* [PATCH 1/6] netfilter: nf_conntrack: make sequence number adjustments usuable without NAT
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-23 7:52 ` [PATCH 2/6] net: syncookies: export cookie_v4_init_sequence/cookie_v4_check Patrick McHardy
` (6 subsequent siblings)
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Split out sequence number adjustments from NAT and move them to the conntrack
core to make them usable for SYN proxying. The sequence number adjustment
information is moved to a seperate extend. The extend is added to new
conntracks when a NAT mapping is set up for a connection using a helper.
As a side effect, this saves 24 bytes per connection with NAT in the common
case that a connection does not have a helper assigned.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
include/linux/netfilter.h | 9 +-
include/net/netfilter/nf_conntrack_extend.h | 2 +
include/net/netfilter/nf_conntrack_seqadj.h | 49 +++++
include/net/netfilter/nf_nat.h | 10 -
include/net/netfilter/nf_nat_helper.h | 19 --
include/uapi/linux/netfilter/nf_conntrack_common.h | 3 +-
include/uapi/linux/netfilter/nfnetlink_conntrack.h | 15 +-
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 7 +-
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 7 +-
net/netfilter/Makefile | 2 +-
net/netfilter/nf_conntrack_core.c | 16 +-
net/netfilter/nf_conntrack_netlink.c | 115 +++++------
net/netfilter/nf_conntrack_proto_tcp.c | 18 +-
net/netfilter/nf_conntrack_seqadj.c | 218 ++++++++++++++++++++
net/netfilter/nf_nat_core.c | 16 +-
net/netfilter/nf_nat_helper.c | 228 +--------------------
net/netfilter/nf_nat_sip.c | 3 +-
net/netfilter/nfnetlink_queue_ct.c | 8 +-
18 files changed, 369 insertions(+), 376 deletions(-)
create mode 100644 include/net/netfilter/nf_conntrack_seqadj.h
create mode 100644 net/netfilter/nf_conntrack_seqadj.c
diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index e2cf786be..708fe72ab9 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -319,6 +319,7 @@ extern void nf_ct_attach(struct sk_buff *, const struct sk_buff *);
extern void (*nf_ct_destroy)(struct nf_conntrack *) __rcu;
struct nf_conn;
+enum ip_conntrack_info;
struct nlattr;
struct nfq_ct_hook {
@@ -327,14 +328,10 @@ struct nfq_ct_hook {
int (*parse)(const struct nlattr *attr, struct nf_conn *ct);
int (*attach_expect)(const struct nlattr *attr, struct nf_conn *ct,
u32 portid, u32 report);
-};
-extern struct nfq_ct_hook __rcu *nfq_ct_hook;
-
-struct nfq_ct_nat_hook {
void (*seq_adjust)(struct sk_buff *skb, struct nf_conn *ct,
- u32 ctinfo, s32 off);
+ enum ip_conntrack_info ctinfo, s32 off);
};
-extern struct nfq_ct_nat_hook __rcu *nfq_ct_nat_hook;
+extern struct nfq_ct_hook __rcu *nfq_ct_hook;
#else
static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {}
#endif
diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h
index 977bc8a..2a22bcb 100644
--- a/include/net/netfilter/nf_conntrack_extend.h
+++ b/include/net/netfilter/nf_conntrack_extend.h
@@ -9,6 +9,7 @@ enum nf_ct_ext_id {
NF_CT_EXT_HELPER,
#if defined(CONFIG_NF_NAT) || defined(CONFIG_NF_NAT_MODULE)
NF_CT_EXT_NAT,
+ NF_CT_EXT_SEQADJ,
#endif
NF_CT_EXT_ACCT,
#ifdef CONFIG_NF_CONNTRACK_EVENTS
@@ -31,6 +32,7 @@ enum nf_ct_ext_id {
#define NF_CT_EXT_HELPER_TYPE struct nf_conn_help
#define NF_CT_EXT_NAT_TYPE struct nf_conn_nat
+#define NF_CT_EXT_SEQADJ_TYPE struct nf_conn_seqadj
#define NF_CT_EXT_ACCT_TYPE struct nf_conn_counter
#define NF_CT_EXT_ECACHE_TYPE struct nf_conntrack_ecache
#define NF_CT_EXT_ZONE_TYPE struct nf_conntrack_zone
diff --git a/include/net/netfilter/nf_conntrack_seqadj.h b/include/net/netfilter/nf_conntrack_seqadj.h
new file mode 100644
index 0000000..30bfbbe
--- /dev/null
+++ b/include/net/netfilter/nf_conntrack_seqadj.h
@@ -0,0 +1,49 @@
+#ifndef _NF_CONNTRACK_SEQADJ_H
+#define _NF_CONNTRACK_SEQADJ_H
+
+#include <net/netfilter/nf_conntrack_extend.h>
+
+/**
+ * struct nf_ct_seqadj - sequence number adjustment information
+ *
+ * @correction_pos: position of the last TCP sequence number modification
+ * @offset_before: sequence number offset before last modification
+ * @offset_after: sequence number offset after last modification
+ */
+struct nf_ct_seqadj {
+ u32 correction_pos;
+ s32 offset_before;
+ s32 offset_after;
+};
+
+struct nf_conn_seqadj {
+ struct nf_ct_seqadj seq[IP_CT_DIR_MAX];
+};
+
+static inline struct nf_conn_seqadj *nfct_seqadj(const struct nf_conn *ct)
+{
+ return nf_ct_ext_find(ct, NF_CT_EXT_SEQADJ);
+}
+
+static inline struct nf_conn_seqadj *nfct_seqadj_ext_add(struct nf_conn *ct)
+{
+ return nf_ct_ext_add(ct, NF_CT_EXT_SEQADJ, GFP_ATOMIC);
+}
+
+extern int nf_ct_seqadj_set(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ __be32 seq, s32 off);
+extern void nf_ct_tcp_seqadj_set(struct sk_buff *skb,
+ struct nf_conn *ct,
+ enum ip_conntrack_info ctinfo,
+ s32 off);
+
+extern int nf_ct_seq_adjust(struct sk_buff *skb,
+ struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ unsigned int protoff);
+extern s32 nf_ct_seq_offset(const struct nf_conn *ct, enum ip_conntrack_dir,
+ u32 seq);
+
+extern int nf_conntrack_seqadj_init(void);
+extern void nf_conntrack_seqadj_fini(void);
+
+#endif /* _NF_CONNTRACK_SEQADJ_H */
diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h
index e244141..59a1924 100644
--- a/include/net/netfilter/nf_nat.h
+++ b/include/net/netfilter/nf_nat.h
@@ -13,15 +13,6 @@ enum nf_nat_manip_type {
#define HOOK2MANIP(hooknum) ((hooknum) != NF_INET_POST_ROUTING && \
(hooknum) != NF_INET_LOCAL_IN)
-/* NAT sequence number modifications */
-struct nf_nat_seq {
- /* position of the last TCP sequence number modification (if any) */
- u_int32_t correction_pos;
-
- /* sequence number offset before and after last modification */
- int32_t offset_before, offset_after;
-};
-
#include <linux/list.h>
#include <linux/netfilter/nf_conntrack_pptp.h>
#include <net/netfilter/nf_conntrack_extend.h>
@@ -39,7 +30,6 @@ struct nf_conn;
/* The structure embedded in the conntrack structure. */
struct nf_conn_nat {
struct hlist_node bysource;
- struct nf_nat_seq seq[IP_CT_DIR_MAX];
struct nf_conn *ct;
union nf_conntrack_nat_help help;
#if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \
diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h
index 194c347..404324d 100644
--- a/include/net/netfilter/nf_nat_helper.h
+++ b/include/net/netfilter/nf_nat_helper.h
@@ -39,28 +39,9 @@ extern int nf_nat_mangle_udp_packet(struct sk_buff *skb,
const char *rep_buffer,
unsigned int rep_len);
-extern void nf_nat_set_seq_adjust(struct nf_conn *ct,
- enum ip_conntrack_info ctinfo,
- __be32 seq, s32 off);
-extern int nf_nat_seq_adjust(struct sk_buff *skb,
- struct nf_conn *ct,
- enum ip_conntrack_info ctinfo,
- unsigned int protoff);
-extern int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
- struct nf_conn *ct,
- enum ip_conntrack_info ctinfo,
- unsigned int protoff);
-
/* Setup NAT on this expected conntrack so it follows master, but goes
* to port ct->master->saved_proto. */
extern void nf_nat_follow_master(struct nf_conn *ct,
struct nf_conntrack_expect *this);
-extern s32 nf_nat_get_offset(const struct nf_conn *ct,
- enum ip_conntrack_dir dir,
- u32 seq);
-
-extern void nf_nat_tcp_seq_adjust(struct sk_buff *skb, struct nf_conn *ct,
- u32 dir, s32 off);
-
#endif
diff --git a/include/uapi/linux/netfilter/nf_conntrack_common.h b/include/uapi/linux/netfilter/nf_conntrack_common.h
index d69483f..8dd8038 100644
--- a/include/uapi/linux/netfilter/nf_conntrack_common.h
+++ b/include/uapi/linux/netfilter/nf_conntrack_common.h
@@ -99,7 +99,8 @@ enum ip_conntrack_events {
IPCT_PROTOINFO, /* protocol information has changed */
IPCT_HELPER, /* new helper has been set */
IPCT_MARK, /* new mark has been set */
- IPCT_NATSEQADJ, /* NAT is doing sequence adjustment */
+ IPCT_SEQADJ, /* sequence adjustment has changed */
+ IPCT_NATSEQADJ = IPCT_SEQADJ,
IPCT_SECMARK, /* new security mark has been set */
IPCT_LABEL, /* new connlabel has been set */
};
diff --git a/include/uapi/linux/netfilter/nfnetlink_conntrack.h b/include/uapi/linux/netfilter/nfnetlink_conntrack.h
index 08fabc6..acad6c5 100644
--- a/include/uapi/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/uapi/linux/netfilter/nfnetlink_conntrack.h
@@ -42,8 +42,10 @@ enum ctattr_type {
CTA_ID,
CTA_NAT_DST,
CTA_TUPLE_MASTER,
- CTA_NAT_SEQ_ADJ_ORIG,
- CTA_NAT_SEQ_ADJ_REPLY,
+ CTA_SEQ_ADJ_ORIG,
+ CTA_NAT_SEQ_ADJ_ORIG = CTA_SEQ_ADJ_ORIG,
+ CTA_SEQ_ADJ_REPLY,
+ CTA_NAT_SEQ_ADJ_REPLY = CTA_SEQ_ADJ_REPLY,
CTA_SECMARK, /* obsolete */
CTA_ZONE,
CTA_SECCTX,
@@ -165,6 +167,15 @@ enum ctattr_protonat {
};
#define CTA_PROTONAT_MAX (__CTA_PROTONAT_MAX - 1)
+enum ctattr_seqadj {
+ CTA_SEQADJ_UNSPEC,
+ CTA_SEQADJ_CORRECTION_POS,
+ CTA_SEQADJ_OFFSET_BEFORE,
+ CTA_SEQADJ_OFFSET_AFTER,
+ __CTA_SEQADJ_MAX
+};
+#define CTA_SEQADJ_MAX (__CTA_SEQADJ_MAX - 1)
+
enum ctattr_natseq {
CTA_NAT_SEQ_UNSPEC,
CTA_NAT_SEQ_CORRECTION_POS,
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index 0a2e0e3..86f5b34 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -25,6 +25,7 @@
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_zones.h>
#include <net/netfilter/nf_conntrack_core.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
#include <net/netfilter/nf_nat_helper.h>
#include <net/netfilter/ipv4/nf_defrag_ipv4.h>
@@ -136,11 +137,7 @@ static unsigned int ipv4_confirm(unsigned int hooknum,
/* adjust seqs for loopback traffic only in outgoing direction */
if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
!nf_is_loopback_packet(skb)) {
- typeof(nf_nat_seq_adjust_hook) seq_adjust;
-
- seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
- if (!seq_adjust ||
- !seq_adjust(skb, ct, ctinfo, ip_hdrlen(skb))) {
+ if (!nf_ct_seq_adjust(skb, ct, ctinfo, ip_hdrlen(skb))) {
NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
return NF_DROP;
}
diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
index c9b6a6e..d6e4dd8 100644
--- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c
@@ -28,6 +28,7 @@
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_zones.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/ipv6/nf_conntrack_ipv6.h>
#include <net/netfilter/nf_nat_helper.h>
#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
@@ -158,11 +159,7 @@ static unsigned int ipv6_confirm(unsigned int hooknum,
/* adjust seqs for loopback traffic only in outgoing direction */
if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
!nf_is_loopback_packet(skb)) {
- typeof(nf_nat_seq_adjust_hook) seq_adjust;
-
- seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
- if (!seq_adjust ||
- !seq_adjust(skb, ct, ctinfo, protoff)) {
+ if (!nf_ct_seq_adjust(skb, ct, ctinfo, protoff)) {
NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
return NF_DROP;
}
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index ebfa7dc..89a9c16 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -1,6 +1,6 @@
netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o
-nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_proto.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o nf_conntrack_extend.o nf_conntrack_acct.o
+nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_proto.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o nf_conntrack_extend.o nf_conntrack_acct.o nf_conntrack_seqadj.o
nf_conntrack-$(CONFIG_NF_CONNTRACK_TIMEOUT) += nf_conntrack_timeout.o
nf_conntrack-$(CONFIG_NF_CONNTRACK_TIMESTAMP) += nf_conntrack_timestamp.o
nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index da6f178..00a7a94 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -39,6 +39,7 @@
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_expect.h>
#include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_extend.h>
#include <net/netfilter/nf_conntrack_acct.h>
@@ -1326,6 +1327,7 @@ void nf_conntrack_cleanup_end(void)
nf_ct_extend_unregister(&nf_ct_zone_extend);
#endif
nf_conntrack_proto_fini();
+ nf_conntrack_seqadj_fini();
nf_conntrack_labels_fini();
nf_conntrack_helper_fini();
nf_conntrack_timeout_fini();
@@ -1531,6 +1533,10 @@ int nf_conntrack_init_start(void)
if (ret < 0)
goto err_labels;
+ ret = nf_conntrack_seqadj_init();
+ if (ret < 0)
+ goto err_seqadj;
+
#ifdef CONFIG_NF_CONNTRACK_ZONES
ret = nf_ct_extend_register(&nf_ct_zone_extend);
if (ret < 0)
@@ -1555,6 +1561,8 @@ err_proto:
nf_ct_extend_unregister(&nf_ct_zone_extend);
err_extend:
#endif
+ nf_conntrack_seqadj_fini();
+err_seqadj:
nf_conntrack_labels_fini();
err_labels:
nf_conntrack_helper_fini();
@@ -1577,9 +1585,6 @@ void nf_conntrack_init_end(void)
/* For use by REJECT target */
RCU_INIT_POINTER(ip_ct_attach, nf_conntrack_attach);
RCU_INIT_POINTER(nf_ct_destroy, destroy_conntrack);
-
- /* Howto get NAT offsets */
- RCU_INIT_POINTER(nf_ct_nat_offset, NULL);
}
/*
@@ -1666,8 +1671,3 @@ err_slabname:
err_stat:
return ret;
}
-
-s32 (*nf_ct_nat_offset)(const struct nf_conn *ct,
- enum ip_conntrack_dir dir,
- u32 seq);
-EXPORT_SYMBOL_GPL(nf_ct_nat_offset);
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index fa61fea..7c55745 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -37,6 +37,7 @@
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_expect.h>
#include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_tuple.h>
@@ -381,9 +382,8 @@ nla_put_failure:
return -1;
}
-#ifdef CONFIG_NF_NAT_NEEDED
static int
-dump_nat_seq_adj(struct sk_buff *skb, const struct nf_nat_seq *natseq, int type)
+dump_ct_seq_adj(struct sk_buff *skb, const struct nf_ct_seqadj *seq, int type)
{
struct nlattr *nest_parms;
@@ -391,12 +391,12 @@ dump_nat_seq_adj(struct sk_buff *skb, const struct nf_nat_seq *natseq, int type)
if (!nest_parms)
goto nla_put_failure;
- if (nla_put_be32(skb, CTA_NAT_SEQ_CORRECTION_POS,
- htonl(natseq->correction_pos)) ||
- nla_put_be32(skb, CTA_NAT_SEQ_OFFSET_BEFORE,
- htonl(natseq->offset_before)) ||
- nla_put_be32(skb, CTA_NAT_SEQ_OFFSET_AFTER,
- htonl(natseq->offset_after)))
+ if (nla_put_be32(skb, CTA_SEQADJ_CORRECTION_POS,
+ htonl(seq->correction_pos)) ||
+ nla_put_be32(skb, CTA_SEQADJ_OFFSET_BEFORE,
+ htonl(seq->offset_before)) ||
+ nla_put_be32(skb, CTA_SEQADJ_OFFSET_AFTER,
+ htonl(seq->offset_after)))
goto nla_put_failure;
nla_nest_end(skb, nest_parms);
@@ -408,27 +408,24 @@ nla_put_failure:
}
static inline int
-ctnetlink_dump_nat_seq_adj(struct sk_buff *skb, const struct nf_conn *ct)
+ctnetlink_dump_ct_seq_adj(struct sk_buff *skb, const struct nf_conn *ct)
{
- struct nf_nat_seq *natseq;
- struct nf_conn_nat *nat = nfct_nat(ct);
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
+ struct nf_ct_seqadj *seq;
- if (!(ct->status & IPS_SEQ_ADJUST) || !nat)
+ if (!(ct->status & IPS_SEQ_ADJUST) || !seqadj)
return 0;
- natseq = &nat->seq[IP_CT_DIR_ORIGINAL];
- if (dump_nat_seq_adj(skb, natseq, CTA_NAT_SEQ_ADJ_ORIG) == -1)
+ seq = &seqadj->seq[IP_CT_DIR_ORIGINAL];
+ if (dump_ct_seq_adj(skb, seq, CTA_SEQ_ADJ_ORIG) == -1)
return -1;
- natseq = &nat->seq[IP_CT_DIR_REPLY];
- if (dump_nat_seq_adj(skb, natseq, CTA_NAT_SEQ_ADJ_REPLY) == -1)
+ seq = &seqadj->seq[IP_CT_DIR_REPLY];
+ if (dump_ct_seq_adj(skb, seq, CTA_SEQ_ADJ_REPLY) == -1)
return -1;
return 0;
}
-#else
-#define ctnetlink_dump_nat_seq_adj(a, b) (0)
-#endif
static inline int
ctnetlink_dump_id(struct sk_buff *skb, const struct nf_conn *ct)
@@ -502,7 +499,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
ctnetlink_dump_id(skb, ct) < 0 ||
ctnetlink_dump_use(skb, ct) < 0 ||
ctnetlink_dump_master(skb, ct) < 0 ||
- ctnetlink_dump_nat_seq_adj(skb, ct) < 0)
+ ctnetlink_dump_ct_seq_adj(skb, ct) < 0)
goto nla_put_failure;
nlmsg_end(skb, nlh);
@@ -707,8 +704,8 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item)
ctnetlink_dump_master(skb, ct) < 0)
goto nla_put_failure;
- if (events & (1 << IPCT_NATSEQADJ) &&
- ctnetlink_dump_nat_seq_adj(skb, ct) < 0)
+ if (events & (1 << IPCT_SEQADJ) &&
+ ctnetlink_dump_ct_seq_adj(skb, ct) < 0)
goto nla_put_failure;
}
@@ -1439,66 +1436,65 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, const struct nlattr * const cda[]
return err;
}
-#ifdef CONFIG_NF_NAT_NEEDED
-static const struct nla_policy nat_seq_policy[CTA_NAT_SEQ_MAX+1] = {
- [CTA_NAT_SEQ_CORRECTION_POS] = { .type = NLA_U32 },
- [CTA_NAT_SEQ_OFFSET_BEFORE] = { .type = NLA_U32 },
- [CTA_NAT_SEQ_OFFSET_AFTER] = { .type = NLA_U32 },
+static const struct nla_policy seqadj_policy[CTA_SEQADJ_MAX+1] = {
+ [CTA_SEQADJ_CORRECTION_POS] = { .type = NLA_U32 },
+ [CTA_SEQADJ_OFFSET_BEFORE] = { .type = NLA_U32 },
+ [CTA_SEQADJ_OFFSET_AFTER] = { .type = NLA_U32 },
};
static inline int
-change_nat_seq_adj(struct nf_nat_seq *natseq, const struct nlattr * const attr)
+change_seq_adj(struct nf_ct_seqadj *seq, const struct nlattr * const attr)
{
int err;
- struct nlattr *cda[CTA_NAT_SEQ_MAX+1];
+ struct nlattr *cda[CTA_SEQADJ_MAX+1];
- err = nla_parse_nested(cda, CTA_NAT_SEQ_MAX, attr, nat_seq_policy);
+ err = nla_parse_nested(cda, CTA_SEQADJ_MAX, attr, seqadj_policy);
if (err < 0)
return err;
- if (!cda[CTA_NAT_SEQ_CORRECTION_POS])
+ if (!cda[CTA_SEQADJ_CORRECTION_POS])
return -EINVAL;
- natseq->correction_pos =
- ntohl(nla_get_be32(cda[CTA_NAT_SEQ_CORRECTION_POS]));
+ seq->correction_pos =
+ ntohl(nla_get_be32(cda[CTA_SEQADJ_CORRECTION_POS]));
- if (!cda[CTA_NAT_SEQ_OFFSET_BEFORE])
+ if (!cda[CTA_SEQADJ_OFFSET_BEFORE])
return -EINVAL;
- natseq->offset_before =
- ntohl(nla_get_be32(cda[CTA_NAT_SEQ_OFFSET_BEFORE]));
+ seq->offset_before =
+ ntohl(nla_get_be32(cda[CTA_SEQADJ_OFFSET_BEFORE]));
- if (!cda[CTA_NAT_SEQ_OFFSET_AFTER])
+ if (!cda[CTA_SEQADJ_OFFSET_AFTER])
return -EINVAL;
- natseq->offset_after =
- ntohl(nla_get_be32(cda[CTA_NAT_SEQ_OFFSET_AFTER]));
+ seq->offset_after =
+ ntohl(nla_get_be32(cda[CTA_SEQADJ_OFFSET_AFTER]));
return 0;
}
static int
-ctnetlink_change_nat_seq_adj(struct nf_conn *ct,
- const struct nlattr * const cda[])
+ctnetlink_change_seq_adj(struct nf_conn *ct,
+ const struct nlattr * const cda[])
{
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
int ret = 0;
- struct nf_conn_nat *nat = nfct_nat(ct);
- if (!nat)
+ if (!seqadj)
return 0;
- if (cda[CTA_NAT_SEQ_ADJ_ORIG]) {
- ret = change_nat_seq_adj(&nat->seq[IP_CT_DIR_ORIGINAL],
- cda[CTA_NAT_SEQ_ADJ_ORIG]);
+ if (cda[CTA_SEQ_ADJ_ORIG]) {
+ ret = change_seq_adj(&seqadj->seq[IP_CT_DIR_ORIGINAL],
+ cda[CTA_SEQ_ADJ_ORIG]);
if (ret < 0)
return ret;
ct->status |= IPS_SEQ_ADJUST;
}
- if (cda[CTA_NAT_SEQ_ADJ_REPLY]) {
- ret = change_nat_seq_adj(&nat->seq[IP_CT_DIR_REPLY],
- cda[CTA_NAT_SEQ_ADJ_REPLY]);
+ if (cda[CTA_SEQ_ADJ_REPLY]) {
+ ret = change_seq_adj(&seqadj->seq[IP_CT_DIR_REPLY],
+ cda[CTA_SEQ_ADJ_REPLY]);
if (ret < 0)
return ret;
@@ -1507,7 +1503,6 @@ ctnetlink_change_nat_seq_adj(struct nf_conn *ct,
return 0;
}
-#endif
static int
ctnetlink_attach_labels(struct nf_conn *ct, const struct nlattr * const cda[])
@@ -1573,13 +1568,12 @@ ctnetlink_change_conntrack(struct nf_conn *ct,
ct->mark = ntohl(nla_get_be32(cda[CTA_MARK]));
#endif
-#ifdef CONFIG_NF_NAT_NEEDED
- if (cda[CTA_NAT_SEQ_ADJ_ORIG] || cda[CTA_NAT_SEQ_ADJ_REPLY]) {
- err = ctnetlink_change_nat_seq_adj(ct, cda);
+ if (cda[CTA_SEQ_ADJ_ORIG] || cda[CTA_SEQ_ADJ_REPLY]) {
+ err = ctnetlink_change_seq_adj(ct, cda);
if (err < 0)
return err;
}
-#endif
+
if (cda[CTA_LABELS]) {
err = ctnetlink_attach_labels(ct, cda);
if (err < 0)
@@ -1684,13 +1678,11 @@ ctnetlink_create_conntrack(struct net *net, u16 zone,
goto err2;
}
-#ifdef CONFIG_NF_NAT_NEEDED
- if (cda[CTA_NAT_SEQ_ADJ_ORIG] || cda[CTA_NAT_SEQ_ADJ_REPLY]) {
- err = ctnetlink_change_nat_seq_adj(ct, cda);
+ if (cda[CTA_SEQ_ADJ_ORIG] || cda[CTA_SEQ_ADJ_REPLY]) {
+ err = ctnetlink_change_seq_adj(ct, cda);
if (err < 0)
goto err2;
}
-#endif
memset(&ct->proto, 0, sizeof(ct->proto));
if (cda[CTA_PROTOINFO]) {
@@ -1804,7 +1796,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
(1 << IPCT_ASSURED) |
(1 << IPCT_HELPER) |
(1 << IPCT_PROTOINFO) |
- (1 << IPCT_NATSEQADJ) |
+ (1 << IPCT_SEQADJ) |
(1 << IPCT_MARK) | events,
ct, NETLINK_CB(skb).portid,
nlmsg_report(nlh));
@@ -1827,7 +1819,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb,
(1 << IPCT_HELPER) |
(1 << IPCT_LABEL) |
(1 << IPCT_PROTOINFO) |
- (1 << IPCT_NATSEQADJ) |
+ (1 << IPCT_SEQADJ) |
(1 << IPCT_MARK),
ct, NETLINK_CB(skb).portid,
nlmsg_report(nlh));
@@ -2082,7 +2074,7 @@ ctnetlink_nfqueue_build(struct sk_buff *skb, struct nf_conn *ct)
goto nla_put_failure;
if ((ct->status & IPS_SEQ_ADJUST) &&
- ctnetlink_dump_nat_seq_adj(skb, ct) < 0)
+ ctnetlink_dump_ct_seq_adj(skb, ct) < 0)
goto nla_put_failure;
#ifdef CONFIG_NF_CONNTRACK_MARK
@@ -2211,6 +2203,7 @@ static struct nfq_ct_hook ctnetlink_nfqueue_hook = {
.build = ctnetlink_nfqueue_build,
.parse = ctnetlink_nfqueue_parse,
.attach_expect = ctnetlink_nfqueue_attach_expect,
+ .seq_adjust = nf_ct_tcp_seqadj_set,
};
#endif /* CONFIG_NETFILTER_NETLINK_QUEUE_CT */
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 8f308d8..3c3401e 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -27,6 +27,7 @@
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_ecache.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/nf_log.h>
#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
#include <net/netfilter/ipv6/nf_conntrack_ipv6.h>
@@ -495,21 +496,6 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff,
}
}
-#ifdef CONFIG_NF_NAT_NEEDED
-static inline s32 nat_offset(const struct nf_conn *ct,
- enum ip_conntrack_dir dir,
- u32 seq)
-{
- typeof(nf_ct_nat_offset) get_offset = rcu_dereference(nf_ct_nat_offset);
-
- return get_offset != NULL ? get_offset(ct, dir, seq) : 0;
-}
-#define NAT_OFFSET(ct, dir, seq) \
- (nat_offset(ct, dir, seq))
-#else
-#define NAT_OFFSET(ct, dir, seq) 0
-#endif
-
static bool tcp_in_window(const struct nf_conn *ct,
struct ip_ct_tcp *state,
enum ip_conntrack_dir dir,
@@ -540,7 +526,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
tcp_sack(skb, dataoff, tcph, &sack);
/* Take into account NAT sequence number mangling */
- receiver_offset = NAT_OFFSET(ct, !dir, ack - 1);
+ receiver_offset = nf_ct_seq_offset(ct, !dir, ack - 1);
ack -= receiver_offset;
sack -= receiver_offset;
diff --git a/net/netfilter/nf_conntrack_seqadj.c b/net/netfilter/nf_conntrack_seqadj.c
new file mode 100644
index 0000000..483eb9c
--- /dev/null
+++ b/net/netfilter/nf_conntrack_seqadj.c
@@ -0,0 +1,218 @@
+#include <linux/types.h>
+#include <linux/netfilter.h>
+#include <net/tcp.h>
+
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_extend.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
+
+int nf_ct_seqadj_set(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ __be32 seq, s32 off)
+{
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
+ enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+ struct nf_ct_seqadj *this_way;
+
+ if (off == 0)
+ return 0;
+
+ set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
+
+ spin_lock_bh(&ct->lock);
+ this_way = &seqadj->seq[dir];
+ if (this_way->offset_before == this_way->offset_after ||
+ before(this_way->correction_pos, seq)) {
+ this_way->correction_pos = seq;
+ this_way->offset_before = this_way->offset_after;
+ this_way->offset_after += off;
+ }
+ spin_unlock_bh(&ct->lock);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(nf_ct_seqadj_set);
+
+void nf_ct_tcp_seqadj_set(struct sk_buff *skb,
+ struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ s32 off)
+{
+ const struct tcphdr *th;
+
+ if (nf_ct_protonum(ct) != IPPROTO_TCP)
+ return;
+
+ th = (struct tcphdr *)(skb_network_header(skb) + ip_hdrlen(skb));
+ nf_ct_seqadj_set(ct, ctinfo, th->seq, off);
+}
+EXPORT_SYMBOL_GPL(nf_ct_tcp_seqadj_set);
+
+/* Adjust one found SACK option including checksum correction */
+static void nf_ct_sack_block_adjust(struct sk_buff *skb,
+ struct tcphdr *tcph,
+ unsigned int sackoff,
+ unsigned int sackend,
+ struct nf_ct_seqadj *seq)
+{
+ while (sackoff < sackend) {
+ struct tcp_sack_block_wire *sack;
+ __be32 new_start_seq, new_end_seq;
+
+ sack = (void *)skb->data + sackoff;
+ if (after(ntohl(sack->start_seq) - seq->offset_before,
+ seq->correction_pos))
+ new_start_seq = htonl(ntohl(sack->start_seq) -
+ seq->offset_after);
+ else
+ new_start_seq = htonl(ntohl(sack->start_seq) -
+ seq->offset_before);
+
+ if (after(ntohl(sack->end_seq) - seq->offset_before,
+ seq->correction_pos))
+ new_end_seq = htonl(ntohl(sack->end_seq) -
+ seq->offset_after);
+ else
+ new_end_seq = htonl(ntohl(sack->end_seq) -
+ seq->offset_before);
+
+ pr_debug("sack_adjust: start_seq: %d->%d, end_seq: %d->%d\n",
+ ntohl(sack->start_seq), new_start_seq,
+ ntohl(sack->end_seq), new_end_seq);
+
+ inet_proto_csum_replace4(&tcph->check, skb,
+ sack->start_seq, new_start_seq, 0);
+ inet_proto_csum_replace4(&tcph->check, skb,
+ sack->end_seq, new_end_seq, 0);
+ sack->start_seq = new_start_seq;
+ sack->end_seq = new_end_seq;
+ sackoff += sizeof(*sack);
+ }
+}
+
+/* TCP SACK sequence number adjustment */
+static unsigned int nf_ct_sack_adjust(struct sk_buff *skb,
+ unsigned int protoff,
+ struct tcphdr *tcph,
+ struct nf_conn *ct,
+ enum ip_conntrack_info ctinfo)
+{
+ unsigned int dir, optoff, optend;
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
+
+ optoff = protoff + sizeof(struct tcphdr);
+ optend = protoff + tcph->doff * 4;
+
+ if (!skb_make_writable(skb, optend))
+ return 0;
+
+ dir = CTINFO2DIR(ctinfo);
+
+ while (optoff < optend) {
+ /* Usually: option, length. */
+ unsigned char *op = skb->data + optoff;
+
+ switch (op[0]) {
+ case TCPOPT_EOL:
+ return 1;
+ case TCPOPT_NOP:
+ optoff++;
+ continue;
+ default:
+ /* no partial options */
+ if (optoff + 1 == optend ||
+ optoff + op[1] > optend ||
+ op[1] < 2)
+ return 0;
+ if (op[0] == TCPOPT_SACK &&
+ op[1] >= 2+TCPOLEN_SACK_PERBLOCK &&
+ ((op[1] - 2) % TCPOLEN_SACK_PERBLOCK) == 0)
+ nf_ct_sack_block_adjust(skb, tcph, optoff + 2,
+ optoff+op[1],
+ &seqadj->seq[!dir]);
+ optoff += op[1];
+ }
+ }
+ return 1;
+}
+
+/* TCP sequence number adjustment. Returns 1 on success, 0 on failure */
+int nf_ct_seq_adjust(struct sk_buff *skb,
+ struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ unsigned int protoff)
+{
+ enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+ struct tcphdr *tcph;
+ __be32 newseq, newack;
+ s32 seqoff, ackoff;
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
+ struct nf_ct_seqadj *this_way, *other_way;
+ int res;
+
+ this_way = &seqadj->seq[dir];
+ other_way = &seqadj->seq[!dir];
+
+ if (!skb_make_writable(skb, protoff + sizeof(*tcph)))
+ return 0;
+
+ tcph = (void *)skb->data + protoff;
+ spin_lock_bh(&ct->lock);
+ if (after(ntohl(tcph->seq), this_way->correction_pos))
+ seqoff = this_way->offset_after;
+ else
+ seqoff = this_way->offset_before;
+
+ if (after(ntohl(tcph->ack_seq) - other_way->offset_before,
+ other_way->correction_pos))
+ ackoff = other_way->offset_after;
+ else
+ ackoff = other_way->offset_before;
+
+ newseq = htonl(ntohl(tcph->seq) + seqoff);
+ newack = htonl(ntohl(tcph->ack_seq) - ackoff);
+
+ inet_proto_csum_replace4(&tcph->check, skb, tcph->seq, newseq, 0);
+ inet_proto_csum_replace4(&tcph->check, skb, tcph->ack_seq, newack, 0);
+
+ pr_debug("Adjusting sequence number from %u->%u, ack from %u->%u\n",
+ ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq),
+ ntohl(newack));
+
+ tcph->seq = newseq;
+ tcph->ack_seq = newack;
+
+ res = nf_ct_sack_adjust(skb, protoff, tcph, ct, ctinfo);
+ spin_unlock_bh(&ct->lock);
+
+ return res;
+}
+EXPORT_SYMBOL_GPL(nf_ct_seq_adjust);
+
+s32 nf_ct_seq_offset(const struct nf_conn *ct,
+ enum ip_conntrack_dir dir,
+ u32 seq)
+{
+ struct nf_conn_seqadj *seqadj = nfct_seqadj(ct);
+ struct nf_ct_seqadj *this_way;
+
+ if (!seqadj)
+ return 0;
+
+ this_way = &seqadj->seq[dir];
+ return after(seq, this_way->correction_pos) ?
+ this_way->offset_after : this_way->offset_before;
+}
+EXPORT_SYMBOL_GPL(nf_ct_seq_offset);
+
+static struct nf_ct_ext_type nf_ct_seqadj_extend __read_mostly = {
+ .len = sizeof(struct nf_conn_seqadj),
+ .align = __alignof__(struct nf_conn_seqadj),
+ .id = NF_CT_EXT_SEQADJ,
+};
+
+int nf_conntrack_seqadj_init(void)
+{
+ return nf_ct_extend_register(&nf_ct_seqadj_extend);
+}
+
+void nf_conntrack_seqadj_fini(void)
+{
+ nf_ct_extend_unregister(&nf_ct_seqadj_extend);
+}
diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c
index 6ff8083..6f0f4f7 100644
--- a/net/netfilter/nf_nat_core.c
+++ b/net/netfilter/nf_nat_core.c
@@ -25,6 +25,7 @@
#include <net/netfilter/nf_nat_core.h>
#include <net/netfilter/nf_nat_helper.h>
#include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_zones.h>
#include <linux/netfilter/nf_nat.h>
@@ -402,6 +403,9 @@ nf_nat_setup_info(struct nf_conn *ct,
ct->status |= IPS_SRC_NAT;
else
ct->status |= IPS_DST_NAT;
+
+ if (nfct_help(ct))
+ nfct_seqadj_ext_add(ct);
}
if (maniptype == NF_NAT_MANIP_SRC) {
@@ -764,10 +768,6 @@ static struct nf_ct_helper_expectfn follow_master_nat = {
.expectfn = nf_nat_follow_master,
};
-static struct nfq_ct_nat_hook nfq_ct_nat = {
- .seq_adjust = nf_nat_tcp_seq_adjust,
-};
-
static int __init nf_nat_init(void)
{
int ret;
@@ -787,14 +787,9 @@ static int __init nf_nat_init(void)
/* Initialize fake conntrack so that NAT will skip it */
nf_ct_untracked_status_or(IPS_NAT_DONE_MASK);
- BUG_ON(nf_nat_seq_adjust_hook != NULL);
- RCU_INIT_POINTER(nf_nat_seq_adjust_hook, nf_nat_seq_adjust);
BUG_ON(nfnetlink_parse_nat_setup_hook != NULL);
RCU_INIT_POINTER(nfnetlink_parse_nat_setup_hook,
nfnetlink_parse_nat_setup);
- BUG_ON(nf_ct_nat_offset != NULL);
- RCU_INIT_POINTER(nf_ct_nat_offset, nf_nat_get_offset);
- RCU_INIT_POINTER(nfq_ct_nat_hook, &nfq_ct_nat);
#ifdef CONFIG_XFRM
BUG_ON(nf_nat_decode_session_hook != NULL);
RCU_INIT_POINTER(nf_nat_decode_session_hook, __nf_nat_decode_session);
@@ -813,10 +808,7 @@ static void __exit nf_nat_cleanup(void)
unregister_pernet_subsys(&nf_nat_net_ops);
nf_ct_extend_unregister(&nat_extend);
nf_ct_helper_expectfn_unregister(&follow_master_nat);
- RCU_INIT_POINTER(nf_nat_seq_adjust_hook, NULL);
RCU_INIT_POINTER(nfnetlink_parse_nat_setup_hook, NULL);
- RCU_INIT_POINTER(nf_ct_nat_offset, NULL);
- RCU_INIT_POINTER(nfq_ct_nat_hook, NULL);
#ifdef CONFIG_XFRM
RCU_INIT_POINTER(nf_nat_decode_session_hook, NULL);
#endif
diff --git a/net/netfilter/nf_nat_helper.c b/net/netfilter/nf_nat_helper.c
index 46b9baa..2840abb 100644
--- a/net/netfilter/nf_nat_helper.c
+++ b/net/netfilter/nf_nat_helper.c
@@ -20,67 +20,13 @@
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_ecache.h>
#include <net/netfilter/nf_conntrack_expect.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <net/netfilter/nf_nat.h>
#include <net/netfilter/nf_nat_l3proto.h>
#include <net/netfilter/nf_nat_l4proto.h>
#include <net/netfilter/nf_nat_core.h>
#include <net/netfilter/nf_nat_helper.h>
-#define DUMP_OFFSET(x) \
- pr_debug("offset_before=%d, offset_after=%d, correction_pos=%u\n", \
- x->offset_before, x->offset_after, x->correction_pos);
-
-/* Setup TCP sequence correction given this change at this sequence */
-static inline void
-adjust_tcp_sequence(u32 seq,
- int sizediff,
- struct nf_conn *ct,
- enum ip_conntrack_info ctinfo)
-{
- enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
- struct nf_conn_nat *nat = nfct_nat(ct);
- struct nf_nat_seq *this_way = &nat->seq[dir];
-
- pr_debug("adjust_tcp_sequence: seq = %u, sizediff = %d\n",
- seq, sizediff);
-
- pr_debug("adjust_tcp_sequence: Seq_offset before: ");
- DUMP_OFFSET(this_way);
-
- spin_lock_bh(&ct->lock);
-
- /* SYN adjust. If it's uninitialized, or this is after last
- * correction, record it: we don't handle more than one
- * adjustment in the window, but do deal with common case of a
- * retransmit */
- if (this_way->offset_before == this_way->offset_after ||
- before(this_way->correction_pos, seq)) {
- this_way->correction_pos = seq;
- this_way->offset_before = this_way->offset_after;
- this_way->offset_after += sizediff;
- }
- spin_unlock_bh(&ct->lock);
-
- pr_debug("adjust_tcp_sequence: Seq_offset after: ");
- DUMP_OFFSET(this_way);
-}
-
-/* Get the offset value, for conntrack. Caller must have the conntrack locked */
-s32 nf_nat_get_offset(const struct nf_conn *ct,
- enum ip_conntrack_dir dir,
- u32 seq)
-{
- struct nf_conn_nat *nat = nfct_nat(ct);
- struct nf_nat_seq *this_way;
-
- if (!nat)
- return 0;
-
- this_way = &nat->seq[dir];
- return after(seq, this_way->correction_pos)
- ? this_way->offset_after : this_way->offset_before;
-}
-
/* Frobs data inside this packet, which is linear. */
static void mangle_contents(struct sk_buff *skb,
unsigned int dataoff,
@@ -135,30 +81,6 @@ static int enlarge_skb(struct sk_buff *skb, unsigned int extra)
return 1;
}
-void nf_nat_set_seq_adjust(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
- __be32 seq, s32 off)
-{
- if (!off)
- return;
- set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
- adjust_tcp_sequence(ntohl(seq), off, ct, ctinfo);
- nf_conntrack_event_cache(IPCT_NATSEQADJ, ct);
-}
-EXPORT_SYMBOL_GPL(nf_nat_set_seq_adjust);
-
-void nf_nat_tcp_seq_adjust(struct sk_buff *skb, struct nf_conn *ct,
- u32 ctinfo, int off)
-{
- const struct tcphdr *th;
-
- if (nf_ct_protonum(ct) != IPPROTO_TCP)
- return;
-
- th = (struct tcphdr *)(skb_network_header(skb)+ ip_hdrlen(skb));
- nf_nat_set_seq_adjust(ct, ctinfo, th->seq, off);
-}
-EXPORT_SYMBOL_GPL(nf_nat_tcp_seq_adjust);
-
/* Generic function for mangling variable-length address changes inside
* NATed TCP connections (like the PORT XXX,XXX,XXX,XXX,XXX,XXX
* command in FTP).
@@ -203,8 +125,8 @@ int __nf_nat_mangle_tcp_packet(struct sk_buff *skb,
datalen, oldlen);
if (adjust && rep_len != match_len)
- nf_nat_set_seq_adjust(ct, ctinfo, tcph->seq,
- (int)rep_len - (int)match_len);
+ nf_ct_seqadj_set(ct, ctinfo, tcph->seq,
+ (int)rep_len - (int)match_len);
return 1;
}
@@ -264,150 +186,6 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb,
}
EXPORT_SYMBOL(nf_nat_mangle_udp_packet);
-/* Adjust one found SACK option including checksum correction */
-static void
-sack_adjust(struct sk_buff *skb,
- struct tcphdr *tcph,
- unsigned int sackoff,
- unsigned int sackend,
- struct nf_nat_seq *natseq)
-{
- while (sackoff < sackend) {
- struct tcp_sack_block_wire *sack;
- __be32 new_start_seq, new_end_seq;
-
- sack = (void *)skb->data + sackoff;
- if (after(ntohl(sack->start_seq) - natseq->offset_before,
- natseq->correction_pos))
- new_start_seq = htonl(ntohl(sack->start_seq)
- - natseq->offset_after);
- else
- new_start_seq = htonl(ntohl(sack->start_seq)
- - natseq->offset_before);
-
- if (after(ntohl(sack->end_seq) - natseq->offset_before,
- natseq->correction_pos))
- new_end_seq = htonl(ntohl(sack->end_seq)
- - natseq->offset_after);
- else
- new_end_seq = htonl(ntohl(sack->end_seq)
- - natseq->offset_before);
-
- pr_debug("sack_adjust: start_seq: %d->%d, end_seq: %d->%d\n",
- ntohl(sack->start_seq), new_start_seq,
- ntohl(sack->end_seq), new_end_seq);
-
- inet_proto_csum_replace4(&tcph->check, skb,
- sack->start_seq, new_start_seq, 0);
- inet_proto_csum_replace4(&tcph->check, skb,
- sack->end_seq, new_end_seq, 0);
- sack->start_seq = new_start_seq;
- sack->end_seq = new_end_seq;
- sackoff += sizeof(*sack);
- }
-}
-
-/* TCP SACK sequence number adjustment */
-static inline unsigned int
-nf_nat_sack_adjust(struct sk_buff *skb,
- unsigned int protoff,
- struct tcphdr *tcph,
- struct nf_conn *ct,
- enum ip_conntrack_info ctinfo)
-{
- unsigned int dir, optoff, optend;
- struct nf_conn_nat *nat = nfct_nat(ct);
-
- optoff = protoff + sizeof(struct tcphdr);
- optend = protoff + tcph->doff * 4;
-
- if (!skb_make_writable(skb, optend))
- return 0;
-
- dir = CTINFO2DIR(ctinfo);
-
- while (optoff < optend) {
- /* Usually: option, length. */
- unsigned char *op = skb->data + optoff;
-
- switch (op[0]) {
- case TCPOPT_EOL:
- return 1;
- case TCPOPT_NOP:
- optoff++;
- continue;
- default:
- /* no partial options */
- if (optoff + 1 == optend ||
- optoff + op[1] > optend ||
- op[1] < 2)
- return 0;
- if (op[0] == TCPOPT_SACK &&
- op[1] >= 2+TCPOLEN_SACK_PERBLOCK &&
- ((op[1] - 2) % TCPOLEN_SACK_PERBLOCK) == 0)
- sack_adjust(skb, tcph, optoff+2,
- optoff+op[1], &nat->seq[!dir]);
- optoff += op[1];
- }
- }
- return 1;
-}
-
-/* TCP sequence number adjustment. Returns 1 on success, 0 on failure */
-int
-nf_nat_seq_adjust(struct sk_buff *skb,
- struct nf_conn *ct,
- enum ip_conntrack_info ctinfo,
- unsigned int protoff)
-{
- struct tcphdr *tcph;
- int dir;
- __be32 newseq, newack;
- s32 seqoff, ackoff;
- struct nf_conn_nat *nat = nfct_nat(ct);
- struct nf_nat_seq *this_way, *other_way;
- int res;
-
- dir = CTINFO2DIR(ctinfo);
-
- this_way = &nat->seq[dir];
- other_way = &nat->seq[!dir];
-
- if (!skb_make_writable(skb, protoff + sizeof(*tcph)))
- return 0;
-
- tcph = (void *)skb->data + protoff;
- spin_lock_bh(&ct->lock);
- if (after(ntohl(tcph->seq), this_way->correction_pos))
- seqoff = this_way->offset_after;
- else
- seqoff = this_way->offset_before;
-
- if (after(ntohl(tcph->ack_seq) - other_way->offset_before,
- other_way->correction_pos))
- ackoff = other_way->offset_after;
- else
- ackoff = other_way->offset_before;
-
- newseq = htonl(ntohl(tcph->seq) + seqoff);
- newack = htonl(ntohl(tcph->ack_seq) - ackoff);
-
- inet_proto_csum_replace4(&tcph->check, skb, tcph->seq, newseq, 0);
- inet_proto_csum_replace4(&tcph->check, skb, tcph->ack_seq, newack, 0);
-
- pr_debug("Adjusting sequence number from %u->%u, ack from %u->%u\n",
- ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq),
- ntohl(newack));
-
- tcph->seq = newseq;
- tcph->ack_seq = newack;
-
- res = nf_nat_sack_adjust(skb, protoff, tcph, ct, ctinfo);
- spin_unlock_bh(&ct->lock);
-
- return res;
-}
-
/* Setup NAT on this expected conntrack so it follows master. */
/* If we fail to get a free NAT slot, we'll get dropped on confirm */
void nf_nat_follow_master(struct nf_conn *ct,
diff --git a/net/netfilter/nf_nat_sip.c b/net/netfilter/nf_nat_sip.c
index dac11f7..f979040 100644
--- a/net/netfilter/nf_nat_sip.c
+++ b/net/netfilter/nf_nat_sip.c
@@ -20,6 +20,7 @@
#include <net/netfilter/nf_nat_helper.h>
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_expect.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
#include <linux/netfilter/nf_conntrack_sip.h>
MODULE_LICENSE("GPL");
@@ -308,7 +309,7 @@ static void nf_nat_sip_seq_adjust(struct sk_buff *skb, unsigned int protoff,
return;
th = (struct tcphdr *)(skb->data + protoff);
- nf_nat_set_seq_adjust(ct, ctinfo, th->seq, off);
+ nf_ct_seqadj_set(ct, ctinfo, th->seq, off);
}
/* Handles expected signalling connections and media streams */
diff --git a/net/netfilter/nfnetlink_queue_ct.c b/net/netfilter/nfnetlink_queue_ct.c
index be89303..96cac50 100644
--- a/net/netfilter/nfnetlink_queue_ct.c
+++ b/net/netfilter/nfnetlink_queue_ct.c
@@ -87,14 +87,14 @@ nla_put_failure:
void nfqnl_ct_seq_adjust(struct sk_buff *skb, struct nf_conn *ct,
enum ip_conntrack_info ctinfo, int diff)
{
- struct nfq_ct_nat_hook *nfq_nat_ct;
+ struct nfq_ct_hook *nfq_ct;
- nfq_nat_ct = rcu_dereference(nfq_ct_nat_hook);
- if (nfq_nat_ct == NULL)
+ nfq_ct = rcu_dereference(nfq_ct_hook);
+ if (nfq_ct == NULL)
return;
if ((ct->status & IPS_NAT_MASK) && diff)
- nfq_nat_ct->seq_adjust(skb, ct, ctinfo, diff);
+ nfq_ct->seq_adjust(skb, ct, ctinfo, diff);
}
int nfqnl_attach_expect(struct nf_conn *ct, const struct nlattr *attr,
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 2/6] net: syncookies: export cookie_v4_init_sequence/cookie_v4_check
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
2013-08-23 7:52 ` [PATCH 1/6] netfilter: nf_conntrack: make sequence number adjustments usuable without NAT Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-23 7:52 ` [PATCH 3/6] netfilter: add SYNPROXY core/target Patrick McHardy
` (5 subsequent siblings)
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Extract the local TCP stack independant parts of tcp_v4_init_sequence()
and cookie_v4_check() and export them for use by the upcoming SYNPROXY
target.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
include/net/tcp.h | 4 ++++
net/ipv4/syncookies.c | 29 ++++++++++++++++++-----------
2 files changed, 22 insertions(+), 11 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 18fc999..63256c6 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -480,9 +480,13 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb);
/* From syncookies.c */
extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
+extern int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
+ u32 cookie);
extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
struct ip_options *opt);
#ifdef CONFIG_SYN_COOKIES
+extern u32 __cookie_v4_init_sequence(const struct iphdr *iph,
+ const struct tcphdr *th, u16 *mssp);
extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb,
__u16 *mss);
#else
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index b05c96e..14a15c4 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -160,26 +160,33 @@ static __u16 const msstab[] = {
* Generate a syncookie. mssp points to the mss, which is returned
* rounded down to the value encoded in the cookie.
*/
-__u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
+u32 __cookie_v4_init_sequence(const struct iphdr *iph, const struct tcphdr *th,
+ u16 *mssp)
{
- const struct iphdr *iph = ip_hdr(skb);
- const struct tcphdr *th = tcp_hdr(skb);
int mssind;
const __u16 mss = *mssp;
- tcp_synq_overflow(sk);
-
for (mssind = ARRAY_SIZE(msstab) - 1; mssind ; mssind--)
if (mss >= msstab[mssind])
break;
*mssp = msstab[mssind];
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
-
return secure_tcp_syn_cookie(iph->saddr, iph->daddr,
th->source, th->dest, ntohl(th->seq),
jiffies / (HZ * 60), mssind);
}
+EXPORT_SYMBOL_GPL(__cookie_v4_init_sequence);
+
+__u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
+{
+ const struct iphdr *iph = ip_hdr(skb);
+ const struct tcphdr *th = tcp_hdr(skb);
+
+ tcp_synq_overflow(sk);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
+
+ return __cookie_v4_init_sequence(iph, th, mssp);
+}
/*
* This (misnamed) value is the age of syncookie which is permitted.
@@ -192,10 +199,9 @@ __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
* Check if a ack sequence number is a valid syncookie.
* Return the decoded mss if it is, or 0 if not.
*/
-static inline int cookie_check(struct sk_buff *skb, __u32 cookie)
+int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th,
+ u32 cookie)
{
- const struct iphdr *iph = ip_hdr(skb);
- const struct tcphdr *th = tcp_hdr(skb);
__u32 seq = ntohl(th->seq) - 1;
__u32 mssind = check_tcp_syn_cookie(cookie, iph->saddr, iph->daddr,
th->source, th->dest, seq,
@@ -204,6 +210,7 @@ static inline int cookie_check(struct sk_buff *skb, __u32 cookie)
return mssind < ARRAY_SIZE(msstab) ? msstab[mssind] : 0;
}
+EXPORT_SYMBOL_GPL(__cookie_v4_check);
static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb,
struct request_sock *req,
@@ -284,7 +291,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
goto out;
if (tcp_synq_no_recent_overflow(sk) ||
- (mss = cookie_check(skb, cookie)) == 0) {
+ (mss = __cookie_v4_check(ip_hdr(skb), th, cookie)) == 0) {
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
goto out;
}
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 3/6] netfilter: add SYNPROXY core/target
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
2013-08-23 7:52 ` [PATCH 1/6] netfilter: nf_conntrack: make sequence number adjustments usuable without NAT Patrick McHardy
2013-08-23 7:52 ` [PATCH 2/6] net: syncookies: export cookie_v4_init_sequence/cookie_v4_check Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-23 7:52 ` [PATCH 4/6] net: syncookies: export cookie_v6_init_sequence/cookie_v6_check Patrick McHardy
` (4 subsequent siblings)
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Add a SYNPROXY for netfilter. The code is split into two parts, the synproxy
core with common functions and an address family specific target.
The SYNPROXY receives the connection request from the client, responds with
a SYN/ACK containing a SYN cookie and announcing a zero window and checks
whether the final ACK from the client contains a valid cookie.
It then establishes a connection to the original destination and, if
successful, sends a window update to the client with the window size
announced by the server.
Support for timestamps, SACK, window scaling and MSS options can be
statically configured as target parameters if the features of the server
are known. If timestamps are used, the timestamp value sent back to
the client in the SYN/ACK will be different from the real timestamp of
the server. In order to now break PAWS, the timestamps are translated in
the direction server->client.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
include/net/netfilter/nf_conntrack_extend.h | 6 +-
include/net/netfilter/nf_conntrack_seqadj.h | 2 +
include/net/netfilter/nf_conntrack_synproxy.h | 74 +++++
include/uapi/linux/netfilter/xt_SYNPROXY.h | 16 +
net/ipv4/netfilter/Kconfig | 13 +
net/ipv4/netfilter/Makefile | 1 +
net/ipv4/netfilter/ipt_SYNPROXY.c | 437 ++++++++++++++++++++++++++
net/netfilter/Kconfig | 3 +
net/netfilter/Makefile | 3 +
net/netfilter/nf_conntrack_core.c | 6 +
net/netfilter/nf_conntrack_seqadj.c | 20 ++
net/netfilter/nf_synproxy_core.c | 429 +++++++++++++++++++++++++
12 files changed, 1009 insertions(+), 1 deletion(-)
create mode 100644 include/net/netfilter/nf_conntrack_synproxy.h
create mode 100644 include/uapi/linux/netfilter/xt_SYNPROXY.h
create mode 100644 net/ipv4/netfilter/ipt_SYNPROXY.c
create mode 100644 net/netfilter/nf_synproxy_core.c
diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h
index 2a22bcb..ff95434 100644
--- a/include/net/netfilter/nf_conntrack_extend.h
+++ b/include/net/netfilter/nf_conntrack_extend.h
@@ -9,8 +9,8 @@ enum nf_ct_ext_id {
NF_CT_EXT_HELPER,
#if defined(CONFIG_NF_NAT) || defined(CONFIG_NF_NAT_MODULE)
NF_CT_EXT_NAT,
- NF_CT_EXT_SEQADJ,
#endif
+ NF_CT_EXT_SEQADJ,
NF_CT_EXT_ACCT,
#ifdef CONFIG_NF_CONNTRACK_EVENTS
NF_CT_EXT_ECACHE,
@@ -27,6 +27,9 @@ enum nf_ct_ext_id {
#ifdef CONFIG_NF_CONNTRACK_LABELS
NF_CT_EXT_LABELS,
#endif
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+ NF_CT_EXT_SYNPROXY,
+#endif
NF_CT_EXT_NUM,
};
@@ -39,6 +42,7 @@ enum nf_ct_ext_id {
#define NF_CT_EXT_TSTAMP_TYPE struct nf_conn_tstamp
#define NF_CT_EXT_TIMEOUT_TYPE struct nf_conn_timeout
#define NF_CT_EXT_LABELS_TYPE struct nf_conn_labels
+#define NF_CT_EXT_SYNPROXY_TYPE struct nf_conn_synproxy
/* Extensions: optional stuff which isn't permanently in struct. */
struct nf_ct_ext {
diff --git a/include/net/netfilter/nf_conntrack_seqadj.h b/include/net/netfilter/nf_conntrack_seqadj.h
index 30bfbbe..f6177a5 100644
--- a/include/net/netfilter/nf_conntrack_seqadj.h
+++ b/include/net/netfilter/nf_conntrack_seqadj.h
@@ -30,6 +30,8 @@ static inline struct nf_conn_seqadj *nfct_seqadj_ext_add(struct nf_conn *ct)
return nf_ct_ext_add(ct, NF_CT_EXT_SEQADJ, GFP_ATOMIC);
}
+extern int nf_ct_seqadj_init(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ s32 off);
extern int nf_ct_seqadj_set(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
__be32 seq, s32 off);
extern void nf_ct_tcp_seqadj_set(struct sk_buff *skb,
diff --git a/include/net/netfilter/nf_conntrack_synproxy.h b/include/net/netfilter/nf_conntrack_synproxy.h
new file mode 100644
index 0000000..cc43895
--- /dev/null
+++ b/include/net/netfilter/nf_conntrack_synproxy.h
@@ -0,0 +1,74 @@
+#ifndef _NF_CONNTRACK_SYNPROXY_H
+#define _NF_CONNTRACK_SYNPROXY_H
+
+#include <net/netns/generic.h>
+
+struct nf_conn_synproxy {
+ u32 isn;
+ u32 tsoff;
+};
+
+static inline struct nf_conn_synproxy *nfct_synproxy(const struct nf_conn *ct)
+{
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+ return nf_ct_ext_find(ct, NF_CT_EXT_SYNPROXY);
+#else
+ return NULL;
+#endif
+}
+
+static inline struct nf_conn_synproxy *nfct_synproxy_ext_add(struct nf_conn *ct)
+{
+#if IS_ENABLED(CONFIG_NETFILTER_SYNPROXY)
+ return nf_ct_ext_add(ct, NF_CT_EXT_SYNPROXY, GFP_ATOMIC);
+#else
+ return NULL;
+#endif
+}
+
+struct synproxy_stats {
+ unsigned int syn_received;
+ unsigned int cookie_invalid;
+ unsigned int cookie_valid;
+};
+
+struct synproxy_net {
+ struct nf_conn *tmpl;
+ struct synproxy_stats __percpu *stats;
+};
+
+extern int synproxy_net_id;
+static inline struct synproxy_net *synproxy_pernet(struct net *net)
+{
+ return net_generic(net, synproxy_net_id);
+}
+
+struct synproxy_options {
+ u8 options;
+ u8 wscale;
+ u16 mss;
+ u32 tsval;
+ u32 tsecr;
+};
+
+struct tcphdr;
+struct xt_synproxy_info;
+extern void synproxy_parse_options(const struct sk_buff *skb, unsigned int doff,
+ const struct tcphdr *th,
+ struct synproxy_options *opts);
+extern unsigned int synproxy_options_size(const struct synproxy_options *opts);
+extern void synproxy_build_options(struct tcphdr *th,
+ const struct synproxy_options *opts);
+
+extern void synproxy_init_timestamp_cookie(const struct xt_synproxy_info *info,
+ struct synproxy_options *opts);
+extern void synproxy_check_timestamp_cookie(struct synproxy_options *opts);
+
+extern unsigned int synproxy_tstamp_adjust(struct sk_buff *skb,
+ unsigned int protoff,
+ struct tcphdr *th,
+ struct nf_conn *ct,
+ enum ip_conntrack_info ctinfo,
+ const struct nf_conn_synproxy *synproxy);
+
+#endif /* _NF_CONNTRACK_SYNPROXY_H */
diff --git a/include/uapi/linux/netfilter/xt_SYNPROXY.h b/include/uapi/linux/netfilter/xt_SYNPROXY.h
new file mode 100644
index 0000000..2d59fba
--- /dev/null
+++ b/include/uapi/linux/netfilter/xt_SYNPROXY.h
@@ -0,0 +1,16 @@
+#ifndef _XT_SYNPROXY_H
+#define _XT_SYNPROXY_H
+
+#define XT_SYNPROXY_OPT_MSS 0x01
+#define XT_SYNPROXY_OPT_WSCALE 0x02
+#define XT_SYNPROXY_OPT_SACK_PERM 0x04
+#define XT_SYNPROXY_OPT_TIMESTAMP 0x08
+#define XT_SYNPROXY_OPT_ECN 0x10
+
+struct xt_synproxy_info {
+ __u8 options;
+ __u8 wscale;
+ __u16 mss;
+};
+
+#endif /* _XT_SYNPROXY_H */
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 4e90280..1657e39 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -110,6 +110,19 @@ config IP_NF_TARGET_REJECT
To compile it as a module, choose M here. If unsure, say N.
+config IP_NF_TARGET_SYNPROXY
+ tristate "SYNPROXY target support"
+ depends on NF_CONNTRACK && NETFILTER_ADVANCED
+ select NETFILTER_SYNPROXY
+ select SYN_COOKIES
+ help
+ The SYNPROXY target allows you to intercept TCP connections and
+ establish them using syncookies before they are passed on to the
+ server. This allows to avoid conntrack and server resource usage
+ during SYN-flood attacks.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
config IP_NF_TARGET_ULOG
tristate "ULOG target support (obsolete)"
default m if NETFILTER_ADVANCED=n
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index 007b128..3622b24 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o
obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o
obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o
obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o
+obj-$(CONFIG_IP_NF_TARGET_SYNPROXY) += ipt_SYNPROXY.o
obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o
# generic ARP tables
diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c
new file mode 100644
index 0000000..d8577e0
--- /dev/null
+++ b/net/ipv4/netfilter/ipt_SYNPROXY.c
@@ -0,0 +1,437 @@
+/*
+ * Copyright (c) 2013 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <net/tcp.h>
+
+#include <linux/netfilter_ipv4/ip_tables.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_SYNPROXY.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
+#include <net/netfilter/nf_conntrack_synproxy.h>
+
+static struct iphdr *
+synproxy_build_ip(struct sk_buff *skb, u32 saddr, u32 daddr)
+{
+ struct iphdr *iph;
+
+ skb_reset_network_header(skb);
+ iph = (struct iphdr *)skb_put(skb, sizeof(*iph));
+ iph->version = 4;
+ iph->ihl = sizeof(*iph) / 4;
+ iph->tos = 0;
+ iph->id = 0;
+ iph->frag_off = htons(IP_DF);
+ iph->ttl = sysctl_ip_default_ttl;
+ iph->protocol = IPPROTO_TCP;
+ iph->check = 0;
+ iph->saddr = saddr;
+ iph->daddr = daddr;
+
+ return iph;
+}
+
+static void
+synproxy_send_tcp(const struct sk_buff *skb, struct sk_buff *nskb,
+ struct nf_conntrack *nfct, enum ip_conntrack_info ctinfo,
+ struct iphdr *niph, struct tcphdr *nth,
+ unsigned int tcp_hdr_size)
+{
+ nth->check = ~tcp_v4_check(tcp_hdr_size, niph->saddr, niph->daddr, 0);
+ nskb->ip_summed = CHECKSUM_PARTIAL;
+ nskb->csum_start = (unsigned char *)nth - nskb->head;
+ nskb->csum_offset = offsetof(struct tcphdr, check);
+
+ skb_dst_set_noref(nskb, skb_dst(skb));
+ nskb->protocol = htons(ETH_P_IP);
+ if (ip_route_me_harder(nskb, RTN_UNSPEC))
+ goto free_nskb;
+
+ nskb->nfct = nfct;
+ nskb->nfctinfo = ctinfo;
+ nf_conntrack_get(nfct);
+
+ ip_local_out(nskb);
+ return;
+
+free_nskb:
+ kfree_skb(nskb);
+}
+
+static void
+synproxy_send_client_synack(const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct iphdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+ u16 mss = opts->mss;
+
+ iph = ip_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, iph->daddr, iph->saddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->dest;
+ nth->dest = th->source;
+ nth->seq = htonl(__cookie_v4_init_sequence(iph, th, &mss));
+ nth->ack_seq = htonl(ntohl(th->seq) + 1);
+ tcp_flag_word(nth) = TCP_FLAG_SYN | TCP_FLAG_ACK;
+ if (opts->options & XT_SYNPROXY_OPT_ECN)
+ tcp_flag_word(nth) |= TCP_FLAG_ECE;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = 0;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_server_syn(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct iphdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ip_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, iph->saddr, iph->daddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->source;
+ nth->dest = th->dest;
+ nth->seq = htonl(ntohl(th->seq) - 1);
+ /* ack_seq is used to relay our ISN to the synproxy hook to initialize
+ * sequence number translation once a connection tracking entry exists.
+ */
+ nth->ack_seq = htonl(ntohl(th->ack_seq) - 1);
+ tcp_flag_word(nth) = TCP_FLAG_SYN;
+ if (opts->options & XT_SYNPROXY_OPT_ECN)
+ tcp_flag_word(nth) |= TCP_FLAG_ECE | TCP_FLAG_CWR;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = th->window;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, &snet->tmpl->ct_general, IP_CT_NEW,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_server_ack(const struct synproxy_net *snet,
+ const struct ip_ct_tcp *state,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct iphdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ip_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, iph->daddr, iph->saddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->dest;
+ nth->dest = th->source;
+ nth->seq = htonl(ntohl(th->ack_seq));
+ nth->ack_seq = htonl(ntohl(th->seq) + 1);
+ tcp_flag_word(nth) = TCP_FLAG_ACK;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = htons(state->seen[IP_CT_DIR_ORIGINAL].td_maxwin);
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_client_ack(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct iphdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ip_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, iph->saddr, iph->daddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->source;
+ nth->dest = th->dest;
+ nth->seq = htonl(ntohl(th->seq) + 1);
+ nth->ack_seq = th->ack_seq;
+ tcp_flag_word(nth) = TCP_FLAG_ACK;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = th->window;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
+ niph, nth, tcp_hdr_size);
+}
+
+static unsigned int
+synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par)
+{
+ const struct xt_synproxy_info *info = par->targinfo;
+ struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+ struct synproxy_options opts = {};
+ struct tcphdr *th, _th;
+
+ if (nf_ip_checksum(skb, par->hooknum, par->thoff, IPPROTO_TCP))
+ return NF_DROP;
+
+ th = skb_header_pointer(skb, par->thoff, sizeof(_th), &_th);
+ if (th == NULL)
+ return NF_DROP;
+
+ synproxy_parse_options(skb, par->thoff, th, &opts);
+
+ if (th->syn && !th->ack) {
+ /* Initial SYN from client */
+ this_cpu_inc(snet->stats->syn_received);
+
+ if (th->ece && th->cwr)
+ opts.options |= XT_SYNPROXY_OPT_ECN;
+
+ opts.options &= info->options;
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_init_timestamp_cookie(info, &opts);
+ else
+ opts.options &= ~(XT_SYNPROXY_OPT_WSCALE |
+ XT_SYNPROXY_OPT_SACK_PERM |
+ XT_SYNPROXY_OPT_ECN);
+
+ synproxy_send_client_synack(skb, th, &opts);
+ } else if (th->ack && !(th->fin || th->rst)) {
+ /* ACK from client */
+ int mss = __cookie_v4_check(ip_hdr(skb), th,
+ ntohl(th->ack_seq) - 1);
+ if (mss == 0) {
+ this_cpu_inc(snet->stats->cookie_invalid);
+ return NF_DROP;
+ }
+
+ this_cpu_inc(snet->stats->cookie_valid);
+ opts.mss = mss;
+
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_check_timestamp_cookie(&opts);
+
+ synproxy_send_server_syn(snet, skb, th, &opts);
+ }
+
+ return NF_DROP;
+}
+
+static unsigned int ipv4_synproxy_hook(unsigned int hooknum,
+ struct sk_buff *skb,
+ const struct net_device *in,
+ const struct net_device *out,
+ int (*okfn)(struct sk_buff *))
+{
+ struct synproxy_net *snet = synproxy_pernet(dev_net(in ? : out));
+ enum ip_conntrack_info ctinfo;
+ struct nf_conn *ct;
+ struct nf_conn_synproxy *synproxy;
+ struct synproxy_options opts = {};
+ const struct ip_ct_tcp *state;
+ struct tcphdr *th, _th;
+ unsigned int thoff;
+
+ ct = nf_ct_get(skb, &ctinfo);
+ if (ct == NULL)
+ return NF_ACCEPT;
+
+ synproxy = nfct_synproxy(ct);
+ if (synproxy == NULL)
+ return NF_ACCEPT;
+
+ if (nf_is_loopback_packet(skb))
+ return NF_ACCEPT;
+
+ thoff = ip_hdrlen(skb);
+ th = skb_header_pointer(skb, thoff, sizeof(_th), &_th);
+ if (th == NULL)
+ return NF_DROP;
+
+ state = &ct->proto.tcp;
+ switch (state->state) {
+ case TCP_CONNTRACK_SYN_SENT:
+ synproxy_parse_options(skb, thoff, th, &opts);
+
+ synproxy->isn = ntohl(th->ack_seq);
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy->tsoff = -opts.tsecr;
+ break;
+ case TCP_CONNTRACK_SYN_RECV:
+ if (!th->syn)
+ break;
+
+ synproxy_parse_options(skb, thoff, th, &opts);
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy->tsoff += opts.tsval;
+
+ opts.options &= ~(XT_SYNPROXY_OPT_MSS |
+ XT_SYNPROXY_OPT_WSCALE |
+ XT_SYNPROXY_OPT_SACK_PERM);
+
+ swap(opts.tsval, opts.tsecr);
+ synproxy_send_server_ack(snet, state, skb, th, &opts);
+
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn - ntohl(th->seq));
+
+ swap(opts.tsval, opts.tsecr);
+ synproxy_send_client_ack(snet, skb, th, &opts);
+
+ consume_skb(skb);
+ return NF_STOLEN;
+ case TCP_CONNTRACK_CLOSE:
+ if (!th->rst)
+ break;
+
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
+ ntohl(th->seq) + 1);
+ break;
+ default:
+ break;
+ }
+
+ synproxy_tstamp_adjust(skb, thoff, th, ct, ctinfo, synproxy);
+ return NF_ACCEPT;
+}
+
+static int synproxy_tg4_check(const struct xt_tgchk_param *par)
+{
+ const struct ipt_entry *e = par->entryinfo;
+
+ if (e->ip.proto != IPPROTO_TCP ||
+ e->ip.invflags & XT_INV_PROTO)
+ return -EINVAL;
+
+ return nf_ct_l3proto_try_module_get(par->family);
+}
+
+static void synproxy_tg4_destroy(const struct xt_tgdtor_param *par)
+{
+ nf_ct_l3proto_module_put(par->family);
+}
+
+static struct xt_target synproxy_tg4_reg __read_mostly = {
+ .name = "SYNPROXY",
+ .family = NFPROTO_IPV4,
+ .target = synproxy_tg4,
+ .targetsize = sizeof(struct xt_synproxy_info),
+ .checkentry = synproxy_tg4_check,
+ .destroy = synproxy_tg4_destroy,
+ .me = THIS_MODULE,
+};
+
+static struct nf_hook_ops ipv4_synproxy_ops[] __read_mostly = {
+ {
+ .hook = ipv4_synproxy_hook,
+ .owner = THIS_MODULE,
+ .pf = NFPROTO_IPV4,
+ .hooknum = NF_INET_LOCAL_IN,
+ .priority = NF_IP_PRI_CONNTRACK_CONFIRM - 1,
+ },
+ {
+ .hook = ipv4_synproxy_hook,
+ .owner = THIS_MODULE,
+ .pf = NFPROTO_IPV4,
+ .hooknum = NF_INET_POST_ROUTING,
+ .priority = NF_IP_PRI_CONNTRACK_CONFIRM - 1,
+ },
+};
+
+static int __init synproxy_tg4_init(void)
+{
+ int err;
+
+ err = nf_register_hooks(ipv4_synproxy_ops,
+ ARRAY_SIZE(ipv4_synproxy_ops));
+ if (err < 0)
+ goto err1;
+
+ err = xt_register_target(&synproxy_tg4_reg);
+ if (err < 0)
+ goto err2;
+
+ return 0;
+
+err2:
+ nf_unregister_hooks(ipv4_synproxy_ops, ARRAY_SIZE(ipv4_synproxy_ops));
+err1:
+ return err;
+}
+
+static void __exit synproxy_tg4_exit(void)
+{
+ xt_unregister_target(&synproxy_tg4_reg);
+ nf_unregister_hooks(ipv4_synproxy_ops, ARRAY_SIZE(ipv4_synproxy_ops));
+}
+
+module_init(synproxy_tg4_init);
+module_exit(synproxy_tg4_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index c45fc1a..62a171a 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -408,6 +408,9 @@ config NF_NAT_TFTP
depends on NF_CONNTRACK && NF_NAT
default NF_NAT && NF_CONNTRACK_TFTP
+config NETFILTER_SYNPROXY
+ tristate
+
endif # NF_CONNTRACK
config NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 89a9c16..c3a0a12 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -61,6 +61,9 @@ obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o
obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o
obj-$(CONFIG_NF_NAT_TFTP) += nf_nat_tftp.o
+# SYNPROXY
+obj-$(CONFIG_NETFILTER_SYNPROXY) += nf_synproxy_core.o
+
# generic X tables
obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 00a7a94..5d892fe 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -48,6 +48,7 @@
#include <net/netfilter/nf_conntrack_timestamp.h>
#include <net/netfilter/nf_conntrack_timeout.h>
#include <net/netfilter/nf_conntrack_labels.h>
+#include <net/netfilter/nf_conntrack_synproxy.h>
#include <net/netfilter/nf_nat.h>
#include <net/netfilter/nf_nat_core.h>
#include <net/netfilter/nf_nat_helper.h>
@@ -799,6 +800,11 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,
if (IS_ERR(ct))
return (struct nf_conntrack_tuple_hash *)ct;
+ if (tmpl && nfct_synproxy(tmpl)) {
+ nfct_seqadj_ext_add(ct);
+ nfct_synproxy_ext_add(ct);
+ }
+
timeout_ext = tmpl ? nf_ct_timeout_find(tmpl) : NULL;
if (timeout_ext)
timeouts = NF_CT_TIMEOUT_EXT_DATA(timeout_ext);
diff --git a/net/netfilter/nf_conntrack_seqadj.c b/net/netfilter/nf_conntrack_seqadj.c
index 483eb9c..5f9bfd0 100644
--- a/net/netfilter/nf_conntrack_seqadj.c
+++ b/net/netfilter/nf_conntrack_seqadj.c
@@ -6,6 +6,26 @@
#include <net/netfilter/nf_conntrack_extend.h>
#include <net/netfilter/nf_conntrack_seqadj.h>
+int nf_ct_seqadj_init(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
+ s32 off)
+{
+ enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+ struct nf_conn_seqadj *seqadj;
+ struct nf_ct_seqadj *this_way;
+
+ if (off == 0)
+ return 0;
+
+ set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
+
+ seqadj = nfct_seqadj(ct);
+ this_way = &seqadj->seq[dir];
+ this_way->offset_before = off;
+ this_way->offset_after = off;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(nf_ct_seqadj_init);
+
int nf_ct_seqadj_set(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
__be32 seq, s32 off)
{
diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c
new file mode 100644
index 0000000..502faae
--- /dev/null
+++ b/net/netfilter/nf_synproxy_core.c
@@ -0,0 +1,429 @@
+/*
+ * Copyright (c) 2013 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <asm/unaligned.h>
+#include <net/tcp.h>
+#include <net/netns/generic.h>
+
+#include <linux/netfilter_ipv4/ip_tables.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_tcpudp.h>
+#include <linux/netfilter/xt_SYNPROXY.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_extend.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
+#include <net/netfilter/nf_conntrack_synproxy.h>
+
+int synproxy_net_id;
+EXPORT_SYMBOL_GPL(synproxy_net_id);
+
+void
+synproxy_parse_options(const struct sk_buff *skb, unsigned int doff,
+ const struct tcphdr *th, struct synproxy_options *opts)
+{
+ int length = (th->doff * 4) - sizeof(*th);
+ u8 buf[40], *ptr;
+
+ ptr = skb_header_pointer(skb, doff + sizeof(*th), length, buf);
+ BUG_ON(ptr == NULL);
+
+ opts->options = 0;
+ while (length > 0) {
+ int opcode = *ptr++;
+ int opsize;
+
+ switch (opcode) {
+ case TCPOPT_EOL:
+ return;
+ case TCPOPT_NOP:
+ length--;
+ continue;
+ default:
+ opsize = *ptr++;
+ if (opsize < 2)
+ return;
+ if (opsize > length)
+ return;
+
+ switch (opcode) {
+ case TCPOPT_MSS:
+ if (opsize == TCPOLEN_MSS) {
+ opts->mss = get_unaligned_be16(ptr);
+ opts->options |= XT_SYNPROXY_OPT_MSS;
+ }
+ break;
+ case TCPOPT_WINDOW:
+ if (opsize == TCPOLEN_WINDOW) {
+ opts->wscale = *ptr;
+ if (opts->wscale > 14)
+ opts->wscale = 14;
+ opts->options |= XT_SYNPROXY_OPT_WSCALE;
+ }
+ break;
+ case TCPOPT_TIMESTAMP:
+ if (opsize == TCPOLEN_TIMESTAMP) {
+ opts->tsval = get_unaligned_be32(ptr);
+ opts->tsecr = get_unaligned_be32(ptr + 4);
+ opts->options |= XT_SYNPROXY_OPT_TIMESTAMP;
+ }
+ break;
+ case TCPOPT_SACK_PERM:
+ if (opsize == TCPOLEN_SACK_PERM)
+ opts->options |= XT_SYNPROXY_OPT_SACK_PERM;
+ break;
+ }
+
+ ptr += opsize - 2;
+ length -= opsize;
+ }
+ }
+}
+EXPORT_SYMBOL_GPL(synproxy_parse_options);
+
+unsigned int synproxy_options_size(const struct synproxy_options *opts)
+{
+ unsigned int size = 0;
+
+ if (opts->options & XT_SYNPROXY_OPT_MSS)
+ size += TCPOLEN_MSS_ALIGNED;
+ if (opts->options & XT_SYNPROXY_OPT_TIMESTAMP)
+ size += TCPOLEN_TSTAMP_ALIGNED;
+ else if (opts->options & XT_SYNPROXY_OPT_SACK_PERM)
+ size += TCPOLEN_SACKPERM_ALIGNED;
+ if (opts->options & XT_SYNPROXY_OPT_WSCALE)
+ size += TCPOLEN_WSCALE_ALIGNED;
+
+ return size;
+}
+EXPORT_SYMBOL_GPL(synproxy_options_size);
+
+void
+synproxy_build_options(struct tcphdr *th, const struct synproxy_options *opts)
+{
+ __be32 *ptr = (__be32 *)(th + 1);
+ u8 options = opts->options;
+
+ if (options & XT_SYNPROXY_OPT_MSS)
+ *ptr++ = htonl((TCPOPT_MSS << 24) |
+ (TCPOLEN_MSS << 16) |
+ opts->mss);
+
+ if (options & XT_SYNPROXY_OPT_TIMESTAMP) {
+ if (options & XT_SYNPROXY_OPT_SACK_PERM)
+ *ptr++ = htonl((TCPOPT_SACK_PERM << 24) |
+ (TCPOLEN_SACK_PERM << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
+ TCPOLEN_TIMESTAMP);
+ else
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
+ TCPOLEN_TIMESTAMP);
+
+ *ptr++ = htonl(opts->tsval);
+ *ptr++ = htonl(opts->tsecr);
+ } else if (options & XT_SYNPROXY_OPT_SACK_PERM)
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_SACK_PERM << 8) |
+ TCPOLEN_SACK_PERM);
+
+ if (options & XT_SYNPROXY_OPT_WSCALE)
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_WINDOW << 16) |
+ (TCPOLEN_WINDOW << 8) |
+ opts->wscale);
+}
+EXPORT_SYMBOL_GPL(synproxy_build_options);
+
+void synproxy_init_timestamp_cookie(const struct xt_synproxy_info *info,
+ struct synproxy_options *opts)
+{
+ opts->tsecr = opts->tsval;
+ opts->tsval = tcp_time_stamp & ~0x3f;
+
+ if (opts->options & XT_SYNPROXY_OPT_WSCALE)
+ opts->tsval |= info->wscale;
+ else
+ opts->tsval |= 0xf;
+
+ if (opts->options & XT_SYNPROXY_OPT_SACK_PERM)
+ opts->tsval |= 1 << 4;
+
+ if (opts->options & XT_SYNPROXY_OPT_ECN)
+ opts->tsval |= 1 << 5;
+}
+EXPORT_SYMBOL_GPL(synproxy_init_timestamp_cookie);
+
+void synproxy_check_timestamp_cookie(struct synproxy_options *opts)
+{
+ opts->wscale = opts->tsecr & 0xf;
+ if (opts->wscale != 0xf)
+ opts->options |= XT_SYNPROXY_OPT_WSCALE;
+
+ opts->options |= opts->tsecr & (1 << 4) ? XT_SYNPROXY_OPT_SACK_PERM : 0;
+
+ opts->options |= opts->tsecr & (1 << 5) ? XT_SYNPROXY_OPT_ECN : 0;
+}
+EXPORT_SYMBOL_GPL(synproxy_check_timestamp_cookie);
+
+unsigned int synproxy_tstamp_adjust(struct sk_buff *skb,
+ unsigned int protoff,
+ struct tcphdr *th,
+ struct nf_conn *ct,
+ enum ip_conntrack_info ctinfo,
+ const struct nf_conn_synproxy *synproxy)
+{
+ unsigned int optoff, optend;
+ u32 *ptr, old;
+
+ if (synproxy->tsoff == 0)
+ return 1;
+
+ optoff = protoff + sizeof(struct tcphdr);
+ optend = protoff + th->doff * 4;
+
+ if (!skb_make_writable(skb, optend))
+ return 0;
+
+ while (optoff < optend) {
+ unsigned char *op = skb->data + optoff;
+
+ switch (op[0]) {
+ case TCPOPT_EOL:
+ return 1;
+ case TCPOPT_NOP:
+ optoff++;
+ continue;
+ default:
+ if (optoff + 1 == optend ||
+ optoff + op[1] > optend ||
+ op[1] < 2)
+ return 0;
+ if (op[0] == TCPOPT_TIMESTAMP &&
+ op[1] == TCPOLEN_TIMESTAMP) {
+ if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) {
+ ptr = (u32 *)&op[2];
+ old = *ptr;
+ *ptr = htonl(ntohl(*ptr) -
+ synproxy->tsoff);
+ } else {
+ ptr = (u32 *)&op[6];
+ old = *ptr;
+ *ptr = htonl(ntohl(*ptr) +
+ synproxy->tsoff);
+ }
+ inet_proto_csum_replace4(&th->check, skb,
+ old, *ptr, 0);
+ return 1;
+ }
+ optoff += op[1];
+ }
+ }
+ return 1;
+}
+EXPORT_SYMBOL_GPL(synproxy_tstamp_adjust);
+
+static struct nf_ct_ext_type nf_ct_synproxy_extend __read_mostly = {
+ .len = sizeof(struct nf_conn_synproxy),
+ .align = __alignof__(struct nf_conn_synproxy),
+ .id = NF_CT_EXT_SYNPROXY,
+};
+
+#ifdef CONFIG_PROC_FS
+static void *synproxy_cpu_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ struct synproxy_net *snet = synproxy_pernet(seq_file_net(seq));
+ int cpu;
+
+ if (*pos == 0)
+ return SEQ_START_TOKEN;
+
+ for (cpu = *pos - 1; cpu < nr_cpu_ids; cpu++) {
+ if (!cpu_possible(cpu))
+ continue;
+ *pos = cpu + 1;
+ return per_cpu_ptr(snet->stats, cpu);
+ }
+
+ return NULL;
+}
+
+static void *synproxy_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ struct synproxy_net *snet = synproxy_pernet(seq_file_net(seq));
+ int cpu;
+
+ for (cpu = *pos; cpu < nr_cpu_ids; cpu++) {
+ if (!cpu_possible(cpu))
+ continue;
+ *pos = cpu + 1;
+ return per_cpu_ptr(snet->stats, cpu);
+ }
+
+ return NULL;
+}
+
+static void synproxy_cpu_seq_stop(struct seq_file *seq, void *v)
+{
+ return;
+}
+
+static int synproxy_cpu_seq_show(struct seq_file *seq, void *v)
+{
+ struct synproxy_stats *stats = v;
+
+ if (v == SEQ_START_TOKEN) {
+ seq_printf(seq, "entries\tsyn_received\t"
+ "cookie_invalid\tcookie_valid\n");
+ return 0;
+ }
+
+ seq_printf(seq, "%08x\t%08x\t%08x\t%08x\n", 0,
+ stats->syn_received,
+ stats->cookie_invalid,
+ stats->cookie_valid);
+
+ return 0;
+}
+
+static const struct seq_operations synproxy_cpu_seq_ops = {
+ .start = synproxy_cpu_seq_start,
+ .next = synproxy_cpu_seq_next,
+ .stop = synproxy_cpu_seq_stop,
+ .show = synproxy_cpu_seq_show,
+};
+
+static int synproxy_cpu_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open_net(inode, file, &synproxy_cpu_seq_ops,
+ sizeof(struct seq_net_private));
+}
+
+static const struct file_operations synproxy_cpu_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = synproxy_cpu_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release_net,
+};
+
+static int __net_init synproxy_proc_init(struct net *net)
+{
+ if (!proc_create("synproxy", S_IRUGO, net->proc_net_stat,
+ &synproxy_cpu_seq_fops))
+ return -ENOMEM;
+ return 0;
+}
+
+static void __net_exit synproxy_proc_exit(struct net *net)
+{
+ remove_proc_entry("synproxy", net->proc_net_stat);
+}
+#else
+static int __net_init synproxy_proc_init(struct net *net)
+{
+ return 0;
+}
+
+static void __net_exit synproxy_proc_exit(struct net *net)
+{
+ return;
+}
+#endif /* CONFIG_PROC_FS */
+
+static int __net_init synproxy_net_init(struct net *net)
+{
+ struct synproxy_net *snet = synproxy_pernet(net);
+ struct nf_conntrack_tuple t;
+ struct nf_conn *ct;
+ int err = -ENOMEM;
+
+ memset(&t, 0, sizeof(t));
+ ct = nf_conntrack_alloc(net, 0, &t, &t, GFP_KERNEL);
+ if (IS_ERR(ct)) {
+ err = PTR_ERR(ct);
+ goto err1;
+ }
+
+ __set_bit(IPS_TEMPLATE_BIT, &ct->status);
+ __set_bit(IPS_CONFIRMED_BIT, &ct->status);
+ if (!nfct_seqadj_ext_add(ct))
+ goto err2;
+ if (!nfct_synproxy_ext_add(ct))
+ goto err2;
+
+ snet->tmpl = ct;
+
+ snet->stats = alloc_percpu(struct synproxy_stats);
+ if (snet->stats == NULL)
+ goto err2;
+
+ err = synproxy_proc_init(net);
+ if (err < 0)
+ goto err3;
+
+ return 0;
+
+err3:
+ free_percpu(snet->stats);
+err2:
+ nf_conntrack_free(ct);
+err1:
+ return err;
+}
+
+static void __net_exit synproxy_net_exit(struct net *net)
+{
+ struct synproxy_net *snet = synproxy_pernet(net);
+
+ nf_conntrack_free(snet->tmpl);
+ synproxy_proc_exit(net);
+ free_percpu(snet->stats);
+}
+
+static struct pernet_operations synproxy_net_ops = {
+ .init = synproxy_net_init,
+ .exit = synproxy_net_exit,
+ .id = &synproxy_net_id,
+ .size = sizeof(struct synproxy_net),
+};
+
+static int __init synproxy_core_init(void)
+{
+ int err;
+
+ err = nf_ct_extend_register(&nf_ct_synproxy_extend);
+ if (err < 0)
+ goto err1;
+
+ err = register_pernet_subsys(&synproxy_net_ops);
+ if (err < 0)
+ goto err2;
+
+ return 0;
+
+err2:
+ nf_ct_extend_unregister(&nf_ct_synproxy_extend);
+err1:
+ return err;
+}
+
+static void __exit synproxy_core_exit(void)
+{
+ unregister_pernet_subsys(&synproxy_net_ops);
+ nf_ct_extend_unregister(&nf_ct_synproxy_extend);
+}
+
+module_init(synproxy_core_init);
+module_exit(synproxy_core_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 4/6] net: syncookies: export cookie_v6_init_sequence/cookie_v6_check
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
` (2 preceding siblings ...)
2013-08-23 7:52 ` [PATCH 3/6] netfilter: add SYNPROXY core/target Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-23 7:52 ` [PATCH 5/6] netfilter: add IPv6 SYNPROXY target Patrick McHardy
` (3 subsequent siblings)
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Extract the local TCP stack independant parts of tcp_v6_init_sequence()
and cookie_v6_check() and export them for use by the upcoming IPv6 SYNPROXY
target.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
include/net/tcp.h | 4 ++++
net/ipv6/syncookies.c | 25 ++++++++++++++++---------
2 files changed, 20 insertions(+), 9 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 63256c6..22872c7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -503,8 +503,12 @@ extern bool cookie_check_timestamp(struct tcp_options_received *opt,
struct net *net, bool *ecn_ok);
/* From net/ipv6/syncookies.c */
+extern int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th,
+ u32 cookie);
extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
#ifdef CONFIG_SYN_COOKIES
+extern u32 __cookie_v6_init_sequence(const struct ipv6hdr *iph,
+ const struct tcphdr *th, u16 *mssp);
extern __u32 cookie_v6_init_sequence(struct sock *sk, const struct sk_buff *skb,
__u16 *mss);
#else
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index d5dda20..bf63ac8 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -112,32 +112,38 @@ static __u32 check_tcp_syn_cookie(__u32 cookie, const struct in6_addr *saddr,
& COOKIEMASK;
}
-__u32 cookie_v6_init_sequence(struct sock *sk, const struct sk_buff *skb, __u16 *mssp)
+u32 __cookie_v6_init_sequence(const struct ipv6hdr *iph,
+ const struct tcphdr *th, __u16 *mssp)
{
- const struct ipv6hdr *iph = ipv6_hdr(skb);
- const struct tcphdr *th = tcp_hdr(skb);
int mssind;
const __u16 mss = *mssp;
- tcp_synq_overflow(sk);
-
for (mssind = ARRAY_SIZE(msstab) - 1; mssind ; mssind--)
if (mss >= msstab[mssind])
break;
*mssp = msstab[mssind];
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
-
return secure_tcp_syn_cookie(&iph->saddr, &iph->daddr, th->source,
th->dest, ntohl(th->seq),
jiffies / (HZ * 60), mssind);
}
+EXPORT_SYMBOL_GPL(__cookie_v6_init_sequence);
-static inline int cookie_check(const struct sk_buff *skb, __u32 cookie)
+__u32 cookie_v6_init_sequence(struct sock *sk, const struct sk_buff *skb, __u16 *mssp)
{
const struct ipv6hdr *iph = ipv6_hdr(skb);
const struct tcphdr *th = tcp_hdr(skb);
+
+ tcp_synq_overflow(sk);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
+
+ return __cookie_v6_init_sequence(iph, th, mssp);
+}
+
+int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th,
+ __u32 cookie)
+{
__u32 seq = ntohl(th->seq) - 1;
__u32 mssind = check_tcp_syn_cookie(cookie, &iph->saddr, &iph->daddr,
th->source, th->dest, seq,
@@ -145,6 +151,7 @@ static inline int cookie_check(const struct sk_buff *skb, __u32 cookie)
return mssind < ARRAY_SIZE(msstab) ? msstab[mssind] : 0;
}
+EXPORT_SYMBOL_GPL(__cookie_v6_check);
struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
{
@@ -167,7 +174,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
goto out;
if (tcp_synq_no_recent_overflow(sk) ||
- (mss = cookie_check(skb, cookie)) == 0) {
+ (mss = __cookie_v6_check(ipv6_hdr(skb), th, cookie)) == 0) {
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESFAILED);
goto out;
}
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 5/6] netfilter: add IPv6 SYNPROXY target
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
` (3 preceding siblings ...)
2013-08-23 7:52 ` [PATCH 4/6] net: syncookies: export cookie_v6_init_sequence/cookie_v6_check Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-23 7:52 ` [PATCH 6/6] netfilter: synproxy: fix handling of retransmissions before established state Patrick McHardy
` (2 subsequent siblings)
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Add an IPv6 version of the SYNPROXY target. The main differences to the
IPv4 version is routing and IP header construction.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
net/ipv6/netfilter/Kconfig | 13 ++
net/ipv6/netfilter/Makefile | 1 +
net/ipv6/netfilter/ip6t_SYNPROXY.c | 460 +++++++++++++++++++++++++++++++++++++
3 files changed, 474 insertions(+)
create mode 100644 net/ipv6/netfilter/ip6t_SYNPROXY.c
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 4433ab40..a7f842b 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -153,6 +153,19 @@ config IP6_NF_TARGET_REJECT
To compile it as a module, choose M here. If unsure, say N.
+config IP6_NF_TARGET_SYNPROXY
+ tristate "SYNPROXY target support"
+ depends on NF_CONNTRACK && NETFILTER_ADVANCED
+ select NETFILTER_SYNPROXY
+ select SYN_COOKIES
+ help
+ The SYNPROXY target allows you to intercept TCP connections and
+ establish them using syncookies before they are passed on to the
+ server. This allows to avoid conntrack and server resource usage
+ during SYN-flood attacks.
+
+ To compile it as a module, choose M here. If unsure, say N.
+
config IP6_NF_MANGLE
tristate "Packet mangling"
default m if NETFILTER_ADVANCED=n
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index 2d11fcc..58f031b 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -37,3 +37,4 @@ obj-$(CONFIG_IP6_NF_MATCH_RT) += ip6t_rt.o
obj-$(CONFIG_IP6_NF_TARGET_MASQUERADE) += ip6t_MASQUERADE.o
obj-$(CONFIG_IP6_NF_TARGET_NPT) += ip6t_NPT.o
obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o
+obj-$(CONFIG_IP6_NF_TARGET_SYNPROXY) += ip6t_SYNPROXY.o
diff --git a/net/ipv6/netfilter/ip6t_SYNPROXY.c b/net/ipv6/netfilter/ip6t_SYNPROXY.c
new file mode 100644
index 0000000..644e6ed
--- /dev/null
+++ b/net/ipv6/netfilter/ip6t_SYNPROXY.c
@@ -0,0 +1,460 @@
+/*
+ * Copyright (c) 2013 Patrick McHardy <kaber@trash.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <net/ip6_checksum.h>
+#include <net/ip6_route.h>
+#include <net/tcp.h>
+
+#include <linux/netfilter_ipv6/ip6_tables.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_SYNPROXY.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
+#include <net/netfilter/nf_conntrack_synproxy.h>
+
+static struct ipv6hdr *
+synproxy_build_ip(struct sk_buff *skb, const struct in6_addr *saddr,
+ const struct in6_addr *daddr)
+{
+ struct ipv6hdr *iph;
+
+ skb_reset_network_header(skb);
+ iph = (struct ipv6hdr *)skb_put(skb, sizeof(*iph));
+ ip6_flow_hdr(iph, 0, 0);
+ iph->hop_limit = 64; //XXX
+ iph->nexthdr = IPPROTO_TCP;
+ iph->saddr = *saddr;
+ iph->daddr = *daddr;
+
+ return iph;
+}
+
+static void
+synproxy_send_tcp(const struct sk_buff *skb, struct sk_buff *nskb,
+ struct nf_conntrack *nfct, enum ip_conntrack_info ctinfo,
+ struct ipv6hdr *niph, struct tcphdr *nth,
+ unsigned int tcp_hdr_size)
+{
+ struct net *net = nf_ct_net((struct nf_conn *)nfct);
+ struct dst_entry *dst;
+ struct flowi6 fl6;
+
+ nth->check = ~tcp_v6_check(tcp_hdr_size, &niph->saddr, &niph->daddr, 0);
+ nskb->ip_summed = CHECKSUM_PARTIAL;
+ nskb->csum_start = (unsigned char *)nth - nskb->head;
+ nskb->csum_offset = offsetof(struct tcphdr, check);
+
+ memset(&fl6, 0, sizeof(fl6));
+ fl6.flowi6_proto = IPPROTO_TCP;
+ fl6.saddr = niph->saddr;
+ fl6.daddr = niph->daddr;
+ fl6.fl6_sport = nth->source;
+ fl6.fl6_dport = nth->dest;
+ security_skb_classify_flow((struct sk_buff *)skb, flowi6_to_flowi(&fl6));
+ dst = ip6_route_output(net, NULL, &fl6);
+ if (dst == NULL || dst->error) {
+ dst_release(dst);
+ goto free_nskb;
+ }
+ dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), NULL, 0);
+ if (IS_ERR(dst))
+ goto free_nskb;
+
+ skb_dst_set(nskb, dst);
+
+ nskb->nfct = nfct;
+ nskb->nfctinfo = ctinfo;
+ nf_conntrack_get(nfct);
+
+ ip6_local_out(nskb);
+ return;
+
+free_nskb:
+ kfree_skb(nskb);
+}
+
+static void
+synproxy_send_client_synack(const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct ipv6hdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+ u16 mss = opts->mss;
+
+ iph = ipv6_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, &iph->daddr, &iph->saddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->dest;
+ nth->dest = th->source;
+ nth->seq = htonl(__cookie_v6_init_sequence(iph, th, &mss));
+ nth->ack_seq = htonl(ntohl(th->seq) + 1);
+ tcp_flag_word(nth) = TCP_FLAG_SYN | TCP_FLAG_ACK;
+ if (opts->options & XT_SYNPROXY_OPT_ECN)
+ tcp_flag_word(nth) |= TCP_FLAG_ECE;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = 0;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_server_syn(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct ipv6hdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ipv6_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, &iph->saddr, &iph->daddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->source;
+ nth->dest = th->dest;
+ nth->seq = htonl(ntohl(th->seq) - 1);
+ /* ack_seq is used to relay our ISN to the synproxy hook to initialize
+ * sequence number translation once a connection tracking entry exists.
+ */
+ nth->ack_seq = htonl(ntohl(th->ack_seq) - 1);
+ tcp_flag_word(nth) = TCP_FLAG_SYN;
+ if (opts->options & XT_SYNPROXY_OPT_ECN)
+ tcp_flag_word(nth) |= TCP_FLAG_ECE | TCP_FLAG_CWR;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = th->window;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, &snet->tmpl->ct_general, IP_CT_NEW,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_server_ack(const struct synproxy_net *snet,
+ const struct ip_ct_tcp *state,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct ipv6hdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ipv6_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, &iph->daddr, &iph->saddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->dest;
+ nth->dest = th->source;
+ nth->seq = htonl(ntohl(th->ack_seq));
+ nth->ack_seq = htonl(ntohl(th->seq) + 1);
+ tcp_flag_word(nth) = TCP_FLAG_ACK;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = htons(state->seen[IP_CT_DIR_ORIGINAL].td_maxwin);
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED,
+ niph, nth, tcp_hdr_size);
+}
+
+static void
+synproxy_send_client_ack(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ const struct synproxy_options *opts)
+{
+ struct sk_buff *nskb;
+ struct ipv6hdr *iph, *niph;
+ struct tcphdr *nth;
+ unsigned int tcp_hdr_size;
+
+ iph = ipv6_hdr(skb);
+
+ tcp_hdr_size = sizeof(*nth) + synproxy_options_size(opts);
+ nskb = alloc_skb(sizeof(*niph) + tcp_hdr_size + MAX_TCP_HEADER,
+ GFP_ATOMIC);
+ if (nskb == NULL)
+ return;
+ skb_reserve(nskb, MAX_TCP_HEADER);
+
+ niph = synproxy_build_ip(nskb, &iph->saddr, &iph->daddr);
+
+ skb_reset_transport_header(nskb);
+ nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
+ nth->source = th->source;
+ nth->dest = th->dest;
+ nth->seq = htonl(ntohl(th->seq) + 1);
+ nth->ack_seq = th->ack_seq;
+ tcp_flag_word(nth) = TCP_FLAG_ACK;
+ nth->doff = tcp_hdr_size / 4;
+ nth->window = th->window;
+ nth->check = 0;
+ nth->urg_ptr = 0;
+
+ synproxy_build_options(nth, opts);
+
+ synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
+ niph, nth, tcp_hdr_size);
+}
+
+static unsigned int
+synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par)
+{
+ const struct xt_synproxy_info *info = par->targinfo;
+ struct synproxy_net *snet = synproxy_pernet(dev_net(par->in));
+ struct synproxy_options opts = {};
+ struct tcphdr *th, _th;
+
+ if (nf_ip6_checksum(skb, par->hooknum, par->thoff, IPPROTO_TCP))
+ return NF_DROP;
+
+ th = skb_header_pointer(skb, par->thoff, sizeof(_th), &_th);
+ if (th == NULL)
+ return NF_DROP;
+
+ synproxy_parse_options(skb, par->thoff, th, &opts);
+
+ if (th->syn) {
+ /* Initial SYN from client */
+ this_cpu_inc(snet->stats->syn_received);
+
+ if (th->ece && th->cwr)
+ opts.options |= XT_SYNPROXY_OPT_ECN;
+
+ opts.options &= info->options;
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_init_timestamp_cookie(info, &opts);
+ else
+ opts.options &= ~(XT_SYNPROXY_OPT_WSCALE |
+ XT_SYNPROXY_OPT_SACK_PERM |
+ XT_SYNPROXY_OPT_ECN);
+
+ synproxy_send_client_synack(skb, th, &opts);
+ } else if (th->ack && !(th->fin || th->rst)) {
+ /* ACK from client */
+ int mss = __cookie_v6_check(ipv6_hdr(skb), th,
+ ntohl(th->ack_seq) - 1);
+ if (mss == 0) {
+ this_cpu_inc(snet->stats->cookie_invalid);
+ return NF_DROP;
+ }
+
+ this_cpu_inc(snet->stats->cookie_valid);
+ opts.mss = mss;
+
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_check_timestamp_cookie(&opts);
+
+ synproxy_send_server_syn(snet, skb, th, &opts);
+ }
+
+ return NF_DROP;
+}
+
+static unsigned int ipv6_synproxy_hook(unsigned int hooknum,
+ struct sk_buff *skb,
+ const struct net_device *in,
+ const struct net_device *out,
+ int (*okfn)(struct sk_buff *))
+{
+ struct synproxy_net *snet = synproxy_pernet(dev_net(in ? : out));
+ enum ip_conntrack_info ctinfo;
+ struct nf_conn *ct;
+ struct nf_conn_synproxy *synproxy;
+ struct synproxy_options opts = {};
+ const struct ip_ct_tcp *state;
+ struct tcphdr *th, _th;
+ __be16 frag_off;
+ u8 nexthdr;
+ int thoff;
+
+ ct = nf_ct_get(skb, &ctinfo);
+ if (ct == NULL)
+ return NF_ACCEPT;
+
+ synproxy = nfct_synproxy(ct);
+ if (synproxy == NULL)
+ return NF_ACCEPT;
+
+ if (nf_is_loopback_packet(skb))
+ return NF_ACCEPT;
+
+ nexthdr = ipv6_hdr(skb)->nexthdr;
+ thoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr,
+ &frag_off);
+ if (thoff < 0)
+ return NF_ACCEPT;
+
+ th = skb_header_pointer(skb, thoff, sizeof(_th), &_th);
+ if (th == NULL)
+ return NF_DROP;
+
+ state = &ct->proto.tcp;
+ switch (state->state) {
+ case TCP_CONNTRACK_SYN_SENT:
+ synproxy_parse_options(skb, thoff, th, &opts);
+
+ synproxy->isn = ntohl(th->ack_seq);
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy->tsoff = -opts.tsecr;
+ break;
+ case TCP_CONNTRACK_SYN_RECV:
+ if (!th->syn)
+ break;
+
+ synproxy_parse_options(skb, thoff, th, &opts);
+ if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy->tsoff += opts.tsval;
+
+ opts.options &= ~(XT_SYNPROXY_OPT_MSS |
+ XT_SYNPROXY_OPT_WSCALE |
+ XT_SYNPROXY_OPT_SACK_PERM);
+
+ swap(opts.tsval, opts.tsecr);
+ synproxy_send_server_ack(snet, state, skb, th, &opts);
+
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn - ntohl(th->seq));
+
+ swap(opts.tsval, opts.tsecr);
+ synproxy_send_client_ack(snet, skb, th, &opts);
+
+ consume_skb(skb);
+ return NF_STOLEN;
+ case TCP_CONNTRACK_CLOSE:
+ if (!th->rst)
+ break;
+
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
+ ntohl(th->seq) + 1);
+ break;
+ default:
+ break;
+ }
+
+ synproxy_tstamp_adjust(skb, thoff, th, ct, ctinfo, synproxy);
+ return NF_ACCEPT;
+}
+
+static int synproxy_tg6_check(const struct xt_tgchk_param *par)
+{
+ const struct ip6t_entry *e = par->entryinfo;
+
+ if (!(e->ipv6.flags & IP6T_F_PROTO) ||
+ e->ipv6.proto != IPPROTO_TCP ||
+ e->ipv6.invflags & XT_INV_PROTO)
+ return -EINVAL;
+
+ return nf_ct_l3proto_try_module_get(par->family);
+}
+
+static void synproxy_tg6_destroy(const struct xt_tgdtor_param *par)
+{
+ nf_ct_l3proto_module_put(par->family);
+}
+
+static struct xt_target synproxy_tg6_reg __read_mostly = {
+ .name = "SYNPROXY",
+ .family = NFPROTO_IPV6,
+ .target = synproxy_tg6,
+ .targetsize = sizeof(struct xt_synproxy_info),
+ .checkentry = synproxy_tg6_check,
+ .destroy = synproxy_tg6_destroy,
+ .me = THIS_MODULE,
+};
+
+static struct nf_hook_ops ipv6_synproxy_ops[] __read_mostly = {
+ {
+ .hook = ipv6_synproxy_hook,
+ .owner = THIS_MODULE,
+ .pf = NFPROTO_IPV6,
+ .hooknum = NF_INET_LOCAL_IN,
+ .priority = NF_IP_PRI_CONNTRACK_CONFIRM - 1,
+ },
+ {
+ .hook = ipv6_synproxy_hook,
+ .owner = THIS_MODULE,
+ .pf = NFPROTO_IPV6,
+ .hooknum = NF_INET_POST_ROUTING,
+ .priority = NF_IP_PRI_CONNTRACK_CONFIRM - 1,
+ },
+};
+
+static int __init synproxy_tg6_init(void)
+{
+ int err;
+
+ err = nf_register_hooks(ipv6_synproxy_ops,
+ ARRAY_SIZE(ipv6_synproxy_ops));
+ if (err < 0)
+ goto err1;
+
+ err = xt_register_target(&synproxy_tg6_reg);
+ if (err < 0)
+ goto err2;
+
+ return 0;
+
+err2:
+ nf_unregister_hooks(ipv6_synproxy_ops, ARRAY_SIZE(ipv6_synproxy_ops));
+err1:
+ return err;
+}
+
+static void __exit synproxy_tg6_exit(void)
+{
+ xt_unregister_target(&synproxy_tg6_reg);
+ nf_unregister_hooks(ipv6_synproxy_ops, ARRAY_SIZE(ipv6_synproxy_ops));
+}
+
+module_init(synproxy_tg6_init);
+module_exit(synproxy_tg6_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 6/6] netfilter: synproxy: fix handling of retransmissions before established state
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
` (4 preceding siblings ...)
2013-08-23 7:52 ` [PATCH 5/6] netfilter: add IPv6 SYNPROXY target Patrick McHardy
@ 2013-08-23 7:52 ` Patrick McHardy
2013-08-26 20:24 ` [PATCH RFC 0/6] SYNPROXY target v2 David Miller
2013-08-26 22:42 ` Pablo Neira Ayuso
7 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-23 7:52 UTC (permalink / raw)
To: pablo; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Fix multiple issues with retransmissions and out of order packets:
- the final ACK to the server was incorrectly manually attached to the
existing conntrack entry, causing conntrack to skip handling the packet
and thus not moving the connection from SYN_RECV to ESTABLISHED state.
- if the final ACK then got lost, a retransmitted SYN/ACK would
reinitialize timestamp adjustments. The initialization was done
incrementally though, so at that point the translation would result
in incorrect timestamp values. To prevent this, the timestamp
adjustments are not initialized incrementally anymore, but the initial
timestamp value is stored and adjustment is initialized based on the
difference between the value stored and the value received in the
SYN/ACK. Additionally the first point makes sure we're in ESTABLISHED
state after initialization is complete.
- when a keep-alive from the client was received before the connection
moved to ESTABLISHED state (SYN to the server was lost) and the client
sent a keep alive, TCP conntrack regarded the keep alive as invalid in
this state, meaning it didn't get associated with the conntrack and
(since INVALID packets are directed to the SYNPROXY target by the
ruleset to catch the final ACK from the client) hit the SYNPROXY target
again.
Since keep alives are sent with SEG.SEQ = SND.NXT-1, a new connection
with a sequence number off by one was established to the server.
A special case is added to TCP conntrack handling of invalid packets
to catch the keep alives and associate them with the correct conntrack
entry so the retransmitted SYN to the server will use the correct
sequence number. A new counter has been introduced for these
retranmissions.
Additionally:
- fix the window size announced in the window update sent to the client
when the connection to the server is established: the window value is
taken from the server's SYN/ACK, meaning it is unscaled, so we have to
scale it down when including it in a pure ACK.
- don't reinitializex sequence number translation when a RST is received
for an establshed connection. This is only supposed to be done when the
first and only reply is a RST.
With these changes, I get 100% success rate at connection establishment
even with drop rates of 10%.
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
include/net/netfilter/nf_conntrack_synproxy.h | 2 +
net/ipv4/netfilter/ipt_SYNPROXY.c | 88 ++++++++++++++++-----------
net/ipv6/netfilter/ip6t_SYNPROXY.c | 88 ++++++++++++++++-----------
net/netfilter/nf_conntrack_proto_tcp.c | 16 +++++
net/netfilter/nf_synproxy_core.c | 10 +--
5 files changed, 132 insertions(+), 72 deletions(-)
diff --git a/include/net/netfilter/nf_conntrack_synproxy.h b/include/net/netfilter/nf_conntrack_synproxy.h
index cc43895..c3de4a2 100644
--- a/include/net/netfilter/nf_conntrack_synproxy.h
+++ b/include/net/netfilter/nf_conntrack_synproxy.h
@@ -5,6 +5,7 @@
struct nf_conn_synproxy {
u32 isn;
+ u32 its;
u32 tsoff;
};
@@ -30,6 +31,7 @@ struct synproxy_stats {
unsigned int syn_received;
unsigned int cookie_invalid;
unsigned int cookie_valid;
+ unsigned int cookie_retrans;
};
struct synproxy_net {
diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c
index d8577e0..8fd6143 100644
--- a/net/ipv4/netfilter/ipt_SYNPROXY.c
+++ b/net/ipv4/netfilter/ipt_SYNPROXY.c
@@ -54,9 +54,11 @@ synproxy_send_tcp(const struct sk_buff *skb, struct sk_buff *nskb,
if (ip_route_me_harder(nskb, RTN_UNSPEC))
goto free_nskb;
- nskb->nfct = nfct;
- nskb->nfctinfo = ctinfo;
- nf_conntrack_get(nfct);
+ if (nfct) {
+ nskb->nfct = nfct;
+ nskb->nfctinfo = ctinfo;
+ nf_conntrack_get(nfct);
+ }
ip_local_out(nskb);
return;
@@ -109,7 +111,7 @@ synproxy_send_client_synack(const struct sk_buff *skb, const struct tcphdr *th,
static void
synproxy_send_server_syn(const struct synproxy_net *snet,
const struct sk_buff *skb, const struct tcphdr *th,
- const struct synproxy_options *opts)
+ const struct synproxy_options *opts, u32 recv_seq)
{
struct sk_buff *nskb;
struct iphdr *iph, *niph;
@@ -131,7 +133,7 @@ synproxy_send_server_syn(const struct synproxy_net *snet,
nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
nth->source = th->source;
nth->dest = th->dest;
- nth->seq = htonl(ntohl(th->seq) - 1);
+ nth->seq = htonl(recv_seq - 1);
/* ack_seq is used to relay our ISN to the synproxy hook to initialize
* sequence number translation once a connection tracking entry exists.
*/
@@ -186,8 +188,7 @@ synproxy_send_server_ack(const struct synproxy_net *snet,
synproxy_build_options(nth, opts);
- synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED,
- niph, nth, tcp_hdr_size);
+ synproxy_send_tcp(skb, nskb, NULL, 0, niph, nth, tcp_hdr_size);
}
static void
@@ -219,14 +220,36 @@ synproxy_send_client_ack(const struct synproxy_net *snet,
nth->ack_seq = th->ack_seq;
tcp_flag_word(nth) = TCP_FLAG_ACK;
nth->doff = tcp_hdr_size / 4;
- nth->window = th->window;
+ nth->window = ntohs(htons(th->window) >> opts->wscale);
nth->check = 0;
nth->urg_ptr = 0;
synproxy_build_options(nth, opts);
- synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
- niph, nth, tcp_hdr_size);
+ synproxy_send_tcp(skb, nskb, NULL, 0, niph, nth, tcp_hdr_size);
+}
+
+static bool
+synproxy_recv_client_ack(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ struct synproxy_options *opts, u32 recv_seq)
+{
+ int mss;
+
+ mss = __cookie_v4_check(ip_hdr(skb), th, ntohl(th->ack_seq) - 1);
+ if (mss == 0) {
+ this_cpu_inc(snet->stats->cookie_invalid);
+ return false;
+ }
+
+ this_cpu_inc(snet->stats->cookie_valid);
+ opts->mss = mss;
+
+ if (opts->options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_check_timestamp_cookie(opts);
+
+ synproxy_send_server_syn(snet, skb, th, opts, recv_seq);
+ return true;
}
static unsigned int
@@ -262,23 +285,9 @@ synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par)
XT_SYNPROXY_OPT_ECN);
synproxy_send_client_synack(skb, th, &opts);
- } else if (th->ack && !(th->fin || th->rst)) {
+ } else if (th->ack && !(th->fin || th->rst))
/* ACK from client */
- int mss = __cookie_v4_check(ip_hdr(skb), th,
- ntohl(th->ack_seq) - 1);
- if (mss == 0) {
- this_cpu_inc(snet->stats->cookie_invalid);
- return NF_DROP;
- }
-
- this_cpu_inc(snet->stats->cookie_valid);
- opts.mss = mss;
-
- if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy_check_timestamp_cookie(&opts);
-
- synproxy_send_server_syn(snet, skb, th, &opts);
- }
+ synproxy_recv_client_ack(snet, skb, th, &opts, ntohl(th->seq));
return NF_DROP;
}
@@ -319,17 +328,30 @@ static unsigned int ipv4_synproxy_hook(unsigned int hooknum,
case TCP_CONNTRACK_SYN_SENT:
synproxy_parse_options(skb, thoff, th, &opts);
+ if (!th->syn && th->ack &&
+ CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) {
+ /* Keep-Alives are sent with SEG.SEQ = SND.NXT-1,
+ * therefore we need to add 1 to make the SYN sequence
+ * number match the one of first SYN.
+ */
+ if (synproxy_recv_client_ack(snet, skb, th, &opts,
+ ntohl(th->seq) + 1))
+ this_cpu_inc(snet->stats->cookie_retrans);
+
+ return NF_DROP;
+ }
+
synproxy->isn = ntohl(th->ack_seq);
if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy->tsoff = -opts.tsecr;
+ synproxy->its = opts.tsecr;
break;
case TCP_CONNTRACK_SYN_RECV:
- if (!th->syn)
+ if (!th->syn || !th->ack)
break;
synproxy_parse_options(skb, thoff, th, &opts);
if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy->tsoff += opts.tsval;
+ synproxy->tsoff = opts.tsval - synproxy->its;
opts.options &= ~(XT_SYNPROXY_OPT_MSS |
XT_SYNPROXY_OPT_WSCALE |
@@ -346,11 +368,9 @@ static unsigned int ipv4_synproxy_hook(unsigned int hooknum,
consume_skb(skb);
return NF_STOLEN;
case TCP_CONNTRACK_CLOSE:
- if (!th->rst)
- break;
-
- nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
- ntohl(th->seq) + 1);
+ if (th->rst && !test_bit(IPS_SEEN_REPLY_BIT, &ct->status))
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
+ ntohl(th->seq) + 1);
break;
default:
break;
diff --git a/net/ipv6/netfilter/ip6t_SYNPROXY.c b/net/ipv6/netfilter/ip6t_SYNPROXY.c
index 644e6ed..b388964 100644
--- a/net/ipv6/netfilter/ip6t_SYNPROXY.c
+++ b/net/ipv6/netfilter/ip6t_SYNPROXY.c
@@ -69,9 +69,11 @@ synproxy_send_tcp(const struct sk_buff *skb, struct sk_buff *nskb,
skb_dst_set(nskb, dst);
- nskb->nfct = nfct;
- nskb->nfctinfo = ctinfo;
- nf_conntrack_get(nfct);
+ if (nfct) {
+ nskb->nfct = nfct;
+ nskb->nfctinfo = ctinfo;
+ nf_conntrack_get(nfct);
+ }
ip6_local_out(nskb);
return;
@@ -124,7 +126,7 @@ synproxy_send_client_synack(const struct sk_buff *skb, const struct tcphdr *th,
static void
synproxy_send_server_syn(const struct synproxy_net *snet,
const struct sk_buff *skb, const struct tcphdr *th,
- const struct synproxy_options *opts)
+ const struct synproxy_options *opts, u32 recv_seq)
{
struct sk_buff *nskb;
struct ipv6hdr *iph, *niph;
@@ -146,7 +148,7 @@ synproxy_send_server_syn(const struct synproxy_net *snet,
nth = (struct tcphdr *)skb_put(nskb, tcp_hdr_size);
nth->source = th->source;
nth->dest = th->dest;
- nth->seq = htonl(ntohl(th->seq) - 1);
+ nth->seq = htonl(recv_seq - 1);
/* ack_seq is used to relay our ISN to the synproxy hook to initialize
* sequence number translation once a connection tracking entry exists.
*/
@@ -201,8 +203,7 @@ synproxy_send_server_ack(const struct synproxy_net *snet,
synproxy_build_options(nth, opts);
- synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED,
- niph, nth, tcp_hdr_size);
+ synproxy_send_tcp(skb, nskb, NULL, 0, niph, nth, tcp_hdr_size);
}
static void
@@ -234,14 +235,36 @@ synproxy_send_client_ack(const struct synproxy_net *snet,
nth->ack_seq = th->ack_seq;
tcp_flag_word(nth) = TCP_FLAG_ACK;
nth->doff = tcp_hdr_size / 4;
- nth->window = th->window;
+ nth->window = ntohs(htons(th->window) >> opts->wscale);
nth->check = 0;
nth->urg_ptr = 0;
synproxy_build_options(nth, opts);
- synproxy_send_tcp(skb, nskb, skb->nfct, IP_CT_ESTABLISHED_REPLY,
- niph, nth, tcp_hdr_size);
+ synproxy_send_tcp(skb, nskb, NULL, 0, niph, nth, tcp_hdr_size);
+}
+
+static bool
+synproxy_recv_client_ack(const struct synproxy_net *snet,
+ const struct sk_buff *skb, const struct tcphdr *th,
+ struct synproxy_options *opts, u32 recv_seq)
+{
+ int mss;
+
+ mss = __cookie_v6_check(ipv6_hdr(skb), th, ntohl(th->ack_seq) - 1);
+ if (mss == 0) {
+ this_cpu_inc(snet->stats->cookie_invalid);
+ return false;
+ }
+
+ this_cpu_inc(snet->stats->cookie_valid);
+ opts->mss = mss;
+
+ if (opts->options & XT_SYNPROXY_OPT_TIMESTAMP)
+ synproxy_check_timestamp_cookie(opts);
+
+ synproxy_send_server_syn(snet, skb, th, opts, recv_seq);
+ return true;
}
static unsigned int
@@ -277,23 +300,9 @@ synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par)
XT_SYNPROXY_OPT_ECN);
synproxy_send_client_synack(skb, th, &opts);
- } else if (th->ack && !(th->fin || th->rst)) {
+ } else if (th->ack && !(th->fin || th->rst))
/* ACK from client */
- int mss = __cookie_v6_check(ipv6_hdr(skb), th,
- ntohl(th->ack_seq) - 1);
- if (mss == 0) {
- this_cpu_inc(snet->stats->cookie_invalid);
- return NF_DROP;
- }
-
- this_cpu_inc(snet->stats->cookie_valid);
- opts.mss = mss;
-
- if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy_check_timestamp_cookie(&opts);
-
- synproxy_send_server_syn(snet, skb, th, &opts);
- }
+ synproxy_recv_client_ack(snet, skb, th, &opts, ntohl(th->seq));
return NF_DROP;
}
@@ -341,17 +350,30 @@ static unsigned int ipv6_synproxy_hook(unsigned int hooknum,
case TCP_CONNTRACK_SYN_SENT:
synproxy_parse_options(skb, thoff, th, &opts);
+ if (!th->syn && th->ack &&
+ CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) {
+ /* Keep-Alives are sent with SEG.SEQ = SND.NXT-1,
+ * therefore we need to add 1 to make the SYN sequence
+ * number match the one of first SYN.
+ */
+ if (synproxy_recv_client_ack(snet, skb, th, &opts,
+ ntohl(th->seq) + 1))
+ this_cpu_inc(snet->stats->cookie_retrans);
+
+ return NF_DROP;
+ }
+
synproxy->isn = ntohl(th->ack_seq);
if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy->tsoff = -opts.tsecr;
+ synproxy->its = opts.tsecr;
break;
case TCP_CONNTRACK_SYN_RECV:
- if (!th->syn)
+ if (!th->syn || !th->ack)
break;
synproxy_parse_options(skb, thoff, th, &opts);
if (opts.options & XT_SYNPROXY_OPT_TIMESTAMP)
- synproxy->tsoff += opts.tsval;
+ synproxy->tsoff = opts.tsval - synproxy->its;
opts.options &= ~(XT_SYNPROXY_OPT_MSS |
XT_SYNPROXY_OPT_WSCALE |
@@ -368,11 +390,9 @@ static unsigned int ipv6_synproxy_hook(unsigned int hooknum,
consume_skb(skb);
return NF_STOLEN;
case TCP_CONNTRACK_CLOSE:
- if (!th->rst)
- break;
-
- nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
- ntohl(th->seq) + 1);
+ if (th->rst && !test_bit(IPS_SEEN_REPLY_BIT, &ct->status))
+ nf_ct_seqadj_init(ct, ctinfo, synproxy->isn -
+ ntohl(th->seq) + 1);
break;
default:
break;
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 3c3401e..5d714db 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -28,6 +28,7 @@
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_ecache.h>
#include <net/netfilter/nf_conntrack_seqadj.h>
+#include <net/netfilter/nf_conntrack_synproxy.h>
#include <net/netfilter/nf_log.h>
#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
#include <net/netfilter/ipv6/nf_conntrack_ipv6.h>
@@ -942,6 +943,21 @@ static int tcp_packet(struct nf_conn *ct,
"state %s ", tcp_conntrack_names[old_state]);
return NF_ACCEPT;
case TCP_CONNTRACK_MAX:
+ /* Special case for SYN proxy: when the SYN to the server or
+ * the SYN/ACK from the server is lost, the client may transmit
+ * a keep-alive packet while in SYN_SENT state. This needs to
+ * be associated with the original conntrack entry in order to
+ * generate a new SYN with the correct sequence number.
+ */
+ if (nfct_synproxy(ct) && old_state == TCP_CONNTRACK_SYN_SENT &&
+ index == TCP_ACK_SET && dir == IP_CT_DIR_ORIGINAL &&
+ ct->proto.tcp.last_dir == IP_CT_DIR_ORIGINAL &&
+ ct->proto.tcp.seen[dir].td_end - 1 == ntohl(th->seq)) {
+ pr_debug("nf_ct_tcp: SYN proxy client keep alive\n");
+ spin_unlock_bh(&ct->lock);
+ return NF_ACCEPT;
+ }
+
/* Invalid packet */
pr_debug("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n",
dir, get_conntrack_index(th), old_state);
diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c
index 502faae..c067845 100644
--- a/net/netfilter/nf_synproxy_core.c
+++ b/net/netfilter/nf_synproxy_core.c
@@ -281,15 +281,17 @@ static int synproxy_cpu_seq_show(struct seq_file *seq, void *v)
struct synproxy_stats *stats = v;
if (v == SEQ_START_TOKEN) {
- seq_printf(seq, "entries\tsyn_received\t"
- "cookie_invalid\tcookie_valid\n");
+ seq_printf(seq, "entries\t\tsyn_received\t"
+ "cookie_invalid\tcookie_valid\t"
+ "cookie_retrans\n");
return 0;
}
- seq_printf(seq, "%08x\t%08x\t%08x\t%08x\n", 0,
+ seq_printf(seq, "%08x\t%08x\t%08x\t%08x\t%08x\n", 0,
stats->syn_received,
stats->cookie_invalid,
- stats->cookie_valid);
+ stats->cookie_valid,
+ stats->cookie_retrans);
return 0;
}
--
1.8.1.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH RFC 0/6] SYNPROXY target v2
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
` (5 preceding siblings ...)
2013-08-23 7:52 ` [PATCH 6/6] netfilter: synproxy: fix handling of retransmissions before established state Patrick McHardy
@ 2013-08-26 20:24 ` David Miller
2013-08-26 22:42 ` Pablo Neira Ayuso
7 siblings, 0 replies; 10+ messages in thread
From: David Miller @ 2013-08-26 20:24 UTC (permalink / raw)
To: kaber; +Cc: pablo, netfilter-devel, netdev, mph, jesper.brouer, as
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 23 Aug 2013 09:52:03 +0200
> The following patches contain the current version of the SYNPROXY target.
> Changes this the last posting are:
>
> - use sysctl_ip_default_ttl instead of hardcoding 64
>
> - use MAX_TCP_HEADER instead of LL_MAX_HEADER
>
> - add some comments requested by Jesper regarding ack_seq initialization
> in the server's SYN packet
>
> - use consume_skb() instead of kfree_skb() in the synproxy hook
>
> - remove a fixme and add explicit check for "-p tcp" in the SYNPROXY rule
> in the IPv6 version
>
> - some whitespace fixes
>
> - a larger number of fixes for properly handling retransmissions and
> out of order packets, please see the changelog included in patch 6/6
> for details.
>
> I've kept those in a seperate patch for now to ease review, we'll do
> some more extensive testing on monday and I'll fold the changes in
> their respective patches before the final submission.
>
> Comments welcome.
No objections to the non-netfilter portions:
Acked-by: David S. Miller <davem@davemloft.net>
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH RFC 0/6] SYNPROXY target v2
2013-08-23 7:52 [PATCH RFC 0/6] SYNPROXY target v2 Patrick McHardy
` (6 preceding siblings ...)
2013-08-26 20:24 ` [PATCH RFC 0/6] SYNPROXY target v2 David Miller
@ 2013-08-26 22:42 ` Pablo Neira Ayuso
2013-08-27 6:56 ` Patrick McHardy
7 siblings, 1 reply; 10+ messages in thread
From: Pablo Neira Ayuso @ 2013-08-26 22:42 UTC (permalink / raw)
To: Patrick McHardy; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
Hi Patrick!
On Fri, Aug 23, 2013 at 09:52:03AM +0200, Patrick McHardy wrote:
> The following patches contain the current version of the SYNPROXY target.
I've been looking at my mailbox and patchwork but I couldn't find the
userspace iptables patches, could you post them?
Thanks!
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH RFC 0/6] SYNPROXY target v2
2013-08-26 22:42 ` Pablo Neira Ayuso
@ 2013-08-27 6:56 ` Patrick McHardy
0 siblings, 0 replies; 10+ messages in thread
From: Patrick McHardy @ 2013-08-27 6:56 UTC (permalink / raw)
To: Pablo Neira Ayuso; +Cc: netfilter-devel, netdev, mph, jesper.brouer, as
On Tue, Aug 27, 2013 at 12:42:51AM +0200, Pablo Neira Ayuso wrote:
> Hi Patrick!
>
> On Fri, Aug 23, 2013 at 09:52:03AM +0200, Patrick McHardy wrote:
> > The following patches contain the current version of the SYNPROXY target.
>
> I've been looking at my mailbox and patchwork but I couldn't find the
> userspace iptables patches, could you post them?
Sure, coming up in a minute.
^ permalink raw reply [flat|nested] 10+ messages in thread