[receiver/statsd] fix data race in TCP server shutdown

Author: asweet-confluent

.chloggen/fix-statsd-tcp-server-data-race.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: bug_fix
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: statsdreceiver
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Fixes a data race that can occur during shutdown of the statsdreceiver's TCP server implementation.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [42986]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [api]

receiver/statsdreceiver/internal/transport/tcp_server.go

@@ -20,7 +20,8 @@ var errTCPServerDone = errors.New("server stopped")
 type tcpServer struct {
 	listener  net.Listener
 	reporter  Reporter
-	wg        sync.WaitGroup
+	wg        *sync.WaitGroup
+	wgMu      sync.Mutex
 	transport Transport
 	stopChan  chan struct{}
 }
@@ -30,7 +31,9 @@ var _ Server = (*tcpServer)(nil)
 
 // NewTCPServer creates a transport.Server using TCP as its transport.
 func NewTCPServer(transport Transport, address string) (Server, error) {
-	var tsrv tcpServer
+	tsrv := tcpServer{
+		wg: &sync.WaitGroup{},
+	}
 	var err error
 
 	if !transport.IsStreamTransport() {
@@ -69,7 +72,10 @@ LOOP:
 
 		select {
 		case conn := <-connChan:
+			// Potential data race here because t.wg.Add is called concurrently with t.wg.Wait in Close().
+			t.wgMu.Lock()
 			t.wg.Add(1)
+			t.wgMu.Unlock()
 			go t.handleConn(conn, transferChan)
 		case <-t.stopChan:
 			break LOOP
@@ -88,7 +94,9 @@ func (t *tcpServer) handleConn(c net.Conn, transferChan chan<- Metric) {
 			if !errors.Is(err, io.EOF) {
 				t.reporter.OnDebugf("TCP transport (%s) Error reading payload: %v", c.LocalAddr(), err)
 			}
+			t.wgMu.Lock()
 			t.wg.Done()
+			t.wgMu.Unlock()
 			return
 		}
 		buf := bytes.NewBuffer(append(remainder, payload[0:n]...))
@@ -111,6 +119,8 @@ func (t *tcpServer) handleConn(c net.Conn, transferChan chan<- Metric) {
 // Close closes the server.
 func (t *tcpServer) Close() error {
 	close(t.stopChan)
+	t.wgMu.Lock()
 	t.wg.Wait()
+	t.wgMu.Unlock()
 	return t.listener.Close()
 }