Find your next learning step
Reclosure proves that a known lineage problem was repaired at one point in time. Monitoring must then show whether the fix survives new scenes, software releases, cache keys, partner exports and workflow changes. Cache-control drift is especially quiet: the official product may be current while a local object, notebook or tile service begins serving an older identifier again. Fleet-recurrence status should retire only after sustained negative evidence across the propagation paths that once made the problem systemic.
Open the chapter contents · Remote Sensing root · Gold Standard of Education · How X Works
Chapter contents
Choose the route closest to the decision, or read from Chapter 1. The chapters remain open; these grouped routes are a navigable index.
Route 1 · Define the decision
Route 2 · Work the ledgers
Route 3 · Transfer the method
Route 4 · Check limits and evidence
Route 5 · Ask and continue
CHAPTER 1 OF 12 · DEFINE THE DECISION
1. Define the post-reclosure claim
State the retired product family, successor identifiers, repaired cause, affected fleet, bounded historical exceptions, reclosure date, monitoring window and reopening triggers. Reclosure is a scoped claim about controlled operational use, not proof that every old copy vanished.
Name the paths that can reintroduce drift: scheduled downloads, cloud caches, desktop folders, notebooks, model registries, partner transfers and derivative stores.
CHAPTER 2 OF 12 · DEFINE THE DECISION
2. Distinguish cache drift from source reprocessing
Source reprocessing changes the authoritative product and usually its processing identifier. Cache drift occurs when local selection, expiry, invalidation or key design serves an object that no longer matches the intended source.
Compare collection, acquisition date, processing date, tier, algorithm version and auxiliary inputs. A friendly filename or current folder name cannot establish identity on its own.
CHAPTER 3 OF 12 · WORK THE LEDGERS
3. Work a reclosure-monitoring register
A flood service monitors four months after closing a P4-to-P5 fleet recertification event.
| Monitoring signal | Month 1 | Month 2 | Month 3 | Month 4 | Interpretation |
|---|---|---|---|---|---|
| Active retired identifiers | 0 | 0 | 0 | 0 | Stable |
| Cache invalidation failures | 0 | 1 | 0 | 0 | Repair and watch |
| Unresolvable processing dates | 2 | 0 | 0 | 0 | Initial cleanup effective |
| Partner receipt gaps | 1 | 0 | 0 | 0 | Closed |
| Derivative lineage sample pass | 98% | 100% | 100% | 100% | Stable after repair |
The failed cache key used a region alias omitted from the original invalidation list. The service expands the test set to every alias rather than treating the event as random noise.
CHAPTER 4 OF 12 · WORK THE LEDGERS
4. Decide when fleet-recurrence status may retire
Retirement removes the special fleet-wide incident state while ordinary provenance controls continue.
| Retirement gate | Required evidence | Observed result | Decision |
|---|---|---|---|
| Sustained clean windows | Three complete cycles after last failure | Three cycles pass | Pass |
| Propagation coverage | All cache classes and partner routes challenged | One new alias added and passed | Pass |
| Historical exceptions | Read-only and labelled | Verified | Pass |
| Detection control | Injected retired identifier raises alert | Alert in 6 minutes | Pass |
| Owner readiness | Routine team accepts monitoring | Handover drill passes | Pass |
Retirement should reduce incident overhead without deleting the evidence, trigger history or ability to reopen the ledger.
CHAPTER 5 OF 12 · WORK THE LEDGERS
5. Test prevention, detection and recovery separately
Prevention rejects or invalidates a stale object. Detection notices a mismatched identifier or lineage gap. Recovery restores the current object and regenerates affected derivatives. Passing one does not prove the other two.
Use safe injected identifiers or isolated test caches so the challenge cannot contaminate production. Record cleanup and verify that the test artefact no longer resolves through any operational path.
CHAPTER 6 OF 12 · TRANSFER THE METHOD
6. Move from incident monitoring to routine control
After retirement, keep product-manifest comparison, sampled derivative traces, cache-age checks and partner receipts at a risk-based frequency. Retain the unusual paths discovered during the incident in the routine inventory.
Set an explicit reopen trigger: an operational retired identifier, repeated cache-control failure, unexplained processing-date mismatch or partner derivative that cannot be traced to the authorised product.
CHAPTER 7 OF 12 · TRANSFER THE METHOD
7. Changed-context transfer: wildfire severity maps
Transfer the method to a wildfire service after reprocessed scenes corrected a calibration issue. The fleet replaces its products, regenerates severity maps and re-closes the ledger.
Two months later, a training notebook republishes an old cached tile into a staging bucket. Detection catches the retired identifier before public release, the route is repaired and the clean-window count restarts because the propagation path was operationally plausible.
CHAPTER 8 OF 12 · TRANSFER THE METHOD
8. Connect recertification, reclosure and debt retirement
Begin with fleet recertification completion, shared-cache tests and reclosure evidence. Use successor-ledger closure and repeated-fleet discovery escalation for the earlier incident boundary.
Continue through debt burn-down, orphaned-copy discovery and amendment-reopen triggers when unresolved copies remain.
CHAPTER 9 OF 12 · CHECK LIMITS AND EVIDENCE
9. Limits and common misconceptions
Several clean scans do not prove that offline exports or unknown partner copies are absent. A cache miss also does not prove the fetched object is current unless the authoritative identifier and processing date are checked.
Fleet-recurrence retirement is not archive deletion. Historical products can remain legitimate evidence when clearly identified and isolated from current operational selection.
CHAPTER 10 OF 12 · CHECK LIMITS AND EVIDENCE
10. Practical evidence and retirement checks
Retain manifests, scan queries, cache inventories, aliases, invalidation logs, processing identifiers, derivative traces, partner receipts, injected tests, alerts, repairs, clean-window counts and the retirement decision.
Before retirement, challenge every known propagation class, trace at least one public output end to end and confirm routine owners can reopen the ledger. Preserve the incident evidence after special monitoring ends.
CHAPTER 11 OF 12 · ASK AND CONTINUE
11. Frequently asked questions
How long should reclosure monitoring continue?
Long enough to cover the operating cycles and propagation paths that could reveal recurrence. Restart or extend the clean window after a material failure.
Does one stale research notebook reopen the fleet?
Not automatically. Reopen when the object can feed an operational path, exposes a failed control or suggests wider recurrence.
What remains after recurrence status retires?
Routine provenance, cache and product-identity controls remain, along with the evidence and trigger needed to reopen the incident.
CHAPTER 12 OF 12 · ASK AND CONTINUE
12. Sources and next reading
USGS explains that when a Landsat scene is reprocessed, the Collection 2 product identifier changes to reflect the latest processing date. The Landsat reprocessing events guidance describes locating replaced products and notes that affected products and derivatives may need replacement. The flood and wildfire cases are original teaching examples.
Continue through the Remote Sensing root, the Gold Standard of Education and How X Works.
Top · Contents · Previous · How X Works