#include <linux/kernel.h>
#include <linux/ip.h>
#include <linux/sctp.h>
#include <net/ip.h>
#include <net/ip6_checksum.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv4.h>
#include <net/sctp/checksum.h>
#include <net/ip_vs.h>
static int
sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd,
int *verdict, struct ip_vs_conn **cpp,
struct ip_vs_iphdr *iph)
{
struct net *net;
struct ip_vs_service *svc;
struct netns_ipvs *ipvs;
sctp_chunkhdr_t _schunkh, *sch;
sctp_sctphdr_t *sh, _sctph;
sh = skb_header_pointer(skb, iph->len, sizeof(_sctph), &_sctph);
if (sh == NULL) {
*verdict = NF_DROP;
return 0;
}
sch = skb_header_pointer(skb, iph->len + sizeof(sctp_sctphdr_t),
sizeof(_schunkh), &_schunkh);
if (sch == NULL) {
*verdict = NF_DROP;
return 0;
}
net = skb_net(skb);
ipvs = net_ipvs(net);
rcu_read_lock();
if ((sch->type == SCTP_CID_INIT || sysctl_sloppy_sctp(ipvs)) &&
(svc = ip_vs_service_find(net, af, skb->mark, iph->protocol,
&iph->daddr, sh->dest))) {
int ignored;
if (ip_vs_todrop(ipvs)) {
/*
* It seems that we are very loaded.
* We have to drop this packet :(
*/
rcu_read_unlock();
*verdict = NF_DROP;
return 0;
}
/*
* Let the virtual server select a real server for the
* incoming connection, and create a connection entry.
*/
*cpp = ip_vs_schedule(svc, skb, pd, &ignored, iph);
if (!*cpp && ignored <= 0) {
if (!ignored)
*verdict = ip_vs_leave(svc, skb, pd, iph);
else
*verdict = NF_DROP;
rcu_read_unlock();
return 0;
}
}
rcu_read_unlock();
/* NF_ACCEPT */
return 1;
}
static void sctp_nat_csum(struct sk_buff *skb, sctp_sctphdr_t *sctph,
unsigned int sctphoff)
{
sctph->checksum = sctp_compute_cksum(skb, sctphoff);
skb->ip_summed = CHECKSUM_UNNECESSARY;
}
static int
sctp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
struct ip_vs_conn *cp, struct ip_vs_iphdr *iph)
{
sctp_sctphdr_t *sctph;
unsigned int sctphoff = iph->len;
bool payload_csum = false;
#ifdef CONFIG_IP_VS_IPV6
if (cp->af == AF_INET6 && iph->fragoffs)
return 1;
#endif
/* csum_check requires unshared skb */
if (!skb_make_writable(skb, sctphoff + sizeof(*sctph)))
return 0;
if (unlikely(cp->app != NULL)) {
int ret;
/* Some checks before mangling */
if (pp->csum_check && !pp->csum_check(cp->af, skb, pp))
return 0;
/* Call application helper if needed */
ret = ip_vs_app_pkt_out(cp, skb);
if (ret == 0)
return 0;
/* ret=2: csum update is needed after payload mangling */
if (ret == 2)
payload_csum = true;
}
sctph = (void *) skb_network_header(skb) + sctphoff;
/* Only update csum if we really have to */
if (sctph->source != cp->vport || payload_csum ||
skb->ip_summed == CHECKSUM_PARTIAL) {
sctph->source = cp->vport;
sctp_nat_csum(skb, sctph, sctphoff);
} else {
skb->ip_summed = CHECKSUM_UNNECESSARY;
}
return 1;
}
static int
sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
struct ip_vs_conn *cp, struct ip_vs_iphdr *iph)
{
sctp_sctphdr_t *sctph;
unsigned int sctphoff = iph->len;
bool payload_csum = false;
#ifdef CONFIG_IP_VS_IPV6
if (cp->af == AF_INET6 && iph->fragoffs)
return 1;
#endif
/* csum_check requires unshared skb */
if (!skb_make_writable(skb, sctphoff + sizeof(*sctph)))
return 0;
if (unlikely(cp->app != NULL)) {
int ret;
/* Some checks before mangling */
if (pp->csum_check && !pp->csum_check(cp->af, skb, pp))
return 0;
/* Call application helper if needed */
ret = ip_vs_app_pkt_in(cp, skb);
if (ret == 0)
return 0;
/* ret=2: csum update is needed after payload mangling */
if (ret == 2)
payload_csum = true;
}
sctph = (void *) skb_network_header(skb) + sctphoff;
/* Only update csum if we really have to */
if (sctph->dest != cp->dport || payload_csum ||
(skb->ip_summed == CHECKSUM_PARTIAL &&
!(skb_dst(skb)->dev->features & NETIF_F_SCTP_CSUM))) {
sctph->dest = cp->dport;
sctp_nat_csum(skb, sctph, sctphoff);
} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
skb->ip_summed = CHECKSUM_UNNECESSARY;
}
return 1;
}
static int
sctp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp)
{
unsigned int sctphoff;
struct sctphdr *sh, _sctph;
__le32 cmp, val;
#ifdef CONFIG_IP_VS_IPV6
if (af == AF_INET6)
sctphoff = sizeof(struct ipv6hdr);
else
#endif
sctphoff = ip_hdrlen(skb);
sh = skb_header_pointer(skb, sctphoff, sizeof(_sctph), &_sctph);
if (sh == NULL)
return 0;
cmp = sh->checksum;
val = sctp_compute_cksum(skb, sctphoff);
if (val != cmp) {
/* CRC failure, dump it. */
IP_VS_DBG_RL_PKT(0, af, pp, skb, 0,
"Failed checksum for");
return 0;
}
return 1;
}
enum ipvs_sctp_event_t {
IP_VS_SCTP_DATA = 0, /* DATA, SACK, HEARTBEATs */
IP_VS_SCTP_INIT,
IP_VS_SCTP_INIT_ACK,
IP_VS_SCTP_COOKIE_ECHO,
IP_VS_SCTP_COOKIE_ACK,
IP_VS_SCTP_SHUTDOWN,
IP_VS_SCTP_SHUTDOWN_ACK,
IP_VS_SCTP_SHUTDOWN_COMPLETE,
IP_VS_SCTP_ERROR,
IP_VS_SCTP_ABORT,
IP_VS_SCTP_EVENT_LAST
};
/* RFC 2960, 3.2 Chunk Field Descriptions */
static __u8 sctp_events[] = {
[SCTP_CID_DATA] = IP_VS_SCTP_DATA,
[SCTP_CID_INIT] = IP_VS_SCTP_INIT,
[SCTP_CID_INIT_ACK] = IP_VS_SCTP_INIT_ACK,
[SCTP_CID_SACK] = IP_VS_SCTP_DATA,
[SCTP_CID_HEARTBEAT] = IP_VS_SCTP_DATA,
[SCTP_CID_HEARTBEAT_ACK] = IP_VS_SCTP_DATA,
[SCTP_CID_ABORT] = IP_VS_SCTP_ABORT,
[SCTP_CID_SHUTDOWN] = IP_VS_SCTP_SHUTDOWN,
[SCTP_CID_SHUTDOWN_ACK] = IP_VS_SCTP_SHUTDOWN_ACK,
[SCTP_CID_ERROR] = IP_VS_SCTP_ERROR,
[SCTP_CID_COOKIE_ECHO] = IP_VS_SCTP_COOKIE_ECHO,
[SCTP_CID_COOKIE_ACK] = IP_VS_SCTP_COOKIE_ACK,
[SCTP_CID_ECN_ECNE] = IP_VS_SCTP_DATA,
[SCTP_CID_ECN_CWR] = IP_VS_SCTP_DATA,
[SCTP_CID_SHUTDOWN_COMPLETE] = IP_VS_SCTP_SHUTDOWN_COMPLETE,
};
/* SCTP States:
* See RFC 2960, 4. SCTP Association State Diagram
*
* New states (not in diagram):
* - INIT1 state: use shorter timeout for dropped INIT packets
* - REJECTED state: use shorter timeout if INIT is rejected with ABORT
* - INIT, COOKIE_SENT, COOKIE_REPLIED, COOKIE states: for better debugging
*
* The states are as seen in real server. In the diagram, INIT1, INIT,
* COOKIE_SENT and COOKIE_REPLIED processing happens in CLOSED state.
*
* States as per packets from client (C) and server (S):
*
* Setup of client connection:
* IP_VS_SCTP_S_INIT1: First C:INIT sent, wait for S:INIT-ACK
* IP_VS_SCTP_S_INIT: Next C:INIT sent, wait for S:INIT-ACK
* IP_VS_SCTP_S_COOKIE_SENT: S:INIT-ACK sent, wait for C:COOKIE-ECHO
* IP_VS_SCTP_S_COOKIE_REPLIED: C:COOKIE-ECHO sent, wait for S:COOKIE-ACK
*
* Setup of server connection:
* IP_VS_SCTP_S_COOKIE_WAIT: S:INIT sent, wait for C:INIT-ACK
* IP_VS_SCTP_S_COOKIE: C:INIT-ACK sent, wait for S:COOKIE-ECHO
* IP_VS_SCTP_S_COOKIE_ECHOED: S:COOKIE-ECHO sent, wait for C:COOKIE-ACK
*/
#define sNO IP_VS_SCTP_S_NONE
#define sI1 IP_VS_SCTP_S_INIT1
#define sIN IP_VS_SCTP_S_INIT
#define sCS IP_VS_SCTP_S_COOKIE_SENT
#define sCR IP_VS_SCTP_S_COOKIE_REPLIED
#define sCW IP_VS_SCTP_S_COOKIE_WAIT
#define sCO IP_VS_SCTP_S_COOKIE
#define sCE IP_VS_SCTP_S_COOKIE_ECHOED
#define sES IP_VS_SCTP_S_ESTABLISHED
#define sSS IP_VS_SCTP_S_SHUTDOWN_SENT
#define sSR IP_VS_SCTP_S_SHUTDOWN_RECEIVED
#define sSA IP_VS_SCTP_S_SHUTDOWN_ACK_SENT
#define sRJ IP_VS_SCTP_S_REJECTED
#define sCL IP_VS_SCTP_S_CLOSED
static const __u8 sctp_states
[IP_VS_DIR_LAST][IP_VS_SCTP_EVENT_LAST][IP_VS_SCTP_S_LAST] = {
{ /* INPUT */
/* sNO, sI1, sIN, sCS, sCR, sCW, sCO, sCE, sES, sSS, sSR, sSA, sRJ, sCL*/
/* d */{sES, sI1, sIN, sCS, sCR, sCW, sCO, sCE, sES, sSS, sSR, sSA, sRJ, sCL},
/* i */{sI1, sIN, sIN, sCS, sCR, sCW, sCO, sCE, sES, sSS, sSR, sSA, sIN, sIN},
/* i_a */{sCW, sCW, sCW, sCS, sCR, sCO, sCO, sCE, sES, sSS, sSR, sSA, sRJ, sCL},
/* c_e */{sCR, sIN, sIN, sCR, sCR, sCW, sCO, sCE, sES, sSS, sSR, sSA, sRJ, sCL},
/* c_a */{sES, sI1, sIN, sCS, sCR, sCW, sCO, sES, sES, sSS, sSR, sSA, sRJ, sCL},
/* s */{sSR, sI1, sIN, sCS, sCR, sCW, sC
评论0