infiniflow/ragflow · error · Exception

No dataset is selected.

Error message

No dataset is selected.

What it means

After collecting kb_ids from plain IDs and canvas-variable lookups, the Retrieval component filters out falsy IDs and calls KnowledgebaseService.get_by_ids. If that returns nothing (agent/tools/retrieval.py:111), it raises Exception('No dataset is selected.'), meaning the resolved ID set is empty or none of the IDs correspond to existing rows.

Solutions

  1. Bind at least one existing dataset in the Retrieval component's configuration before running the canvas.
  2. If dataset_ids uses a variable, debug-run the canvas and confirm the variable produces a non-empty list of valid dataset names/IDs.
  3. Add a guard branch upstream: skip or short-circuit the Retrieval node when the selection is empty instead of letting it raise.
  4. Confirm the datasets still exist and are accessible to the agent's tenant.

Example fix

// before
"dataset_ids": []  // nothing bound

// after
"dataset_ids": ["6a1f2e...dataset-id..."]
Defensive patterns

Strategy: validation

Validate before calling

resolved = [kb_id for kb_id in dataset_ids if kb_id]
# for variable-based entries, resolve each '@' reference first
if not resolved:
    raise ValueError("Retrieval component has no dataset bound - select at least one")

Try / catch

try:
    retrieval.invoke()
except Exception as e:
    if "No dataset is selected" in str(e):
        skip_or_warn("retrieval skipped: empty dataset selection")
    else:
        raise

Prevention

When it happens

Trigger: dataset_ids is an empty list; the canvas variable feeding dataset_ids resolves to an empty list, empty string, or None (whose .id access or filtering yields nothing); all resolved IDs are empty strings and get filtered out; every looked-up dataset was deleted between authoring and run.

Common situations: A template agent whose dataset binding was cleared; an upstream component (e.g. a Switch or LLM branch) outputs an empty selection; datasets deleted after the agent was configured.

Related errors

AI-assisted analysis of infiniflow/ragflow@554fb1133a (2026-08-15). Data as JSON: /api/errors/85f0be7b8e45d4f5. Report an issue: GitHub.

Appendix: source

Thrown at agent/tools/retrieval.py:111

            if id.find("@") < 0:
                kb_ids.append(id)
                continue
            kb_nm = self._canvas.get_variable_value(id)
            # if kb_nm is a list
            kb_nm_list = kb_nm if isinstance(kb_nm, list) else [kb_nm]
            for nm_or_id in kb_nm_list:
                e, kb = KnowledgebaseService.get_by_name(nm_or_id, self._canvas._tenant_id)
                if not e:
                    e, kb = KnowledgebaseService.get_by_id(nm_or_id)
                    if not e:
                        raise Exception(f"Dataset({nm_or_id}) does not exist.")
                kb_ids.append(kb.id)

        filtered_kb_ids: list[str] = list(set([kb_id for kb_id in kb_ids if kb_id]))

        kbs = KnowledgebaseService.get_by_ids(filtered_kb_ids)
        if not kbs:
            raise Exception("No dataset is selected.")

        embd_nms = list(set([kb.embd_id for kb in kbs]))
        assert len(embd_nms) == 1, "Knowledge bases use different embedding models."

        embd_mdl = None
        if embd_nms:
            tenant_id = self._canvas.get_tenant_id()
            embd_model_config = resolve_model_config(tenant_id, LLMType.EMBEDDING, embd_nms[0])
            embd_mdl = LLMBundle(tenant_id, embd_model_config)

        rerank_mdl = None
        if self._param.rerank_id:
            rerank_model_config = resolve_model_config(kbs[0].tenant_id, LLMType.RERANK, self._param.rerank_id)
            rerank_mdl = LLMBundle(kbs[0].tenant_id, rerank_model_config)

        vars = self.get_input_elements_from_text(query_text)
        vars = {k: o["value"] for k, o in vars.items()}
        query = self.string_format(query_text, vars)

View on GitHub (pinned to 554fb1133a)