Skip to contentSkip to conversation

Patch "netfilter: core: only allow one nat hook per hook point" has been added to the 4.15-stable tree

PATCH NAK [email protected] 2 participants 3 messages Apr 9, 2018

Subsystems (3)
lore
Share & tools
lore

[email protected]

author maintainer Applied
#0 lore raw

This is a note to let you know that I've just added the patch titled

netfilter: core: only allow one nat hook per hook point

to the 4.15-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
netfilter-core-only-allow-one-nat-hook-per-hook-point.patch
and it can be found in the queue-4.15 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

Quoted 1 line

From foo@baz Mon Apr 9 10:16:32 CEST 2018

From: Florian Westphal <[email protected]>
Date: Fri, 8 Dec 2017 17:01:54 +0100
Subject: netfilter: core: only allow one nat hook per hook point

From: Florian Westphal <[email protected]>

[ Upstream commit f92b40a8b2645af38bd6814651c59c1e690db53d ]

The netfilter NAT core cannot deal with more than one NAT hook per hook location (prerouting, input ...), because the NAT hooks install a NAT null binding in case the iptables nat table (iptable_nat hooks) or the corresponding nftables chain (nft nat hooks) doesn't specify a nat transformation.

Null bindings are needed to detect port collsisions between NAT-ed and non-NAT-ed connections.

This causes nftables NAT rules to not work when iptable_nat module is loaded, and vice versa because nat binding has already been attached when the second nat hook is consulted.

The netfilter core is not really the correct location to handle this (hooks are just hooks, the core has no notion of what kinds of side
effects a hook implements), but its the only place where we can check
for conflicts between both iptables hooks and nftables hooks without adding dependencies.

So add nat annotation to hook_ops to describe those hooks that will add NAT bindings and then make core reject if such a hook already exists. The annotation fills a padding hole, in case further restrictions appar we might change this to a 'u8 type' instead of bool.

iptables error if nft nat hook active:
iptables -t nat -A POSTROUTING -j MASQUERADE
iptables v1.4.21: can't initialize iptables table `nat': File exists Perhaps iptables or your kernel needs to be upgraded.

nftables error if iptables nat table present:
nft -f /etc/nftables/ipv4-nat
/usr/etc/nftables/ipv4-nat:3:1-2: Error: Could not process rule: File exists table nat {
^^

---
 include/linux/netfilter.h         |    1 +
 net/ipv4/netfilter/iptable_nat.c  |    4 ++++
 net/ipv6/netfilter/ip6table_nat.c |    4 ++++
 net/netfilter/core.c              |    6 ++++++
 net/netfilter/nf_tables_api.c     |    2 ++
 5 files changed, 17 insertions(+)
include/linux/netfilter.h +1 -0
@@ -67,6 +67,7 @@ struct nf_hook_ops {
 	struct net_device	*dev;
 	void			*priv;
 	u_int8_t		pf;
+	bool			nat_hook;
 	unsigned int		hooknum;
 	/* Hooks are ordered in ascending priority. */
 	int			priority;
net/ipv4/netfilter/iptable_nat.c +4 -0
@@ -72,6 +72,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= iptable_nat_ipv4_in,
 		.pf		= NFPROTO_IPV4,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP_PRI_NAT_DST,
 	},
@@ -79,6 +80,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= iptable_nat_ipv4_out,
 		.pf		= NFPROTO_IPV4,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
@@ -86,6 +88,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= iptable_nat_ipv4_local_fn,
 		.pf		= NFPROTO_IPV4,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP_PRI_NAT_DST,
 	},
@@ -93,6 +96,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= iptable_nat_ipv4_fn,
 		.pf		= NFPROTO_IPV4,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP_PRI_NAT_SRC,
 	},
net/ipv6/netfilter/ip6table_nat.c +4 -0
@@ -74,6 +74,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= ip6table_nat_in,
 		.pf		= NFPROTO_IPV6,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_PRE_ROUTING,
 		.priority	= NF_IP6_PRI_NAT_DST,
 	},
@@ -81,6 +82,7 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= ip6table_nat_out,
 		.pf		= NFPROTO_IPV6,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_POST_ROUTING,
 		.priority	= NF_IP6_PRI_NAT_SRC,
 	},
@@ -88,12 +90,14 @@ static const struct nf_hook_ops nf_nat_i
 	{
 		.hook		= ip6table_nat_local_fn,
 		.pf		= NFPROTO_IPV6,
+		.nat_hook	= true,
 		.hooknum	= NF_INET_LOCAL_OUT,
 		.priority	= NF_IP6_PRI_NAT_DST,
 	},
 	/* After packet filtering, change source */
 	{
 		.hook		= ip6table_nat_fn,
+		.nat_hook	= true,
 		.pf		= NFPROTO_IPV6,
 		.hooknum	= NF_INET_LOCAL_IN,
 		.priority	= NF_IP6_PRI_NAT_SRC,
net/netfilter/core.c +6 -0
@@ -135,6 +135,12 @@ nf_hook_entries_grow(const struct nf_hoo
 			++i;
 			continue;
 		}
+
+		if (reg->nat_hook && orig_ops[i]->nat_hook) {
+			kvfree(new);
+			return ERR_PTR(-EEXIST);
+		}
+
 		if (inserted || reg->priority > orig_ops[i]->priority) {
 			new_ops[nhooks] = (void *)orig_ops[i];
 			new->hooks[nhooks] = old->hooks[i];
net/netfilter/nf_tables_api.c +2 -0
@@ -1400,6 +1400,8 @@ static int nf_tables_addchain(struct nft
 				ops->hook = hookfn;
 			if (afi->hook_ops_init)
 				afi->hook_ops_init(ops, i);
+			if (basechain->type->type == NFT_CHAIN_T_NAT)
+				ops->nat_hook = true;
 		}
 
 		chain->flags |= NFT_BASE_CHAIN;

Patches currently in stable-queue which might be from [email protected] are

queue-4.15/netfilter-core-only-allow-one-nat-hook-per-hook-point.patch

maintainer NAK
#1 lore raw
Quoted 16 lines · [email protected] <[email protected]> wrote:

This is a note to let you know that I've just added the patch titled

netfilter: core: only allow one nat hook per hook point

to the 4.15-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
netfilter-core-only-allow-one-nat-hook-per-hook-point.patch
and it can be found in the queue-4.15 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

Quoted 6 lines

From foo@baz Mon Apr 9 10:16:32 CEST 2018
From: Florian Westphal <[email protected]>
Date: Fri, 8 Dec 2017 17:01:54 +0100
Subject: netfilter: core: only allow one nat hook per hook point

From: Florian Westphal <[email protected]>

NAK.

This also needs
netfilter: nf_tables: permit second nat hook if colliding hook is going away

I would not add these to stable.

author maintainer
#2 lore raw
Quoted 22 lines · Florian Westphal
Quoted 21 lines · [email protected] <[email protected]> wrote:

This is a note to let you know that I've just added the patch titled

netfilter: core: only allow one nat hook per hook point

to the 4.15-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
netfilter-core-only-allow-one-nat-hook-per-hook-point.patch
and it can be found in the queue-4.15 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From foo@baz Mon Apr 9 10:16:32 CEST 2018
From: Florian Westphal <[email protected]>
Date: Fri, 8 Dec 2017 17:01:54 +0100
Subject: netfilter: core: only allow one nat hook per hook point

Quoted 8 lines
Quoted 1 line

From: Florian Westphal <[email protected]>

NAK.

This also needs
netfilter: nf_tables: permit second nat hook if colliding hook is going away

I would not add these to stable.

Thank you for letting me know, now dropped from my tree.

greg k-h