Skip to content
KitploitKITPLOIT
HerramientasExploitsBlog
Log in
Enviar
HerramientasExploitsBlog
Enviar

¡Herramientas de Hacking, PenTest y Ciberseguridad para tu Arsenal de Seguridad!

Kitploit es un directorio de herramientas de hacking, ciberseguridad y pentesting. Descubre las últimas actualizaciones de proyectos para encontrar vulnerabilidades, analizar sistemas, automatizar pruebas y fortalecer tu seguridad.

FeedsContactoPrivacidad© 2026 Kitploit

Directorio de Herramientas

Categorías

Ver todas las categorías
Loading categories
Herramientas/GitHubGitHub/ersinkoc/sshole-check
Defensive ToolsVulnerability ScannersContainer SecurityConfiguration AuditingPenetration TestingAuthenticationMisconfigurationIncident ResponseLog Analysis
GitHubersinkoc/sshole-check

sshole-check

31hace 14 díasAún no revisado
Compartir

Audit & interim mitigation for CVE-2026-123158 (*SSHole) — OpenSSH password bypass on musl/Alpine — across Docker containers, images and host

Ver Repositorio
Contenido no disponible en el idioma solicitado. Mostrando versión en inglés.

sshole-check

Audit and interim-mitigation tool for CVE-2026-123158 ("*SSHole"), an OpenSSH password authentication bypass on musl-based Linux (Alpine). It covers your running Docker containers, local images and the host.

Example output (abridged, from a test machine):

$ sudo ./sshole-check.sh --images
CVE-2026-123158 (*SSHole) audit · sshole-check 1.0.0 · 2026-11-05 22:45:52 UTC

[VULN  ] container sshole-real ghcr.io/requarks/wiki:2.5
                   OpenSSH 10.3p1 · openssh-server-10.3_p1-r1 openssh-server-common-10.3_p1-r1 · alpine 3.24.3
                   PasswordAuthentication=yes · PAM=off · PermitRootLogin=prohibit-password · port: 22
                   '*' accounts: root carol · $5$/$6$ hashed: bob
                   sshd running: yes · published: 22→0.0.0.0:33776 [::]:33776
                   → unpatched (package older than 10.3_p1-r2, the Alpine 3.24 fix) and all conditions are met
                   · root is '*' but PermitRootLogin=prohibit-password, root cannot log in with a password
                   · sshd is running and published to the outside
[WARN  ] image     docker.gitea.com/gitea:1.24.4 dd29b53ab011
                   OpenSSH 10.0p2 · openssh-server-10.0_p1-r7 openssh-server-common-10.0_p1-r7 · alpine 3.22.1
                   PasswordAuthentication=yes · PAM=off · PermitRootLogin=without-password · port: 22
                   '*' accounts: root git · $5$/$6$ hashed: -
                   EXPOSE 22/tcp 3000/tcp
                   → unpatched (package older than 10.0_p1-r11, the Alpine 3.22 fix); conditions not met right now, becomes exploitable once an account gets a password
[OK    ] container postgres postgres:16-alpine
                   no sshd

Summary: OK=339 WARN=8 VULN=1 BREACH=0 SKIPPED=11

It never tries to log in. The probe only reads the sshd binary, sshd -T / sshd_config, /etc/shadow, the apk database and log files inside each target. Read the script before running it; it is a single file of plain bash and POSIX sh.


Quick start

curl -fsSLO https://raw.githubusercontent.com/ersinkoc/sshole-check/main/sshole-check.sh
less sshole-check.sh            # read it first
chmod +x sshole-check.sh
sudo ./sshole-check.sh          # running containers + host
sudo ./sshole-check.sh --images # also every local image

Requirements: bash on the machine that runs it, plus access to the Docker (or Podman) CLI. The targets only need /bin/sh. Images with no shell, such as distroless or scratch images, are reported as SKIP; they cannot run sshd anyway.


About CVE-2026-123158

Name*SSHole / "sshhole"
ComponentOpenSSH sshd, password authentication without PAM
PlatformLinux + musl libc (Alpine, also OpenWrt / Buildroot / Yocto builds)
Fixed inOpenSSH 10.6p2 (2026-11-05), plus Alpine backports to every branch from 3.16 on
CVSS 3.18.1 High, AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
CWECWE-394 (Unexpected Status Code or Return Value)
GHSAGHSA-4fcc-vm97-gx3x
Reported byÖzgün Kültekin (Wakefield)

What goes wrong

Three behaviours that are each reasonable on their own add up to a login bypass:

  1. musl crypt() returns "*" on failure. Since 2012 musl returns a failure string instead of NULL. Its SHA-crypt implementation refuses keys longer than 256 bytes (KEY_MAX, a DoS guard), so a password longer than 256 bytes hashes to the literal string *.
  2. OpenSSH hashes the password even when the account has no usable hash. Since the CVE-2016-6210 user-enumeration fix, when the stored password field is too short to contain a salt, sshd borrows the salt and settings of another account's real $5$/$6$ hash so that timing looks the same. That is why a second, password-protected account is needed.
  3. Alpine base images ship root:*. Since Alpine 3.16 (2022), * is the "no password, keys only" marker for root. The same marker is used for key-only users, directory-service users (nslcd, sssd, winbind) and the alpine user in tiny-cloud images.

Without PAM, sshd compares crypt(password, borrowed_salt) with the stored field. For a 257 to 1024 byte password that is "*" == "*", and the login is accepted.

OpenSSH 10.6p2 works around it by rejecting stored hashes that are too short to be valid. The underlying behaviour in musl is unchanged; the fix is in sshd.

You are vulnerable only if all of these hold

  • sshd is OpenSSH < 10.6p2 without the Alpine backport, linked against musl
  • Password checks don't go through PAM: the openssh-server package, or openssh-server-pam with UsePAM no
  • PasswordAuthentication yes (the default)
  • At least one account whose shadow password field is exactly * (a ! lock is not affected)
  • At least one other local account with a $5$ or $6$ hash
  • The attacker can reach the SSH port

Root itself is only reachable with PermitRootLogin yes; the default prohibit-password blocks password logins for root. On Alpine's AWS/Azure tiny-cloud images, though, the * user alpine has passwordless doas, so getting in as alpine means getting root.

Not affected: glibc distributions (Debian, Ubuntu, RHEL, Fedora …), UsePAM yes, PasswordAuthentication no, containers that don't run sshd, Dropbear, and OpenBSD's sshd.

Fixed versions

Alpine branchFirst fixed opensshFixed point release
edge10.6_p2-r0–
3.2410.3_p1-r23.24.4
3.2310.2_p1-r13.23.8
3.2210.0_p1-r113.22.8
3.219.9_p2-r13.21.10
3.209.7_p1-r6(EOL, package only)
3.199.6_p1-r3(EOL, package only)
3.189.3_p2-r4(EOL, package only)
3.179.1_p1-r7(EOL, package only)
3.169.0_p1-r6(EOL, package only)

Alpine backported the fix, so most branches keep their old upstream version: a fixed 3.24 box still reports OpenSSH_10.3p1. sshd -V alone cannot tell you whether you are patched. That is why this tool compares the apk package release against the table above.


Why Docker is a special case

  • apk upgrade on a running container is lost when the container is recreated. The image is what you have to fix.
  • Base images pinned to an old tag (alpine:3.22.1, FROM node:20-alpine@sha256:…) keep the vulnerable package forever, and a rebuild with cache can reuse the old apk add layer.
  • An image can be "not exploitable right now" (no $6$ account yet) and still become exploitable as soon as an entrypoint script runs chpasswd. sshole-check reports this as WARN.
  • A container on --network host, or with -p 22:22 / -p 2222:22, puts the bug on the network. sshole-check shows the published mapping for each container.

What it checks

For every target it runs a small POSIX sh probe inside it. Containers are probed with docker exec -u 0. Images are started with docker run --rm --network none --entrypoint sh -u 0. The host is probed directly.

Descargar herramienta

Más Populares

Ver todos →

Descubre las herramientas más usadas por nuestra comunidad.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 El kit de herramientas OSINT definitivo: grafo de investigación, puntuación de confianza, familias de alias, mapa, línea de tiempo e informes. 49 módulos, más de 400 sitios, sin necesidad de claves API.

    1.0k
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    # Aplicación de Ejecución con un Solo Toque GhostLock (CVE-2026-43499)

    593
  • 3

    looking-glass

    anonghosty/looking-glass

    Plataforma CSIRT que ingiere informes de Shadowserver en MongoDB y los reproduce como mapas interactivos, análisis, hallazgos, avisos y paneles ejecutivos.

    371
  • 4

    redforge

    cybersecurity91/redforge

    Kit de herramientas de red team del lado del navegador con 102 herramientas offline para reconocimiento, generación de payloads, ataques de credenciales, criptografía, codificación y elaboración de informes de compromiso.

    336
  • 5

    ARTEX

    baarai/artex

    Plataforma de pruebas de penetración autónoma multiagente basada en LLM con agentes planificadores/trabajadores, grafos de activos y de exploración, registro de tráfico MITM y reevaluación de vulnerabilidades.

    326
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    El sistema de acción para la ciberseguridad nativa de IA—donde la intención se convierte en ejecución gobernada, la evidencia se convierte en memoria operativa y cada operación mejora la siguiente.

    322

Explora todas las herramientas

Explora nuestra colección de herramientas

Ver todas las herramientas →