What do others use for ensuring the authenticity of images after downloading them with docker pull?

We’ve setup our CI build process to use docker for consistent, cross-platform builds. To ensure that our builds don’t use a malicious docker image (because the surface area of attack with TLS is enormous if you’re using X.509), we’ve been using DCT (Docker Content Trust).

Unfortunately, I just discovered that the official docker documentation says that DCT is being deprecated. Apparently this was announced last year, and in June this blog post was published with advice:

• https://www.docker.com/blog/docker-content-trust-retirement-and-migration-guidance/

Cosign is not secure

We spent some time looking into cosign, but we discovered that the private keys aren't actually in the hands of the developer.

Rather, they use this complicated setup using very insecure X.509 to issue temporary certificates.

The result is that the OIDC identity provider (e.g., GitHub) extends the vector of attack significantly -- to probably tens of thousands of people -- that can publish a malicious image that will be accepted by cosign as "trusted"

Notation (Notary v2)

I also looked at Notation (aka “Notary v2”), but there’s no way to bootstrap the software safely, since (perplexingly) their tool for verifying the authenticity of images using cryptographic signatures itself can’t be verified using a cryptographic signature.

Alternatives

Are there any other alternatives that I can use to replace DCT to ensure the authenticity (using cryptography) of the container images that I download -- where the keys are actually held by the developer (thus significantly reducing the "insider threat" risk)?

What do you (or does your org do) to ensure that you’re not using maliciously-modified containers after pulling a new docker image?

PWNED Welcome, once again, to PWNED, the weekly column where we recount the adventures of IT explorers who found their own pile of quicksand and then jumped right into it. This week's story involves keeping sensitive information in a very vulnerable place and then not protecting it adequately.

The tale comes to us courtesy of Stanislav Kazanov, head of strategic practices at Innowise, a software development firm. A few years ago Kazanov and his group were hired to perform compliance and data architecture audits on a fintech startup where execs had invested more than $1 million to develop a "military grade" security system complete with biometric MFA, endpoint detection, and a ton of physical security.

During the audit, Kazanov logged onto the company's SharePoint site and found a folder called "DevOps_Handoff" on the company-wide intranet that any employee could access. Within that folder was a spreadsheet with the very obscure and deceptive filename Prod_DB_Root_Creds_DO_NOT_SHARE.xlsx. Clearly, this naming convention would throw off any would-be hackers.

In a cautionary tale of agentic AI, AWS reportedly suffered service outages caused by its own AI coding tools in December - though the company insists the downtime was ultimately due to human error.

Amazon's cloud platform suffered a 13-hour disruption affecting one service after resident engineers allowed Kiro to make changes, according to four sources who spoke to the Financial Times.

IPFire has launched IPFire DBL (Domain Blocklist), a community-driven system that lets administrators fine-tune network filtering policies.

IPFire DBL moves away from the old model of a single large list and instead uses categories. It groups millions of domains so administrators can choose only the categories that suit their needs.

Categories include Malware, Phishing, Advertising, Pornography, Gambling, Games, and DNS-over-HTTPS, among others. This method saves resources and gives operators more control over their policies.

Some tech support jobs are sweet, and others go sour. Whatever taste they leave in your mouth, The Register celebrates them all each week in On Call – the reader-contributed column that shares your support experiences.

This week, meet a reader we'll Regomize as "Parker" who told us he once had a mostly delicious job in a candy factory.

Who, Me? Welcome to Monday morning and another instalment of “Who, Me?” - the weekly reader-contributed column in which we share your stories of what not to do at work, and how to get away with it.

This week, meet a reader we’ll Regomize as “Patrick” who told us he once installed an extra shelf of storage for a NAS at a local council office in Australia.

The job initially went well.

DAOS has been a great success in the traditional HPC/supercomputing world, but is nowhere in the new, AI-focused, GPU supercomputing arena. What will it take for DAOS to find customers outside its high-end, legacy supercomputing niche?

The DAOS parallel filesystem has a strong IO500 presence, holding positions 1 (Argonne) and 2 (LRZ) in the current Production SC25 list. The two, according to HPE, combined have four times the storage benchmark score of the next 30 storage systems. DAOS also appears at number 13 (Zuse Institute, Berlin), and 17 (China Telecom Research Institute). The software appears more often in the full IO500 list, with 16 of the top 30 submissions using DAOS, and 26 of the top 45 being DAOS devotees.

An Ohio IT contractor has pleaded guilty to breaking into his former employer's systems and causing nearly $1 million worth of damage after being fired.

Maxwell Schultz, 35, impersonated another contractor to gain access to the company's network after his credentials were revoked.

Announcing the news, US attorney Nicholas J. Ganjei did not specify the company in question, which is typical in these malicious insider cases, although local media reported it to be Houston-based Waste Management. The Register contacted Waste Management for clarity, but it did not respond.

The US Securities and Exchange Commission (SEC) has abandoned the lawsuit it pursued against SolarWinds and its chief infosec officer for misleading investors about security practices that led to the 2020 SUNBURST attack.

In a joint motion filed Thursday, the SEC along with SolarWinds and its Chief Information Security Officer Timothy G. Brown, asked the court dismiss [PDF] the commission's ongoing civil enforcement action.

The market for server virtualization tools is about to fragment, according to analyst firm Gartner.

“The server virtualization market is facing the most significant disruption in decades,” states the analyst’s October market guide to server virtualization platforms. Broadcom’s acquisition of VMware – and subsequent decision to focus its efforts on a broad public cloud platform – is the reason for the shifting market.

“For many Gartner clients, the aftereffects of VMware’s acquisition by Broadcom represent a turning point in the market for server virtualization,” the market guide states. “Disruption is being driven by customer concerns about increases in the total cost of ownership, the quality of support, and changes to product roadmaps with limited perceived benefit.”

A faulty HP OneAgent update (version 1.2.50.9581) silently deleted Microsoft Entra ID certificates on HP's AI-enabled devices, breaking cloud authentication for affected organizations[^1]. The issue stemmed from a cleanup script in package SP161710 that indiscriminately removed certificates containing "1E" in their identifiers, inadvertently deleting critical MS-Organization-Access certificates needed for Entra ID authentication[^1].

The problem specifically affected HP's Next Gen AI models like the EliteBook X Flip G1i, with the update pushed through HP's AWS IoT backend without proper testing[^1]. While HP has pulled the problematic update, affected devices require manual intervention - either logging in with local admin credentials to rejoin Entra ID or using Microsoft Defender's Live Response for remote fixes[^1][^2].

According to HP, "The update is no longer available and will not affect more AI PCs. We're investigating the issue and working closely with impacted customers on mitigation"[^12].

[^1]: PatchMyPC - HP OneAgent Update Broke Entra Trust on HP AI Devices [^2]: CyberSecurityNews - HP OneAgent Update Brokes Trust And Disconnect Devices From Entra ID [^12]: BleepingComputer - HP pulls update that broke Microsoft Entra ID auth on some AI PCs