linux/net/ipv4/xfrm4_input.c
Eric Dumazet c074da2810 ipv4: tcp: dont cache unconfirmed intput dst
DDOS synflood attacks hit badly IP route cache.

On typical machines, this cache is allowed to hold up to 8 Millions dst
entries, 256 bytes for each, for a total of 2GB of memory.

rt_garbage_collect() triggers and tries to cleanup things.

Eventually route cache is disabled but machine is under fire and might
OOM and crash.

This patch exploits the new TCP early demux, to set a nocache
boolean in case incoming TCP frame is for a not yet ESTABLISHED or
TIMEWAIT socket.

This 'nocache' boolean is then used in case dst entry is not found in
route cache, to create an unhashed dst entry (DST_NOCACHE)

SYN-cookie-ACK sent use a similar mechanism (ipv4: tcp: dont cache
output dst for syncookies), so after this patch, a machine is able to
absorb a DDOS synflood attack without polluting its IP route cache.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Hans Schillstrom <hans.schillstrom@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-27 15:34:24 -07:00

167 lines
4.2 KiB
C

/*
* xfrm4_input.c
*
* Changes:
* YOSHIFUJI Hideaki @USAGI
* Split up af-specific portion
* Derek Atkins <derek@ihtfp.com>
* Add Encapsulation support
*
*/
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/string.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv4.h>
#include <net/ip.h>
#include <net/xfrm.h>
int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb)
{
return xfrm4_extract_header(skb);
}
static inline int xfrm4_rcv_encap_finish(struct sk_buff *skb)
{
if (skb_dst(skb) == NULL) {
const struct iphdr *iph = ip_hdr(skb);
if (ip_route_input_noref(skb, iph->daddr, iph->saddr,
iph->tos, skb->dev, false))
goto drop;
}
return dst_input(skb);
drop:
kfree_skb(skb);
return NET_RX_DROP;
}
int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi,
int encap_type)
{
XFRM_SPI_SKB_CB(skb)->family = AF_INET;
XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
return xfrm_input(skb, nexthdr, spi, encap_type);
}
EXPORT_SYMBOL(xfrm4_rcv_encap);
int xfrm4_transport_finish(struct sk_buff *skb, int async)
{
struct iphdr *iph = ip_hdr(skb);
iph->protocol = XFRM_MODE_SKB_CB(skb)->protocol;
#ifndef CONFIG_NETFILTER
if (!async)
return -iph->protocol;
#endif
__skb_push(skb, skb->data - skb_network_header(skb));
iph->tot_len = htons(skb->len);
ip_send_check(iph);
NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, skb->dev, NULL,
xfrm4_rcv_encap_finish);
return 0;
}
/* If it's a keepalive packet, then just eat it.
* If it's an encapsulated packet, then pass it to the
* IPsec xfrm input.
* Returns 0 if skb passed to xfrm or was dropped.
* Returns >0 if skb should be passed to UDP.
* Returns <0 if skb should be resubmitted (-ret is protocol)
*/
int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb)
{
struct udp_sock *up = udp_sk(sk);
struct udphdr *uh;
struct iphdr *iph;
int iphlen, len;
__u8 *udpdata;
__be32 *udpdata32;
__u16 encap_type = up->encap_type;
/* if this is not encapsulated socket, then just return now */
if (!encap_type)
return 1;
/* If this is a paged skb, make sure we pull up
* whatever data we need to look at. */
len = skb->len - sizeof(struct udphdr);
if (!pskb_may_pull(skb, sizeof(struct udphdr) + min(len, 8)))
return 1;
/* Now we can get the pointers */
uh = udp_hdr(skb);
udpdata = (__u8 *)uh + sizeof(struct udphdr);
udpdata32 = (__be32 *)udpdata;
switch (encap_type) {
default:
case UDP_ENCAP_ESPINUDP:
/* Check if this is a keepalive packet. If so, eat it. */
if (len == 1 && udpdata[0] == 0xff) {
goto drop;
} else if (len > sizeof(struct ip_esp_hdr) && udpdata32[0] != 0) {
/* ESP Packet without Non-ESP header */
len = sizeof(struct udphdr);
} else
/* Must be an IKE packet.. pass it through */
return 1;
break;
case UDP_ENCAP_ESPINUDP_NON_IKE:
/* Check if this is a keepalive packet. If so, eat it. */
if (len == 1 && udpdata[0] == 0xff) {
goto drop;
} else if (len > 2 * sizeof(u32) + sizeof(struct ip_esp_hdr) &&
udpdata32[0] == 0 && udpdata32[1] == 0) {
/* ESP Packet with Non-IKE marker */
len = sizeof(struct udphdr) + 2 * sizeof(u32);
} else
/* Must be an IKE packet.. pass it through */
return 1;
break;
}
/* At this point we are sure that this is an ESPinUDP packet,
* so we need to remove 'len' bytes from the packet (the UDP
* header and optional ESP marker bytes) and then modify the
* protocol to ESP, and then call into the transform receiver.
*/
if (skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
goto drop;
/* Now we can update and verify the packet length... */
iph = ip_hdr(skb);
iphlen = iph->ihl << 2;
iph->tot_len = htons(ntohs(iph->tot_len) - len);
if (skb->len < iphlen + len) {
/* packet is too small!?! */
goto drop;
}
/* pull the data buffer up to the ESP header and set the
* transport header to point to ESP. Keep UDP on the stack
* for later.
*/
__skb_pull(skb, len);
skb_reset_transport_header(skb);
/* process ESP */
return xfrm4_rcv_encap(skb, IPPROTO_ESP, 0, encap_type);
drop:
kfree_skb(skb);
return 0;
}
int xfrm4_rcv(struct sk_buff *skb)
{
return xfrm4_rcv_spi(skb, ip_hdr(skb)->protocol, 0);
}
EXPORT_SYMBOL(xfrm4_rcv);