Make · Help

Make: linked feeder was not found when resuming?

An incomplete Make execution fails at an aggregator? Inspect saved run state, Source Module and data completeness before restarting.

Source date: 5 November 2026Editorial team: KAN KAI WORKSEditorially updated:

Quick answer

Linked feeder was not found during Make resumption concerns the relationship between an aggregator source and saved execution state. A successful fresh run does not prove the older execution can resume. Preserve the execution and its data, then compare module IDs, Source Module and aggregation boundaries with the saved configuration. Before restarting, reconcile existing external results and account for every expected input item.

Source basis: Q1 · Q2 · Q3

Applies to
The original report distinguishes working fresh runs from failed resumption. It confirms no general platform defect; inspect saved execution configuration separately from the current scenario.
How to verify the result
The affected business operation completes fully, every expected item ID appears in the intended aggregate, and external actions already performed are not duplicated.
Your first check

Does only the saved resumption fail, or also a new controlled run?

A Make scenario normally completes, but an incomplete execution cannot resume after a failure: it reports that a linked feeder was not found for an aggregator. This matters particularly when the final result must include every attachment or component record belonging to one business operation.

Distinguish a new run from saved execution state

An original report describes this between an iterator, processing step and aggregator. Fresh runs worked, while resuming failed despite an apparently appropriate Source Module setting. It does not establish a universal fix. Q1 Make documents aggregation as beginning at a specific source, and incomplete executions retain configuration that matters to recovery. Q2 Q3

A successful new test therefore does not prove that an older saved execution can resume with its original state. Similarly, a green final result does not establish that every required input was included.

Preserve and inspect the saved state

  1. Retain execution and data. Record the incomplete execution ID, original failure, module IDs and intended business operation. Do not delete the entry while its unfinished work is still required.
  2. Define the aggregation. Which module should begin collection, which ends it, and how many uniquely identified inputs belong to the operation? Compare that expectation with Source Module and the highlighted aggregation area.
  3. Compare configurations. Have modules been replaced, recreated or rewired since the failure? Compare the saved execution with the current scenario. Identical display names need not identify the same internal module.
  4. Test fresh and resumed paths separately. Use a copy with harmless data and disconnected external writes. Record whether only resumption fails or ordinary aggregation already produces an incorrect item count.
  5. Prepare a narrow support case. If the feeder error persists despite a traceable configuration, provide execution ID, affected modules, redacted blueprint and minimal reproduction steps. Do not label it a confirmed platform defect before investigation.

Recover a complete business result

Decide whether each saved operation needs controlled continuation or a reconciled restart. Include any external results already created. An Ignore route or substitute data is unsuitable if it silently omits required content while marking the scenario successful.

Preserve expected input IDs alongside the final collection. This exposes missing or repeated elements that a simple successful execution indicator would not reveal. If a process is redesigned for per-item recovery, retain a clear association with the parent business operation.

Frequently asked questions

Is creating a new aggregator enough?

It may change a fresh run, but does not automatically repair the state retained by an incomplete execution. Preserve that execution first.

When is the incident resolved?

When the specific business operation completes fully without duplicate effects. Compare expected element IDs with the actual aggregated result, rather than relying only on a green scenario status.

Help with this case

We can help isolate the affected process and prepare a controlled next step. Discuss case A0371 with KAN KAI WORKS.

Support with this issue

We investigate Make using your error message, versions and a narrowly scoped example.

Describe your case →Related service: AI & automation

Sources

Sources checked on 5 November 2026. Individual reports do not establish general prevalence.