The updated technical errors in the US-EAST-1 area of ​​Amazon Web Services have caused widespread troubles for customers, including difficulty in accessing the management console and some other service issues.

The problem seems to be concentrated in the US-EAST-1 region, which is the oldest AWS region, located in Northern Virginia. As AWS pointed out in its status report, this could have a global impact:

"This issue is affecting the Global Console landing page, which is also hosted on US-EAST-1."

The company stated that customers can access the region-specific console by directly accessing the region's URL.

But in US-EAST-1, the affected services are not only the console, but also EC2 (Elastic Computing Cloud), DynamoDB and Amazon Connect. In fact, if EC2 does not work properly, hundreds of other services may be affected because they run on EC2 in the background.

North American AWS status report shows issues with critical services

Twitter was crowded with frustrated customers, and suppliers apologized to customers for the blackout. Even cloud service providers have been affected because many of these services are also running on AWS. For example, Elastic Cloud reports: "We have encountered capacity related to the increase in error rates in us-east-1 (AWS N. Virginia) Expanding the problem and monitoring the situation."

Major companies believed to be affected include Amazon's own Alexa, Music and Ring, Netflix, Disney, Discourse (they report "AWS Route 53, one of our DNS providers", Tinder and Roku.

A developer said: "AWS has failed, and I spent 2 hours trying to debug why my code can't run", which illustrates the extent to which it is assumed that the public cloud service is up and running.

Although this is a serious outage, except for the management console, other areas generally seem to be unaffected. However, super-large-scale services have a common problem, that is, although the overall flexibility is very good, due to the interdependence of services, there is the possibility of cascading failures.

AWS stated in its console and EC2 status report, "We have identified the root cause, and we are actively working to recover", hoping that this interruption will not exist for a long time. ®

For the RISC-V team currently holding a virtual summit, this interruption is very bad news.

"We know and work closely with the technical team to resolve this issue, and once it is fully operational again, we will update everyone," a spokesperson told The Register.

"For those who are already in the conversation, we suggest not to refresh the screen, as this may disconnect your streaming. All conversations will be recorded and will be available to you on demand shortly after the virtual event platform goes online again ."

Smartish vacuum manufacturer iRobot also reported that the service of its app was affected.

Amazon said at 1404 Pacific Time (2204 UTC): "We have implemented a mitigation measure that shows a significant recovery in the US-EAST-1 area."

"We will continue to closely monitor the health of network equipment, and we hope to continue to make progress in full recovery. At present we still do not have an estimated time of arrival for full recovery."

By 1635 PST (0035 UTC) things looked much better, even though the user hadn't gotten out of the predicament yet.

"With the resolution of network equipment issues, we are now working to restore any damaged services," the AWS report said. "We will provide additional updates for the damaged service in the corresponding entry on the service health dashboard."

At the time of this writing, the AWS status page only records that one service is experiencing problems-Amazon Elastic Container Service in Northern Virginia-which means that "task size less than 4vCPU is unlikely to have an insufficient capacity error."

AWS did not explain what went wrong with its network equipment.

Maybe one day—maybe—zero trust will solve many of our network security problems. But now, if you want to ensure that there are no eavesdroppers on your network, you need a virtual private network (VPN).

Commercial VPNs have only one small problem: many of them are not trustworthy. So, what can you do? Well, running your own is of course the open source answer. And, today, the VPN you choose is the built-in VPN of Linux: WireGuard.

Why is WireGuard instead of OpenVPN or IKEv2? Because it is easier to implement while maintaining security and providing faster speeds. Moreover, when it comes to VPN, the most important thing is to balance speed and security.

Those of us who are afraid of being enslaved by the robot overlord in the future may have another reason not to sleep at night: The engineers have shown some legless, soft breeds and have made some major leaps.

Animated pancake-like robots have demonstrated their ability to perform a series of failures in a way that their creators-Chinese software robotics engineers-describe as "fast, continuous, and directional jumps".

"Jumping is an important motor function that can expand the navigation range, overcome obstacles, and adapt to unstructured environments," said Chen Rui of Chongqing University and Pu Huayan of Shanghai University.

Intel has updated what it says allows the implementation of "software-defined silicon" (SDSi) code.

Chipzilla put some of the SDSi code into the Linux kernel in September 2021, describing it as a technology that allows users to activate the sleep function in silicon. This code outlines the process of enabling new features by verifying a cryptographically signed license.

David E. Box, a Linux developer working at Intel, published a new article on the Linux kernel mailing list explaining the second version of the Intel SDSi code.

The Ministry of Science and Information and Communication Technology of Korea has provided some suggestions to large technology companies on how to make their services appropriately flexible, and has increased the obligation to notify users when they fail—in Korean.

This guide applies to Google, Meta (Facebook's parent company), Netflix, Naver, Kakao and Wavve. Everyone was told to improve response to failures by strengthening preemptive error detection and verification systems, and to create a backup storage system that supports fast content recovery.

The guide provides methods that Big Tech can use to measure user load and then plan accordingly to ensure that their services remain available. The uptime requirements are not specified.

Microsoft has extended its Secure-core concept applied to PCs in 2019 to include servers, Windows Server, and Azure Stack HCI.

Secured-core believes that Microsoft cooperates with hardware manufacturers to ensure that their products include TPM 2.0 modules, secure boot is enabled by default in the BIOS, and dynamic root of trust is used for measurement technology, allowing the use of Intel’s Trusted Execution Technology (TXT) and AMD's Secure Virtual Machine (SVM).

Once these elements are in place, Microsoft believes that hardware is more difficult to attack based on firmware and less prone to running unverified code.

Microsoft has released an update to its ancient Notepad text editor.

Dave Grochocki, lead project manager for Windows Inbox Apps, wrote that Microsoft's favorite feature is the new dark mode.

"This is a hot issue in the community, and we hope you like this gorgeous new theme as much as we do," Grochocki said with a smile, explaining that you can enable and fiddle with fonts on the new settings page. If you have done this elsewhere, Notepad will also record your favorite system themes and use a dark mode.

Alibaba's DAMO Academy has demonstrated a chip that stacks logic and memory in 3D, and Chinese media believe it represents an architecture that can break the bottleneck of von Neumann.

Dharma Academy is the blue sky research institution of Alibaba. Founder Jack Ma dedicated it to "discovery, adventure, motivation and prospects" when it was launched in 2017.

The Chinese media Pandaily and ScienceNet both reported on the chip-the former linked to the now-deleted post on the Weibo website Weibo as evidence.

NASA announced on Tuesday that, nearly two decades later, it has upgraded its near-Earth asteroid monitoring algorithm to more accurately simulate dangerous space rocks.

The new system called Sentry-II is more powerful than its predecessor Sentry. Astronomers working at the Space Agency's Near-Earth Object Research Center can now automatically calculate the thermal effects that propel the asteroid's orbit, which may rush the asteroid toward our parent star.

The so-called Yakovsky effect describes the subtle and gradual changes in the movement of an asteroid as it is heated by the sun. When the asteroid spins, the side of its surface exposed to the star will be heated. As it continues to spin, the hot zone enters the shade and cools down. Infrared energy radiates outward; photons carry momentum and generate tiny thrust on the asteroid. For a long time, these little kicks can change their path and knock them out of their original tracks.

Mehta was sued in a class action lawsuit on Tuesday for allegedly amplifying hate speech and helping the Myanmar military to genocide the Rohingya. The compensation amounted to US$150 billion.

The case was led by an anonymous Rohingya refugee living in the United States and accused the entity, formerly Facebook, of inciting hatred and causing actual harm to the main Muslim group for many years. Not only did the social media platform ignore hate speech posts, but the service’s algorithm allegedly actively promoted anti-Rohingya propaganda as hundreds of thousands of people fled Myanmar to escape persecution.

Facebook has already acknowledged its role in this movement, with an estimated 25,000 people killed and 700,000 forced to leave the country. The lawsuit also occurred after former employee and whistleblower Frances Haugen leaked internal documents that showed how its algorithm puts participation above safety.

Trendforce said in a study this week that the shortage of power management chips is worsening and is preventing companies from manufacturing cars, personal computers, and items with batteries or switches.

The cost of a power management IC is only a few cents, and it is a cheap chip that contains display drivers and USB-C components that are in short supply. These chips are as important as CPU or memory for personal computers and other electronic products.

In the past 20-odd months, with the advent of electric vehicles and the growing demand for personal computers and consumer electronics, the demand for PMICs has reached its peak. Trendforce expects prices to rise by 10% to a six-year high of $0.23.

After a sharp increase in sales in the third quarter and better-than-expected forecasts for the last three months of the fiscal year, NoSQL database giant MongoDB's share price rebounded by 15%.

Revenue for the period ending October 31, 2021 was US$226.9 million, a year-on-year increase of 50%. The stock price rise is also based on MongoDB's forecast of its fourth quarter revenue as of January of 242 million U.S. dollars, while analysts estimate it is 226 million U.S. dollars.

By category, the database business reported a subscription revenue of US$217.9 million in the third quarter, an increase of 51% year-on-year. Service revenue reached US$9 million, an increase of US$35 year-on-year.

The Register-Independent news and opinions from the technical community. Partially released

Hand biting IT © 1998–2021