run inspectable.
The local state store tracks projects, runs, stages, tables, and work status while logs capture summaries, warnings, errors, SQL, and troubleshooting detail.
Retry work that finished with an explicit failure.
Recovery can reconstruct supported failed direct loads, staged streams, comparisons, indexes, and foreign keys, then rerun configured validation.
Failed-task recovery is not a distributed durable queue.
A host or power loss before a task records its failure may not leave the serialized payload required for replay. Plan process-level restarts and validation accordingly.
Take the next step
Plan recovery around explicit evidence.
We will identify retryable work, cleanup behavior, validation, and the operational response to a process-level interruption.
Plan my migration