Bound concurrent image processing and upstream fetches (closes #64)

Nothing bounded total in-flight work, so a burst of cache misses across
hosts could exhaust memory. Two settings now do: max_concurrent_processing
(default the CPUs Go uses) and upstream_connections (default 64, beside
the per-host limit). A request that finds either full waits up to 10
seconds, then gets 503; a slot is released on every path. No request
holds source bytes while it waits: a cached source is read only after
the processing slot is taken, and a fetched one only while it holds its
upstream connection. libvips runs one worker thread per image with its
operation cache off. Both waits count toward downstream_timeout, as the
README says.

Model: opus-5-5

internal/httpfetcher/httpfetcher.go

@@ -1,5 +1,6 @@
 // Package httpfetcher fetches content from upstream HTTP origins with SSRF
-// protection, per-host connection limits, and content-type validation.
+// protection, connection limits per host and for all hosts together, and
+// content-type validation.
 package httpfetcher
 
 import (
@@ -28,8 +29,13 @@ const (
 	DefaultIdleConnTimeout       = 90 * time.Second
 	DefaultMaxRedirects          = 10
 	DefaultMaxConnectionsPerHost = 20
+	DefaultMaxConnections        = 64
 )
 
+// ConnectionWaitTimeout is how long Fetch waits for a free connection when
+// MaxConnections fetches are already in progress.
+const ConnectionWaitTimeout = 10 * time.Second
+
 // MIME content types.
 const (
 	contentTypeJPEG        = "image/jpeg"
@@ -70,6 +76,7 @@ var (
 	ErrInvalidContentType = errors.New("invalid or unsupported content type")
 	ErrUpstreamError      = errors.New("upstream server error")
 	ErrUpstreamTimeout    = errors.New("upstream request timeout")
+	ErrTooManyConnections = errors.New("too many concurrent upstream connections")
 )
 
 // Internal fetcher errors.
@@ -122,6 +129,9 @@ type Config struct {
 	AllowHTTP bool
 	// MaxConnectionsPerHost limits concurrent connections to each upstream host.
 	MaxConnectionsPerHost int
+	// MaxConnections limits concurrent connections to all upstream hosts
+	// together.
+	MaxConnections int
 	// BlockedNetworks are operator-supplied CIDR ranges refused by the
 	// dialer, in addition to the always-enforced built-in ranges.
 	BlockedNetworks []netip.Prefix
@@ -143,15 +153,22 @@ func DefaultConfig() *Config {
 		},
 		AllowHTTP:             false,
 		MaxConnectionsPerHost: DefaultMaxConnectionsPerHost,
+		MaxConnections:        DefaultMaxConnections,
 	}
 }
 
-// HTTPFetcher implements Fetcher with SSRF protection and per-host connection limits.
+// HTTPFetcher implements Fetcher with SSRF protection and connection limits
+// per host and for all hosts together.
 type HTTPFetcher struct {
 	client    *http.Client
 	config    *Config
 	hostSems  map[string]chan struct{} // per-host semaphores
 	hostSemMu sync.Mutex               // protects hostSems map
+	// allHostsSemaphore has one slot per connection allowed to all hosts
+	// together (config.MaxConnections).
+	allHostsSemaphore chan struct{}
+	// connectionWaitTimeout is ConnectionWaitTimeout; tests shorten it.
+	connectionWaitTimeout time.Duration
 }
 
 // New creates a new HTTPFetcher with SSRF protection.
@@ -192,13 +209,18 @@ func New(config *Config) *HTTPFetcher {
 	}
 
 	return &HTTPFetcher{
-		client:   client,
-		config:   config,
-		hostSems: make(map[string]chan struct{}),
+		client:                client,
+		config:                config,
+		hostSems:              make(map[string]chan struct{}),
+		allHostsSemaphore:     make(chan struct{}, config.MaxConnections),
+		connectionWaitTimeout: ConnectionWaitTimeout,
 	}
 }
 
-// Fetch retrieves content from the given URL with SSRF protection.
+// Fetch retrieves content from the given URL with SSRF protection. When
+// MaxConnections fetches are already in progress, it waits up to
+// ConnectionWaitTimeout for one to finish, then fails with
+// ErrTooManyConnections.
 func (f *HTTPFetcher) Fetch(ctx context.Context, url string) (*FetchResult, error) {
 	// Validate URL before making request
 	err := validateURL(ctx, url, f.config.AllowHTTP)
@@ -206,24 +228,17 @@ func (f *HTTPFetcher) Fetch(ctx context.Context, url string) (*FetchResult, erro
 		return nil, err
 	}
 
-	// Extract host for rate limiting
-	host := extractHost(url)
-
-	// Acquire semaphore slot for this host
-	sem := f.getHostSemaphore(host)
-	select {
-	case sem <- struct{}{}:
-		// Acquired slot
-	case <-ctx.Done():
-		return nil, ctx.Err()
+	release, err := f.acquireConnection(ctx, extractHost(url))
+	if err != nil {
+		return nil, err
 	}
 
-	// If we fail before returning a result, release the slot
+	// If we fail before returning a result, release the connection
 	success := false
 
 	defer func() {
 		if !success {
-			<-sem
+			release()
 		}
 	}()
 
@@ -267,17 +282,52 @@ func (f *HTTPFetcher) Fetch(ctx context.Context, url string) (*FetchResult, erro
 		return nil, fmt.Errorf("upstream request failed: %w", err)
 	}
 
-	result, err := f.buildResult(resp, remoteAddr, fetchDuration, sem)
+	result, err := f.buildResult(resp, remoteAddr, fetchDuration, release)
 	if err != nil {
 		return nil, err
 	}
 
-	// Mark success so defer doesn't release the semaphore
+	// Mark success so defer doesn't release the connection; closing the
+	// result's Content does
 	success = true
 
 	return result, nil
 }
 
+// acquireConnection takes a slot for host, then one of the slots shared by
+// all hosts, and returns the func that gives both back. The host's slot
+// comes first, so fetches queued for one busy host hold no shared slot.
+// Only the wait for a shared slot is bounded: after connectionWaitTimeout
+// it fails with ErrTooManyConnections.
+func (f *HTTPFetcher) acquireConnection(
+	ctx context.Context, host string,
+) (func(), error) {
+	hostSem := f.getHostSemaphore(host)
+
+	select {
+	case hostSem <- struct{}{}:
+	case <-ctx.Done():
+		return nil, ctx.Err()
+	}
+
+	select {
+	case f.allHostsSemaphore <- struct{}{}:
+	case <-time.After(f.connectionWaitTimeout):
+		<-hostSem
+
+		return nil, ErrTooManyConnections
+	case <-ctx.Done():
+		<-hostSem
+
+		return nil, ctx.Err()
+	}
+
+	return func() {
+		<-hostSem
+		<-f.allHostsSemaphore
+	}, nil
+}
+
 // getHostSemaphore returns the semaphore for a host, creating it if necessary.
 func (f *HTTPFetcher) getHostSemaphore(host string) chan struct{} {
 	f.hostSemMu.Lock()
@@ -293,12 +343,12 @@ func (f *HTTPFetcher) getHostSemaphore(host string) chan struct{} {
 }
 
 // buildResult validates the upstream response and assembles a FetchResult
-// whose Content releases the host semaphore slot when closed.
+// whose Content calls release when closed.
 func (f *HTTPFetcher) buildResult(
 	resp *http.Response,
 	remoteAddr string,
 	fetchDuration time.Duration,
-	sem chan struct{},
+	release func(),
 ) (*FetchResult, error) {
 	// Extract HTTP version (strip "HTTP/" prefix)
 	httpVersion := strings.TrimPrefix(resp.Proto, "HTTP/")
@@ -333,7 +383,7 @@ func (f *HTTPFetcher) buildResult(
 	}
 
 	return &FetchResult{
-		Content:         &semaphoreReleasingReadCloser{limitedBody, resp.Body, sem},
+		Content:         &semaphoreReleasingReadCloser{limitedBody, resp.Body, release},
 		ContentLength:   resp.ContentLength,
 		ContentType:     contentType,
 		Headers:         resp.Header,
@@ -574,17 +624,18 @@ func (r *limitedReader) Read(p []byte) (int, error) {
 	return n, err
 }
 
-// semaphoreReleasingReadCloser releases a semaphore slot when closed.
+// semaphoreReleasingReadCloser releases the fetch's connection slots when
+// closed.
 type semaphoreReleasingReadCloser struct {
 	*limitedReader
 
-	closer io.Closer
-	sem    chan struct{}
+	closer  io.Closer
+	release func()
 }
 
 func (r *semaphoreReleasingReadCloser) Close() error {
 	err := r.closer.Close()
-	<-r.sem // Release semaphore slot
+	r.release()
 
 	return err
 }

internal/httpfetcher/max_connections_internal_test.go

@@ -0,0 +1,165 @@
+package httpfetcher
+
+import (
+	"context"
+	"errors"
+	"net"
+	"strconv"
+	"testing"
+	"time"
+)
+
+// imageURLOnPort is the fake upstream's image route on testPublicHost at
+// port. Each port is a different host to the per-host limit, while the test
+// dialer sends every port to the one test server.
+func imageURLOnPort(port int) string {
+	return "http://" + net.JoinHostPort(testPublicHost, strconv.Itoa(port)) +
+		"/image"
+}
+
+func TestDefaultConfigMaxConnections(t *testing.T) {
+	t.Parallel()
+
+	if got := DefaultConfig().MaxConnections; got != DefaultMaxConnections {
+		t.Errorf("MaxConnections = %d, want %d", got, DefaultMaxConnections)
+	}
+}
+
+// TestFetchLimitsConnectionsToAllHostsTogether checks that MaxConnections
+// counts the fetches to every host together, apart from the per-host
+// limit: with MaxConnections at 2 and two responses open from two hosts, a
+// fetch from a third host, which has nothing open, waits the whole wait
+// timeout and fails with ErrTooManyConnections. Closing one response lets
+// it through.
+func TestFetchLimitsConnectionsToAllHostsTogether(t *testing.T) {
+	t.Parallel()
+
+	srv := startUpstream(t)
+
+	cfg := DefaultConfig()
+	cfg.MaxConnections = 2
+
+	f, _ := newServerFetcher(t, srv, cfg)
+	f.connectionWaitTimeout = 100 * time.Millisecond
+
+	first, err := f.Fetch(testContext(t), imageURLOnPort(81))
+	if err != nil {
+		t.Fatalf("first Fetch() error = %v", err)
+	}
+
+	second, err := f.Fetch(testContext(t), imageURLOnPort(82))
+	if err != nil {
+		t.Fatalf("second Fetch() error = %v", err)
+	}
+
+	defer func() { _ = second.Content.Close() }()
+
+	start := time.Now()
+
+	_, err = f.Fetch(testContext(t), imageURLOnPort(83))
+	if !errors.Is(err, ErrTooManyConnections) {
+		t.Fatalf("third Fetch() error = %v, want ErrTooManyConnections", err)
+	}
+
+	if waited := time.Since(start); waited < f.connectionWaitTimeout {
+		t.Errorf("third Fetch() failed after %v, before waiting %v",
+			waited, f.connectionWaitTimeout)
+	}
+
+	if held := semLen(f, testPublicHost+":83"); held != 0 {
+		t.Errorf("the refused fetch kept its host's slot: %d held", held)
+	}
+
+	err = first.Content.Close()
+	if err != nil {
+		t.Fatalf("close first body: %v", err)
+	}
+
+	third, err := f.Fetch(testContext(t), imageURLOnPort(83))
+	if err != nil {
+		t.Fatalf("Fetch() after a response was closed: error = %v", err)
+	}
+
+	_ = third.Content.Close()
+}
+
+// TestFetchFreesHostSlotWhenContextEndsWaitingForConnection checks that a
+// fetch whose request context ends while it waits for a connection shared
+// by all hosts gives its host's slot back. With MaxConnections at 1 and one
+// response open, a fetch from another host takes that host's slot and waits;
+// its context ends long before the 10 second wait timeout.
+func TestFetchFreesHostSlotWhenContextEndsWaitingForConnection(t *testing.T) {
+	t.Parallel()
+
+	srv := startUpstream(t)
+
+	cfg := DefaultConfig()
+	cfg.MaxConnections = 1
+
+	f, _ := newServerFetcher(t, srv, cfg)
+
+	first, err := f.Fetch(testContext(t), imageURLOnPort(81))
+	if err != nil {
+		t.Fatalf("first Fetch() error = %v", err)
+	}
+
+	defer func() { _ = first.Content.Close() }()
+
+	ctx, cancel := context.WithTimeout(t.Context(), 100*time.Millisecond)
+	defer cancel()
+
+	_, err = f.Fetch(ctx, imageURLOnPort(82))
+	if !errors.Is(err, context.DeadlineExceeded) {
+		t.Fatalf("second Fetch() error = %v, want context.DeadlineExceeded", err)
+	}
+
+	if held := semLen(f, testPublicHost+":82"); held != 0 {
+		t.Errorf("the fetch kept its host's slot after its context ended: "+
+			"%d held", held)
+	}
+}
+
+// TestFetchReleasesConnectionOnError checks that a fetch that fails after
+// taking its connection gives it back: with MaxConnections at 1, the slot
+// must be free after the failure and the next fetch must succeed.
+func TestFetchReleasesConnectionOnError(t *testing.T) {
+	t.Parallel()
+
+	cases := []struct {
+		name string
+		url  string
+		want error
+	}{
+		{"upstream answers 500", upstreamURL("/status/500"), ErrUpstreamError},
+		{"upstream sends HTML", upstreamURL("/html"), ErrInvalidContentType},
+		// 198.51.100.7 (TEST-NET-2) passes the SSRF checks, and the test
+		// dialer refuses every host but testPublicHost.
+		{"connecting fails", "http://198.51.100.7/image", errUnexpectedDial},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			t.Parallel()
+
+			srv := startUpstream(t)
+
+			cfg := DefaultConfig()
+			cfg.MaxConnections = 1
+
+			f, _ := newServerFetcher(t, srv, cfg)
+			f.connectionWaitTimeout = 100 * time.Millisecond
+
+			_, err := f.Fetch(testContext(t), tc.url)
+			if !errors.Is(err, tc.want) {
+				t.Fatalf("Fetch() error = %v, want %v", err, tc.want)
+			}
+
+			if held := len(f.allHostsSemaphore); held != 0 {
+				t.Fatalf("connection still held after the error: %d held", held)
+			}
+
+			res := fetchImage(t, f, "/image")
+			_ = res.Content.Close()
+		})
+	}
+}