Skip to content
NewA calmer, faster ZeroHour
Advisorymedium

USN-8820-1: curl vulnerabilities

Ubuntu Security Notices ()3 · 1 read
AI summary · grok-4.7Summaries by Vani AI

Ubuntu patched curl flaws that may bypass LDAP peer checks or allow HTTP/2 crashes and code execution.

Ubuntu Security Notice USN-8820-1 covers multiple curl vulnerabilities. CVE-2026-13608, found by Eunsoo Kim, involves incorrect SASL negotiation during LDAP authentication; a machine-in-the-middle attacker could bypass peer validation, and it affects only Ubuntu 24.04 LTS and 26.04 LTS. CVE-2026-18924, found by Stephan Zeisberg, concerns mishandled HTTP/2 server-push streams on shared connections, which a remote attacker could use to crash curl or possibly execute arbitrary code. Stanislav Fort also reported incorrect lifetime management of pooled TLS connections when using curl's multi interface; the published notice text is truncated.

Key facts
  • CVE-2026-13608: LDAP SASL handling may let a MITM bypass peer validation.
  • That issue affects only Ubuntu 24.04 LTS and Ubuntu 26.04 LTS.
  • CVE-2026-18924: HTTP/2 server push on shared connections may crash curl or allow code execution.
  • Stanislav Fort reported a pooled TLS connection lifetime bug in curl's multi interface.
Analyze CVE-2026-13608 with Vani AIImpact, affected versions, exploitation, mitigations and detection

Vulnerabilities mentioned

All
CVEVulnerabilityCVSSEPSSSignalsPublished
CVE-2026-18924
+1 related CVE
Use-After-Free in libcurl HTTP/2 Server Push Handling with Shared Connections
9.1
group max
<1%PoC

From the original article

Eunsoo Kim discovered that curl incorrectly handled SASL negotiation for LDAP authentication in certain circumstances. A machine-in-the-middle attacker could possibly use this issue to bypass peer validation. This issue only affected Ubuntu 24.

Ubuntu Security Notices

ubuntu.com

Read the full article at Ubuntu Security Notices

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.