iptables: don't NAT existing connections

Fixes a NAT bug that manifested as:
- A SYN was sent from gVisor to another host, unaffected by iptables.
- The corresponding SYN/ACK was NATted by a PREROUTING REDIRECT rule
  despite being part of the existing connection.
- The socket that sent the SYN never received the SYN/ACK and thus a
  connection could not be established.

We handle this (as Linux does) by tracking all connections, inserting a
no-op conntrack rule for new connections with no rules of their own.

Needed for istio support (#170).

1 files changed, 1 insertions, 1 deletions

diff --git a/pkg/tcpip/stack/iptables_targets.go b/pkg/tcpip/stack/iptables_targets.go
index d43f60c67..dc88033c7 100644
--- a/pkg/tcpip/stack/iptables_targets.go
+++ b/pkg/tcpip/stack/iptables_targets.go
@@ -153,7 +153,7 @@ func (rt RedirectTarget) Action(pkt *PacketBuffer, ct *ConnTrack, hook Hook, gso
 		// Set up conection for matching NAT rule. Only the first
 		// packet of the connection comes here. Other packets will be
 		// manipulated in connection tracking.
-		if conn := ct.createConnFor(pkt, hook, rt); conn != nil {
+		if conn := ct.insertRedirectConn(pkt, hook, rt); conn != nil {
 			ct.handlePacket(pkt, hook, gso, r)
 		}
 	default: