2023-02-01 19:23:20

by Alok Tiagi

[permalink] [raw]
Subject: [RFC] net: add new socket option SO_SETNETNS

This socket option provides a mechanism for users to switch a sockets network
namespace. This enables use cases where multiple IPv6 only network namespaces
can use a single IPv4 network namespace for IPv4 only egress connectivity by
switching their sockets from IPv6 to IPv4 network namespace. This allows for
migration of systems to IPv6 only while keeping their connectivity to IPv4 only
destinations intact.

Today, we achieve this by setting up seccomp filter to intercept network system
calls like connect() from a container in a container manager which runs in an
IPv4 only network namespace. The container manager creates a new IPv4 connection
and injects the new file descriptor through SECCOMP_NOTIFY_IOCTL_ADDFD replacing
the original file descriptor from the connect() call. This does not work for
cases where the original file descriptor is handed off to a system like epoll
before the connect() call. After a new file descriptor is injected the original
file descriptor being referenced by the epoll fd is not longer valid leading to
failures. As a workaround the container manager when intercepting connect()
loops through all open socket file descriptors to check if they are referencing
the socket attempting the connect() and replace the reference with the to be
injected file descriptor. This workaround is cumbersome and makes the solution
prone to similar yet to be discovered issues.

With SO_SETNETNS, the container manager can simply switch the original
unconnected socket’s network namespace to the IPv4 only network namespace
without the need for injecting any new socket. The container can then proceed
with the connect() call and establish connectivity to the IPv4 only destination.

This socket option is only allowed for sockets that have never been connected
since connected or recently disconnected sockets maybe bound to their network
namespaces network device and switching their namespace may lead to undefined
behavior.

Signed-off-by: aloktiagi <[email protected]>
---
include/uapi/asm-generic/socket.h | 2 +
net/core/sock.c | 46 +++++
tools/testing/selftests/net/Makefile | 1 +
tools/testing/selftests/net/so_set_netns.c | 208 +++++++++++++++++++++
4 files changed, 257 insertions(+)
create mode 100644 tools/testing/selftests/net/so_set_netns.c

diff --git a/include/uapi/asm-generic/socket.h b/include/uapi/asm-generic/socket.h
index 638230899e98..dc9498233fe5 100644
--- a/include/uapi/asm-generic/socket.h
+++ b/include/uapi/asm-generic/socket.h
@@ -132,6 +132,8 @@

#define SO_RCVMARK 75

+#define SO_SETNETNS 76
+
#if !defined(__KERNEL__)

#if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))
diff --git a/net/core/sock.c b/net/core/sock.c
index f954d5893e79..34cb72b211a6 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
WRITE_ONCE(sk->sk_txrehash, (u8)val);
break;

+ case SO_SETNETNS:
+ {
+ struct net *other_ns, *my_ns;
+
+ if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
+ ret = -EOPNOTSUPP;
+ break;
+ }
+
+ if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
+ ret = -EOPNOTSUPP;
+ break;
+ }
+
+ other_ns = get_net_ns_by_fd(val);
+ if (IS_ERR(other_ns)) {
+ ret = PTR_ERR(other_ns);
+ break;
+ }
+
+ if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
+ ret = -EPERM;
+ goto out_err;
+ }
+
+ /* check that the socket has never been connected or recently disconnected */
+ if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
+ ret = -EOPNOTSUPP;
+ goto out_err;
+ }
+
+ /* check that the socket is not bound to an interface*/
+ if (sk->sk_bound_dev_if != 0) {
+ ret = -EOPNOTSUPP;
+ goto out_err;
+ }
+
+ my_ns = sock_net(sk);
+ sock_net_set(sk, other_ns);
+ put_net(my_ns);
+ break;
+out_err:
+ put_net(other_ns);
+ break;
+ }
+
default:
ret = -ENOPROTOOPT;
break;
diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 3007e98a6d64..c2e7679e31bb 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -75,6 +75,7 @@ TEST_GEN_PROGS += so_incoming_cpu
TEST_PROGS += sctp_vrf.sh
TEST_GEN_FILES += sctp_hello
TEST_GEN_FILES += csum
+TEST_GEN_PROGS += so_set_netns

TEST_FILES := settings

diff --git a/tools/testing/selftests/net/so_set_netns.c b/tools/testing/selftests/net/so_set_netns.c
new file mode 100644
index 000000000000..cc7767d23a5d
--- /dev/null
+++ b/tools/testing/selftests/net/so_set_netns.c
@@ -0,0 +1,208 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <error.h>
+#include <fcntl.h>
+#include <sched.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+
+#include <linux/tcp.h>
+#include <linux/socket.h>
+
+#include <sys/types.h>
+#include <sys/sendfile.h>
+#include <sys/socket.h>
+#include <sys/stat.h>
+
+#include "../kselftest_harness.h"
+
+#ifndef SO_SETNETNS
+#define SO_SETNETNS 76
+#endif
+
+static int unshare_open(void)
+{
+ const char *netns_path = "/proc/self/ns/net";
+ int fd, ret;
+
+ if (unshare(CLONE_NEWNET) != 0)
+ return -1;
+
+ fd = open(netns_path, O_RDONLY);
+ if (fd <= 0)
+ return -1;
+
+ ret = system("ip link set lo up");
+ if (ret < 0)
+ return -1;
+
+ return fd;
+}
+
+static int switch_ns(int fd)
+{
+ if (setns(fd, CLONE_NEWNET))
+ return -1;
+ return 0;
+}
+
+static void init_namespaces(struct __test_metadata *_metadata,
+ int *netns_client, int *netns_server)
+{
+ *netns_client = unshare_open();
+ ASSERT_GE(*netns_client, 0);
+
+ *netns_server = unshare_open();
+ ASSERT_GE(*netns_server, 0);
+}
+
+static void setup_network(struct __test_metadata *_metadata,
+ int *netns_client, int *netns_server)
+{
+ int ret;
+
+ ret = switch_ns(*netns_client);
+ ASSERT_EQ(ret, 0);
+
+ ret = system("ip addr add fd::1/64 dev lo");
+ ASSERT_EQ(ret, 0);
+
+ ret = switch_ns(*netns_server);
+ ASSERT_EQ(ret, 0);
+
+ ret = system("ip addr add 192.168.1.1/24 dev lo");
+ ASSERT_EQ(ret, 0);
+}
+
+static void setup_client_server(struct __test_metadata *_metadata,
+ int *netns_client, int *netns_server,
+ int *client_fd, int *server_fd)
+{
+ struct sockaddr_in addr;
+ int ret;
+
+ ret = switch_ns(*netns_client);
+ ASSERT_EQ(ret, 0);
+
+ *client_fd = socket(AF_INET, SOCK_STREAM, 0);
+
+ ret = switch_ns(*netns_server);
+ ASSERT_EQ(ret, 0);
+
+ addr.sin_family = AF_INET;
+ addr.sin_addr.s_addr = inet_addr("192.168.1.1");
+ addr.sin_port = htons(80);
+
+ *server_fd = socket(AF_INET, SOCK_STREAM, 0);
+ ret = bind(*server_fd, &addr, sizeof(addr));
+ ASSERT_EQ(ret, 0);
+ ret = listen(*server_fd, 10);
+ ASSERT_EQ(ret, 0);
+}
+
+FIXTURE(so_set_netns)
+{
+ int netns_client, netns_server;
+ int client_fd, server_fd;
+};
+
+FIXTURE_SETUP(so_set_netns)
+{
+ init_namespaces(_metadata, &self->netns_client, &self->netns_server);
+ setup_network(_metadata, &self->netns_client, &self->netns_server);
+ setup_client_server(_metadata,
+ &self->netns_client, &self->netns_server,
+ &self->client_fd, &self->server_fd);
+}
+
+FIXTURE_TEARDOWN(so_set_netns)
+{
+ close(self->client_fd);
+ close(self->server_fd);
+ close(self->netns_client);
+ close(self->netns_server);
+}
+
+TEST_F(so_set_netns, test_socket_ns_switch_unconnected) {
+ struct sockaddr_in addr;
+ int ret;
+
+ addr.sin_family = AF_INET;
+ addr.sin_addr.s_addr = inet_addr("192.168.1.1");
+ addr.sin_port = htons(80);
+
+ ret = switch_ns(self->netns_client);
+ ASSERT_EQ(ret, 0);
+
+ ret = setsockopt(self->client_fd,
+ SOL_SOCKET, SO_SETNETNS,
+ &self->netns_server,
+ sizeof(self->netns_server));
+ ASSERT_EQ(ret, 0);
+
+ ret = connect(self->client_fd, &addr, sizeof(addr));
+ ASSERT_EQ(ret, 0);
+}
+
+TEST_F(so_set_netns, test_socket_ns_switch_connected) {
+ struct sockaddr_in addr;
+ int ret;
+
+ addr.sin_family = AF_INET;
+ addr.sin_addr.s_addr = inet_addr("192.168.1.1");
+ addr.sin_port = htons(80);
+
+ ret = setsockopt(self->client_fd,
+ SOL_SOCKET, SO_SETNETNS,
+ &self->netns_server,
+ sizeof(self->netns_server));
+ ASSERT_EQ(ret, 0);
+
+ ret = connect(self->client_fd, &addr, sizeof(addr));
+ ASSERT_EQ(ret, 0);
+
+ // switching network namespace of connected
+ // socket should fail
+ ret = setsockopt(self->client_fd,
+ SOL_SOCKET, SO_SETNETNS,
+ &self->netns_client,
+ sizeof(self->netns_client));
+ ASSERT_EQ(ret, -1);
+ ASSERT_EQ(errno, EOPNOTSUPP);
+}
+
+TEST_F(so_set_netns, test_socket_ns_switch_disconnected) {
+ struct sockaddr_in addr;
+ int ret;
+
+ addr.sin_family = AF_INET;
+ addr.sin_addr.s_addr = inet_addr("192.168.1.1");
+ addr.sin_port = htons(80);
+
+ ret = setsockopt(self->client_fd,
+ SOL_SOCKET, SO_SETNETNS,
+ &self->netns_server,
+ sizeof(self->netns_server));
+ ASSERT_EQ(ret, 0);
+
+ ret = connect(self->client_fd, &addr, sizeof(addr));
+ ASSERT_EQ(ret, 0);
+
+ close(self->server_fd);
+
+ // switching network namespace of recently disconnected
+ // socket should fail
+ ret = setsockopt(self->client_fd,
+ SOL_SOCKET, SO_SETNETNS,
+ &self->netns_client,
+ sizeof(self->netns_client));
+ ASSERT_EQ(ret, -1);
+ ASSERT_EQ(errno, EOPNOTSUPP);
+}
+
+TEST_HARNESS_MAIN
--
2.34.1



2023-02-02 19:55:38

by Alok Tiagi

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> > break;
> >
> > + case SO_SETNETNS:
> > + {
> > + struct net *other_ns, *my_ns;
> > +
> > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> > + ret = -EOPNOTSUPP;
> > + break;
> > + }
> > +
> > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> > + ret = -EOPNOTSUPP;
> > + break;
> > + }
> > +
> > + other_ns = get_net_ns_by_fd(val);
> > + if (IS_ERR(other_ns)) {
> > + ret = PTR_ERR(other_ns);
> > + break;
> > + }
> > +
> > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> > + ret = -EPERM;
> > + goto out_err;
> > + }
> > +
> > + /* check that the socket has never been connected or recently disconnected */
> > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> > + ret = -EOPNOTSUPP;
> > + goto out_err;
> > + }
> > +
> > + /* check that the socket is not bound to an interface*/
> > + if (sk->sk_bound_dev_if != 0) {
> > + ret = -EOPNOTSUPP;
> > + goto out_err;
> > + }
> > +
> > + my_ns = sock_net(sk);
> > + sock_net_set(sk, other_ns);
> > + put_net(my_ns);
> > + break;
>
> cpu 0 cpu 2
> --- ---
> ns = sock_net(sk);
> my_ns = sock_net(sk);
> sock_net_set(sk, other_ns);
> put_net(my_ns);
> ns is invalid ?

That is the reason we want the socket to be in an un-connected state. That
should help us avoid this situation.

>
> > +out_err:
> > + put_net(other_ns);
> > + break;
> > + }
> > +
> > default:
> > ret = -ENOPROTOOPT;
> > break;

2023-02-02 20:10:46

by Eric Dumazet

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
>
> On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> > > break;
> > >
> > > + case SO_SETNETNS:
> > > + {
> > > + struct net *other_ns, *my_ns;
> > > +
> > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> > > + ret = -EOPNOTSUPP;
> > > + break;
> > > + }
> > > +
> > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> > > + ret = -EOPNOTSUPP;
> > > + break;
> > > + }
> > > +
> > > + other_ns = get_net_ns_by_fd(val);
> > > + if (IS_ERR(other_ns)) {
> > > + ret = PTR_ERR(other_ns);
> > > + break;
> > > + }
> > > +
> > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> > > + ret = -EPERM;
> > > + goto out_err;
> > > + }
> > > +
> > > + /* check that the socket has never been connected or recently disconnected */
> > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> > > + ret = -EOPNOTSUPP;
> > > + goto out_err;
> > > + }
> > > +
> > > + /* check that the socket is not bound to an interface*/
> > > + if (sk->sk_bound_dev_if != 0) {
> > > + ret = -EOPNOTSUPP;
> > > + goto out_err;
> > > + }
> > > +
> > > + my_ns = sock_net(sk);
> > > + sock_net_set(sk, other_ns);
> > > + put_net(my_ns);
> > > + break;
> >
> > cpu 0 cpu 2
> > --- ---
> > ns = sock_net(sk);
> > my_ns = sock_net(sk);
> > sock_net_set(sk, other_ns);
> > put_net(my_ns);
> > ns is invalid ?
>
> That is the reason we want the socket to be in an un-connected state. That
> should help us avoid this situation.

This is not enough....

Another thread might look at sock_net(sk), for example from inet_diag
or tcp timers
(which can be fired even in un-connected state)

Even UDP sockets can receive packets while being un-connected,
and they need to deref the net pointer.

Currently there is no protection about sock_net(sk) being changed on the fly,
and the struct net could disappear and be freed.

There are ~1500 uses of sock_net(sk) in the kernel, I do not think
you/we want to audit all
of them to check what could go wrong...

2023-02-02 23:59:06

by Alok Tiagi

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Thu, Feb 02, 2023 at 09:10:23PM +0100, Eric Dumazet wrote:
> On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
> >
> > On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> > > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> > > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> > > > break;
> > > >
> > > > + case SO_SETNETNS:
> > > > + {
> > > > + struct net *other_ns, *my_ns;
> > > > +
> > > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> > > > + ret = -EOPNOTSUPP;
> > > > + break;
> > > > + }
> > > > +
> > > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> > > > + ret = -EOPNOTSUPP;
> > > > + break;
> > > > + }
> > > > +
> > > > + other_ns = get_net_ns_by_fd(val);
> > > > + if (IS_ERR(other_ns)) {
> > > > + ret = PTR_ERR(other_ns);
> > > > + break;
> > > > + }
> > > > +
> > > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> > > > + ret = -EPERM;
> > > > + goto out_err;
> > > > + }
> > > > +
> > > > + /* check that the socket has never been connected or recently disconnected */
> > > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> > > > + ret = -EOPNOTSUPP;
> > > > + goto out_err;
> > > > + }
> > > > +
> > > > + /* check that the socket is not bound to an interface*/
> > > > + if (sk->sk_bound_dev_if != 0) {
> > > > + ret = -EOPNOTSUPP;
> > > > + goto out_err;
> > > > + }
> > > > +
> > > > + my_ns = sock_net(sk);
> > > > + sock_net_set(sk, other_ns);
> > > > + put_net(my_ns);
> > > > + break;
> > >
> > > cpu 0 cpu 2
> > > --- ---
> > > ns = sock_net(sk);
> > > my_ns = sock_net(sk);
> > > sock_net_set(sk, other_ns);
> > > put_net(my_ns);
> > > ns is invalid ?
> >
> > That is the reason we want the socket to be in an un-connected state. That
> > should help us avoid this situation.
>
> This is not enough....
>
> Another thread might look at sock_net(sk), for example from inet_diag
> or tcp timers
> (which can be fired even in un-connected state)
>
> Even UDP sockets can receive packets while being un-connected,
> and they need to deref the net pointer.
>
> Currently there is no protection about sock_net(sk) being changed on the fly,
> and the struct net could disappear and be freed.
>
> There are ~1500 uses of sock_net(sk) in the kernel, I do not think
> you/we want to audit all
> of them to check what could go wrong...

I agree, auditing all the uses of sock_net(sk) is not a feasible option. From my
exploration of the usage of sock_net(sk) it appeared that it might be safe to
swap a sockets net ns if it had never been connected but I looked at only a
subset of such uses.

Introducing a ref counting logic to every access of sock_net(sk) may help get
around this but invovles a bigger change to increment and decrement the count at
every use of sock_net().

Any suggestions if this could be achieved in another way much close to the
socket creation time or any comments on our workaround for injecting sockets using
seccomp addfd?

2023-02-03 15:13:22

by Eric Dumazet

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Fri, Feb 3, 2023 at 12:59 AM Alok Tiagi <[email protected]> wrote:
>
> On Thu, Feb 02, 2023 at 09:10:23PM +0100, Eric Dumazet wrote:
> > On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
> > >
> > > On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> > > > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> > > > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > > > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> > > > > break;
> > > > >
> > > > > + case SO_SETNETNS:
> > > > > + {
> > > > > + struct net *other_ns, *my_ns;
> > > > > +
> > > > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> > > > > + ret = -EOPNOTSUPP;
> > > > > + break;
> > > > > + }
> > > > > +
> > > > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> > > > > + ret = -EOPNOTSUPP;
> > > > > + break;
> > > > > + }
> > > > > +
> > > > > + other_ns = get_net_ns_by_fd(val);
> > > > > + if (IS_ERR(other_ns)) {
> > > > > + ret = PTR_ERR(other_ns);
> > > > > + break;
> > > > > + }
> > > > > +
> > > > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> > > > > + ret = -EPERM;
> > > > > + goto out_err;
> > > > > + }
> > > > > +
> > > > > + /* check that the socket has never been connected or recently disconnected */
> > > > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> > > > > + ret = -EOPNOTSUPP;
> > > > > + goto out_err;
> > > > > + }
> > > > > +
> > > > > + /* check that the socket is not bound to an interface*/
> > > > > + if (sk->sk_bound_dev_if != 0) {
> > > > > + ret = -EOPNOTSUPP;
> > > > > + goto out_err;
> > > > > + }
> > > > > +
> > > > > + my_ns = sock_net(sk);
> > > > > + sock_net_set(sk, other_ns);
> > > > > + put_net(my_ns);
> > > > > + break;
> > > >
> > > > cpu 0 cpu 2
> > > > --- ---
> > > > ns = sock_net(sk);
> > > > my_ns = sock_net(sk);
> > > > sock_net_set(sk, other_ns);
> > > > put_net(my_ns);
> > > > ns is invalid ?
> > >
> > > That is the reason we want the socket to be in an un-connected state. That
> > > should help us avoid this situation.
> >
> > This is not enough....
> >
> > Another thread might look at sock_net(sk), for example from inet_diag
> > or tcp timers
> > (which can be fired even in un-connected state)
> >
> > Even UDP sockets can receive packets while being un-connected,
> > and they need to deref the net pointer.
> >
> > Currently there is no protection about sock_net(sk) being changed on the fly,
> > and the struct net could disappear and be freed.
> >
> > There are ~1500 uses of sock_net(sk) in the kernel, I do not think
> > you/we want to audit all
> > of them to check what could go wrong...
>
> I agree, auditing all the uses of sock_net(sk) is not a feasible option. From my
> exploration of the usage of sock_net(sk) it appeared that it might be safe to
> swap a sockets net ns if it had never been connected but I looked at only a
> subset of such uses.
>
> Introducing a ref counting logic to every access of sock_net(sk) may help get
> around this but invovles a bigger change to increment and decrement the count at
> every use of sock_net().
>
> Any suggestions if this could be achieved in another way much close to the
> socket creation time or any comments on our workaround for injecting sockets using
> seccomp addfd?

Maybe the existing BPF hook in inet_create() could be used ?

err = BPF_CGROUP_RUN_PROG_INET_SOCK(sk);

The BPF program might be able to switch the netns, because at this
time the new socket is not
yet visible from external threads.

Although it is not going to catch dual stack uses (open a V6 socket,
then use a v4mapped address at bind()/connect()/...

2023-02-03 17:51:01

by Alok Tiagi

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Fri, Feb 03, 2023 at 04:09:12PM +0100, Eric Dumazet wrote:
> On Fri, Feb 3, 2023 at 12:59 AM Alok Tiagi <[email protected]> wrote:
> >
> > On Thu, Feb 02, 2023 at 09:10:23PM +0100, Eric Dumazet wrote:
> > > On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
> > > >
> > > > On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> > > > > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> > > > > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > > > > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> > > > > > break;
> > > > > >
> > > > > > + case SO_SETNETNS:
> > > > > > + {
> > > > > > + struct net *other_ns, *my_ns;
> > > > > > +
> > > > > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> > > > > > + ret = -EOPNOTSUPP;
> > > > > > + break;
> > > > > > + }
> > > > > > +
> > > > > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> > > > > > + ret = -EOPNOTSUPP;
> > > > > > + break;
> > > > > > + }
> > > > > > +
> > > > > > + other_ns = get_net_ns_by_fd(val);
> > > > > > + if (IS_ERR(other_ns)) {
> > > > > > + ret = PTR_ERR(other_ns);
> > > > > > + break;
> > > > > > + }
> > > > > > +
> > > > > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> > > > > > + ret = -EPERM;
> > > > > > + goto out_err;
> > > > > > + }
> > > > > > +
> > > > > > + /* check that the socket has never been connected or recently disconnected */
> > > > > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> > > > > > + ret = -EOPNOTSUPP;
> > > > > > + goto out_err;
> > > > > > + }
> > > > > > +
> > > > > > + /* check that the socket is not bound to an interface*/
> > > > > > + if (sk->sk_bound_dev_if != 0) {
> > > > > > + ret = -EOPNOTSUPP;
> > > > > > + goto out_err;
> > > > > > + }
> > > > > > +
> > > > > > + my_ns = sock_net(sk);
> > > > > > + sock_net_set(sk, other_ns);
> > > > > > + put_net(my_ns);
> > > > > > + break;
> > > > >
> > > > > cpu 0 cpu 2
> > > > > --- ---
> > > > > ns = sock_net(sk);
> > > > > my_ns = sock_net(sk);
> > > > > sock_net_set(sk, other_ns);
> > > > > put_net(my_ns);
> > > > > ns is invalid ?
> > > >
> > > > That is the reason we want the socket to be in an un-connected state. That
> > > > should help us avoid this situation.
> > >
> > > This is not enough....
> > >
> > > Another thread might look at sock_net(sk), for example from inet_diag
> > > or tcp timers
> > > (which can be fired even in un-connected state)
> > >
> > > Even UDP sockets can receive packets while being un-connected,
> > > and they need to deref the net pointer.
> > >
> > > Currently there is no protection about sock_net(sk) being changed on the fly,
> > > and the struct net could disappear and be freed.
> > >
> > > There are ~1500 uses of sock_net(sk) in the kernel, I do not think
> > > you/we want to audit all
> > > of them to check what could go wrong...
> >
> > I agree, auditing all the uses of sock_net(sk) is not a feasible option. From my
> > exploration of the usage of sock_net(sk) it appeared that it might be safe to
> > swap a sockets net ns if it had never been connected but I looked at only a
> > subset of such uses.
> >
> > Introducing a ref counting logic to every access of sock_net(sk) may help get
> > around this but invovles a bigger change to increment and decrement the count at
> > every use of sock_net().
> >
> > Any suggestions if this could be achieved in another way much close to the
> > socket creation time or any comments on our workaround for injecting sockets using
> > seccomp addfd?
>
> Maybe the existing BPF hook in inet_create() could be used ?
>
> err = BPF_CGROUP_RUN_PROG_INET_SOCK(sk);
>
> The BPF program might be able to switch the netns, because at this
> time the new socket is not
> yet visible from external threads.
>
> Although it is not going to catch dual stack uses (open a V6 socket,
> then use a v4mapped address at bind()/connect()/...

We thought of a similar approach by intercepting the socket() call in seccomp
and injecting a new file descritpor much earlier but as you said we run into the
issue of handling dual stack sockets since we do not know in advance if its
going to be used for a v4mapped address.


2023-02-03 22:35:35

by Eric W. Biederman

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

Alok Tiagi <[email protected]> writes:

> On Fri, Feb 03, 2023 at 04:09:12PM +0100, Eric Dumazet wrote:
>> On Fri, Feb 3, 2023 at 12:59 AM Alok Tiagi <[email protected]> wrote:
>> >
>> > On Thu, Feb 02, 2023 at 09:10:23PM +0100, Eric Dumazet wrote:
>> > > On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
>> > > >
>> > > > On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
>> > > > > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
>> > > > > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
>> > > > > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
>> > > > > > break;
>> > > > > >
>> > > > > > + case SO_SETNETNS:
>> > > > > > + {
>> > > > > > + struct net *other_ns, *my_ns;
>> > > > > > +
>> > > > > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
>> > > > > > + ret = -EOPNOTSUPP;
>> > > > > > + break;
>> > > > > > + }
>> > > > > > +
>> > > > > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
>> > > > > > + ret = -EOPNOTSUPP;
>> > > > > > + break;
>> > > > > > + }
>> > > > > > +
>> > > > > > + other_ns = get_net_ns_by_fd(val);
>> > > > > > + if (IS_ERR(other_ns)) {
>> > > > > > + ret = PTR_ERR(other_ns);
>> > > > > > + break;
>> > > > > > + }
>> > > > > > +
>> > > > > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
>> > > > > > + ret = -EPERM;
>> > > > > > + goto out_err;
>> > > > > > + }
>> > > > > > +
>> > > > > > + /* check that the socket has never been connected or recently disconnected */
>> > > > > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
>> > > > > > + ret = -EOPNOTSUPP;
>> > > > > > + goto out_err;
>> > > > > > + }
>> > > > > > +
>> > > > > > + /* check that the socket is not bound to an interface*/
>> > > > > > + if (sk->sk_bound_dev_if != 0) {
>> > > > > > + ret = -EOPNOTSUPP;
>> > > > > > + goto out_err;
>> > > > > > + }
>> > > > > > +
>> > > > > > + my_ns = sock_net(sk);
>> > > > > > + sock_net_set(sk, other_ns);
>> > > > > > + put_net(my_ns);
>> > > > > > + break;
>> > > > >
>> > > > > cpu 0 cpu 2
>> > > > > --- ---
>> > > > > ns = sock_net(sk);
>> > > > > my_ns = sock_net(sk);
>> > > > > sock_net_set(sk, other_ns);
>> > > > > put_net(my_ns);
>> > > > > ns is invalid ?
>> > > >
>> > > > That is the reason we want the socket to be in an un-connected state. That
>> > > > should help us avoid this situation.
>> > >
>> > > This is not enough....
>> > >
>> > > Another thread might look at sock_net(sk), for example from inet_diag
>> > > or tcp timers
>> > > (which can be fired even in un-connected state)
>> > >
>> > > Even UDP sockets can receive packets while being un-connected,
>> > > and they need to deref the net pointer.
>> > >
>> > > Currently there is no protection about sock_net(sk) being changed on the fly,
>> > > and the struct net could disappear and be freed.
>> > >
>> > > There are ~1500 uses of sock_net(sk) in the kernel, I do not think
>> > > you/we want to audit all
>> > > of them to check what could go wrong...
>> >
>> > I agree, auditing all the uses of sock_net(sk) is not a feasible option. From my
>> > exploration of the usage of sock_net(sk) it appeared that it might be safe to
>> > swap a sockets net ns if it had never been connected but I looked at only a
>> > subset of such uses.
>> >
>> > Introducing a ref counting logic to every access of sock_net(sk) may help get
>> > around this but invovles a bigger change to increment and decrement the count at
>> > every use of sock_net().
>> >
>> > Any suggestions if this could be achieved in another way much close to the
>> > socket creation time or any comments on our workaround for injecting sockets using
>> > seccomp addfd?
>>
>> Maybe the existing BPF hook in inet_create() could be used ?
>>
>> err = BPF_CGROUP_RUN_PROG_INET_SOCK(sk);
>>
>> The BPF program might be able to switch the netns, because at this
>> time the new socket is not
>> yet visible from external threads.
>>
>> Although it is not going to catch dual stack uses (open a V6 socket,
>> then use a v4mapped address at bind()/connect()/...
>
> We thought of a similar approach by intercepting the socket() call in seccomp
> and injecting a new file descritpor much earlier but as you said we run into the
> issue of handling dual stack sockets since we do not know in advance if its
> going to be used for a v4mapped address.

I would suggest adding a default ipv4 route from your ipv6 network
namespaces to your ipv4 network namespace, but that only works for
outbound traffic. The inbound traffic problem is classically solved
via nat.

That you are not suggesting using nat has me thinking there is something
subtle in what you are trying to do that I am missing.

Perhaps your userspace can do:

previous_netns = open("/proc/self/ns/net");
setns(ipv4_netns);
socket();
setns(previous_netns);


As the network namespace is per thread this is atomic if you add
the logic to block signals around it.

Eric

2023-02-04 18:44:19

by Alok Tiagi

[permalink] [raw]
Subject: Re: [RFC] net: add new socket option SO_SETNETNS

On Fri, Feb 03, 2023 at 03:17:06PM -0600, Eric W. Biederman wrote:
> Alok Tiagi <[email protected]> writes:
>
> > On Fri, Feb 03, 2023 at 04:09:12PM +0100, Eric Dumazet wrote:
> >> On Fri, Feb 3, 2023 at 12:59 AM Alok Tiagi <[email protected]> wrote:
> >> >
> >> > On Thu, Feb 02, 2023 at 09:10:23PM +0100, Eric Dumazet wrote:
> >> > > On Thu, Feb 2, 2023 at 8:55 PM Alok Tiagi <[email protected]> wrote:
> >> > > >
> >> > > > On Thu, Feb 02, 2023 at 09:48:10AM +0800, Hillf Danton wrote:
> >> > > > > On Wed, 1 Feb 2023 19:22:57 +0000 aloktiagi <[email protected]>
> >> > > > > > @@ -1535,6 +1535,52 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> >> > > > > > WRITE_ONCE(sk->sk_txrehash, (u8)val);
> >> > > > > > break;
> >> > > > > >
> >> > > > > > + case SO_SETNETNS:
> >> > > > > > + {
> >> > > > > > + struct net *other_ns, *my_ns;
> >> > > > > > +
> >> > > > > > + if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6) {
> >> > > > > > + ret = -EOPNOTSUPP;
> >> > > > > > + break;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_DGRAM) {
> >> > > > > > + ret = -EOPNOTSUPP;
> >> > > > > > + break;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + other_ns = get_net_ns_by_fd(val);
> >> > > > > > + if (IS_ERR(other_ns)) {
> >> > > > > > + ret = PTR_ERR(other_ns);
> >> > > > > > + break;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + if (!ns_capable(other_ns->user_ns, CAP_NET_ADMIN)) {
> >> > > > > > + ret = -EPERM;
> >> > > > > > + goto out_err;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + /* check that the socket has never been connected or recently disconnected */
> >> > > > > > + if (sk->sk_state != TCP_CLOSE || sk->sk_shutdown & SHUTDOWN_MASK) {
> >> > > > > > + ret = -EOPNOTSUPP;
> >> > > > > > + goto out_err;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + /* check that the socket is not bound to an interface*/
> >> > > > > > + if (sk->sk_bound_dev_if != 0) {
> >> > > > > > + ret = -EOPNOTSUPP;
> >> > > > > > + goto out_err;
> >> > > > > > + }
> >> > > > > > +
> >> > > > > > + my_ns = sock_net(sk);
> >> > > > > > + sock_net_set(sk, other_ns);
> >> > > > > > + put_net(my_ns);
> >> > > > > > + break;
> >> > > > >
> >> > > > > cpu 0 cpu 2
> >> > > > > --- ---
> >> > > > > ns = sock_net(sk);
> >> > > > > my_ns = sock_net(sk);
> >> > > > > sock_net_set(sk, other_ns);
> >> > > > > put_net(my_ns);
> >> > > > > ns is invalid ?
> >> > > >
> >> > > > That is the reason we want the socket to be in an un-connected state. That
> >> > > > should help us avoid this situation.
> >> > >
> >> > > This is not enough....
> >> > >
> >> > > Another thread might look at sock_net(sk), for example from inet_diag
> >> > > or tcp timers
> >> > > (which can be fired even in un-connected state)
> >> > >
> >> > > Even UDP sockets can receive packets while being un-connected,
> >> > > and they need to deref the net pointer.
> >> > >
> >> > > Currently there is no protection about sock_net(sk) being changed on the fly,
> >> > > and the struct net could disappear and be freed.
> >> > >
> >> > > There are ~1500 uses of sock_net(sk) in the kernel, I do not think
> >> > > you/we want to audit all
> >> > > of them to check what could go wrong...
> >> >
> >> > I agree, auditing all the uses of sock_net(sk) is not a feasible option. From my
> >> > exploration of the usage of sock_net(sk) it appeared that it might be safe to
> >> > swap a sockets net ns if it had never been connected but I looked at only a
> >> > subset of such uses.
> >> >
> >> > Introducing a ref counting logic to every access of sock_net(sk) may help get
> >> > around this but invovles a bigger change to increment and decrement the count at
> >> > every use of sock_net().
> >> >
> >> > Any suggestions if this could be achieved in another way much close to the
> >> > socket creation time or any comments on our workaround for injecting sockets using
> >> > seccomp addfd?
> >>
> >> Maybe the existing BPF hook in inet_create() could be used ?
> >>
> >> err = BPF_CGROUP_RUN_PROG_INET_SOCK(sk);
> >>
> >> The BPF program might be able to switch the netns, because at this
> >> time the new socket is not
> >> yet visible from external threads.
> >>
> >> Although it is not going to catch dual stack uses (open a V6 socket,
> >> then use a v4mapped address at bind()/connect()/...
> >
> > We thought of a similar approach by intercepting the socket() call in seccomp
> > and injecting a new file descritpor much earlier but as you said we run into the
> > issue of handling dual stack sockets since we do not know in advance if its
> > going to be used for a v4mapped address.
>
> I would suggest adding a default ipv4 route from your ipv6 network
> namespaces to your ipv4 network namespace, but that only works for
> outbound traffic. The inbound traffic problem is classically solved
> via nat.
>
> That you are not suggesting using nat has me thinking there is something
> subtle in what you are trying to do that I am missing.
>
> Perhaps your userspace can do:
>
> previous_netns = open("/proc/self/ns/net");
> setns(ipv4_netns);
> socket();
> setns(previous_netns);
>
>
> As the network namespace is per thread this is atomic if you add
> the logic to block signals around it.
>
> Eric

That is correct, we are not using nat, but we are providing a mechanism for the
users of our container platform to move to ipv6 only while keeping egress
connectivity to their ipv4 destinations. We are doing this transparently without
any change in user code, but by intercept networking syscalls in a container
manager running in a dedicated ipv4 only network namespace. Our current solution
as described in my original commit message has limitations and we are looking
for a way to switch a sockets namespace from the ipv6 only container network
namespace to the dedicated ipv4 network namespace which really simplifies our
design.

Since our userspace is the container workload we have no control over how they
instantiate their sockets.