Add retry on image push 5xx errors

Some registries can be flaky and return intermittent 5xx errors. This
change allows those errors to be retried, similarly to network-level
errors.

Note that this needs the upstream containerd fix
https://github.com/containerd/containerd/pull/5276 to work reliably.

This was tested with a registry that was modified to return 504 on every
other manifest PUT. Without the change, exports to the registry fail
every other attempt.  With the change and the related containerd change,
exports to the registry always succeed.

Signed-off-by: Aaron Lehmann <alehmann@netflix.com>

util/contentutil/copy.go

@@ -65,7 +65,7 @@ func CopyChain(ctx context.Context, ingester content.Ingester, provider content.
 	handlers := []images.Handler{
 		images.ChildrenHandler(provider),
 		filterHandler,
-		retryhandler.New(remotes.FetchHandler(ingester, &localFetcher{provider}), nil),
+		retryhandler.New(remotes.FetchHandler(ingester, &localFetcher{provider}), func(_ []byte) {}),
 	}
 
 	if err := images.Dispatch(ctx, images.Handlers(handlers...), nil, desc); err != nil {

util/imageutil/config.go

@@ -101,7 +101,7 @@ func Config(ctx context.Context, str string, resolver remotes.Resolver, cache Co
 	children := childrenConfigHandler(cache, platform)
 
 	handlers := []images.Handler{
-		retryhandler.New(remotes.FetchHandler(cache, fetcher), nil),
+		retryhandler.New(remotes.FetchHandler(cache, fetcher), func(_ []byte) {}),
 		children,
 	}
 	if err := images.Dispatch(ctx, images.Handlers(handlers...), nil, desc); err != nil {

util/progress/logs/logs.go

@@ -133,6 +133,7 @@ func (sw *streamWriter) Close() error {
 func LoggerFromContext(ctx context.Context) func([]byte) {
 	return func(dt []byte) {
 		pw, _, _ := progress.FromContext(ctx)
+		defer pw.Close()
 		pw.Write(identity.NewID(), client.VertexLog{
 			Stream: stderr,
 			Data:   []byte(dt),

util/resolver/retryhandler/retry.go

@@ -10,6 +10,7 @@ import (
 	"time"
 
 	"github.com/containerd/containerd/images"
+	remoteserrors "github.com/containerd/containerd/remotes/errors"
 	ocispec "github.com/opencontainers/image-spec/specs-go/v1"
 	"github.com/pkg/errors"
 )
@@ -48,6 +49,14 @@ func New(f images.HandlerFunc, logger func([]byte)) images.HandlerFunc {
 }
 
 func retryError(err error) bool {
+	// Retry on 5xx errors
+	var errUnexpectedStatus remoteserrors.ErrUnexpectedStatus
+	if errors.As(err, &errUnexpectedStatus) &&
+		errUnexpectedStatus.StatusCode >= 500 &&
+		errUnexpectedStatus.StatusCode <= 599 {
+		return true
+	}
+
 	if errors.Is(err, io.EOF) || errors.Is(err, syscall.ECONNRESET) || errors.Is(err, syscall.EPIPE) {
 		return true
 	}