Reliability Engineering
199 episodes
#6188: Every Way Linux Relieves Memory Pressure
Zram, zswap, swappiness, and the OOM killers — the full ladder Linux climbs when RAM runs out, and what to actually run.
#6187: Can Your Laptop Borrow Your Desktop's RAM?
Two gigabytes free on the laptop, sixty-four idle on the desktop next door. Can swap over the network actually help?
#6096: How Long Does an M-Disc Actually Last?
A thousand years, or a hundred and eighty? What a burned disc on a Jerusalem patio can — and can't — tell us about archival storage.
#5970: Swapping Your Linux Kernel: What Actually Changes
Your kernel is swappable without reinstalling anything. But do Liquorix and XanMod actually beat the default? The benchmarks say less than the mark...
#5966: How Much of Your Linux Desktop Is Actually Misconfigured?
Most "Linux tweaks" are folklore. A few are real — and one toggle most people never open is worth up to 42%.
#5961: What Your GPU Is Actually Waiting On
Your GPU says 100% utilized. It's multiplying zeros. Here's what the counters actually measure during inference.
#5954: Layered Safeguards and the Math Behind Them
Three safeguards at 90% each seems bulletproof—until shared failure modes collapse the math.
#5942: Service Workers: The Proxy That Never Sleeps
A JavaScript file with no tab, no DOM, and the power to intercept every request your site makes — and why refreshing never updates it.
#5889: How Pipeline Pigs Actually Work
A bundle of straw that squealed through the pipe gave the industry its weirdest word. Now pigs weigh tons and hunt cracks underground.
#5885: The World's Oldest Software Still Running
From a 1958 Pentagon contract system to Voyager's 68KB computer, some software refuses to die. What's the oldest code still in production?
#5870: Homebox Fork's Two Backend Scripts Need a Real Job Runner
A Homebox fork has two incremental scripts — WebP conversion and AI image enrichment — and no scheduler built to run them safely.
#5844: Five of the Buggiest Software Ever Shipped
Windows ME, Sonic '06, Big Rigs, and more — the software where the bugs weren't a problem, they were the product.
#5766: SNMP: The 40-Year-Old Stopgap Running Your Rack
SNMP was designed as a temporary fix. Four decades later it's on every switch, UPS, and router you own. Here's how it actually works.
#5759: Why Servers Have Two Power Cords
Dual-corded servers, transfer switches, and the layered engineering behind keeping data centers online.
#5637: Lux Sensors for Home Assistant: The Feedback Loop Nobody Warns You About
Lux sensors sound simple until the lights you turn on trip the sensor that turned them on. Here's the full chain, from Aqara T1 to Matter's missing...
#5553: When 52 Volts Meets a 12-Volt Modem
A fried fiber modem raises the real question: where do you draw the line between preparedness and "done enough"?
#5540: Why Your Trackpad Clicks Stopped Working
Your trackpad surface works but the clicks don't? That's not a trackpad failure — it's a swollen battery pushing the click plate out of reach.
#5522: Building a Podcast Backup Appliance
One listener wants a single-purpose box that pulls every episode to local storage he controls. Turns out the software exists — the packaging doesn't.
#5376: Why a Zello Pager Wearable Doesn't Exist Yet
Pagers blink until acknowledged. Zello beeps until you tap. But nobody's shipped the wearable that bridges them.
#5286: Soldering Isn't Dead — It's a Hiring Bottleneck
Robots build the boards, but humans fix them — and the industry can't hire enough people who can hold an iron.
#5285: Two Queues, One Cron Tick: Serverless Batch Pipelines
What does a queue look like when your process is disposable? Two queues, one cron tick, and a 50% batch discount.
#5252: Paper in a Safe: The Most Rugged Key Backup
Hardware tokens protect your key from being copied. Paper in a safe protects it from being lost. You need both.
#5232: Zello, PTT Buttons, and the Kitchen Intercom Question
Why your Bluetooth speaker ignores Zello's push-to-talk — and what to buy instead for a hands-full, phone-off fallback.
#4996: Memory Leaks: The Full Taxonomy
From C to Rust to Kubernetes OOM kills — the real taxonomy of memory leaks and how to hunt them.