Fix too_many_pings during DPU SetPackage by coordinating keepalive settings

The DPU proxy client was sending keepalive pings every 10s, but the gRPC
server uses the default EnforcementPolicy with MinTime=5m. During long-running
SetPackage operations (sonic-installer can take minutes), this mismatch
causes the server to terminate the connection with GOAWAY/ENHANCE_YOUR_CALM
and debug data 'too_many_pings'.

Fix:
1. Increase proxy client keepalive interval from 10s to 30s and timeout
   from 3s to 10s for better resilience during long operations.
2. Add explicit KeepaliveEnforcementPolicy to the gNMI server allowing
   pings as frequent as every 20s, with a safety margin above the client's
   30s interval.

Fixes: #619
Related: sonic-net/sonic-mgmt#23054

Signed-off-by: Dawei Huang <daweihuang@microsoft.com>

Author: hdwhdw

pkg/interceptors/dpuproxy/proxy.go

@@ -177,13 +177,18 @@ func (p *DPUProxy) getConnection(ctx context.Context, dpuIndex, ipAddress string
 		target := fmt.Sprintf("%s:%s", ipAddress, port)
 		glog.Infof("[DPUProxy] Trying to connect to DPU%s at %s (attempt %d/%d)", dpuIndex, target, i+1, len(portsToTry))
 
-		// Create connection with keepalive settings for long-lived connections
+		// Create connection with keepalive settings for long-lived connections.
+		// Use a conservative ping interval to avoid triggering the server's
+		// default EnforcementPolicy (MinTime=5m). Operations like SetPackage
+		// can block for minutes during image installation; aggressive pinging
+		// causes the server to send GOAWAY with "too_many_pings".
+		// See: https://github.com/sonic-net/sonic-gnmi/issues/619
 		conn, err := grpc.NewClient(
 			target,
 			grpc.WithTransportCredentials(insecure.NewCredentials()),
 			grpc.WithKeepaliveParams(keepalive.ClientParameters{
-				Time:                10 * time.Second, // Send keepalive ping every 10s
-				Timeout:             3 * time.Second,  // Wait 3s for ping ack before considering connection dead
+				Time:                30 * time.Second, // Send keepalive ping every 30s
+				Timeout:             10 * time.Second, // Wait 10s for ping ack before considering connection dead
 				PermitWithoutStream: true,             // Send pings even when no active RPCs
 			}),
 		)

telemetry/telemetry.go

@@ -527,6 +527,17 @@ func startGNMIServer(telemetryCfg *TelemetryConfig, cfg *gnmi.Config, serverCont
 				MaxConnectionIdle: time.Duration(*telemetryCfg.IdleConnDuration) * time.Second, // duration in which idle connection will be closed, default is inf
 			}
 
+			// Allow clients (e.g. DPU proxy) to send keepalive pings at a
+			// reasonable rate. Without this the default MinTime is 5 minutes,
+			// causing "too_many_pings" GOAWAY for clients that ping more
+			// frequently during long-running operations like SetPackage.
+			// See: https://github.com/sonic-net/sonic-gnmi/issues/619
+			keep_alive_policy := keepalive.EnforcementPolicy{
+				MinTime:             20 * time.Second, // Allow pings as frequent as every 20s
+				PermitWithoutStream: true,             // Allow pings when there are no active streams
+			}
+			commonOpts = append(commonOpts, grpc.KeepaliveEnforcementPolicy(keep_alive_policy))
+
 			tlsOpts = []grpc.ServerOption{grpc.Creds(credentials.NewTLS(tlsCfg))}
 
 			if *telemetryCfg.IdleConnDuration > 0 { // non inf case