Skip to content
VulniPulse
High7.5Red Hat Linux

High [CVE-2026-14362] HashiCorp memberlist: Denial of Service via push/pull state handling

This high-severity Red Hat Linux advisory covers CVE-2026-14362 affecting Red Hat Advanced Cluster Management for Kubernetes 2.15, Red Hat Advanced Cluster Management for Kubernetes 2.16, Red Hat OpenShift Container Platform 4.21.

CVE-2026-14362 Published Jul 8, 2026Updated by vendor Jul 8, 2026
Affected products & platforms
Red Hat LinuxUnclassified
Open vendor 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 32 official vendor sources and 160+ reviewed platform categories.

Matching phone alertsOptional email delivery

Summary

HashiCorp memberlist before version 0.6.0 is vulnerable to a denial-of-service issue in its push/pull state handling that may allow an attacker with network access to the gossip port to exhaust memory on a receiving node and cause the process to terminate. This vulnerability (CVE-2026-14362) is fixed in memberlist 0.6.0.

This flaw results in a Denial of Service (DoS). Only Red Hat products that run the affected component as a live memberlist gossip agent (a process that binds and listens on a UDP/TCP gossip port for cluster membership) are exposed to this denial-of-service issue.

Products that pull in github.com/hashicorp/memberlist only as a transitive Go dependency, without starting a gossip listener, do not exercise the vulnerable push/pull state-handling code path and are not affected.

Grafana links github.com/hashicorp/memberlist solely for its optional unified-alerting high-availability (HA) clustering feature; the gossip listener only starts when unified_alerting.ha_peers (plus ha_listen_address/ha_advertise_address) is explicitly configured.

Red Hat Advanced Cluster Management for Kubernetes ships its acm-grafana component with unified alerting disabled by default (unified_alerting.enabled = false) and never sets ha_peers, so it never starts the gossip listener and does not exercise the vulnerable code path; it is not affected.

Affected versions
  • < 0.6.0

Official advisory · high-confidence parse· fetched 12 days ago·verify at source

Fixed versions
  • 0.6.0
  • rhacm2/prometheus-alertmanager-rhel9:1786547824
  • rhacm2/prometheus-alertmanager-rhel9:1786909156
  • openshift4/ose-prometheus-alertmanager-rhel9:1786500210
  • openshift4/ose-prometheus-alertmanager-rhel9:1786584932
  • RHSA-2026:54427
  • RHSA-2026:57191
  • RHSA-2026:54602
  • RHSA-2026:54770

Official advisory · high-confidence parse· fetched 12 days ago·verify at source

Mitigation checklist

Recommended fix / mitigation
  • Upgrade github.com/hashicorp/memberlist to version 0.6.0 or later, which fixes the push/pull state handling issue. As a temporary mitigation, restrict network access to the gossip port (UDP/TCP, commonly 7946 or 9094) to trusted cluster members only, e.g. via network policy, firewall rules, or security groups, since the flaw requires network access to the gossip listener to trigger memory exhaustion.

Official advisory · high-confidence parse· fetched 12 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.