Skip to contentSkip to conversation

netfilter: nf_conntrack_tcp: re-init for syn packets only

PATCH Applied Florian Westphal 5 participants 5 messages Apr 25, 2022 – Jun 10, 2022

Subsystems (2)
lore
Share & tools
lore

Florian Westphal

author maintainer
#0 lore raw

PATCH 1/1 netfilter: nf_conntrack_tcp: re-init for syn packets only

Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell pinpointed to nf_conntrack tcp_in_window() bug.

tcp trace shows following sequence:

I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]> R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..] R > I Flags [P.], seq 1:89, ack 1, [..]

Note 3rd ACK is from responder to initiator so following branch is taken:

    } else if (((state->state == TCP_CONNTRACK_SYN_SENT
               && dir == IP_CT_DIR_ORIGINAL)
               || (state->state == TCP_CONNTRACK_SYN_RECV
               && dir == IP_CT_DIR_REPLY))
               && after(end, sender->td_end)) {

... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY. This causes the scaling factor to be reset to 0: window scale option is only present in syn(ack) packets. This in turn makes nf_conntrack mark valid packets as out-of-window.

This was always broken, it exists even in original commit where window tracking was added to ip_conntrack (nf_conntrack predecessor) in 2.6.9-rc1 kernel.

Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").

Upon closer look, those conditionals/branches can be merged:

Because earlier checks prevent syn-ack from showing up in original direction, the 'dir' checks in the conditional quoted above are redundant, remove them. Return early for pure syn retransmitted in reply direction (simultaneous open).

Fixes 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.")
Cc Jozsef Kadlecsik <[email protected]>
---
 net/netfilter/nf_conntrack_proto_tcp.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)
net/netfilter/nf_conntrack_proto_tcp.c +6 -15
index 8ec55cd72572..204a5cdff5b1 100644
@@ -556,24 +556,14 @@ static bool tcp_in_window(struct nf_conn *ct,
 			}
 
 		}
-	} else if (((state->state == TCP_CONNTRACK_SYN_SENT
-		     && dir == IP_CT_DIR_ORIGINAL)
-		   || (state->state == TCP_CONNTRACK_SYN_RECV
-		     && dir == IP_CT_DIR_REPLY))
-		   && after(end, sender->td_end)) {
+	} else if (tcph->syn &&
+		   after(end, sender->td_end) &&
+		   (state->state == TCP_CONNTRACK_SYN_SENT ||
+		    state->state == TCP_CONNTRACK_SYN_RECV)) {
 		/*
 		 * RFC 793: "if a TCP is reinitialized ... then it need
 		 * not wait at all; it must only be sure to use sequence
 		 * numbers larger than those recently used."
-		 */
-		sender->td_end =
-		sender->td_maxend = end;
-		sender->td_maxwin = (win == 0 ? 1 : win);
-
-		tcp_options(skb, dataoff, tcph, sender);
-	} else if (tcph->syn && dir == IP_CT_DIR_REPLY &&
-		   state->state == TCP_CONNTRACK_SYN_SENT) {
-		/* Retransmitted syn-ack, or syn (simultaneous open).
 		 *
 		 * Re-init state for this direction, just like for the first
 		 * syn(-ack) reply, it might differ in seq, ack or tcp options.
@@ -581,7 +571,8 @@ static bool tcp_in_window(struct nf_conn *ct,
 		tcp_init_sender(sender, receiver,
 				skb, dataoff, tcph,
 				end, win);
-		if (!tcph->ack)
+
+		if (dir == IP_CT_DIR_REPLY && !tcph->ack)
 			return true;
 	}
 
Signature · 1 line
2.35.1
Acked-by on v1 1/1
#1 lore raw
Quoted 35 lines · Florian Westphal

Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell pinpointed to nf_conntrack tcp_in_window() bug.

tcp trace shows following sequence:

I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]> R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..] R > I Flags [P.], seq 1:89, ack 1, [..]

Note 3rd ACK is from responder to initiator so following branch is taken:

    } else if (((state->state == TCP_CONNTRACK_SYN_SENT
               && dir == IP_CT_DIR_ORIGINAL)
               || (state->state == TCP_CONNTRACK_SYN_RECV
               && dir == IP_CT_DIR_REPLY))
               && after(end, sender->td_end)) {

... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY. This causes the scaling factor to be reset to 0: window scale option is only present in syn(ack) packets. This in turn makes nf_conntrack mark valid packets as out-of-window.

This was always broken, it exists even in original commit where window tracking was added to ip_conntrack (nf_conntrack predecessor) in 2.6.9-rc1 kernel.

Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").

Upon closer look, those conditionals/branches can be merged:

Because earlier checks prevent syn-ack from showing up in original direction, the 'dir' checks in the conditional quoted above are redundant, remove them. Return early for pure syn retransmitted in reply direction (simultaneous open).

Quoted 4 lines
Fixes 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.")
Cc Jozsef Kadlecsik <[email protected]>

[Sorry, I was away whole last week as well.]

Best regards,
Jozsef

Quoted 51 lines
---
 net/netfilter/nf_conntrack_proto_tcp.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)
net/netfilter/nf_conntrack_proto_tcp.c +6 -15
index 8ec55cd72572..204a5cdff5b1 100644
@@ -556,24 +556,14 @@ static bool tcp_in_window(struct nf_conn *ct,
 			}
 
 		}
-	} else if (((state->state == TCP_CONNTRACK_SYN_SENT
-		     && dir == IP_CT_DIR_ORIGINAL)
-		   || (state->state == TCP_CONNTRACK_SYN_RECV
-		     && dir == IP_CT_DIR_REPLY))
-		   && after(end, sender->td_end)) {
+	} else if (tcph->syn &&
+		   after(end, sender->td_end) &&
+		   (state->state == TCP_CONNTRACK_SYN_SENT ||
+		    state->state == TCP_CONNTRACK_SYN_RECV)) {
 		/*
 		 * RFC 793: "if a TCP is reinitialized ... then it need
 		 * not wait at all; it must only be sure to use sequence
 		 * numbers larger than those recently used."
-		 */
-		sender->td_end =
-		sender->td_maxend = end;
-		sender->td_maxwin = (win == 0 ? 1 : win);
-
-		tcp_options(skb, dataoff, tcph, sender);
-	} else if (tcph->syn && dir == IP_CT_DIR_REPLY &&
-		   state->state == TCP_CONNTRACK_SYN_SENT) {
-		/* Retransmitted syn-ack, or syn (simultaneous open).
 		 *
 		 * Re-init state for this direction, just like for the first
 		 * syn(-ack) reply, it might differ in seq, ack or tcp options.
@@ -581,7 +571,8 @@ static bool tcp_in_window(struct nf_conn *ct,
 		tcp_init_sender(sender, receiver,
 				skb, dataoff, tcph,
 				end, win);
-		if (!tcph->ack)
+
+		if (dir == IP_CT_DIR_REPLY && !tcph->ack)
 			return true;
 	}
 
Signature · 1 line
2.35.1

-
E-mail : [email protected], [email protected] PGP key : https://wigner.hu/~kadlec/pgp_public_key.txt
Address : Wigner Research Centre for Physics

          H-1525 Budapest 114, POB. 49, Hungary
#2 lore raw
Quoted 38 lines · Jozsef Kadlecsik
Quoted 35 lines · Florian Westphal

Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell pinpointed to nf_conntrack tcp_in_window() bug.

tcp trace shows following sequence:

I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]> R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..] R > I Flags [P.], seq 1:89, ack 1, [..]

Note 3rd ACK is from responder to initiator so following branch is taken:

    } else if (((state->state == TCP_CONNTRACK_SYN_SENT
               && dir == IP_CT_DIR_ORIGINAL)
               || (state->state == TCP_CONNTRACK_SYN_RECV
               && dir == IP_CT_DIR_REPLY))
               && after(end, sender->td_end)) {

... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY. This causes the scaling factor to be reset to 0: window scale option is only present in syn(ack) packets. This in turn makes nf_conntrack mark valid packets as out-of-window.

This was always broken, it exists even in original commit where window tracking was added to ip_conntrack (nf_conntrack predecessor) in 2.6.9-rc1 kernel.

Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").

Upon closer look, those conditionals/branches can be merged:

Because earlier checks prevent syn-ack from showing up in original direction, the 'dir' checks in the conditional quoted above are redundant, remove them. Return early for pure syn retransmitted in reply direction (simultaneous open).

Quoted 9 lines
Quoted 4 lines
Fixes 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.")
Cc Jozsef Kadlecsik <[email protected]>

[Sorry, I was away whole last week as well.]

Thanks a lot Florian and Jozsef !

Quoted 60 lines

Best regards,
Jozsef

Quoted 48 lines
---
 net/netfilter/nf_conntrack_proto_tcp.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)
net/netfilter/nf_conntrack_proto_tcp.c +6 -16
index 8ec55cd72572..204a5cdff5b1 100644
@@ -556,24 +556,14 @@ static bool tcp_in_window(struct nf_conn *ct,
                      }
 
              }
-     } else if (((state->state == TCP_CONNTRACK_SYN_SENT
-                  && dir == IP_CT_DIR_ORIGINAL)
-                || (state->state == TCP_CONNTRACK_SYN_RECV
-                  && dir == IP_CT_DIR_REPLY))
-                && after(end, sender->td_end)) {
+     } else if (tcph->syn &&
+                after(end, sender->td_end) &&
+                (state->state == TCP_CONNTRACK_SYN_SENT ||
+                 state->state == TCP_CONNTRACK_SYN_RECV)) {
              /*
               * RFC 793: "if a TCP is reinitialized ... then it need
               * not wait at all; it must only be sure to use sequence
               * numbers larger than those recently used."
-              */
-             sender->td_end =
-             sender->td_maxend = end;
-             sender->td_maxwin = (win == 0 ? 1 : win);
-
-             tcp_options(skb, dataoff, tcph, sender);
-     } else if (tcph->syn && dir == IP_CT_DIR_REPLY &&
-                state->state == TCP_CONNTRACK_SYN_SENT) {
-             /* Retransmitted syn-ack, or syn (simultaneous open).
               *
               * Re-init state for this direction, just like for the first
               * syn(-ack) reply, it might differ in seq, ack or tcp options.
@@ -581,7 +571,8 @@ static bool tcp_in_window(struct nf_conn *ct,
              tcp_init_sender(sender, receiver,
                              skb, dataoff, tcph,
                              end, win);
-             if (!tcph->ack)
+
+             if (dir == IP_CT_DIR_REPLY && !tcph->ack)
                      return true;
      }
 
--
Quoted 3 lines

2.35.1

-
E-mail : [email protected], [email protected] PGP key : https://wigner.hu/~kadlec/pgp_public_key.txt
Address : Wigner Research Centre for Physics

          H-1525 Budapest 114, POB. 49, Hungary
maintainer Applied
#3 lore raw
In reply to Florian Westphal
Quoted 28 lines · Florian Westphal

Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell pinpointed to nf_conntrack tcp_in_window() bug.

tcp trace shows following sequence:

I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]> R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..] R > I Flags [P.], seq 1:89, ack 1, [..]

Note 3rd ACK is from responder to initiator so following branch is taken:

    } else if (((state->state == TCP_CONNTRACK_SYN_SENT
               && dir == IP_CT_DIR_ORIGINAL)
               || (state->state == TCP_CONNTRACK_SYN_RECV
               && dir == IP_CT_DIR_REPLY))
               && after(end, sender->td_end)) {

... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY. This causes the scaling factor to be reset to 0: window scale option is only present in syn(ack) packets. This in turn makes nf_conntrack mark valid packets as out-of-window.

This was always broken, it exists even in original commit where window tracking was added to ip_conntrack (nf_conntrack predecessor) in 2.6.9-rc1 kernel.

Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").

Quoted 6 lines

Upon closer look, those conditionals/branches can be merged:

Because earlier checks prevent syn-ack from showing up in original direction, the 'dir' checks in the conditional quoted above are redundant, remove them. Return early for pure syn retransmitted in reply direction (simultaneous open).

Applied, thanks

#4 lore raw

Thank you all.  Got entangled in other stuff again, just wanted to say I appreciate the effort that went into this.

Kind Regards,
Jaco

Quoted 36 lines · Pablo Neira Ayuso
Quoted 28 lines · Florian Westphal

Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell pinpointed to nf_conntrack tcp_in_window() bug.

tcp trace shows following sequence:

I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]> R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..] R > I Flags [P.], seq 1:89, ack 1, [..]

Note 3rd ACK is from responder to initiator so following branch is taken:

    } else if (((state->state == TCP_CONNTRACK_SYN_SENT
               && dir == IP_CT_DIR_ORIGINAL)
               || (state->state == TCP_CONNTRACK_SYN_RECV
               && dir == IP_CT_DIR_REPLY))
               && after(end, sender->td_end)) {

... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY. This causes the scaling factor to be reset to 0: window scale option is only present in syn(ack) packets. This in turn makes nf_conntrack mark valid packets as out-of-window.

This was always broken, it exists even in original commit where window tracking was added to ip_conntrack (nf_conntrack predecessor) in 2.6.9-rc1 kernel.

Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").

Quoted 6 lines

Upon closer look, those conditionals/branches can be merged:

Because earlier checks prevent syn-ack from showing up in original direction, the 'dir' checks in the conditional quoted above are redundant, remove them. Return early for pure syn retransmitted in reply direction (simultaneous open).

Applied, thanks