| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
inet: reject unhashed sockets during established lookups
TCP established ehash lookups are lockless. The socket caches involved
use SLAB_TYPESAFE_BY_RCU, so memory can be reused before an RCU reader
that retained its old address has finished.
After a tuple match, __inet_lookup_established() and
__inet6_lookup_established() acquire a reference and repeat the tuple
check. They do not check whether the new object is hashed. If the
address is reused for a socket with the same tuple, the lookup can
acquire a reference to the new object and return it while its nulls node
is unhashed.
A PoC that drives reuse with a SYN cookie request socket reproduced the
crash. The new request socket is unhashed and has a NULL rsk_listener.
A lookup that had loaded the address of the old request socket acquired
the reference published by tcp_get_cookie_sock() and returned the new
object. tcp_v4_rcv() then read sk_state through the NULL listener.
The resulting crash was:
BUG: kernel NULL pointer dereference, address: 0000000000000012
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
...
Oops: Oops: 0000 [#1] SMP NOPTI
CPU: 4 UID: 0 PID: 0 Comm: swapper/4 Not tainted 7.3.0-00591-g0e135910deec #1 PREEMPT(lazy)
...
RIP: 0010:tcp_v4_rcv+0x44e/0x1280
...
R10: 0000000000000001 R11: 00000000024ca8c0 R12: 0000000000000000
...
CR2: 0000000000000012 CR3: 0000000103d50004 CR4: 0000000000770ef0
...
Call Trace:
<IRQ>
? raw_local_deliver+0xae/0x240
ip_protocol_deliver_rcu+0x2d/0x180
ip_local_deliver_finish+0x84/0x100
ip_sublist_rcv_finish+0x5f/0x70
ip_sublist_rcv+0x12a/0x1d0
ip_list_rcv+0xfd/0x130
__netif_receive_skb_list_core+0x223/0x250
netif_receive_skb_list_internal+0x1d2/0x2c0
? try_fill_recv+0x39/0x7b0
napi_complete_done+0x74/0x170
virtnet_poll+0x445/0x600
__napi_poll+0x27/0x1e0
net_rx_action+0x2fe/0x3b0
handle_softirqs+0xc2/0x2c0
__irq_exit_rcu+0x74/0xd0
common_interrupt+0x7f/0x90
</IRQ>
...
Kernel panic - not syncing: Fatal exception in interrupt
After taking a reference, check whether the nulls node is unhashed in
both IPv4 and IPv6 established lookups. If it is unhashed or the tuple
no longer matches, drop the reference and restart the lookup.
Lookups racing with the TIME_WAIT hashdance, or with a request being
replaced by its child, now restart and find the new socket instead of
returning the old one.
Request sockets that were never inserted into ehash can have a nonzero
refcount, and a stale lookup can hold a transient reference. Therefore,
when SYN cookie, Fast Open, and request migration paths fail after
setting a nonzero refcount, release the references held by those paths
with reqsk_put() instead of freeing the request directly. |
| A repository created with auto-init accepted pushes before its initialization had finished, and the initial commit was built in a checkout of whatever the repository already contained. A user who is allowed to create repositories could push a crafted tree during that window so that writing the README followed a symbolic link and overwrote a server-side git hook, which runs commands as the Gitea system user. The initial commit is now built in a freshly initialized directory, and a new repository only becomes ready, and accepts pushes, once its initialization is complete. |
| Docmost is open-source collaborative wiki and documentation software. Prior to 0.96.0, two concurrent role-demotion requests from a space admin or workspace owner can each pass the last-admin or last-owner count check before either update is committed because the validation and role update are not serialized in one transaction. The requests can therefore demote the final two space admins or workspace owners and leave the affected space without an admin or the workspace without an owner, preventing normal administrative recovery. This issue is fixed in version 0.96.0. |
| Next.js is a React framework for building full-stack web applications. From 16.3.0 until 16.4.5, Next.js applications using the same 'use cache' function directly during rendering and through unstable_cache() can disclose cached data across root params when that function reads a root param. During the first fill after a cold process start, cross-request deduplication allows a concurrent unstable_cache() request to join the direct call's in-flight fill and receive its result before the root param mismatch is rejected. unstable_cache() then persists that result under a key that omits root params, causing later requests for other root param values to receive the same entry until revalidation or eviction. Exploitation requires the unstable_cache() request to arrive while the first fill is in flight. Outside this race, nesting 'use cache' inside unstable_cache() raises an unsupported-operation error, and avoiding that nesting prevents the issue. This issue is fixed in version 16.4.5. |
| Next.js is a React framework for building full-stack web applications. From 16.2.0 until 16.4.5, Next.js applications with Cache Components enabled can reuse an outer 'use cache' entry across root param values when a nested 'use cache' call reads a root param through next/root-params and returns a value containing pending promises. React Flight can deliver the ready portion to the outer function before the inner cache finishes publishing its root-param dependencies. The outer entry can then complete and be persisted under a key that omits the root param, causing the first written value to be served for other root param values and allowing the outer entry to ignore nested cache tag invalidation and lifetimes. The leaked value is not attacker controlled and depends on which invocation writes the entry first. This variant does not require an existing inner cache entry or a slow cache handler and is not covered by the fix for CVE-2026-103004. This issue is fixed in version 16.4.5. |
| For HTTP/1 servers, when MaxBytesReader's limit is exceeded, "Connection: close" header is automatically set without synchronization, which might race with another header being set explicitly in the server handler. As a result, an HTTP/1 server that utilizes MaxBytesReader or MaxBytesHandler can crash due to runtime panic if its handler reads the request body concurrently with writing the response header. |
| A vulnerability was found in vllm-project vllm 0.30.0/0.31.0. This vulnerability affects unknown code of the file single_type_kv_cache_manager.py of the component Common Prefix Handler. The manipulation results in race condition. The attack may be performed from remote. Attacks of this nature are highly complex. It is stated that the exploitability is difficult. The exploit has been made public and could be used. Upgrading to version 0.32.0 is able to resolve this issue. The patch is identified as e687a251bb056fa527846633f7a1b0a6c0f42374. You should upgrade the affected component. |
| In the Linux kernel, the following vulnerability has been resolved:
nfsd: hold cl_lock in client_has_openowners()
client_has_openowners() walks clp->cl_openowners and reads so_stateids
without clp->cl_lock. nfs4_put_stateowner() unhashes that openowner under
cl_lock and then frees it, so a concurrent EXCHANGE_ID with mismatched
creds can use-after-free the nfs4_openowner.
BUG: KASAN: slab-use-after-free in client_has_state+0x10a/0x140
fs/nfsd/nfs4state.c:3718 client_has_openowners()
nfsd4_exchange_id
nfsd4_proc_compound
nfsd_dispatch
svc_process
Take clp->cl_lock while client_has_openowners() walks the openowner
list. |
| This CVE ID has been rejected or withdrawn by its CVE Numbering Authority. |
| pgEdge RAG Server is an API server for retrieval-augmented generation over PostgreSQL content using pgvector. Prior to 2.0.0, the BM25 keyword index is a single per-pipeline object shared by the unauthenticated query endpoint, and concurrent requests lock its clear, refill, and search operations separately. This allows one request to search a corpus populated through another request's search filter, which can return documents belonging to a different tenant or customer without malformed input. The affected responses cannot be identified retrospectively from the server logs. This issue is fixed in version 2.0.0. |
| A flaw was found in the login-failure storage component of Keycloak, the identity and access management solution. This issue specifically affects deployments using MySQL or MariaDB databases in a multi-node configuration. When multiple failed login attempts occur simultaneously for the same account, a synchronization error can cause the system to enter a repetitive loop that locks the user's record in the database. This results in legitimate login attempts for that user being blocked and failing with an error. Furthermore, because some failures are not correctly recorded during this loop, the system's brute-force protection may not accurately count all failed attempts. |
| In the Linux kernel, the following vulnerability has been resolved:
ALSA: usb-audio: avoid kobject path lookup in DualSense match
The DualSense jack-detection input handler verifies that a matching input
device belongs to the same physical controller by building kobject path
strings for both the input device and the USB audio device, then comparing
the path prefix.
This was observed when a weak physical connection caused the controller
to rapidly disconnect and reconnect. During that repeated hotplug,
snd_dualsense_ih_match() can run while the controller's USB device is
being disconnected. kobject_get_path() walks ancestor kobjects and
dereferences their names; if the USB device kobject name is no longer
valid, this can fault in strlen():
RIP: 0010:strlen+0x10/0x30
Call Trace:
kobject_get_path+0x34/0x150
snd_dualsense_ih_match+0x49/0xd0 [snd_usb_audio]
input_register_device+0x566/0x6a0
ps_probe+0xb89/0x1590 [hid_playstation]
The same ownership check can be done without building kobject path
strings. The input device is parented below the HID device, USB interface
and USB device, so walking the input device parent chain and comparing
against the mixer USB device preserves the check without dereferencing
kobject names during disconnect. |
| Out-of-bounds read in Windows Graphics Kernel allows an authorized attacker to elevate privileges locally. |
| Concurrent Execution using Shared Resource with Improper Synchronization vulnerability in Apache HTTP Server's mod_http2 in assign_headers() can lead to RCE.
This issue affects Apache HTTP Server: from 2.4.55 through 2.4.70.
Affected releases of mod_h2 from GitHub: from 2.0.8 through 2.0.42. |
| Double free in Windows Error Reporting allows an authorized attacker to elevate privileges locally. |
| In port_shutdown of ipc.c, there is a possible Use After Free due to a race condition. This could lead to local escalation of privilege with execution privileges needed. User interaction is for exploitation. |
| Use after free in vhdmp.sys allows an authorized attacker to elevate privileges locally. |
| Use of uninitialized resource in Windows Graphics Kernel allows an authorized attacker to disclose information locally. |
| rails-html-sanitizer is responsible for sanitizing HTML fragments in Rails applications. Before 1.7.2, the HTML4 and HTML5 variants of Rails::HTML::SafeListSanitizer reuse one Rails::HTML::PermitScrubber across calls and mutate its tags and attributes without synchronization. In a multithreaded process, concurrent sanitize calls using different tags: or attributes: options can interleave and cause one call to use the other call's allowlist. If a permissive call admits a script element or event-handler attribute, output expected to be sanitized under a stricter policy can retain it and cause cross-site scripting when rendered. Applications that are single-threaded, use one identical allowlist everywhere, or provide a separate scrubber: are not affected. This issue is fixed in version 1.7.2. |
| Use after free in Windows Graphics Kernel allows an authorized attacker to elevate privileges locally. |