IPVS: Remove incorrect ip_route_me_harder(), fix IPv6
authorJulius Volz <juliusv@google.com>
Mon, 8 Sep 2008 12:55:42 +0000 (14:55 +0200)
committerSimon Horman <horms@verge.net.au>
Mon, 8 Sep 2008 23:43:13 +0000 (09:43 +1000)
Remove an incorrect ip_route_me_harder() that was probably a result of
merging my IPv6 patches with the local client patches. With this, IPv6+NAT
are working again.

Signed-off-by: Julius Volz <juliusv@google.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
net/ipv4/ipvs/ip_vs_core.c

index f5180ac56be38c53a9f7babefcf02e5d09bcc3d1..bdc92d73fbee3eb19874a2c70b53c11cf1419aec 100644 (file)
@@ -904,15 +904,6 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp,
                if (ip_route_me_harder(skb, RTN_LOCAL) != 0)
                        goto drop;
 
-       /* For policy routing, packets originating from this
-        * machine itself may be routed differently to packets
-        * passing through.  We want this packet to be routed as
-        * if it came from this machine itself.  So re-compute
-        * the routing information.
-        */
-       if (ip_route_me_harder(skb, RTN_LOCAL) != 0)
-               goto drop;
-
        IP_VS_DBG_PKT(10, pp, skb, 0, "After SNAT");
 
        ip_vs_out_stats(cp, skb);