| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| Summary
Cloudflare tokio-quiche was found to contain an uncontrolled resource consumption vulnerability in its HTTP/3 DATAGRAM handling. For every previously unseen flow identifier (quarter stream ID) received in a DATAGRAM frame, tokio-quiche creates and retains new, per-flow state. A remote peer can send DATAGRAMs with many distinct identifiers, causing the per-connection flow map and associated channel state to grow without bound.
Impact
Exploitation can exhaust process memory and cause a denial of service. The vulnerable code is used in both client and server configurations, so either role can be affected by a malicious peer. QUIC DATAGRAM support in tokio-quiche is enabled by default through QuicSettings::enable_dgram; applications can disable it by setting this field to false. The vulnerable HTTP/3 path is reachable only when DATAGRAM support is negotiated with the peer. Core quiche and HTTP/3 deployments without DATAGRAM support are not affected.
Mitigation
Disable DATAGRAMs in tokio-quiche’s QuicSettings::enable_dgram.
Remediation
Upgrade to tokio-quiche version 0.23.0 or later. The fixed code creates flow state only for valid DATAGRAM-capable request streams, which are bounded by QUIC’s stream limits. To tolerate packet reordering, servers retain early DATAGRAMs in a bounded, expiring buffer, while clients discard DATAGRAMs for unknown flows. |
| JSON::XS versions before 4.05 for Perl leak the decoded key of an object rejected for a missing colon in decode_hv.
Unless it is short plain ASCII, a quoted key is decoded into a new scalar. Nothing frees that scalar when the object is rejected for a missing ':' after the key.
A process that keeps running after a failed decode retains one key per rejected document, with its size chosen by the sender. Repeated requests exhaust process memory. |
| JSON::XS versions before 4.05 for Perl leak the decoded part of a rejected long string in decode_str.
A string is decoded through a fixed-size buffer (16 KiB in 4.04), and each time the buffer fills its contents move into a new scalar. Nothing frees that scalar when the string is then rejected for a control character, an invalid escape, malformed UTF-8 or the end of the input.
A process that keeps running after a failed decode retains the decoded part of each rejected string, with its size chosen by the sender. Repeated requests exhaust process memory. |
| In the Linux kernel, the following vulnerability has been resolved:
iio: buffer: hw-consumer: free scan_mask on buffer release
The scan_mask lifetime changed in commit 9a2e1233d38c ("iio: buffer:
hw-consumer: remove redundant scan_mask flexible array").
Before that change, the scan mask storage was embedded in struct
hw_consumer_buffer, so iio_hw_buf_release() could free the whole
allocation with a single kfree(hw_buf).
That commit moved the scan mask to a separate bitmap_zalloc() allocation
stored in buffer.scan_mask, but left iio_hw_buf_release() unchanged.
Free the scan mask in iio_hw_buf_release() before freeing the buffer
wrapper. |
| In the Linux kernel, the following vulnerability has been resolved:
iio: adc: ad_sigma_delta: fix CS held asserted and state leaks
In ad_sigma_delta_single_conversion(), set_mode(AD_SD_MODE_IDLE) and
disable_one() were called from the out: block while keep_cs_asserted
was still true. This caused any SPI transfer issued by those callbacks
to carry cs_change=1, leaving CS permanently asserted after the
conversion. Fix by moving both calls into the out_unlock: block, after
keep_cs_asserted is cleared, matching the pattern already used in
ad_sd_calibrate().
In the error path of ad_sd_buffer_postenable(), if an operation fails
after set_mode(AD_SD_MODE_CONTINUOUS) has already succeeded (e.g.
spi_offload_trigger_enable()), the device is left in continuous
conversion mode with CS physically asserted. Additionally,
bus_locked remaining true after spi_bus_unlock() causes subsequent
SPI operations to call spi_sync_locked() without the bus lock actually
held, allowing concurrent SPI access.
Fix the error path by clearing keep_cs_asserted first, then calling
set_mode(AD_SD_MODE_IDLE) to revert the device mode and deassert CS,
then clearing bus_locked before releasing the bus.
For devices that implement neither set_mode nor disable_one (such as
MAX11205, which has no physical CS pin), no SPI transfer is issued
during cleanup and the cs_change flag has no effect on any physical
line. |
| Uncontrolled resource consumption in Windows Secure Socket Tunneling Protocol (SSTP) allows an unauthorized attacker to deny service over a network. |
| Missing release of memory after effective lifetime in Windows Layer-2 Bridge Network Driver allows an unauthorized attacker to deny service over an adjacent network. |
| Issue summary: A QUIC server that rejects a stream opened by the peer may
retain that stream's state for the lifetime of the connection if the peer
resets the stream before the rejection is signalled, even though the peer's
stream credit is returned.
Impact summary: A remote client that completes a QUIC handshake can repeatedly
open and reset streams, causing server memory to grow without bound until the
process runs out of memory and every connection on it is lost. This is a
Denial of Service.
CWE: CWE-401: Missing Release of Memory after Effective Lifetime
Description: QUIC servers track each stream the peer opens. When the incoming
stream policy rejects a stream, OpenSSL schedules a STOP_SENDING frame, marks
the stream deleted and returns the peer's stream credit. A deleted stream is
collected only once its receive part is fully drained, which for a rejected
stream means the STOP_SENDING has been acknowledged. If the peer sends
RESET_STREAM first, the reset cancels the pending STOP_SENDING, so it is never
sent or acknowledged and the stream is never collected.
The default configuration reaches this path. Once a connection has a default
stream, meaning the application has read from or written to the connection as
the tutorial server does, the effective incoming stream policy rejects streams
the application does not accept. Applications that set the REJECT policy
explicitly are affected too.
Each retained stream costs about 300 bytes, and the peer's stream credit
window paces how quickly they accumulate. A client that keeps the connection
open and keeps opening and resetting streams grows server memory without
bound. The severity is Low because growth is slow, needs a long-lived
connection, and causes no memory corruption or disclosure.
FIPS impact: no
The affected code is outside the FIPS module boundary. |
| In the Linux kernel, the following vulnerability has been resolved:
net: psample: fix info leak in PSAMPLE_ATTR_DATA
psample open codes nla_put() presumably to avoid wiping
the data with 0s just to override it with packet data.
This open coding is missing clearing the pad, however,
each netlink attr is padded to 4B and data_len may
not be divisible by 4B. |
| In the Linux kernel, the following vulnerability has been resolved:
xfrm: move policy_bydst RCU sync from per-netns .exit to .pre_exit
The struct pernet_operations docstring in include/net/net_namespace.h
explicitly warns against blocking RCU primitives in .exit handlers:
Exit methods using blocking RCU primitives, such as
synchronize_rcu(), should be implemented via exit_batch.
[...]
Please, avoid synchronize_rcu() at all, where it's possible.
Note that a combination of pre_exit() and exit() can
be used, since a synchronize_rcu() is guaranteed between
the calls.
xfrm_policy_fini() violates this: it calls synchronize_rcu() before
freeing the policy_bydst hash tables (so no RCU reader is mid-
traversal at free time), but runs from xfrm_net_ops.exit -- once per
namespace -- so a cleanup_net() of N namespaces pays N full RCU
grace periods serially.
Use the documented pre_exit/exit split. Move the policy flush (and
the workqueue drains it depends on) into a new .pre_exit handler;
xfrm_policy_fini() then runs in .exit and frees the hash tables
after the synchronize_rcu_expedited() that cleanup_net() guarantees
between the two phases. Providing O(1) RCU grace periods per batch
instead of O(N).
Observed on Linux 6.18 with a workload doing unshare(CLONE_NEWNET)
at ~13/sec sustained: cleanup_net() and the netns_wq rescuer kthread
both stuck in xfrm_policy_fini()'s synchronize_rcu(), >300k struct
net accumulated in the cleanup queue, Percpu in /proc/meminfo climbed
to 130+ GB on 256-CPU hosts, and memcg OOMs followed. setup_net and
__put_net counts were balanced, ruling out a refcount leak. |
| In affected versions of Progress® Telerik® UI for ASP.NET MVC `2012.2.710` through `2026.3.811`, ClassFactory updates a shared class dictionary without exclusive write protection and emits non-collectible dynamic assemblies that remain loaded for the AppDomain's lifetime. The dictionary lookup and insertion are not atomic, and case variations in aggregate descriptors can produce distinct retained types even though member resolution is case-insensitive. An unauthenticated remote attacker can exploit this if the application exposes an aggregation endpoint that is reachable without authentication, by submitting concurrent or repeated crafted requests. These requests may corrupt dictionary state or increase CPU and memory use, potentially degrading availability. The issue is fixed in version `2026.3.1015`. |
| Missing release of memory after effective lifetime vulnerability in openEuler iSulad on Linux.
This vulnerability is associated with program file src/daemon/modules/spec/specs_mount.c.
This issue affects iSulad: through 2.1.6.
The memory leak vulnerability causes the iSulad daemon to be triggered frequently, exhausting the host machine's memory and ultimately leading to an OOM crash of the daemon, resulting in a denial of service (DoS). |
| Missing release of memory after effective lifetime in the MigTD virtio-serial driver for the Intel(R) TDX Migration Trust Domain (MigTD) within Ring 0: Hypervisor may allow a denial of service. System software adversary with a privileged user combined with a low complexity attack may enable denial of service. This result may potentially occur via local access when attack requirements are not present without special internal knowledge and requires no user interaction. The potential vulnerability may impact the confidentiality (none), integrity (none) and availability (low) of the vulnerable system, resulting in subsequent system confidentiality (none), integrity (none) and availability (none) impacts. |
| Missing release of memory after effective lifetime in Windows Kerberos allows an unauthorized attacker to deny service over a network. |
| Missing release of memory after effective lifetime in Windows DNS allows an unauthorized attacker to deny service over a network. |
| Uncontrolled resource consumption in Windows Internet Information Services allows an unauthorized attacker to deny service over a network. |
| Uncontrolled resource consumption in Windows Hyper-V allows an authorized attacker to deny service locally. |
| FFmpeg through 9.0.2 contains a memory leak vulnerability in mediaformat_jni_setBuffer() in libavcodec/mediacodec_wrapper.c that never frees buffers wrapped with NewDirectByteBuffer(). Attackers can supply crafted media files with large extradata, leaking native memory on each MediaCodec decoder configuration in Android applications and causing resource exhaustion. |
| In the Linux kernel, the following vulnerability has been resolved:
usb: typec: tcpci_rt1711h: unregister TCPCI port with devres
rt1711h_probe() registers the TCPCI port before requesting the interrupt
and enabling alert interrupts. If either of those later steps fails, the
probe function returns without unregistering the TCPCI port. The explicit
unregister currently only happens from the remove callback.
Register a devres action immediately after tcpci_register_port() succeeds,
so tcpci_unregister_port() runs on later probe failures and on driver
detach. Drop the remove callback to avoid unregistering the same port
twice.
This issue was identified during our ongoing static-analysis research while
reviewing kernel code. |
| In the Linux kernel, the following vulnerability has been resolved:
batman-adv: v: prevent OGM aggregation on disabled hardif
When an interface gets disabled, the worker is correctly disabled by
batadv_hardif_disable_interface() -> ... -> batadv_v_ogm_iface_disable().
In this process, the skb aggr_list is also freed.
But batadv_v_ogm_send_meshif() can still queue new skbs (via
batadv_v_ogm_queue_on_if()) to the aggr_list. This will only stop after all
cores can no longer find the RCU protected list of hard interfaces. These
queued skbs will never be freed or consumed by batadv_v_ogm_aggr_work.
The batadv_v_ogm_iface_disable() function must block
batadv_v_ogm_queue_on_if() to avoid leak of skbs. |