Skip to content

Topic

Container Security

Securing container images and runtimes, including image composition, hardening and vulnerability management.

Current stories

buildOne report1 publisher

Docker and Kubernetes default seccomp profiles allow the AF_UNIX socket call behind CVE-2026-80521

DepthFirst published a working Ubuntu 26.04 exploit for CVE-2026-80521 on 2026-09-22, before distribution kernels carried a fix, according to a dev.to post. The flaw sits in AF_UNIX, which default Docker and Kubernetes seccomp profiles permit.

Publishers:dev.to

Reality

Evidence40
Adoption
Not scored
Hype gap
Not scored
Incentives
Not scored
Confidence45
buildOne report1 publisher

CVE-2026-40281 lets a newline in a Gotenberg metadata value move files as root

Gotenberg's 8.30.1 patch validated metadata keys but not values, so CVE-2026-40281 lets unauthenticated callers move files as root. The dev.to write-up of the bug does not name a fixed release, so operators have no confirmed version to move to.

Publishers:dev.to

Reality

Evidence58
Adoption
Not scored
Hype gap
Not scored
Incentives
Not scored
Confidence62
buildOne report1 publisher

Kubernetes stops StatefulSet rollbacks from carrying a tenant's pod into another namespace

Kubernetes fixed CVE-2026-2270 on September 23. The StatefulSet controller bug let a user confined to one namespace get a pod created in another. Shared clusters need the upgrade and a review of cluster-wide controllers that copy fields from objects tenants can write.

Publishers:dev.to

Reality

Evidence55
Adoption
Not scored
Hype gap+5
Incentives
Not scored
Confidence60
securityOne report1 publisher

EU's Cyber Resilience Act puts Helm chart and Kubernetes operator vendors on a 24-hour exploit clock

EU Cyber Resilience Act rules have required 24-hour ENISA warnings on exploited flaws in commercial container images since Sept. 11, 2026. Vendors of supported Kubernetes operators and Helm charts are on the same clock, well before the rest of the law is enforced in December 2027.

Reality

Evidence50
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence55
securityOne report1 publisher

Carbonato worms across unauthenticated Docker hosts to plant a Telegram-run AI agent

ThreatDown says the Carbonato worm breaks into Docker daemons open on port 2375 and installs the open-source Hermes AI agent, then rescans nearby networks every five minutes to spread. An operator drives each infected host over Telegram.

Reality

Evidence50
Adoption
Not scored
Hype gap+10
Incentives35
Confidence50
securityConfirmed3 publishers

CARBONATO botnet infects Docker hosts, prioritizing theft of AI provider API keys

CARBONATO, a Docker botnet running since October 2024, hijacks hosts on open port 2375 and steals keys from 14 AI providers to fund its own LLM gateway. ThreatDown found the crew's own container registry exposed, handing defenders 4.3 GB of its toolchain.

Perspective Coverage

3 publishers
Builder
Builder 33%
Operator
Operator 57%
Investor
Investor 10%

Reality

Evidence60
Adoption
Not scored
Hype gap+20
Incentives35
Confidence60
buildOne report1 publisher

Docker's alpine cold start swings fourfold between two 2026 benchmark writeups

Two 2026 writeups measured Docker's alpine cold start at 0.31 and 1.23 seconds, a fourfold spread on the same test. The rootless networking backend a host runs moves throughput from about 40 Gbps to 3.

Publishers:dev.to

Reality

Evidence40
Adoption38
Hype gap+12
Incentives
Not scored
Confidence45
buildOne report1 publisher

Docker 29.8 extends the container umask to the execs the entrypoint trick never reached

Engine 29.8.0's --umask flag sets one mask for the main process, execs and healthchecks. A side-by-side test on a second daemon also shows it accepting a four-digit value and applying only the last three.

Publishers:dev.to

Reality

Evidence66
Adoption25
Hype gap+8
Incentives22
Confidence58
buildOne report1 publisher

Red Hat's interim mitigations cover two of the three kernel flaws CISA lists as exploited

CISA has confirmed exploitation of three Linux kernel flaws without publishing how, and the three entries do not triage the same way, because Red Hat's interim advice covers kTLS and ebtables but not the AF_ALG race.

Publishers:dev.to

Reality

Evidence58
Adoption35
Hype gap+12
Incentives45
Confidence52
securityConfirmed5 publishers

CISA gives agencies one business day to patch three exploited Linux kernel flaws

The three kernel CVEs CISA added to its exploited-bugs catalog on Friday all need local access, and the lowest-scored of them is the one STAR Labs used for privilege escalation and container escape. Red Hat has confirmed public exploit code.

Perspective Coverage

6 publishers
Builder
Builder 30%
Operator
Operator 57%
Investor
Investor 13%

Reality

Evidence74
Adoption68
Hype gap−8
Incentives38
Confidence76
buildOne report1 publisher

Chromium's user-namespace sandbox survives the hardening that kills its setuid helper

PDFik dropped --no-sandbox from its Playwright workers and kept rendering customer-submitted HTML in non-root, capability-stripped pods. Its own security auditor later wrote that the flag was mandatory.

Publishers:dev.to

Reality

Evidence58
Adoption25
Hype gap−5
Incentives60
Confidence52
buildOne report1 publisher

Buildpacks move the base image choice out of every application repository

In an account from The New Stack, a patched runtime image still misses production a week later because every service builds its own way, and buildpacks are offered as the shared path that closes the gap.

Publishers:thenewstack.io

Reality

Evidence32
Adoption
Not scored
Hype gap+28
Incentives
Not scored
Confidence44
buildOne report1 publisher

xz refuses to unpack a PHP extension when Docker's seccomp answers Landlock with EACCES

A dev.to write-up traces a build that fails only on enterprise-kernel servers to three correct components meeting badly. The seccomp profile that fixes it stays inert until the build runs on Docker's classic engine.

Publishers:dev.to

Reality

Evidence50
Adoption12
Hype gap−15
Incentives55
Confidence55
buildOne report1 publisher

Dropping curl, git and vim explains 74 MB of a 385 MB Docker image cut

A dev.to writeup takes a 15-line Flask app from 442 MB to 56.6 MB and publishes its layer sizes. The measured layers cover 74.4 MB of the cut, and the base image swap has to account for the rest.

Publishers:dev.to

Reality

Evidence45
Adoption
Not scored
Hype gap+30
Incentives30
Confidence50
buildOne report1 publisher

Two agent sandboxes ship an unauthenticated shell endpoint on every interface

Cua's auth check keyed on a container variable while its bind default listened everywhere, and the pair scored 9.8. AutoAgent's blunter version had no fixed release when the records went out.

Publishers:dev.to

Reality

Evidence62
Adoption32
Hype gap+8
Incentives35
Confidence55
buildOne report1 publisher

A UDP socket carries Frag Gap from inside a container into host kernel memory

CISA added CVE-2026-53362 to its exploited-vulnerabilities catalog on August 27. The single published account of the bug says the escape needs only unprivileged code on the node plus permission to open a UDP socket, which almost nothing is denied.

Publishers:dev.to

Reality

Evidence20
Adoption12
Hype gap+40
Incentives55
Confidence30
securityOne report1 publisher

Wiz research: base images account for 39 percent of critical container CVE findings; hardened images cut CVEs by 94 percent

The 39 percent comes from Wiz research with no published sample or method, and it bounds the hardened-image pitch as much as it supports it, since most critical container findings sit above the base layer.

Publishers:wiz.io

Reality

Evidence20
Adoption
Not scored
Hype gap+40
Incentives92
Confidence62
investOne report1 publisher

Twistlock's founders raised $51M for Minimus, then handed the leftovers back

The container security pioneers behind a $410 million exit could not sell fewer CVEs to buyers whose budgets were already committed. The registry goes dark on October 22.

Publishers:calcalistech.com

Reality

Evidence60
Adoption14
Hype gap+22
Incentives58
Confidence55
productOne report1 publisher

Container security's cheapest control is shipping less software, not scanning faster

Traefik Labs argues most CVEs in a container come from the OS packaging around the application, not the application. Its pitch is attack surface reduction rather than faster detection.

Publishers:siliconangle.com

Reality

Evidence27
Adoption
Not scored
Hype gap+38
Incentives82
Confidence34