Описание
In the Linux kernel, the following vulnerability has been resolved:
tcp: properly terminate timers for kernel sockets
We had various syzbot reports about tcp timers firing after
the corresponding netns has been dismantled.
Fortunately Josef Bacik could trigger the issue more often,
and could test a patch I wrote two years ago.
When TCP sockets are closed, we call inet_csk_clear_xmit_timers()
to 'stop' the timers.
inet_csk_clear_xmit_timers() can be called from any context,
including when socket lock is held.
This is the reason it uses sk_stop_timer(), aka del_timer().
This means that ongoing timers might finish much later.
For user sockets, this is fine because each running timer
holds a reference on the socket, and the user socket holds
a reference on the netns.
For kernel sockets, we risk that the netns is freed before
timer can complete, because kernel sockets do not hold
reference on the netns.
This patch adds inet_csk_clear_xmit_timers_sync() function
that using sk_stop_timer_sync() to make sure all timers
are terminated before the kernel socket is released.
Modules using kernel sockets close them in their netns exit()
handler.
Also add sock_not_owned_by_me() helper to get LOCKDEP
support : inet_csk_clear_xmit_timers_sync() must not be called
while socket lock is held.
It is very possible we can revert in the future commit
3a58f13a881e ("net: rds: acquire refcount on TCP sockets")
which attempted to solve the issue in rds only.
(net/smc/af_smc.c and net/mptcp/subflow.c have similar code)
We probably can remove the check_net() tests from
tcp_out_of_resources() and __tcp_close() in the future.
A flaw was found in the Linux kernel's TCP subsystem. When kernel-managed TCP sockets are closed, active network timers may continue executing after the associated network namespace has been dismantled and freed. A local attacker can exploit this timing condition to trigger a use-after-free memory error, leading to a system crash and Denial of Service (DoS).
Отчет
A use-after-free flaw was found in the Linux kernel's TCP subsystem, rated by Red Hat as Moderate severity. When a kernel-managed TCP socket is closed, its network timers may continue running after the associated network namespace has been dismantled and freed. A local attacker able to trigger this timing condition could cause a use-after-free memory error, crashing the kernel and resulting in a denial of service. Red Hat Enterprise Linux 9 is not affected because the fix is already present in that release stream. In Red Hat Enterprise Linux 8, the kernel and kernel-rt packages in the 8.6, 8.8, and 8.10 update streams are affected, and fixes are being pursued through the responsible delegated update streams; the RHEL 6 and RHEL 7 Extended Lifecycle Support streams are out of support scope. Fedora and Hosted OpenShift (ARO) offerings that ship the affected kernel are also in scope. Red Hat rates this flaw Moderate (CVSS 3.1 score 5.8) rather than the upstream CVE.org rating of Important (7.8) because exploitation requires winning a timing race with high attack complexity, and Red Hat's analysis found it unlikely that the issue could be triggered remotely.
Меры по смягчению последствий
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
Затронутые пакеты
| Платформа | Пакет | Состояние | Рекомендация | Релиз |
|---|---|---|---|---|
| Red Hat Enterprise Linux 6 | kernel | Out of support scope | ||
| Red Hat Enterprise Linux 7 | kernel | Out of support scope | ||
| Red Hat Enterprise Linux 7 | kernel-rt | Out of support scope | ||
| Red Hat Enterprise Linux 9 | kernel | Not affected | ||
| Red Hat Enterprise Linux 9 | kernel-rt | Not affected | ||
| Red Hat Enterprise Linux 8 | kernel-rt | Fixed | RHSA-2024:5102 | 08.08.2024 |
| Red Hat Enterprise Linux 8 | kernel | Fixed | RHSA-2024:5101 | 08.08.2024 |
| Red Hat Enterprise Linux 8.8 Extended Update Support | kernel | Fixed | RHSA-2024:5255 | 13.08.2024 |
Показывать по
Дополнительная информация
Статус:
EPSS
5.8 Medium
CVSS3
Связанные уязвимости
In the Linux kernel, the following vulnerability has been resolved: tcp: properly terminate timers for kernel sockets We had various syzbot reports about tcp timers firing after the corresponding netns has been dismantled. Fortunately Josef Bacik could trigger the issue more often, and could test a patch I wrote two years ago. When TCP sockets are closed, we call inet_csk_clear_xmit_timers() to 'stop' the timers. inet_csk_clear_xmit_timers() can be called from any context, including when socket lock is held. This is the reason it uses sk_stop_timer(), aka del_timer(). This means that ongoing timers might finish much later. For user sockets, this is fine because each running timer holds a reference on the socket, and the user socket holds a reference on the netns. For kernel sockets, we risk that the netns is freed before timer can complete, because kernel sockets do not hold reference on the netns. This patch adds inet_csk_clear_xmit_timers_sync() function that using sk_stop_timer_s...
In the Linux kernel, the following vulnerability has been resolved: tcp: properly terminate timers for kernel sockets We had various syzbot reports about tcp timers firing after the corresponding netns has been dismantled. Fortunately Josef Bacik could trigger the issue more often, and could test a patch I wrote two years ago. When TCP sockets are closed, we call inet_csk_clear_xmit_timers() to 'stop' the timers. inet_csk_clear_xmit_timers() can be called from any context, including when socket lock is held. This is the reason it uses sk_stop_timer(), aka del_timer(). This means that ongoing timers might finish much later. For user sockets, this is fine because each running timer holds a reference on the socket, and the user socket holds a reference on the netns. For kernel sockets, we risk that the netns is freed before timer can complete, because kernel sockets do not hold reference on the netns. This patch adds inet_csk_clear_xmit_timers_sync() function that using sk_stop_ti
In the Linux kernel, the following vulnerability has been resolved: t ...
In the Linux kernel, the following vulnerability has been resolved: tcp: properly terminate timers for kernel sockets We had various syzbot reports about tcp timers firing after the corresponding netns has been dismantled. Fortunately Josef Bacik could trigger the issue more often, and could test a patch I wrote two years ago. When TCP sockets are closed, we call inet_csk_clear_xmit_timers() to 'stop' the timers. inet_csk_clear_xmit_timers() can be called from any context, including when socket lock is held. This is the reason it uses sk_stop_timer(), aka del_timer(). This means that ongoing timers might finish much later. For user sockets, this is fine because each running timer holds a reference on the socket, and the user socket holds a reference on the netns. For kernel sockets, we risk that the netns is freed before timer can complete, because kernel sockets do not hold reference on the netns. This patch adds inet_csk_clear_xmit_timers_sync() function that using sk_stop...
Уязвимость компонента tcp ядра операционной системы Linux, позволяющая нарушителю вызвать отказ в обслуживании
EPSS
5.8 Medium
CVSS3