All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ido Schimmel <idosch@nvidia.com>
To: <stable@vger.kernel.org>
Cc: <davem@davemloft.net>, <kuba@kernel.org>, <pabeni@redhat.com>,
	<edumazet@google.com>, <nhorman@tuxdriver.com>,
	<yotam.gi@gmail.com>, <sashal@kernel.org>, <fw@strlen.de>,
	<jacob.e.keller@intel.com>, <jiri@nvidia.com>
Subject: [PATCH 5.10 3/4] psample: Require 'CAP_NET_ADMIN' when joining "packets" group
Date: Mon, 11 Dec 2023 14:37:39 +0200	[thread overview]
Message-ID: <20231211123740.822802-4-idosch@nvidia.com> (raw)
In-Reply-To: <20231211123740.822802-1-idosch@nvidia.com>

commit 44ec98ea5ea9cfecd31a5c4cc124703cb5442832 upstream.

The "psample" generic netlink family notifies sampled packets over the
"packets" multicast group. This is problematic since by default generic
netlink allows non-root users to listen to these notifications.

Fix by marking the group with the 'GENL_UNS_ADMIN_PERM' flag. This will
prevent non-root users or root without the 'CAP_NET_ADMIN' capability
(in the user namespace owning the network namespace) from joining the
group.

Tested using [1].

Before:

 # capsh -- -c ./psample_repo
 # capsh --drop=cap_net_admin -- -c ./psample_repo

After:

 # capsh -- -c ./psample_repo
 # capsh --drop=cap_net_admin -- -c ./psample_repo
 Failed to join "packets" multicast group

[1]
 $ cat psample.c
 #include <stdio.h>
 #include <netlink/genl/ctrl.h>
 #include <netlink/genl/genl.h>
 #include <netlink/socket.h>

 int join_grp(struct nl_sock *sk, const char *grp_name)
 {
 	int grp, err;

 	grp = genl_ctrl_resolve_grp(sk, "psample", grp_name);
 	if (grp < 0) {
 		fprintf(stderr, "Failed to resolve \"%s\" multicast group\n",
 			grp_name);
 		return grp;
 	}

 	err = nl_socket_add_memberships(sk, grp, NFNLGRP_NONE);
 	if (err) {
 		fprintf(stderr, "Failed to join \"%s\" multicast group\n",
 			grp_name);
 		return err;
 	}

 	return 0;
 }

 int main(int argc, char **argv)
 {
 	struct nl_sock *sk;
 	int err;

 	sk = nl_socket_alloc();
 	if (!sk) {
 		fprintf(stderr, "Failed to allocate socket\n");
 		return -1;
 	}

 	err = genl_connect(sk);
 	if (err) {
 		fprintf(stderr, "Failed to connect socket\n");
 		return err;
 	}

 	err = join_grp(sk, "config");
 	if (err)
 		return err;

 	err = join_grp(sk, "packets");
 	if (err)
 		return err;

 	return 0;
 }
 $ gcc -I/usr/include/libnl3 -lnl-3 -lnl-genl-3 -o psample_repo psample.c

Fixes: 6ae0a6286171 ("net: Introduce psample, a new genetlink channel for packet sampling")
Reported-by: "The UK's National Cyber Security Centre (NCSC)" <security@ncsc.gov.uk>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/20231206213102.1824398-2-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/psample/psample.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/psample/psample.c b/net/psample/psample.c
index 482c07f2766b..5913da77c71d 100644
--- a/net/psample/psample.c
+++ b/net/psample/psample.c
@@ -30,7 +30,8 @@ enum psample_nl_multicast_groups {
 
 static const struct genl_multicast_group psample_nl_mcgrps[] = {
 	[PSAMPLE_NL_MCGRP_CONFIG] = { .name = PSAMPLE_NL_MCGRP_CONFIG_NAME },
-	[PSAMPLE_NL_MCGRP_SAMPLE] = { .name = PSAMPLE_NL_MCGRP_SAMPLE_NAME },
+	[PSAMPLE_NL_MCGRP_SAMPLE] = { .name = PSAMPLE_NL_MCGRP_SAMPLE_NAME,
+				      .flags = GENL_UNS_ADMIN_PERM },
 };
 
 static struct genl_family psample_nl_family __ro_after_init;
-- 
2.40.1


  parent reply	other threads:[~2023-12-11 12:38 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-11 12:37 [PATCH 5.10 0/4] Generic netlink multicast fixes Ido Schimmel
2023-12-11 12:37 ` [PATCH 5.10 1/4] netlink: don't call ->netlink_bind with table lock held Ido Schimmel
2023-12-11 12:37 ` [PATCH 5.10 2/4] genetlink: add CAP_NET_ADMIN test for multicast bind Ido Schimmel
2023-12-11 12:37 ` Ido Schimmel [this message]
2023-12-11 12:37 ` [PATCH 5.10 4/4] drop_monitor: Require 'CAP_SYS_ADMIN' when joining "events" group Ido Schimmel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231211123740.822802-4-idosch@nvidia.com \
    --to=idosch@nvidia.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=fw@strlen.de \
    --cc=jacob.e.keller@intel.com \
    --cc=jiri@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=nhorman@tuxdriver.com \
    --cc=pabeni@redhat.com \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=yotam.gi@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.