openvswitch: support asymmetric conntrack
Subsystems (4)
Aaron Conole
maintainerPATCH 1/2 openvswitch: support asymmetric conntrack
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
--- net/openvswitch/conntrack.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
net/openvswitch/conntrack.c +11 -0
@@ -903,6 +903,17 @@ static int ovs_ct_nat(struct net *net, struct sw_flow_key *key, } err =ovs_ct_nat_execute (skb, ct, ctinfo, &info->range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ovs_ct_nat_execute (skb, ct, ctinfo, &info->range, + maniptype); + } +/* Mark NAT done if successful and update the flow key. */ if (err ==NF_ACCEPT )ovs_nat_update_key (key, skb, maniptype);
Signature · 1 line
2.21.0
Aaron Conole
maintainerPATCH 2/2 act_ct: support asymmetric conntrack
The act_ct TC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The act_ct action doesn't have such capability.
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)
net/sched/act_ct.c +12 -1
@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))
@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb, return NF_ACCEPT; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + } +return err;#else return NF_ACCEPT ;#endif
Signature · 1 line
2.21.0
Re: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 38 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +5
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC )
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 376
Quoted 10 lines
net/sched/act_ct.c +7
-0+ maniptype =NF_NAT_MANIP_DST ; + else + maniptype =NF_NAT_MANIP_SRC ; + + err = ct_nat_execute(skb, ct, ctinfo, range, maniptype); + } +return err;#else return NF_ACCEPT ;#endif
+paul
Re: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 43 lines · Roi Dayan
Quoted 41 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +9
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; +
Quoted 8 lines
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 376
Quoted 7 lines
net/sched/act_ct.c +3
-0+ err = ct_nat_execute(skb, ct, ctinfo, range, maniptype); + } +return err;#else return NF_ACCEPT ;#endif +paul
Hi Aaron,
I think I understand the issue and this looks good,
Can you describe the scenario to reproduce this?
Thanks,
Paul.
Aaron Conole
maintainerRe: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 53 lines · Paul Blakey
Quoted 43 lines · Roi Dayan
Quoted 41 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +9
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; +Quoted 8 lines
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 376
Quoted 7 lines
net/sched/act_ct.c +3
-0+ err = ct_nat_execute(skb, ct, ctinfo, range, maniptype); + } +return err;#else return NF_ACCEPT ;#endif +paul
Quoted 5 lines
Hi Aaron,
I think I understand the issue and this looks good,
Can you describe the scenario to reproduce this?
It reproduces with OpenShift 3.10, which makes forward direction packets between namespaces pump through a tun device that applies NAT rules to rewrite the dest. Limit the namespace number of ephemeral sockets using by editing net.ipv4.ip_local_port_range in the client namespace, and connect to the server namespace. That's the mechanism for OvS. But for TC I guess there wouldn't be anything convenient avaiable.
I'll try to script up something that doesn't use openshift.
Quoted 4 lines
Thanks,
Paul.
Re: [PATCH net 2/2] act_ct: support asymmetric conntrack
In reply to Aaron ConoleQuoted 33 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +1
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype);
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 372
Quoted 10 lines
net/sched/act_ct.c +10
-0+ err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + }
I keep thinking about this and I'm not entirely convinced that this shouldn't be simpler. More like:
if (DNAT)
DNAT
if (SNAT)
SNAT
So it always does DNAT before SNAT, similarly to what iptables would do on PRE/POSTROUTING chains.
Quoted 7 lines
net/sched/act_ct.c +1
-0+return err;#else return NF_ACCEPT ;#endif Signature · 1 line
2.21.0
Aaron Conole
maintainerRe: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 48 lines · Marcelo Ricardo Leitner
Quoted 33 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +1
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype);net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 372
Quoted 10 lines
net/sched/act_ct.c +10
-0+ err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + }I keep thinking about this and I'm not entirely convinced that this shouldn't be simpler. More like:
Quoted 7 lines
if (DNAT)
DNAT
if (SNAT)
SNATSo it always does DNAT before SNAT, similarly to what iptables would do on PRE/POSTROUTING chains.
I can rewrite the whole function, but I wanted to start with the smaller fix that worked. I also think it needs more testing then (since it's something of a rewrite of the function).
I guess it's not too important - do you think it gives any readability to do it this way? If so, I can respin the patch changing it like you describe.
Quoted 7 lines
Quoted 7 lines
net/sched/act_ct.c +1
-0+return err;#else return NF_ACCEPT ;#endif Signature · 1 line
2.21.0
Re: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 58 lines · Aaron Conole
Quoted 48 lines · Marcelo Ricardo Leitner
Quoted 33 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +1
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype);net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 372
Quoted 10 lines
net/sched/act_ct.c +10
-0+ err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + }I keep thinking about this and I'm not entirely convinced that this shouldn't be simpler. More like:
Quoted 7 lines
if (DNAT)
DNAT
if (SNAT)
SNATSo it always does DNAT before SNAT, similarly to what iptables would do on PRE/POSTROUTING chains.
Quoted 7 lines
I can rewrite the whole function, but I wanted to start with the smaller fix that worked. I also think it needs more testing then (since it's something of a rewrite of the function).
I guess it's not too important - do you think it gives any readability to do it this way? If so, I can respin the patch changing it like you describe.
I didn't mean a rewrite, but just to never handle SNAT before DNAT. So the fix here would be like:
- return ct_nat_execute(skb, ct, ctinfo, range, maniptype);
+ err = ct_nat_execute(skb, ct, ctinfo, range, maniptype);
+ if (err == NF_ACCEPT && maniptype == NF_NAT_MANIP_DST &&
+ ct->status & IPS_SRC_NAT && ct->status & IPS_DST_NAT) {
+ maniptype = NF_NAT_MANIP_SRC;
+ err = ct_nat_execute(skb, ct, ctinfo, range, maniptype);
+ }
+ return err;
Quoted 8 lines
Quoted 7 lines
Quoted 7 lines
net/sched/act_ct.c +1
-0+return err;#else return NF_ACCEPT ;#endif Signature · 1 line
2.21.0
Aaron Conole
maintainerRe: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 70 lines · Marcelo Ricardo Leitner
Quoted 58 lines · Aaron Conole
Quoted 48 lines · Marcelo Ricardo Leitner
Quoted 33 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +1
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype);net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 372
Quoted 10 lines
net/sched/act_ct.c +10
-0+ err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + }I keep thinking about this and I'm not entirely convinced that this shouldn't be simpler. More like:
Quoted 7 lines
if (DNAT)
DNAT
if (SNAT)
SNATSo it always does DNAT before SNAT, similarly to what iptables would do on PRE/POSTROUTING chains.
Quoted 7 lines
I can rewrite the whole function, but I wanted to start with the smaller fix that worked. I also think it needs more testing then (since it's something of a rewrite of the function).
I guess it's not too important - do you think it gives any readability to do it this way? If so, I can respin the patch changing it like you describe.
I didn't mean a rewrite, but just to never handle SNAT before DNAT. So the fix here would be like:
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 373
Quoted 8 lines
net/sched/act_ct.c +7
-1-return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && maniptype ==NF_NAT_MANIP_DST && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { + maniptype =NF_NAT_MANIP_SRC ; + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + } +return err;
But the maniptype of the first call could be NAT_MANIP_SRC. In fact, that's what I see if the packet is reply direction && !related.
So, we need the block to invert the manipulation type. Otherwise, we miss the DNAT manipulation.
So I don't think I can use that block.
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 7 lines
net/sched/act_ct.c +1
-0+return err;#else return NF_ACCEPT ;#endif Signature · 1 line
2.21.0
Marcelo Ricardo Leitner
Acked-byRe: [PATCH net 2/2] act_ct: support asymmetric conntrack
Quoted 72 lines · Aaron Conole
Quoted 70 lines · Marcelo Ricardo Leitner
Quoted 58 lines · Aaron Conole
Quoted 48 lines · Marcelo Ricardo Leitner
Quoted 33 lines · Aaron Conole
The
act_ctTC module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. Theact_ctaction doesn't have such capability.Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes b57dc7c13ea9 ("net/sched: Introduce action ct")Signed-off-by Aaron Conole <[email protected]>--- net/sched/act_ct.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)net/sched/act_ct.c +1
-1@@ -329,6 +329,7 @@ static int tcf_ct_act_nat(struct sk_buff *skb, bool commit) {#if IS_ENABLED (CONFIG_NF_NAT ) +int err;enum nf_nat_manip_type maniptype;if (!(ct_action &TCA_CT_ACT_NAT ))@@ -359,7 +360,17 @@ static int tcf_ct_act_nat(struct sk_buff *skb,return NF_ACCEPT ; } -return ct_nat_execute (skb, ct, ctinfo, range, maniptype);net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 372
Quoted 10 lines
net/sched/act_ct.c +10
-0+ err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { +if (maniptype ==NF_NAT_MANIP_SRC ) + maniptype =NF_NAT_MANIP_DST ; +else + maniptype =NF_NAT_MANIP_SRC ; + + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + }I keep thinking about this and I'm not entirely convinced that this shouldn't be simpler. More like:
Quoted 7 lines
if (DNAT)
DNAT
if (SNAT)
SNATSo it always does DNAT before SNAT, similarly to what iptables would do on PRE/POSTROUTING chains.
Quoted 7 lines
I can rewrite the whole function, but I wanted to start with the smaller fix that worked. I also think it needs more testing then (since it's something of a rewrite of the function).
I guess it's not too important - do you think it gives any readability to do it this way? If so, I can respin the patch changing it like you describe.
I didn't mean a rewrite, but just to never handle SNAT before DNAT. So the fix here would be like:
Quoted 11 lines
net/sched/act_ct.c static int tcf_ct_act_nat(struct sk_buff *skb, · line 373
Quoted 8 lines
net/sched/act_ct.c +7
-1-return ct_nat_execute (skb, ct, ctinfo, range, maniptype); + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); +if (err ==NF_ACCEPT && maniptype ==NF_NAT_MANIP_DST && + ct->status &IPS_SRC_NAT && ct->status &IPS_DST_NAT ) { + maniptype =NF_NAT_MANIP_SRC ; + err =ct_nat_execute (skb, ct, ctinfo, range, maniptype); + } +return err;But the maniptype of the first call could be
NAT_MANIP_SRC. In fact, that's what I see if the packet is reply direction && !related.
Interesting, ok.
Quoted 5 lines
So, we need the block to invert the manipulation type. Otherwise, we miss the DNAT manipulation.
So I don't think I can use that block.
Thanks for digging on it.
Quoted 10 lines
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
Quoted 8 lines
net/sched/act_ct.c +1
-0+return err;#else return NF_ACCEPT ;#endif Signature · 1 line
2.21.0
Quoted 7 lines · Aaron Conole
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Quoted 5 lines
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>
The patch looks ok. But I am not able apply it. can you fix the encoding.
Aaron Conole
maintainerQuoted 8 lines · Pravin Shelar
Quoted 7 lines · Aaron Conole
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Quoted 7 lines
Quoted 5 lines
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>The patch looks ok. But I am not able apply it. can you fix the encoding.
Hrrm. I didn't make any special changes (just used git send-email). I will look at spinning a second patch.
Aaron Conole
maintainerQuoted 13 lines · Aaron Conole
Quoted 11 lines · Pravin Shelar
Quoted 10 lines · Aaron Conole
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>The patch looks ok. But I am not able apply it. can you fix the encoding.
Hrrm. I didn't make any special changes (just used git send-email). I will look at spinning a second patch.
Pravin,
I tried the following:
10:36:59 aconole@dhcp-25 {(312434617cb1...)} ~/git/linux$ curl http://patchwork.ozlabs.org/patch/1192219/mbox/ > test.patch
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed 100 4827 100 4827 0 0 8824 0 --:--:-- --:--:-- --:--:-- 8808
10:37:21 aconole@dhcp-25 {(312434617cb1...)} ~/git/linux$ git am test.patch
Applying: openvswitch: support asymmetric conntrack
10:37:24 aconole@dhcp-25 {(f759cc2b7323...)} ~/git/linux$
Can you check your mailer settings? The patchwork mbox worked fine, and I was able to apply from my own mbox as well.
-Aaron
Downloading from patchwork is working for me. Its strange other patches in my mailbox does not has this issue.
Thanks.
Quoted 42 lines · Aaron Conole
Quoted 13 lines · Aaron Conole
Quoted 11 lines · Pravin Shelar
Quoted 10 lines · Aaron Conole
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>The patch looks ok. But I am not able apply it. can you fix the encoding.
Hrrm. I didn't make any special changes (just used git send-email). I will look at spinning a second patch.
Pravin,
I tried the following:
10:36:59 aconole@dhcp-25 {(
312434617cb1...)} ~/git/linux$ curl http://patchwork.ozlabs.org/patch/1192219/mbox/ > test.patch% Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed100 4827 100 4827 0 0 8824 0 --:--:-- --:--:-- --:--:-- 8808
10:37:21 aconole@dhcp-25 {(312434617cb1...)} ~/git/linux$ git am test.patch
Applying: openvswitch: support asymmetric conntrack
10:37:24 aconole@dhcp-25 {(f759cc2b7323...)} ~/git/linux$Can you check your mailer settings? The patchwork mbox worked fine, and I was able to apply from my own mbox as well.
-Aaron
Le 08/11/2019 à 22:07, Aaron Conole a écrit :
Quoted 11 lines
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>
In this case, ovs_ct_find_existing() won't be able to find the conntrack, right? Inverting the tuple to find the conntrack doesn't work anymore with double NAT. Am I wrong?
Regards,
Nicolas
Aaron Conole
maintainerQuoted 7 lines · Nicolas Dichtel
Le 08/11/2019 à 22:07, Aaron Conole a écrit :
Quoted 6 lines
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Quoted 7 lines
Quoted 5 lines
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>In this case,
ovs_ct_find_existing()won't be able to find the conntrack, right?
vswitchd normally won't allow both actions to get programmed. Even the kernel module won't allow it, so this really will only happen when the connection gets established via the nf_hook path, and then needs to be processed via openvswitch. In those cases, the tuple lookup should be correct, because the nf_nat table should contain the correct tuple data, and the skbuff should have the correct tuples in the packet data to begin with.
Quoted 2 lines
Inverting the tuple to find the conntrack doesn't work anymore with double NAT. Am I wrong?
I think since the packet was double-NAT on the way out (via nf_hook path), then the incoming reply will have the correct NAT tuples and the lookup will happen just fine. Just that during processing, both transformations aren't applied.
Makes sense?
Quoted 2 lines
Regards,
Nicolas
Le 18/11/2019 à 22:19, Aaron Conole a écrit :
Quoted 25 lines
Quoted 7 lines · Nicolas Dichtel
Le 08/11/2019 à 22:07, Aaron Conole a écrit :
Quoted 6 lines
The openvswitch module shares a common conntrack and NAT infrastructure exposed via netfilter. It's possible that a packet needs both SNAT and DNAT manipulation, due to e.g. tuple collision. Netfilter can support this because it runs through the NAT table twice - once on ingress and again after egress. The openvswitch module doesn't have such capability.
Quoted 7 lines
Quoted 5 lines
Like netfilter hook infrastructure, we should run through NAT twice to keep the symmetry.
Fixes 05752523e565 ("openvswitch: Interface with NAT.")Signed-off-by Aaron Conole <[email protected]>In this case,
ovs_ct_find_existing()won't be able to find the conntrack, right?vswitchd normally won't allow both actions to get programmed. Even the kernel module won't allow it, so this really will only happen when the connection gets established via the
nf_hookpath, and then needs to be processed via openvswitch. In those cases, the tuple lookup should be correct, because thenf_nattable should contain the correct tuple data, and the skbuff should have the correct tuples in the packet data to begin with.
Quoted 7 lines
Quoted 2 lines
Inverting the tuple to find the conntrack doesn't work anymore with double NAT. Am I wrong?
I think since the packet was double-NAT on the way out (via
nf_hookpath), then the incoming reply will have the correct NAT tuples and the lookup will happen just fine. Just that during processing, both transformations aren't applied.
Ok, I didn't look deeply, thank you for the explanation.
Regards,
Nicolas