* kernel panic TPROXY , vanilla 4.7.1
@ 2016-08-17 14:31 Denys Fedoryshchenko
2016-08-17 15:42 ` Eric Dumazet
0 siblings, 1 reply; 7+ messages in thread
From: Denys Fedoryshchenko @ 2016-08-17 14:31 UTC (permalink / raw)
To: Linux Kernel Network Developers
Hi!
Tried to run squid on latest kernel, and hit a panic
Sometimes it just shows warning in dmesg (but doesnt work properly)
[ 75.701666] IPv4: Attempt to release TCP socket in state 10
ffff88102d430780
[ 83.866974] squid (2700) used greatest stack depth: 12912 bytes left
[ 87.506644] IPv4: Attempt to release TCP socket in state 10
ffff880078a48780
[ 114.704295] IPv4: Attempt to release TCP socket in state 10
ffff881029f8ad00
I cannot catch yet oops/panic message, netconsole not working.
After triggering warning message 3 times, i am unable to run squid
anymore (without reboot), and in netstat it doesnt show port running.
firewall is:
*mangle
-A PREROUTING -p tcp -m socket -j DIVERT
-A PREROUTING -p tcp -m tcp --dport 80 -i eno1 -j TPROXY --on-port 3129
--on-ip 0.0.0.0 --tproxy-mark 0x1/0x1
-A DIVERT -j MARK --set-xmark 0x1/0xffffffff
-A DIVERT -j ACCEPT
routing
ip rule add fwmark 1 lookup 100
ip route add local default dev eno1 table 100
squid config is default with tproxy option
http_port 3129 tproxy
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: kernel panic TPROXY , vanilla 4.7.1
2016-08-17 14:31 kernel panic TPROXY , vanilla 4.7.1 Denys Fedoryshchenko
@ 2016-08-17 15:42 ` Eric Dumazet
2016-08-17 16:04 ` Eric Dumazet
0 siblings, 1 reply; 7+ messages in thread
From: Eric Dumazet @ 2016-08-17 15:42 UTC (permalink / raw)
To: Denys Fedoryshchenko; +Cc: Linux Kernel Network Developers
On Wed, 2016-08-17 at 17:31 +0300, Denys Fedoryshchenko wrote:
> Hi!
>
> Tried to run squid on latest kernel, and hit a panic
> Sometimes it just shows warning in dmesg (but doesnt work properly)
> [ 75.701666] IPv4: Attempt to release TCP socket in state 10
> ffff88102d430780
> [ 83.866974] squid (2700) used greatest stack depth: 12912 bytes left
> [ 87.506644] IPv4: Attempt to release TCP socket in state 10
> ffff880078a48780
> [ 114.704295] IPv4: Attempt to release TCP socket in state 10
> ffff881029f8ad00
>
> I cannot catch yet oops/panic message, netconsole not working.
>
> After triggering warning message 3 times, i am unable to run squid
> anymore (without reboot), and in netstat it doesnt show port running.
>
> firewall is:
> *mangle
> -A PREROUTING -p tcp -m socket -j DIVERT
> -A PREROUTING -p tcp -m tcp --dport 80 -i eno1 -j TPROXY --on-port 3129
> --on-ip 0.0.0.0 --tproxy-mark 0x1/0x1
> -A DIVERT -j MARK --set-xmark 0x1/0xffffffff
> -A DIVERT -j ACCEPT
>
> routing
> ip rule add fwmark 1 lookup 100
> ip route add local default dev eno1 table 100
>
>
> squid config is default with tproxy option
> http_port 3129 tproxy
>
Hmppff... sorry for this, I will send a fix.
Thanks for the report !
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: kernel panic TPROXY , vanilla 4.7.1
2016-08-17 15:42 ` Eric Dumazet
@ 2016-08-17 16:04 ` Eric Dumazet
2016-08-17 16:44 ` Denys Fedoryshchenko
0 siblings, 1 reply; 7+ messages in thread
From: Eric Dumazet @ 2016-08-17 16:04 UTC (permalink / raw)
To: Denys Fedoryshchenko; +Cc: Linux Kernel Network Developers, netfilter-devel
On Wed, 2016-08-17 at 08:42 -0700, Eric Dumazet wrote:
> On Wed, 2016-08-17 at 17:31 +0300, Denys Fedoryshchenko wrote:
> > Hi!
> >
> > Tried to run squid on latest kernel, and hit a panic
> > Sometimes it just shows warning in dmesg (but doesnt work properly)
> > [ 75.701666] IPv4: Attempt to release TCP socket in state 10
> > ffff88102d430780
> > [ 83.866974] squid (2700) used greatest stack depth: 12912 bytes left
> > [ 87.506644] IPv4: Attempt to release TCP socket in state 10
> > ffff880078a48780
> > [ 114.704295] IPv4: Attempt to release TCP socket in state 10
> > ffff881029f8ad00
> >
> > I cannot catch yet oops/panic message, netconsole not working.
> >
> > After triggering warning message 3 times, i am unable to run squid
> > anymore (without reboot), and in netstat it doesnt show port running.
> >
> > firewall is:
> > *mangle
> > -A PREROUTING -p tcp -m socket -j DIVERT
> > -A PREROUTING -p tcp -m tcp --dport 80 -i eno1 -j TPROXY --on-port 3129
> > --on-ip 0.0.0.0 --tproxy-mark 0x1/0x1
> > -A DIVERT -j MARK --set-xmark 0x1/0xffffffff
> > -A DIVERT -j ACCEPT
> >
> > routing
> > ip rule add fwmark 1 lookup 100
> > ip route add local default dev eno1 table 100
> >
> >
> > squid config is default with tproxy option
> > http_port 3129 tproxy
> >
>
> Hmppff... sorry for this, I will send a fix.
>
> Thanks for the report !
>
Could you try the following ?
Thanks !
net/netfilter/xt_TPROXY.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c
index 7f4414d26a66..663c4c3c9072 100644
--- a/net/netfilter/xt_TPROXY.c
+++ b/net/netfilter/xt_TPROXY.c
@@ -127,6 +127,8 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp,
daddr, dport,
in->ifindex);
+ if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
+ sk = NULL;
/* NOTE: we return listeners even if bound to
* 0.0.0.0, those are filtered out in
* xt_socket, since xt_TPROXY needs 0 bound
@@ -195,6 +197,8 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp,
daddr, ntohs(dport),
in->ifindex);
+ if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
+ sk = NULL;
/* NOTE: we return listeners even if bound to
* 0.0.0.0, those are filtered out in
* xt_socket, since xt_TPROXY needs 0 bound
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: kernel panic TPROXY , vanilla 4.7.1
2016-08-17 16:04 ` Eric Dumazet
@ 2016-08-17 16:44 ` Denys Fedoryshchenko
2016-08-17 16:48 ` Eric Dumazet
2016-08-17 16:56 ` [PATCH net] netfilter: tproxy: properly refcount tcp listeners Eric Dumazet
0 siblings, 2 replies; 7+ messages in thread
From: Denys Fedoryshchenko @ 2016-08-17 16:44 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Linux Kernel Network Developers, netfilter-devel
On 2016-08-17 19:04, Eric Dumazet wrote:
> On Wed, 2016-08-17 at 08:42 -0700, Eric Dumazet wrote:
>> On Wed, 2016-08-17 at 17:31 +0300, Denys Fedoryshchenko wrote:
>> > Hi!
>> >
>> > Tried to run squid on latest kernel, and hit a panic
>> > Sometimes it just shows warning in dmesg (but doesnt work properly)
>> > [ 75.701666] IPv4: Attempt to release TCP socket in state 10
>> > ffff88102d430780
>> > [ 83.866974] squid (2700) used greatest stack depth: 12912 bytes left
>> > [ 87.506644] IPv4: Attempt to release TCP socket in state 10
>> > ffff880078a48780
>> > [ 114.704295] IPv4: Attempt to release TCP socket in state 10
>> > ffff881029f8ad00
>> >
>> > I cannot catch yet oops/panic message, netconsole not working.
>> >
>> > After triggering warning message 3 times, i am unable to run squid
>> > anymore (without reboot), and in netstat it doesnt show port running.
>> >
>> > firewall is:
>> > *mangle
>> > -A PREROUTING -p tcp -m socket -j DIVERT
>> > -A PREROUTING -p tcp -m tcp --dport 80 -i eno1 -j TPROXY --on-port 3129
>> > --on-ip 0.0.0.0 --tproxy-mark 0x1/0x1
>> > -A DIVERT -j MARK --set-xmark 0x1/0xffffffff
>> > -A DIVERT -j ACCEPT
>> >
>> > routing
>> > ip rule add fwmark 1 lookup 100
>> > ip route add local default dev eno1 table 100
>> >
>> >
>> > squid config is default with tproxy option
>> > http_port 3129 tproxy
>> >
>>
>> Hmppff... sorry for this, I will send a fix.
>>
>> Thanks for the report !
>>
>
>
> Could you try the following ?
>
> Thanks !
>
> net/netfilter/xt_TPROXY.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c
> index 7f4414d26a66..663c4c3c9072 100644
> --- a/net/netfilter/xt_TPROXY.c
> +++ b/net/netfilter/xt_TPROXY.c
> @@ -127,6 +127,8 @@ nf_tproxy_get_sock_v4(struct net *net, struct
> sk_buff *skb, void *hp,
> daddr, dport,
> in->ifindex);
>
> + if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
> + sk = NULL;
> /* NOTE: we return listeners even if bound to
> * 0.0.0.0, those are filtered out in
> * xt_socket, since xt_TPROXY needs 0 bound
> @@ -195,6 +197,8 @@ nf_tproxy_get_sock_v6(struct net *net, struct
> sk_buff *skb, int thoff, void *hp,
> daddr, ntohs(dport),
> in->ifindex);
>
> + if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
> + sk = NULL;
> /* NOTE: we return listeners even if bound to
> * 0.0.0.0, those are filtered out in
> * xt_socket, since xt_TPROXY needs 0 bound
Yes, everything fine after patch!
Thanks a lot
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: kernel panic TPROXY , vanilla 4.7.1
2016-08-17 16:44 ` Denys Fedoryshchenko
@ 2016-08-17 16:48 ` Eric Dumazet
2016-08-17 16:56 ` [PATCH net] netfilter: tproxy: properly refcount tcp listeners Eric Dumazet
1 sibling, 0 replies; 7+ messages in thread
From: Eric Dumazet @ 2016-08-17 16:48 UTC (permalink / raw)
To: Denys Fedoryshchenko; +Cc: Linux Kernel Network Developers, netfilter-devel
On Wed, 2016-08-17 at 19:44 +0300, Denys Fedoryshchenko wrote:
> Yes, everything fine after patch!
> Thanks a lot
Perfect, thanks for testing, I am sending the official patch.
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH net] netfilter: tproxy: properly refcount tcp listeners
2016-08-17 16:44 ` Denys Fedoryshchenko
2016-08-17 16:48 ` Eric Dumazet
@ 2016-08-17 16:56 ` Eric Dumazet
2016-08-17 22:51 ` Pablo Neira Ayuso
1 sibling, 1 reply; 7+ messages in thread
From: Eric Dumazet @ 2016-08-17 16:56 UTC (permalink / raw)
To: Denys Fedoryshchenko, Pablo Neira Ayuso, David Miller
Cc: Linux Kernel Network Developers, netfilter-devel
From: Eric Dumazet <edumazet@google.com>
inet_lookup_listener() and inet6_lookup_listener() no longer
take a reference on the found listener.
This minimal patch adds back the refcounting, but we might do
this differently in net-next later.
Fixes: 3b24d854cb35 ("tcp/dccp: do not touch listener sk_refcnt under synflood")
Reported-and-tested-by: Denys Fedoryshchenko <nuclearcat@nuclearcat.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
Note: bug added in 4.7, stable candidate.
net/netfilter/xt_TPROXY.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c
index 7f4414d26a66..663c4c3c9072 100644
--- a/net/netfilter/xt_TPROXY.c
+++ b/net/netfilter/xt_TPROXY.c
@@ -127,6 +127,8 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp,
daddr, dport,
in->ifindex);
+ if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
+ sk = NULL;
/* NOTE: we return listeners even if bound to
* 0.0.0.0, those are filtered out in
* xt_socket, since xt_TPROXY needs 0 bound
@@ -195,6 +197,8 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp,
daddr, ntohs(dport),
in->ifindex);
+ if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
+ sk = NULL;
/* NOTE: we return listeners even if bound to
* 0.0.0.0, those are filtered out in
* xt_socket, since xt_TPROXY needs 0 bound
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH net] netfilter: tproxy: properly refcount tcp listeners
2016-08-17 16:56 ` [PATCH net] netfilter: tproxy: properly refcount tcp listeners Eric Dumazet
@ 2016-08-17 22:51 ` Pablo Neira Ayuso
0 siblings, 0 replies; 7+ messages in thread
From: Pablo Neira Ayuso @ 2016-08-17 22:51 UTC (permalink / raw)
To: Eric Dumazet
Cc: Denys Fedoryshchenko, David Miller,
Linux Kernel Network Developers, netfilter-devel
On Wed, Aug 17, 2016 at 09:56:46AM -0700, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> inet_lookup_listener() and inet6_lookup_listener() no longer
> take a reference on the found listener.
>
> This minimal patch adds back the refcounting, but we might do
> this differently in net-next later.
Applied, thanks Eric!
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2016-08-17 22:51 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-08-17 14:31 kernel panic TPROXY , vanilla 4.7.1 Denys Fedoryshchenko
2016-08-17 15:42 ` Eric Dumazet
2016-08-17 16:04 ` Eric Dumazet
2016-08-17 16:44 ` Denys Fedoryshchenko
2016-08-17 16:48 ` Eric Dumazet
2016-08-17 16:56 ` [PATCH net] netfilter: tproxy: properly refcount tcp listeners Eric Dumazet
2016-08-17 22:51 ` Pablo Neira Ayuso
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).