Skip to content
VulniPulse
Advisory severityHigh7.5Red Hat Linux

High [CVE-2026-59204] Denial of Service via crafted JPEG2000 image

This high-severity Red Hat Linux advisory covers CVE-2026-59204 affecting Red Hat AI Inference Server 3.2, Red Hat AI Inference Server 3.3, Red Hat AI Inference Server 3.4.

Aggregated and source-linked by VulniPulse. Data sources, validation and limitations.

CVE-2026-59204 Source published Source updated

VulniPulse record published Record updated

Affected products & platforms
Red Hat LinuxUnclassified
Open source advisory

Android app · Google Play

Monitor future Red Hat Linux CVEs from your phone.

Choose a whole vendor or a precise platform, then receive matching security advisories by phone notification, email, or both. Coverage follows 34 official vendor sources and 160+ reviewed platform categories.

Matching phone alertsOptional email delivery

Summary

Pillow is a Python imaging library. From 8.2.0 through 12.2.0, src/libImaging/Jpeg2KDecode.c accumulates total_component_width across every tile in a JPEG2000 image instead of recomputing it per tile, allowing a crafted tiled JPEG2000 file to force substantially higher transient memory usage and trigger out-of-memory failures during decoding.

This issue is fixed in version 12.3.0. A remote attacker could exploit this vulnerability by providing a specially crafted JPEG2000 image file.

Due to incorrect calculation of memory requirements for image tiles, processing this file can lead to excessive memory consumption, resulting in a denial of service (DoS) through out-of-memory failures. The impact is limited to denial of service via excessive memory allocation.

A crafted JPEG2000 file can trigger out-of-memory failures, but this does not lead to code execution or information disclosure. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H).

Weakness: CWE-131.

Affected products named by the advisory: Red Hat AI Inference Server 3.2; Red Hat AI Inference Server 3.3; Red Hat AI Inference Server 3.4; Red Hat Enterprise Linux AI 3.3; and 13 more.

Affected versions
  • 8.2.0
  • 12.2.0

Official advisory · high-confidence parse· fetched 18 days ago·verify at source

Fixed versions
  • 12.3.0
  • rhaiis/vllm-cuda-rhel9:1787860580
  • rhaiis/model-opt-cuda-rhel9:1787772157
  • rhaiis/vllm-rocm-rhel9:1787884873
  • rhaiis/model-opt-cuda-rhel9:1787601159
  • rhaii/vllm-spyre-rhel9:1789681201
  • rhaii/vllm-cpu-rhel9:1789681128
  • rhaii/vllm-cuda-rhel9:1789681126
  • rhaii/model-opt-cuda-rhel9:1789681126
  • rhaii/vllm-rocm-rhel9:1789681126
  • rhaii/vllm-cpu-rhel9:1790075793
  • rhaii/vllm-spyre-rhel9:1790076141
  • rhaii/vllm-cuda-rhel9:1790090131
  • rhaii/vllm-rocm-rhel9:1790109620
  • rhaii/model-opt-cuda-rhel9:1790076498
  • rhelai3/disk-image-cuda-rhel9:1788290314
  • rhelai3/bootc-aws-cuda-rhel9:1788273908
  • rhelai3/bootc-azure-cuda-rhel9:1788273909
  • rhelai3/bootc-cuda-rhel9:1788260684
  • rhelai3/bootc-gcp-cuda-rhel9:1788273977
  • rhoai/odh-llm-d-inference-scheduler-rhel9:1789662430
  • rhoai/odh-mlflow-rhel9:1789653606
  • rhoai/odh-automl-rhel9:1786612415
  • rhoai/odh-autorag-rhel9:1786612637
  • rhoai/odh-llm-d-kv-cache-rhel9:1787169432
  • rhoai/odh-mlflow-rhel9:1787226790
  • rhoai/odh-openvino-model-server-rhel9:1787303251
  • rhoai/odh-pipeline-runtime-datascience-cpu-py312-rhel9:1787073866
  • rhoai/odh-pipeline-runtime-pytorch-cuda-py312-rhel9:1787073873
  • rhoai/odh-pipeline-runtime-pytorch-llmcompressor-cuda-py312-rhel9:1787073459
  • rhoai/odh-pipeline-runtime-pytorch-rocm-py312-rhel9:1787073611
  • rhoai/odh-pipeline-runtime-tensorflow-cuda-py312-rhel9:1787073451
  • rhoai/odh-pipeline-runtime-tensorflow-rocm-py312-rhel9:1787073451
  • rhoai/odh-th06-cpu-torch210-py312-rhel9:1787076778
  • rhoai/odh-th06-cuda130-torch210-py312-rhel9:1787077779
  • rhoai/odh-th06-rocm64-torch291-py312-rhel9:1787076481
  • rhoai/odh-training-cuda128-torch29-py312-rhel9:1786611803
  • rhoai/odh-training-rocm64-torch29-py312-rhel9:1786611435
  • rhoai/odh-trustyai-garak-lls-provider-dsp-rhel9:1786613209
  • rhoai/odh-workbench-codeserver-datascience-cpu-py312-rhel9:1787121387
  • rhoai/odh-workbench-jupyter-datascience-cpu-py312-rhel9:1787074331
  • RHSA-2026:61627
  • RHSA-2026:61628
  • RHSA-2026:61629
  • RHSA-2026:59518
  • RHSA-2026:69464
  • RHSA-2026:69466
  • RHSA-2026:69467
  • RHSA-2026:69468
  • RHSA-2026:69469

Official advisory · high-confidence parse· fetched 18 days ago·verify at source

Mitigation checklist

Recommended fix / mitigation
  • If the application does not need JPEG2000 support, block .jp2, .j2k, .jpf, and .jpx uploads at the input layer. For services that do process JPEG2000, set memory limits on the process or container (LimitAS= in systemd, or memory limits in Kubernetes/Podman) so a crafted image can only crash the worker, not the whole host. Add automatic restarts (Restart=always in systemd, or container restart policies) so the service recovers from OOM kills without someone having to intervene.

Official advisory · high-confidence parse· fetched 18 days ago·verify at source

Discussion(0)

No comments yet. Share field notes, upgrade gotchas, or questions — verify against the vendor advisory before acting on community advice.

Sign in to join the discussion.