Skip to main content
These rules are why stamp retry is deliberate instead of a casual “try again” button. They protect against duplicate CFDIs, silent remaps, and correcting the wrong problem.

​
Non-negotiables

Retry never edits the original accounting XML. Remapping always starts from the stored capture. The capture is evidence; the invoice row is Lambda’s claim.
A stamped or completed CFDI is finished. Active or completed cancellation blocks stamp retry. Replacing a Lambda-owned stamp is a separate, support-owned operation — not an everyday workspace path.
Documents that are validating, ready, stamping, pending, workflow-owned, or waiting on a parent are not open for a second stamp lane. Lambda returns the active operation instead of stacking another.
If Lambda cannot prove create never happened, it resumes the exact request and key. Remapping into a fresh create is blocked.
An unchanged rejected request cannot open another attempt. Corrective retry on failed requires proof the fiscal projection changed, or that a policy-approved prerequisite was corrected. Most of that lane is support-owned.
Apply recomputes the plan. A stale hash fails closed. Hashes are not hand-edited, and one plan stays scoped to one invoice (or an explicit bulk fiscal-correction plan that still previews first).
Retry records resume or reprocess claims. Terminal fiscal truth remains stamp completed, webhook stamped, or stamp failed — not a second “retry succeeded” fiscal event.
Retry CargoWise delivery only re-sends artifacts. It never issues another CFDI. See How stamping works.

​
What automatic recovery may do

Automatic recovery may:
  • Re-enqueue work that never started, from the persisted intent
  • Resume the current durable attempt with the same request and key
  • Replay a terminal attempt proven known not created by cloning that exact private request under a new key
It may not:
  • Remap a newer CargoWise capture into new fiscal intent
  • Invent a corrective category or skip preview/apply for support corrections
  • Quietly re-stamp to match post-stamp CargoWise drift
When recovery budgets are exhausted, the document moves to system error. Support then uses a fresh corrective plan and reason.

​
What corrective retry may do

Corrective retry may change course only under categories Lambda derives from evidence: Operators do not free-pick a category to force a stamp.

​
External stamps

Documents observed as stamped outside Lambda (Stamped externally) have no Lambda stamp retry or cancellation action here. They are recorded evidence, not Lambda-owned CFDIs to remake.

​
How this reads in the workspace

  • If Lambda offers Retry stamp, policy already allows a safe continuation for that classification.
  • If the workspace withholds retry, that is usually the product working — see Failures, System errors, and Recovery paths.
  • Post-stamp CargoWise disagreement is not a silent re-stamp. Matching evidence or deliberate support review owns that path — see Rules.