dgraph-io/dgraph · error

while fetching schema

Error message

while fetching schema

What it means

When the request includes a schema clause (schema {} or schema(pred: [...]) DQL), Dgraph fetches schema metadata from the Alpha/Zero cluster over the network via worker.GetSchemaOverNetwork. Failures of that internal RPC surface as this wrapped error. It means schema metadata could not be retrieved, though the main query may have succeeded.

Source

Thrown at query/query.go:3067

		return er, err
	}
	er.Subgraphs = req.Subgraphs
	// calculate metrics.
	metrics := make(map[string]uint64)
	for _, sg := range er.Subgraphs {
		calculateMetrics(sg, metrics)
	}
	er.Metrics = metrics
	namespace, err := x.ExtractNamespace(ctx)
	if err != nil {
		return er, errors.Wrapf(err, "While processing query")
	}
	schemaProcessingStart := time.Now()
	if req.DqlQuery.Schema != nil {
		preds := x.NamespaceAttrList(namespace, req.DqlQuery.Schema.Predicates)
		req.DqlQuery.Schema.Predicates = preds
		if er.SchemaNode, err = worker.GetSchemaOverNetwork(ctx, req.DqlQuery.Schema); err != nil {
			return er, errors.Wrapf(err, "while fetching schema")
		}
		typeNames := x.NamespaceAttrList(namespace, req.DqlQuery.Schema.Types)
		req.DqlQuery.Schema.Types = typeNames
		if er.Types, err = worker.GetTypes(ctx, req.DqlQuery.Schema); err != nil {
			return er, errors.Wrapf(err, "while fetching types")
		}
	}

	if !x.IsRootNsOperation(ctx) {
		// Filter the schema nodes for the given namespace.
		er.SchemaNode = filterSchemaNodeForNamespace(namespace, er.SchemaNode)
		// Filter the types for the given namespace.
		er.Types = filterTypesForNamespace(namespace, er.Types)
	}
	req.Latency.Processing += time.Since(schemaProcessingStart)

	return er, nil
}

View on GitHub (pinned to 759e242be6)

Solutions

1. Check cluster health: curl localhost:8080/health and verify all Alphas and Zero are up
2. Retry once network issues are resolved; this is often transient during rebalancing
3. Verify the requested predicates exist in the schema (curl -XPOST localhost:8080/query -d 'schema {}')
4. Check Dgraph server logs for gRPC timeouts or unreachable peers and fix membership (e.g. remove/re-add dead replicas)

Example fix

// before: request dropped predicates
schema(pred: [oldpred]) { type index }
// after: query existing predicates only
schema(pred: [name]) { type index }

// Probe cluster health before schema queries
resp, err := http.Get("http://localhost:8080/health")
if err != nil || resp.StatusCode != 200 {
    return errors.New("cluster unhealthy; skip schema fetch")
}

er, err := q.ProcessQuery(ctx, req)
if err != nil && strings.Contains(err.Error(), "while fetching schema") {
    time.Sleep(backoff)
    er, err = q.ProcessQuery(ctx, req) // retry after transient RPC failure
}

When it happens

Trigger: Query string containing a schema{} block while the cluster is unhealthy: unreachable peers, the schema-carrying node down, request timeouts, or the requested predicate not existing and being rejected.

Common situations: Cluster during node replacement/rebalance, network partition between Alphas and Zero, misconfigured Zero addresses, or client asking for predicates that were dropped.

Related errors

• unique validation failed to fetch schema for predicates %v
• while fetching types
• cannot retrieve predicate information
• illegal rune found "%c", expecting {
• JSON map is followed by illegal rune "%c"

AI-assisted analysis of dgraph-io/dgraph@759e242be6 (2026-09-01). Data as JSON: /api/errors/921601f61836dc2e. Report an issue: GitHub.