#ifdef CONFIG_IP_VS_DEBUG
EXPORT_SYMBOL(ip_vs_get_debug_level);
#endif
-EXPORT_SYMBOL(ip_vs_make_skb_writable);
/* ID used in ICMP lookups */
}
-int ip_vs_make_skb_writable(struct sk_buff **pskb, int writable_len)
-{
- struct sk_buff *skb = *pskb;
-
- /* skb is already used, better copy skb and its payload */
- if (unlikely(skb_shared(skb) || skb->sk))
- goto copy_skb;
-
- /* skb data is already used, copy it */
- if (unlikely(skb_cloned(skb)))
- goto copy_data;
-
- return pskb_may_pull(skb, writable_len);
-
- copy_data:
- if (unlikely(writable_len > skb->len))
- return 0;
- return !pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
-
- copy_skb:
- if (unlikely(writable_len > skb->len))
- return 0;
- skb = skb_copy(skb, GFP_ATOMIC);
- if (!skb)
- return 0;
- BUG_ON(skb_is_nonlinear(skb));
-
- /* Rest of kernel will get very unhappy if we pass it a
- suddenly-orphaned skbuff */
- if ((*pskb)->sk)
- skb_set_owner_w(skb, (*pskb)->sk);
- kfree_skb(*pskb);
- *pskb = skb;
- return 1;
-}
-
/*
* IPVS persistent scheduling function
* It creates a connection entry according to its template if exists,
* for VS/NAT.
*/
static unsigned int ip_vs_post_routing(unsigned int hooknum,
- struct sk_buff **pskb,
+ struct sk_buff *skb,
const struct net_device *in,
const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
- if (!((*pskb)->ipvs_property))
+ if (!skb->ipvs_property)
return NF_ACCEPT;
/* The packet was sent from IPVS, exit this chain */
return NF_STOP;
return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0));
}
-static inline struct sk_buff *
-ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user)
+static inline int ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user)
{
- skb = ip_defrag(skb, user);
- if (skb)
+ int err = ip_defrag(skb, user);
+
+ if (!err)
ip_send_check(ip_hdr(skb));
- return skb;
+
+ return err;
}
/*
* Currently handles error types - unreachable, quench, ttl exceeded.
* (Only used in VS/NAT)
*/
-static int ip_vs_out_icmp(struct sk_buff **pskb, int *related)
+static int ip_vs_out_icmp(struct sk_buff *skb, int *related)
{
- struct sk_buff *skb = *pskb;
struct iphdr *iph;
struct icmphdr _icmph, *ic;
struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */
/* reassemble IP fragments */
if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
- skb = ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT);
- if (!skb)
+ if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT))
return NF_STOLEN;
- *pskb = skb;
}
iph = ip_hdr(skb);
verdict = NF_DROP;
if (IP_VS_FWD_METHOD(cp) != 0) {
- IP_VS_ERR("shouldn't reach here, because the box is on the"
+ IP_VS_ERR("shouldn't reach here, because the box is on the "
"half connection in the tun/dr module.\n");
}
if (IPPROTO_TCP == cih->protocol || IPPROTO_UDP == cih->protocol)
offset += 2 * sizeof(__u16);
- if (!ip_vs_make_skb_writable(pskb, offset))
+ if (!skb_make_writable(skb, offset))
goto out;
- skb = *pskb;
ip_vs_nat_icmp(skb, pp, cp, 1);
* rewrite addresses of the packet and send it on its way...
*/
static unsigned int
-ip_vs_out(unsigned int hooknum, struct sk_buff **pskb,
+ip_vs_out(unsigned int hooknum, struct sk_buff *skb,
const struct net_device *in, const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
- struct sk_buff *skb = *pskb;
struct iphdr *iph;
struct ip_vs_protocol *pp;
struct ip_vs_conn *cp;
iph = ip_hdr(skb);
if (unlikely(iph->protocol == IPPROTO_ICMP)) {
- int related, verdict = ip_vs_out_icmp(pskb, &related);
+ int related, verdict = ip_vs_out_icmp(skb, &related);
if (related)
return verdict;
- skb = *pskb;
iph = ip_hdr(skb);
}
/* reassemble IP fragments */
if (unlikely(iph->frag_off & htons(IP_MF|IP_OFFSET) &&
!pp->dont_defrag)) {
- skb = ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT);
- if (!skb)
+ if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT))
return NF_STOLEN;
iph = ip_hdr(skb);
- *pskb = skb;
}
ihl = iph->ihl << 2;
IP_VS_DBG_PKT(11, pp, skb, 0, "Outgoing packet");
- if (!ip_vs_make_skb_writable(pskb, ihl))
+ if (!skb_make_writable(skb, ihl))
goto drop;
/* mangle the packet */
- if (pp->snat_handler && !pp->snat_handler(pskb, pp, cp))
+ if (pp->snat_handler && !pp->snat_handler(skb, pp, cp))
goto drop;
- skb = *pskb;
ip_hdr(skb)->saddr = cp->vaddr;
ip_send_check(ip_hdr(skb));
* if it came from this machine itself. So re-compute
* the routing information.
*/
- if (ip_route_me_harder(pskb, RTN_LOCAL) != 0)
+ if (ip_route_me_harder(skb, RTN_LOCAL) != 0)
goto drop;
- skb = *pskb;
IP_VS_DBG_PKT(10, pp, skb, 0, "After SNAT");
drop:
ip_vs_conn_put(cp);
- kfree_skb(*pskb);
+ kfree_skb(skb);
return NF_STOLEN;
}
* Currently handles error types - unreachable, quench, ttl exceeded.
*/
static int
-ip_vs_in_icmp(struct sk_buff **pskb, int *related, unsigned int hooknum)
+ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
{
- struct sk_buff *skb = *pskb;
struct iphdr *iph;
struct icmphdr _icmph, *ic;
struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */
/* reassemble IP fragments */
if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) {
- skb = ip_vs_gather_frags(skb,
- hooknum == NF_IP_LOCAL_IN ?
- IP_DEFRAG_VS_IN : IP_DEFRAG_VS_FWD);
- if (!skb)
+ if (ip_vs_gather_frags(skb, hooknum == NF_IP_LOCAL_IN ?
+ IP_DEFRAG_VS_IN : IP_DEFRAG_VS_FWD))
return NF_STOLEN;
- *pskb = skb;
}
iph = ip_hdr(skb);
* and send it on its way...
*/
static unsigned int
-ip_vs_in(unsigned int hooknum, struct sk_buff **pskb,
+ip_vs_in(unsigned int hooknum, struct sk_buff *skb,
const struct net_device *in, const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
- struct sk_buff *skb = *pskb;
struct iphdr *iph;
struct ip_vs_protocol *pp;
struct ip_vs_conn *cp;
iph = ip_hdr(skb);
if (unlikely(iph->protocol == IPPROTO_ICMP)) {
- int related, verdict = ip_vs_in_icmp(pskb, &related, hooknum);
+ int related, verdict = ip_vs_in_icmp(skb, &related, hooknum);
if (related)
return verdict;
- skb = *pskb;
iph = ip_hdr(skb);
}
ret = NF_ACCEPT;
}
- /* increase its packet counter and check if it is needed
- to be synchronized */
+ /* Increase its packet counter and check if it is needed
+ * to be synchronized
+ *
+ * Sync connection if it is about to close to
+ * encorage the standby servers to update the connections timeout
+ */
atomic_inc(&cp->in_pkts);
if ((ip_vs_sync_state & IP_VS_STATE_MASTER) &&
- (cp->protocol != IPPROTO_TCP ||
- cp->state == IP_VS_TCP_S_ESTABLISHED) &&
- (atomic_read(&cp->in_pkts) % sysctl_ip_vs_sync_threshold[1]
- == sysctl_ip_vs_sync_threshold[0]))
+ (((cp->protocol != IPPROTO_TCP ||
+ cp->state == IP_VS_TCP_S_ESTABLISHED) &&
+ (atomic_read(&cp->in_pkts) % sysctl_ip_vs_sync_threshold[1]
+ == sysctl_ip_vs_sync_threshold[0])) ||
+ ((cp->protocol == IPPROTO_TCP) && (cp->old_state != cp->state) &&
+ ((cp->state == IP_VS_TCP_S_FIN_WAIT) ||
+ (cp->state == IP_VS_TCP_S_CLOSE)))))
ip_vs_sync_conn(cp);
+ cp->old_state = cp->state;
ip_vs_conn_put(cp);
return ret;
* and send them to ip_vs_in_icmp.
*/
static unsigned int
-ip_vs_forward_icmp(unsigned int hooknum, struct sk_buff **pskb,
+ip_vs_forward_icmp(unsigned int hooknum, struct sk_buff *skb,
const struct net_device *in, const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
int r;
- if (ip_hdr(*pskb)->protocol != IPPROTO_ICMP)
+ if (ip_hdr(skb)->protocol != IPPROTO_ICMP)
return NF_ACCEPT;
- return ip_vs_in_icmp(pskb, &r, hooknum);
+ return ip_vs_in_icmp(skb, &r, hooknum);
}