2026-01-01 03:30:40
Courtesy of the complex routing and network configurations that Cloudflare uses, their engineers like to push the Linux network stack to its limits and ideally beyond. In a blog article [Chris Branch] details how they ran into limitations while expanding their use of soft-unicast functionality that fits with their extensive use of anycast to push as much redundancy onto the external network as possible.
The particular issue that they ran into had to do with the Netfilter connection tracking (conntrack) module and the Linux socket subsystem when you use packet rewriting. For soft-unicast it is important that multiple processes are aware of the same connection, yet due to how Linux works this made it impossible to use packet rewriting. Instead they had to use a local proxy initially, but this creates overhead.
To work around this the solution appeared to be to abuse the TCP_REPAIR socket option in Linux, which normally exists to e.g. migrate VM network connections. This enables one to describe the entire socket connection state, thus ‘repairing’ it. Combined with TCP Fast Open to skip the whole handshake bit with a TFO ‘cookie’. This still left a few more issues to fix, with an early demux providing a potential solution.
Ironically, ultimately it was decided to not break the Linux networking stack that much and stick with the much less complicated local proxy to terminate TCP connections and redirect traffic to a local socket. Unfortunately escaping the Linux networking stack isn’t that straightforward.
2026-01-01 02:00:00
Humans have lots of basic requirements that need to be met in order to stay alive. Food is a necessary one, though it’s possible to go without for great stretches of time. Water is more important, with survival becoming difficult beyond a few days in its absence. Most of all, though, we crave oxygen. Without an air supply, death arrives in mere minutes.
The importance of oxygen is why airway management is such a key part of emergency medicine. It can be particularly challenging in cases where there is significant trauma to the head, neck, or surrounding areas. In these cases, new research suggests there may be an alternative route to oxygenating the body—through the rear.
Most of us are familiar with the usual route of human respiration. We take in air through the mouth and nose, and it passes through the windpipe and into the lungs, where oxygen diffuses into the blood. When everything in the body is functional, this system works well. However, when things go wrong, it can suddenly become very difficult to keep a body alive.
Head or neck injuries can block the airway entirely, or infections can fill the lungs with fluid, preventing the transfer of oxygen to the blood. Supportive ventilation methods can help, but can often damage the lungs themselves while in use. When the lungs themselves cease to function at all, often the only real option is the use of a technique called extracorporeal membrane oxygenation, or ECMO. This is where complicated machinery is used to manually oxygenate the blood outside the body. It’s a complex method that can result in major complications, and comes with a wide range of potential side effects, some of which can be fatal.
In these life-or-death situations, it would be desirable to have an alternative oxygenation technique that could be used when the lungs or airway are badly compromised. New research has suggested that enteral ventilation could be just the ticket. It’s a rather out of the box method, involving the use of a special oxygen-carrying liquid called perfluorodecalin. By administering this fluid rectally, it may be possible to deliver oxygen to a patient without having to rely on the function of the lungs themselves.
As you might guess by the name, perfluorodecalin is a flurocarbon. Its molecules are made up of 10 carbon and 18 fluorine atoms, and it exists as a liquid at room temperature. It’s considered chemically and biologically inert, which is key to its use in a medical context. Beyond that, it’s capable of dissolving a great deal of oxygen, with 100 mL of perfluorodecalin able to dissolve 49 mL of oxygen at a temperature of 25 C. The fluid can also carry carbon dioxide, too. Historically, it’s been used as a method to supply oxygen to specific areas of the body in a topical application, and also used as a way to preserve organs or other tissues in an oxygen-rich environment.
Thus far, research remains at an early stage. Initial testing focused on supplying a rectal dose of non-oxygenated perfluorodecalin of 25 to 1,500 mL for up to 60 minutes, at which point patients would excrete the fluid on their own terms. Patients had their vital signs monitored and were studied for any possible adverse effects. The study found that only mild side effects occurred, specifically involving abdominal bloating and pain at higher levels which resolved without further intervention after the procedure was completed. No perfluorodecalin or related compounds were detected in the bloodstream in the immediate aftermath.
The first stage of clinical testing was focused on establishing safety profiles rather than outright testing the efficacy of rectal oxygenation. Nonetheless, even in testing with non-oxygenated perfluorodecalin, the study showed a “modest increase” in oxygen saturation in patients dosed with higher amounts of the fluid (500 mL and 1000 mL). This is a positive sign that this could be a viable route for oxygenation, but more research will be needed to verify the findings and develop the technique into something that could have actual clinical applications. That can be a particularly slow process due to the extensive safety requirements of new medical treatments, but such regulations exist for good reason.
It’s not the first time that physicians have explored alternate methods of delivering oxygen to the body. Other methods of liquid ventilation have been developed, albeit with a focus on delivering oxygen-rich liquids to the lungs themselves. The aim is generally to avoid the lung damage that is often caused by traditional positive-pressure ventilation systems, which can be particularly harmful to patients who are already badly unwell. Similarly, these methods typically use oxygen-rich flurocarbons to do the job. While there have been some promising studies, ultimately the technique remains experimental and challenging to implement.
Enteral ventilation has one major benefit over liquid ventilation using the lungs, precisely because it doesn’t involve the lungs at all. The body’s main airway can remain entirely unobstructed during such a treatment, and does not have to be filled with fluids or tubes that could cause damage on their own. In cases where the airway or lungs are badly damaged or compromised, these techniques could potentially help where liquid ventilation via the lungs would simply not be possible. There can be immediate risks in delivering any kind of liquid to a patient’s lungs, particularly if the transition to liquid breathing doesn’t go to plan. The same simply isn’t true of doing so via the enteral pathway, as the regular airway remains untouched and as functional as it ever was.
As it stands, you’re unlikely to be breathing via the rectum any time soon. However, some years down the line, your local emergency room or ICU might just have another route to administer oxygen when all the standard methods fail. It might be weird and unconventional, but it could help save lives.
2026-01-01 00:30:24
By far, the most widely used psychoactive substance in the world is caffeine. It’s farmed around the world in virtually every place that it has cropped up, most commonly on coffee plants, tea plants, and cocoa plants. But is also found in other less common plants like the yaupon holly in the southeastern United States and yerba maté holly in South America. For how common it is and how long humans have been consuming it, it’s always been a bit difficult to quantify exactly how much is in any given beverage, but [Johnowhitaker] has a solution to that.
This build uses a practice called thin layer chromatography, which separates the components of a mixture by allowing them to travel at different rates across a thin adsorbent layer using a solvent. Different components will move to different places allowing them to be individually measured. In this case, the solvent is ethyl acetate and when the samples of various beverages are exposed to it on a thin strip, the caffeine will move to a predictable location and will show up as a dark smudge under UV light. The smudge’s dimensions can then be accurately measured to indicate the caffeine quantity, and compared against known reference samples.
Although this build does require a few specialized compounds and equipment, it’s by far a simpler and less expensive way of figuring out how much caffeine is in a product than other methods like high-performance liquid chromatography or gas chromatography, both of which can require extremely expensive setups. Plus [Johnowhitaker]’s results all match the pure samples as well as the amounts reported in various beverages so he’s pretty confident in his experimental results on beverages which haven’t provided that information directly.
If you need a sample for your own lab, we covered a method on how to make pure caffeine at home a while back.
2025-12-31 23:00:56
A friend of mine and I both have a similar project in mind, the manufacture of custom footwear with our hackerspace’s shiny new multi-material 3D printer. It seems like a match made in heaven, a machine that can seamlessly integrate components made with widely differing materials into a complex three-dimensional structure. As is so often the case though, there are limits to what can be done with the tool in hand, and here I’ve met one of them.
I can’t get a good range of footwear for my significantly oversized feet, and I want a set of extra grippy soles for a particular sporting application. For that the best material is a rubber, yet the types of rubber that are best for the job can unfortunately not be 3D printed. In understanding why that is the case I’ve followed a fascinating path which has taught me stuff about 3D printing that I certainly didn’t know.
A friend of mine from way back is a petrochemist, so I asked him about the melting points of various rubbers to see if I could find an appropriate filament His answer, predictably, was that it’s not that simple, because rubbers don’t behave in the same way as the polymers I am used to. With a conventional 3D printer filament, as the polymer is fed into the extruder and heated up, it turns to liquid and flows out of the nozzle to the print. It ‘s then hot enough to fuse with the layer below as it solidifies, which is how our 3D prints retain their shape. This property is where we get the term “plastic” from, which loosely means “Able to be moulded”.
My problem is that rubber doesn’t behave that way. As any casual glance at a motor vehicle will tell you, rubber can be moulded, but it doesn’t neatly liquefy and flow in the way my PLA or PET does. It’s a non-Newtonian fluid, a term which I was familiar with from such things as non-drip paint, tomato ketchup, or oobleck, but had never as an electronic engineer directly encountered in something I am working on.
A Newtonian fluid has a linear relationship between shear stress and shear rate. That’s dry language for saying that when you press it, it moves, if you press it more, it moves more, and the readiness with which it moves, or its viscosity, is the same across all pressures.
I’m used to viscosity, having run all manner of dodgy old cars I’m particularly familiar with selecting the correct oil by viscosity figure. A non-Newtonian fluid doesn’t have this linear relationship, and its viscosity changes with pressure. For example the non-drip paint has a high viscosity until you press it with a paint brush, at which point its viscosity falls and it becomes liquid enough to spread around. Rubber does this too, and were I to attempt to squeeze rubber filament through my extruder, it would become very viscous and block it up. The closest thing to a rubber I could reasonably use is TPU, or Thermoplastic PolyUrethane, but as you might guess from its name, it’s not a rubber in the same sense as the rubbers I’m looking at, even though it’s what many people use for shoes. It’s flexible, but not grippy.
So if rubber is non-Newtonian and I can’t print with it, how do they mould it? An online search finds specialist plants for rubber extrusion and moulding so it’s possible, but in fact those rubber moulded items you’re familiar with won’t be made with liquid rubber. Instead they press shredded rubber into a mould and heat it so that it fuses, resulting in a moulded shape. I was fascinated to find that the process doesn’t require excessive temperatures, though whether that makes it achievable in a hackerspace is yet to be determined. Has anyone out there experimented with real rubber? Meanwhile I have those multi-material uppers to work on.
2025-12-31 20:00:58
Although flying well under the radar of the average Linux user, D-Bus has been an integral part of Linux distributions for nearly two decades and counting. Rather than using faster point-to-point interprocess communication via a Unix socket or such, an IPC bus allows for IP communication in a bus-like manner for convenience reasons. D-Bus replaced a few existing IPC buses in the Gnome and KDE desktop environments and became since that time the de-facto standard. Which isn’t to say that D-Bus is well-designed or devoid of flaws, hence attracting the ire of people like [Vaxry] who recently wrote an article on why D-Bus should die and proposes using hyprwire instead.
The broader context is provided by [Brodie Robertson], whose video adds interesting details, such as that Arch Linux wrote its own D-Bus implementation rather than use the reference one. Then there’s CVE-2018-19358 pertaining to the security risk of using an unlocked keyring on D-Bus, as any application on said bus can read the contents. The response by the Gnome developers responsible for D-Bus was very Wayland-like in that they dismissed the CVE as ‘works as designed’.
One reason why the proposed hyperwire/hyprtavern IPC bus would be better is on account of having actual security permissions, real validation of messages and purportedly also solid documentation. Even after nearly twenty years the documentation for D-Bus consists mostly out of poorly documented code, lots of TODOs in ‘documentation’ files along with unfinished drafts. Although [Vaxry] isn’t expecting this hyprwire alternative to be picked up any time soon, it’s hoped that it’ll at least make some kind of improvement possible, rather than Linux limping on with D-Bus for another few decades.
2025-12-31 17:00:30
Having an enclosed 3D printer can make a huge difference when printing certain filaments that are prone to warping. It’s easy enough to build an enclosure to stick your own printer in, but it can get tricky when you want to actively control the conditions inside the chamber. That’s where [Jayant Bhatia]’s Chamber Master project comes in.
This system is built around the ESP32 microcontroller, which provides control to various elements as well as hosts a web dashboard letting you monitor the chamber status remotely. The ESP32 is connected to an SSD1306 OLED display and a rotary encoder, allowing for navigating menus and functions right at the printer, letting you select filament type presets and set custom ones of your own. A DHT11 humidity sensor and a pair of DS18B20 temperature sensors are used to sense the chamber’s environment and intake temperatures.
One of the eye-catching features of the Chamber Master is the iris-controlled 120 mm fan mounted to the side of the chamber, allowing for an adjustable-size opening for air to flow. When paired with PWM fan control, the amount of airflow can be precisely controlled.
