CVE-2026-18313
CBL Mariner vulnerability analysis and mitigation

Overview

CVE-2026-18313 is a memory leak vulnerability in the rpcapd daemon component of libpcap (The Tcpdump Group) that allows an authenticated network attacker to accelerate memory exhaustion and cause a denial of service. The daemon_unpackapplyfilter() function allocates up to 65,536 bytes per received RPCAP_MSG_UPDATEFILTER_REQ or RPCAP_MSG_STARTCAP_REQ message but never frees the allocated buffer, causing memory leaks even under normal use. Affected versions include libpcap 1.9.x and libpcap 1.10.x prior to 1.10.7; version 1.10.7 contains the fix. It carries a CVSS v3.1 base score of 4.3 (Medium) (GitHub Advisory, Feedly). The vulnerability was publicly disclosed and patched on September 5, 2026.

Technical details

The root cause is classified as CWE-401 (Missing Release of Memory after Effective Lifetime). Within rpcapd/daemon.c, the function daemon_unpackapplyfilter() calls malloc() to allocate a temporary buffer for up to RPCAP_BPF_MAXINSNS (8,192) BPF instructions (totaling up to 65,536 bytes) each time an RPCAP_MSG_UPDATEFILTER_REQ or RPCAP_MSG_STARTCAP_REQ message is processed, but the buffer is never freed on any code path — including both successful and some failed validation paths. The fix introduces a free_and_return_status label that ensures free(bf_prog.bf_insns) is called exactly once after a successful malloc() return, regardless of the exit path (libpcap commit). Exploitation requires low-privilege authenticated access to the rpcapd service over the network, with no user interaction needed (GitHub Advisory).

Impact

Successful exploitation results in progressive memory exhaustion on the rpcapd server, ultimately degrading or crashing the packet capture service (availability impact). There is no confidentiality or integrity impact — the vulnerability is limited to a denial-of-service condition. Because rpcapd is used for remote packet capture, its unavailability can disrupt network monitoring and forensic capabilities on affected systems (GitHub Advisory, Feedly).

Exploitability

There is no public proof-of-concept exploit and no evidence of in-the-wild exploitation as of the disclosure date (Feedly). The EPSS score is 0.0, indicating a very low probability of exploitation in the near term. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. Exploitation requires authenticated access to the rpcapd service, which limits the attack surface compared to unauthenticated vulnerabilities.

Exploitation steps

  1. Reconnaissance: Identify hosts running rpcapd (typically on TCP port 2002) using network scanning tools such as Nmap. Confirm the libpcap version is 1.9.x or 1.10.x < 1.10.7.
  2. Authenticate: Obtain valid credentials for the rpcapd service (rpcapd supports both null and password authentication depending on configuration).
  3. Send repeated filter update messages: Using a custom RPCAP client or modified libpcap tooling, repeatedly send RPCAP_MSG_UPDATEFILTER_REQ or RPCAP_MSG_STARTCAP_REQ messages to the server. Each message causes the server to allocate up to 65,536 bytes that are never freed.
  4. Exhaust server memory: Continue sending messages at a high rate to accelerate memory consumption beyond what normal use would cause, eventually exhausting available system memory.
  5. Achieve denial of service: The rpcapd process becomes unresponsive or crashes due to memory exhaustion, disrupting remote packet capture services (libpcap commit, GitHub Advisory).

Indicators of compromise

  • Network: Unusually high volume of RPCAP_MSG_UPDATEFILTER_REQ or RPCAP_MSG_STARTCAP_REQ messages from a single client IP to the rpcapd port (default TCP 2002).
  • Process: Steadily increasing memory consumption of the rpcapd process observable via system monitoring tools (e.g., top, ps, /proc/<pid>/status); process crash or OOM-killer events involving rpcapd.
  • Logs: System logs (e.g., /var/log/syslog, /var/log/messages) showing OOM (Out of Memory) killer events or rpcapd process termination; Valgrind-style memory leak reports referencing daemon_unpackapplyfilter in daemon.c:2372 if diagnostic builds are in use (libpcap commit).

Mitigation and workarounds

Upgrade libpcap to version 1.10.7 or later, which contains the fix backported from commit 26a1c75 and applied in commit f9775af (libpcap commit). As interim workarounds, consider implementing rate limiting on RPCAP_MSG_UPDATEFILTER_REQ and RPCAP_MSG_STARTCAP_REQ messages at the network or application level, restricting rpcapd access to trusted clients only via firewall rules, and configuring process restart policies (e.g., systemd Restart=on-failure) to automatically recover from crashes caused by memory exhaustion (Feedly).

Additional resources

Linux Distribution fix status

Fix availability across major Linux distributions and their releases.

Debian

Fixed

bookworm

libpcap

Affected

sid

libpcap: 1.10.7-1

Fixed

trixie

libpcap

Affected

Ubuntu

Unknown

bionic (esm-infra)

libpcap

Unknown

devel

libpcap

Unknown

focal (esm-infra)

libpcap

Unknown

jammy

libpcap

Unknown

noble

libpcap

Unknown

resolute

libpcap

Unknown

trusty (esm-infra-legacy)

libpcap

Unknown

xenial (esm-infra-legacy)

libpcap

Unknown

RHEL / CentOS

Affected

OpenShift

openshift/ose-rhel-coreos-8

Affected

RHEL 8

libpcap.src

Affected

RHEL 9

libpcap.src

Affected

RHEL 10

libpcap.src

Affected

SourceThis report was generated using AI

Related CBL Mariner vulnerabilities:

CVE ID

Severity

Score

Technologies

Component name

CISA KEV exploit

Has fix

Published date

CVE-2026-6554MEDIUM5.5
  • CBL Mariner logoCBL Mariner
  • libpcap.src
NoYesSep 05, 2026
CVE-2026-6244MEDIUM5.5
  • CBL Mariner logoCBL Mariner
  • libpcap
NoYesSep 05, 2026
CVE-2026-31912MEDIUM5.5
  • CBL Mariner logoCBL Mariner
  • libpcap-devel
NoYesSep 05, 2026
CVE-2026-31911MEDIUM5.5
  • CBL Mariner logoCBL Mariner
  • libpcap
NoYesSep 05, 2026
CVE-2026-18313MEDIUM4.3
  • CBL Mariner logoCBL Mariner
  • libpcap
NoYesSep 05, 2026

Free Vulnerability Assessment

Benchmark your Cloud Security Posture

Evaluate your cloud security practices across 9 security domains to benchmark your risk level and identify gaps in your defenses.

Request assessment

Get a personalized demo

Ready to see Wiz in action?

"Best User Experience I have ever seen, provides full visibility to cloud workloads."
David EstlickCISO
"Wiz provides a single pane of glass to see what is going on in our cloud environments."
Adam FletcherChief Security Officer
"We know that if Wiz identifies something as critical, it actually is."
Greg PoniatowskiHead of Threat and Vulnerability Management