Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessIdentify published TCP port on container host
-
DeliveryEstablish multiple slow TCP connections
-
ExploitTransmit data at minimal rate to stall forwarding proxy
-
ExecutionProxy buffers unbounded client data in host memory
-
PersistHost memory exhausted
-
ImpactContainer runtime and hosted services become unresponsive
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A06:2025 - Insecure Design
ASVS 5.0
V2 - Validation and Business Logic Resource allocation limits
MITRE ATT&CK
1 technique across 1 tactic: Impact The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Exploitation requires that the target container has at least one TCP port published and reachable by the attacker - containers with no published ports are entirely unaffected. The attacker must be able to establish TCP connections to the published port, which in typical deployments means either internet access to a publicly exposed port or network access within a private segment. No authentication is required (CVSS PR:N). No user interaction is needed. There is no requirement for a specific application running inside the container - the vulnerability is in the host-side forwarding proxy layer, not the containerized application. Exploitation is limited to denial-of-service against host memory; there is no confidentiality or integrity impact. The vulnerability does not affect containers using host networking or other network modes that bypass the published-port forwarding proxy. |
| Impact | CVSS 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) accurately reflects the network-reachable, unauthenticated, low-complexity nature of this DoS. However, real-world risk is moderated by deployment context: the vulnerability is only exploitable if the container has published TCP ports accessible to the attacker - containers without published ports are not affected. EPSS of 0.14% at the 4th percentile signals very low observed or predicted exploitation activity, consistent with a niche container runtime and a pure DoS class that attracts fewer opportunistic attackers than RCE or credential-theft vulnerabilities. There is no CISA KEV listing and no POC is known. The primary realistic threat is against internet-exposed containerized services running on Apple's container runtime - a relatively narrow deployment scenario. The combination of high CVSS and very low EPSS is consistent with a theoretically straightforward but currently unweaponized vulnerability in a product with limited enterprise footprint. |
| Exploit ScenarioDerived | An attacker identifies an internet-facing host running Apple's container runtime with one or more published TCP ports serving a containerized application. The attacker opens many simultaneous connections to a published port and begins transmitting data very slowly, artificially prolonging the forwarding proxy's buffer-accumulation phase by either throttling sends or interfering with backend connection establishment timing. Over time, the host's forwarding process accumulates gigabytes of buffered data across these connections, exhausting available RAM and causing the host OS to kill processes or become unresponsive, resulting in a denial of service for all hosted containers. No authentication or exploit code is required - the attack is a deliberate slow-client pattern. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 28, 2026
Unbounded memory buffering in Apple's container runtime exposes host systems running versions prior to 1.2.0 to remote denial-of-service attacks via any published TCP port. The forwarding proxy that bridges client connections to backend containers accumulates client-supplied data without any cap on size or time, meaning a remote unauthenticated attacker can deliberately hold a connection open - sending data slowly or stalling backend establishment - until host memory is exhausted. No public exploit has been identified at time of analysis and EPSS is very low (0.14%, 4th percentile), but the attack is conceptually simple and requires no authentication.
RemediationAI
Upgrade to Apple container version 1.2.0 or later, which is the vendor-released patch addressing this vulnerability per the advisory at https://github.com/apple/container/security/advisories/GHSA-wg28-286f-56v6. If immediate upgrade is not feasible, a targeted compensating control is to remove or restrict published TCP port mappings for containers that do not strictly require external network exposure, as the vulnerability is only reachable via published ports. Additionally, deploying a reverse proxy or load balancer in front of published ports that enforces per-connection read timeouts and maximum request body sizes (e.g., nginx proxy_read_timeout and client_max_body_size directives) can prevent unbounded buffering at the network boundary without modifying the container runtime itself - note this adds a proxy component that must itself be kept patched. Rate-limiting inbound connections at the network perimeter (firewall or cloud security group rules) reduces but does not eliminate the risk.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Apple's container runtime (cpe:2.3:a:apple:container:*:*:*:*:*:*:*:*) implements TCP port publishing by operating a userspace forwarding proxy on the host that accepts inbound connections on published ports and relays traffic to the corresponding container endpoint. CWE-770 (Allocation of Resources Without Limits or Throttling) identifies the root cause: the proxy's accept-and-forward logic buffers inbound client data in memory while the backend container connection is being established, but imposes no limit on either the volume of buffered bytes or the duration the buffering phase is allowed to run. An attacker who can reach the published port can exploit the absence of a read deadline and a buffer ceiling to cause the host process to hold arbitrarily large buffers for arbitrarily long periods, consuming host RAM proportional to connection count and data rate.
Know when Container gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Network-reachable published port, no authentication required, no user interaction; impact is exclusively host memory exhaustion with no confidentiality or integrity effect.
Primary rating from Vendor (apple).
CVSS VectorVendor: apple
Lifecycle Timeline
4DescriptionCVE.org
An attacker that can reach a container's published TCP port may be able to force the host's forwarding process to buffer an unbounded amount of that client's data in memory, for as long as the backend container connection takes to complete - with no cap on how much accumulates or how long the wait can be stretched. This vulnerability is addressed in container version 1.2.0.
Executive ViewAI
Apple's container runtime versions prior to 1.2.0 contain a vulnerability enabling remote denial-of-service attacks without authentication, potentially exhausting system memory and disrupting containerized services. Organizations using the affected runtime with exposed container ports face immediate availability risk.
Apply vendor patch.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.
Path traversal in the ArchiveReader.extractContents() function used by container image load operations allows local atta
Path traversal in Apple's open-source container runtime (versions before 1.2.0) enables a malicious builder peer to requ
An attacker who controls the registry, a mirror, or the network path to it may be able to serve arbitrary image content
If an attacker can write to the filesystem of a container k8s cluster container, they can plant an exec credential plugi
EUVD-2026-63666
NVD
CVE.org
EPSS
GitHub
Exploit-DB