Mintplex-Labs/anything-llm · error · Error

Invalid source property provided

Error message

Invalid source property provided

What it means

After the required-namespace check passes, the pinecone 'namespace-stats' handler verifies the namespace exists in the index (describeIndexStats shows no such namespace) before fetching stats. The throw means the name parsed fine as a request but Pinecone has no vectors under it — the workspace was never embedded, or its embeddings were cleared.

Solutions

1. Embed at least one document into the workspace (upload via UI or data connector) so the namespace is created in Pinecone, then retry stats.
2. List what actually exists: GET /api/ext/vectors/namespaces, or check the Pinecone console's namespace list for the index.
3. If you recently recreated or switched indexes (PINECONE_INDEX), re-embed the workspaces — namespaces do not carry over.
4. Verify the slug character-for-character, including dashes from spaces and lowercase conversion.

Example fix

# before - stats for a never-embedded workspace
curl -X POST .../api/ext/vectors/namespace-stats -d '{"namespace": "empty-ws"}'
# -> Namespace by that name does not exist.

# after - list real namespaces, then query one
# 1) embed a document into the workspace via UI/collector
# 2) curl -H "Authorization: Bearer $KEY" .../api/ext/vectors/namespaces
# 3) curl -X POST .../namespace-stats -d '{"namespace": "populated-ws"}'

const namespaces = await client.get('/api/ext/vectors/namespaces'); // slugs that actually exist
const slugs = namespaces.map((n) => n.name);
if (!slugs.includes(targetNamespace)) {
  return { error: `Namespace '${targetNamespace}' has no embeddings — embed the workspace first` };
}
const stats = await client.post('/api/ext/vectors/namespace-stats', { namespace: targetNamespace });

async function namespaceIsPopulated(vectorDB, namespace) {
  if (!namespace) return false;
  try {
    return await vectorDB.hasNamespace(namespace);
  } catch {
    return false;
  }
}

try {
  const stats = await vectorDB['namespace-stats']({ namespace });
} catch (e) {
  if (/does not exist/i.test(e.message)) {
    // nothing to report on — surface 'workspace not embedded yet' to the caller
    return { message: `Workspace '${namespace}' has no embedded documents yet.` };
  }
  throw e;
}

When it happens

Trigger: Requesting stats for a workspace where no documents have been embedded yet; after embeddings were reset or the Pinecone index was wiped/recreated; querying a slug that exists as a workspace but was never populated; embed jobs that all failed earlier (e.g., empty-document errors) so nothing was ever upserted.

Common situations: Fresh install where the user checks stats before uploading/embedding anything; pointing PINECONE_INDEX at a new empty index after recreating it; failed ingestion runs that silently left the namespace empty; slug typo that matches no populated namespace.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

• res.statusText || "Error generating api key."
• e.message
• e.message
• Flow not found
• Flow not found

AI-assisted analysis of Mintplex-Labs/anything-llm@f92433b4ea (2026-08-18). Data as JSON: /api/errors/118c25afe05e8dbd. Report an issue: GitHub.

Appendix: source

Thrown at collector/extensions/resync/index.js:72

      throw new Error(`Failed to sync YouTube video transcript. ${reason}`);
    response.status(200).json({ success, content });
  } catch (e) {
    console.error(e);
    response.status(200).json({
      success: false,
      content: null,
    });
  }
}

/**
 * Fetches the content of a specific confluence page via its chunkSource.
 * Returns the content as a text string of the page in question and only that page.
 * @param {object} data - metadata from document (eg: chunkSource)
 * @param {import("../../middleware/setDataSigner").ResponseWithSigner} response
 */
async function resyncConfluence({ chunkSource }, response) {
  if (!chunkSource) throw new Error("Invalid source property provided");
  try {
    // Confluence data is `payload` encrypted. So we need to expand its
    // encrypted payload back into query params so we can reFetch the page with same access token/params.
    const source = response.locals.encryptionWorker.expandPayload(chunkSource);
    const {
      fetchConfluencePage,
    } = require("../../utils/extensions/Confluence");
    const baseUrl = source.searchParams.get("baseUrl");
    const { success, reason, content } = await fetchConfluencePage({
      // The stored pathname carries no scheme. Use the one from baseUrl so the
      // page url matches the one the loader builds from the same baseUrl.
      pageUrl: `${protocolOf(baseUrl)}${source.pathname}`,
      baseUrl,
      spaceKey: source.searchParams.get("spaceKey"),
      accessToken: source.searchParams.get("token"),
      username: source.searchParams.get("username"),
      personalAccessToken: source.searchParams.get("personalAccessToken"),
      cloud: source.searchParams.get("cloud") === "true",

View on GitHub (pinned to f92433b4ea)