Added body close on retry for downloader round trip (#10008)

Add missing body close method when webseed roundtrip is retried

Author: mh0lt

erigon-lib/downloader/downloader.go

@@ -24,6 +24,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"math"
 	"math/rand"
 	"net/http"
 	"net/url"
@@ -33,6 +34,7 @@ import (
 	"reflect"
 	"runtime"
 	"sort"
+	"strconv"
 	"strings"
 	"sync"
 	"sync/atomic"
@@ -148,15 +150,52 @@ func insertCloudflareHeaders(req *http.Request) {
 	}
 }
 
+// retryBackoff performs exponential backoff based on the attempt number and limited
+// by the provided minimum and maximum durations.
+//
+// It also tries to parse Retry-After response header when a http.StatusTooManyRequests
+// (HTTP Code 429) is found in the resp parameter. Hence it will return the number of
+// seconds the server states it may be ready to process more requests from this client.
+func calcBackoff(min, max time.Duration, attemptNum int, resp *http.Response) time.Duration {
+	if resp != nil {
+		if resp.StatusCode == http.StatusTooManyRequests || resp.StatusCode == http.StatusServiceUnavailable {
+			if s, ok := resp.Header["Retry-After"]; ok {
+				if sleep, err := strconv.ParseInt(s[0], 10, 64); err == nil {
+					return time.Second * time.Duration(sleep)
+				}
+			}
+		}
+	}
+
+	mult := math.Pow(2, float64(attemptNum)) * float64(min)
+	sleep := time.Duration(mult)
+	if float64(sleep) != mult || sleep > max {
+		sleep = max
+	}
+
+	return sleep
+}
+
 func (r *requestHandler) RoundTrip(req *http.Request) (resp *http.Response, err error) {
+	defer func() {
+		if r := recover(); r != nil {
+			if resp != nil && resp.Body != nil {
+				resp.Body.Close()
+				resp.Body = nil
+			}
+
+			err = fmt.Errorf("http client panic: %s", r)
+		}
+	}()
+
 	insertCloudflareHeaders(req)
 
 	resp, err = r.Transport.RoundTrip(req)
 
-	delay := 500 * time.Millisecond
 	attempts := 1
 	retry := true
 
+	const minDelay = 500 * time.Millisecond
 	const maxDelay = 5 * time.Second
 	const maxAttempts = 10
 
@@ -180,26 +219,35 @@ func (r *requestHandler) RoundTrip(req *http.Request) (resp *http.Response, err
 			r.downloader.stats.WebseedBytesDownload.Add(resp.ContentLength)
 			retry = false
 
-		case http.StatusInternalServerError, http.StatusBadGateway:
+		// the first two statuses here have been observed from cloudflare
+		// during testing.  The remainder are generally understood to be
+		// retriable http responses, calcBackoff will use the Retry-After
+		// header if its availible
+		case http.StatusInternalServerError, http.StatusBadGateway,
+			http.StatusRequestTimeout, http.StatusTooEarly,
+			http.StatusTooManyRequests, http.StatusServiceUnavailable,
+			http.StatusGatewayTimeout:
+
 			r.downloader.stats.WebseedServerFails.Add(1)
 
+			if resp.Body != nil {
+				resp.Body.Close()
+				resp.Body = nil
+			}
+
 			attempts++
-			delayTimer := time.NewTimer(delay)
+			delayTimer := time.NewTimer(calcBackoff(minDelay, maxDelay, attempts, resp))
 
 			select {
 			case <-delayTimer.C:
 				// Note this assumes the req.Body is nil
 				resp, err = r.Transport.RoundTrip(req)
 				r.downloader.stats.WebseedTripCount.Add(1)
 
-				if err == nil && delay < maxDelay {
-					delay = delay + (time.Duration(rand.Intn(200-75)+75)*delay)/100
-				}
-
 			case <-req.Context().Done():
 				err = req.Context().Err()
 			}
-			retry = attempts > maxAttempts
+			retry = attempts < maxAttempts
 
 		default:
 			r.downloader.stats.WebseedBytesDownload.Add(resp.ContentLength)