What causes "canceled: %v with %w" in amir20/dozzle?

Check errors.Is(err, context.Canceled) and log at debug level; it is usually expected during shutdown/navigation

canceled: with

Error message

canceled: %v with %w

What it means

rpcErrToErr maps gRPC status code Canceled to "canceled: %v with %w", wrapping context.Canceled so callers can use errors.Is(err, context.Canceled). It indicates the RPC was cancelled, typically by the caller's own context or a client disconnect, not a server failure.

Solutions

Check errors.Is(err, context.Canceled) and log at debug level; it is usually expected during shutdown/navigation
If cancellations are unexpected, audit who cancels the parent ctx (http request lifecycle, timeout wrappers)
Do not reconnect on Canceled; only resubscribe on genuine stream failures
Increase per-request deadlines if intentional timeouts are cutting streams short

Example fix

// before
if err := <-streamErr; err != nil { log.Error().Err(err).Msg("stream error") }
// after
if err := <-streamErr; err != nil {
    if errors.Is(err, context.Canceled) { return } // expected on disconnect
    log.Error().Err(err).Msg("stream error")
}

Defensive patterns

Strategy: try-catch

Try / catch

if err != nil && errors.Is(err, context.Canceled) {
    return // expected on client disconnect or shutdown; do not log as error
}

Prevention

Never treat Canceled as a failure; it accompanies normal SSE teardown and shutdown
Tie stream contexts to request lifetimes so cancellation is deterministic
Avoid wrapping stream contexts in additional timeouts you do not control

When it happens

Trigger: Streaming calls (StreamEvents, StreamStats, sendLogs, etc.) whose ctx is cancelled: page navigation closing the SSE stream, client-side timeout, or explicit context cancel/shutdown.

Common situations: User closes a browser tab, cancelling the SSE request context; server shutdown cancelling all in-flight streams; a wrapped HTTP handler timing out and cancelling the request context.

Related errors

AI-assisted analysis of amir20/dozzle@d9463cbe21 (2026-09-07). Data as JSON: /api/errors/ee0d3a94e931011a. Report an issue: GitHub.

Appendix: source

Thrown at internal/agent/client.go:123

}

func rpcErrToErr(err error) error {
	if err == nil {
		return nil
	}

	status, ok := status.FromError(err)
	if !ok {
		return err
	}

	if status.Code() == codes.Unknown && status.Message() == "EOF" {
		return fmt.Errorf("EOF error while converting gRPC to error: %w", io.EOF)
	}

	switch status.Code() {
	case codes.Canceled:
		return fmt.Errorf("canceled: %v with %w", status.Message(), context.Canceled)
	case codes.DeadlineExceeded:
		return fmt.Errorf("deadline exceeded: %v with %w", status.Message(), context.DeadlineExceeded)
	case codes.Unknown:
		return fmt.Errorf("unknown error: %v with %w", status.Message(), err)
	case codes.OK:
		return nil
	default:
		return fmt.Errorf("unknown code: %v with %w", status.Code(), err)
	}
}

func (c *Client) LogsBetweenDates(ctx context.Context, containerID string, since time.Time, until time.Time, std container.StdType) (<-chan *container.LogEvent, error) {
	stream, err := c.client.LogsBetweenDates(ctx, &pb.LogsBetweenDatesRequest{
		ContainerId: containerID,
		Since:       timestamppb.New(since),
		Until:       timestamppb.New(until),
		StreamTypes: int32(std),
	})

View on GitHub (pinned to d9463cbe21)