diff --git a/internal/write/service_test.go b/internal/write/service_test.go
index ce062095..2bc5ce23 100644
--- a/internal/write/service_test.go
+++ b/internal/write/service_test.go
@@ -497,3 +497,91 @@ func TestErrorCallback(t *testing.T) {
 	assert.Equal(t, 2, srv.retryQueue.list.Len())
 
 }
+
+func minInt(a, b int) int {
+	if a > b {
+		return b
+	}
+	return a
+}
+
+func TestRetryIntervalAccumulation(t *testing.T) {
+	// log.Log.SetLogLevel(log.DebugLevel)
+	log.Log.SetLogLevel(log.InfoLevel)
+
+	// Setup test service with scenario's configuration
+	hs := test.NewTestService(t, "http://localhost:8086")
+	opts := write.DefaultOptions().
+		SetRetryInterval(20).
+		SetMaxRetryInterval(300).
+		SetMaxRetryTime(100)
+	ctx := context.Background()
+	srv := NewService("my-org", "my-bucket", hs, opts)
+	writeInterval := time.Duration(opts.RetryInterval()) * time.Millisecond
+
+	// Set permanent reply error to force writes fail and retry
+	hs.SetReplyError(&http.Error{StatusCode: 429})
+
+	lastInterval := uint(0)
+	assert.Equal(t, uint(0), srv.RetryDelay) // Should initialize to zero
+	i := 1
+	for ; i <= 45; i++ {
+		b := NewBatch(fmt.Sprintf("%d\n", i), opts.MaxRetryTime())
+		err := srv.HandleWrite(ctx, b)
+		assert.Equal(t, minInt(i, 5), srv.retryQueue.list.Len())
+		assert.GreaterOrEqual(t, srv.RetryDelay, lastInterval)         // Should not decrease while writes failing
+		assert.LessOrEqual(t, srv.RetryDelay, opts.MaxRetryInterval()) // Should not grow larger than max
+		if err != nil {
+			if lastInterval == opts.MaxRetryInterval() {
+				// Write attempt failed, and interval was already at max, so should stay there
+				assert.Equal(t, srv.RetryDelay, opts.MaxRetryInterval())
+				log.Log.Infof("Retry interval capped at %d ms", srv.RetryDelay)
+			} else {
+				// A write attempt was made and failed, so retry interval should have increased
+				assert.Greater(t, srv.RetryDelay, lastInterval)
+				log.Log.Infof("Retry interval increased to %d ms", srv.RetryDelay)
+			}
+		} else {
+			// Write attempt was not made, so retry interval should remain the same
+			assert.Equal(t, srv.RetryDelay, lastInterval)
+			log.Log.Infof("Retry interval still at %d ms", srv.RetryDelay)
+		}
+		lastInterval = srv.RetryDelay
+
+		<-time.After(writeInterval)
+	}
+
+	// Clear error and let write pass
+	hs.SetReplyError(nil)
+
+	// Wait until write queue is ready to retry; in meantime, keep writing and confirming queue state
+	retryTimeout := srv.lastWriteAttempt.Add(time.Millisecond * time.Duration(srv.RetryDelay))
+	log.Log.Infof("Continuing to write for %d ms until flushing write attempt", time.Until(retryTimeout).Milliseconds())
+	for ; time.Until(retryTimeout) >= 0; i++ {
+		b := NewBatch(fmt.Sprintf("%d\n", i), opts.MaxRetryTime())
+		err := srv.HandleWrite(ctx, b)
+		assert.Nil(t, err) // There should be no write attempt
+		assert.Equal(t, minInt(i, 5), srv.retryQueue.list.Len())
+		assert.Equal(t, srv.RetryDelay, opts.MaxRetryInterval()) // Should remain the same
+		log.Log.Infof("Retry interval still at %d ms", srv.RetryDelay)
+		<-time.After(writeInterval)
+	}
+
+	// Retry interval should now have expired, so this write attempt should succeed and cause retry queue to flush
+	b := NewBatch(fmt.Sprintf("%d\n", i), opts.MaxRetryTime())
+	err := srv.HandleWrite(ctx, b)
+	assert.Nil(t, err)
+	assert.Equal(t, 0, srv.retryQueue.list.Len())
+	assert.Equal(t, srv.RetryDelay, uint(0)) // Should reset to zero
+
+	// Ensure proper batches got written to server
+	require.Len(t, hs.Lines(), 5)
+	assert.Equal(t, fmt.Sprintf("%d", i-4), hs.Lines()[0])
+	assert.Equal(t, fmt.Sprintf("%d", i-3), hs.Lines()[1])
+	assert.Equal(t, fmt.Sprintf("%d", i-2), hs.Lines()[2])
+	assert.Equal(t, fmt.Sprintf("%d", i-1), hs.Lines()[3])
+	assert.Equal(t, fmt.Sprintf("%d", i-0), hs.Lines()[4])
+
+	// Debug line to capture output of successful test
+	// assert.True(t, false)
+}