type is not registered. Ensure that beam.RegisterType( ) is…

Error message

type %v is not registered. Ensure that beam.RegisterType(%v) is called before beam.Init().

What it means

checkTypeRegistered panics when the named struct type used with bigqueryio has not been registered with beam.RegisterType before beam.Init(). Beam's runtime needs a global type registry so worker binaries can decode/encode elements across processes; an unregistered type would fail during pipeline serialization/execution.

Solutions

Add beam.RegisterType(reflect.TypeOf(MyRow{})) in an init() function or before beam.Init() in main.
Verify the registration happens in the same binary that builds/runs the pipeline.
Check that the type passed to Read/Write is exactly the registered type (not a type alias with a different identity).

Example fix

// before
func main() {
    beam.Init()
    bigqueryio.Write(s, proj, ds, table, col) // col elem type MyRow
}

// after
type MyRow struct{ N int }
func init() {
    beam.RegisterType(reflect.TypeOf(MyRow{}))
}
func main() {
    beam.Init()
    bigqueryio.Write(s, proj, ds, table, col)
}

Defensive patterns

Strategy: validation

Validate before calling

if _, ok := runtime.LookupType(runtime.TypeKey(reflectx.SkipPtr(reflect.TypeOf(MyRow{})))); !ok {
    beam.RegisterType(reflect.TypeOf(MyRow{}))
}

Try / catch

defer func() {
    if r := recover(); r != nil {
        if strings.Contains(fmt.Sprint(r), "is not registered") {
            log.Fatalf("register the type with beam.RegisterType before beam.Init(): %v", r)
        }
        panic(r)
    }
}()

Prevention

Put beam.RegisterType calls in each type's package init().
Register every struct used as a PCollection element in distributed pipelines.
Keep beam.Init() after all registrations in main().

When it happens

Trigger: Calling bigqueryio.Read or bigqueryio.Write with a named struct type T for which beam.RegisterType(reflect.TypeOf(T{})) was never called, then running with a distributed runner that requires registration.

Common situations: Forgetting the beam.RegisterType line in main() before beam.Init(); adding a new row struct to an existing pipeline; refactoring that moves the type into another package without re-registering it.

Related errors

AI-assisted analysis of apache/beam@12126d8942 (2026-09-13). Data as JSON: /api/errors/2626647ca42922c2. Report an issue: GitHub.

Appendix: source

Thrown at sdks/go/pkg/beam/io/bigqueryio/bigquery.go:242

	checkTypeRegistered(t)

	schema, err := bigquery.InferSchema(reflect.Zero(t).Interface())
	if err != nil {
		panic(errors.Wrapf(err, "invalid schema type: %v", t))
	}
	return schema
}

func checkTypeRegistered(t reflect.Type) {
	t = reflectx.SkipPtr(t)
	key, ok := runtime.TypeKey(t)
	if !ok {
		panic(fmt.Sprintf("type %v must be a named type (not anonymous) for registration", t))
	}

	if _, registered := runtime.LookupType(key); !registered {
		panic(fmt.Sprintf("type %v is not registered. Ensure that beam.RegisterType(%v) "+
			"is called before beam.Init().", t, t))
	}
}

func mustParseTable(table string) QualifiedTableName {
	qn, err := NewQualifiedTableName(table)
	if err != nil {
		panic(err)
	}
	return qn
}

// TODO(herohde) 7/14/2017: allow WriteDispositions. The default
// is not quite what the Dataflow examples do.

// writeOptions represents additional options for executing a write
type writeOptions struct {
	// CreateDisposition specifies the circumstances under which destination table will be created

View on GitHub (pinned to 12126d8942)