t.Logf("Unwanted exception for 'istioctl %s': %v. Stderr: %v", strings.Join(args, " "), fErr, stderr)
		return fErr
	}

	// output will be something like
	//      WORKLOAD    TOTAL RPS    ERROR RPS  P50 LATENCY  P90 LATENCY  P99 LATENCY
	//        server        0.182        0.000         40ms         74ms         97ms
	//
	lines := strings.Split(output, "\n")
	if len(lines) != 3 {
		t.Logf("Expected 2 lines of output, got %q", output)

	}

	var out bytes.Buffer
	printHeader(&out)
	printMetrics(&out, sm)
	output := out.String()

	expectedOutput := `                                  WORKLOAD    TOTAL RPS    ERROR RPS  P50 LATENCY  P90 LATENCY  P99 LATENCY
                                   details        0.040        0.000          2ms          4ms          4ms
`
	if output != expectedOutput {
		t.Fatalf("Unexpected output; got:\n %q\nwant:\n %q", output, expectedOutput)

namespace. It then executes a series of queries per requested workload to
find the following top-level workload metrics: total requests per second,
error rate, and request latency at p50, p90, and p99 percentiles. The
query results are printed to the console, organized by workload name.

All metrics returned are from server-side reports. This means that latencies