January 2024 - wlug - list.wlug.org.nz

New systemd update will bring Windows’ infamous Blue Screen of Death to Linux
by Peter Reutemann 18 Nov '24

18 Nov '24

'Windows' infamous "Blue Screen of Death" is a bit of a punchline. People have made a hobby of spotting them out in the wild, and in some circles, they remain a byword for the supposed flakiness and instability of PCs. To this day, networked PCs in macOS are represented by beige CRT monitors displaying a BSOD. But the BSOD is supposed to be a diagnostic tool, an informational screen that technicians can use to begin homing in on the problem that caused the crash in the first place; that old Windows' BSOD error codes were often so broad and vague as to be useless doesn't make the idea a bad one. Today, version 255 of the Linux systemd project honors that original intent by adding a systemd-bsod component that generates a full-screen display of some error messages when a Linux system crashes. The systemd-bsod component is currently listed as "experimental" and "subject to change." But the functionality is simple: any logged error message that reaches the LOG_EMERG level will be displayed full-screen to allow people to take a photo or write it down. Phoronix reports that, as with BSODs in modern Windows, the Linux version will also generate a QR code to make it easier to look up information on your phone.' -- source: https://arstechnica.com/gadgets/2023/12/linux-distros-are-about-to-get-a-ki… Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, Hamilton, NZ Mobile +64 22 190 2375 https://www.cs.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

2 3

Juice TV Hiccup
by Lawrence D'Oliveiro 15 Oct '24

15 Oct '24

Currently when I switch to FreeView Channel 200, it shows a frozen image with the text “Cinegy” and “Demo Version” on it. Searching for that name takes me to a site that talks about “software defined television”. Either their product licence has expired, or they never had one in the first place ...

2 3

List Of IPTV Resources
by Lawrence D'Oliveiro 22 Sep '24

22 Sep '24

This user <https://github.com/iptv-org> just popped up in my GitHub feed. One repo <https://github.com/iptv-org/iptv> claims to be a list of links to over 8000 streaming TV channels from nations all over the world, while the other is a list of tools*, providers and other resources. Do be aware it’s a bit of a mixed bag: yes, there are legitimate channels (no encryption so far that I have noticed), but also some dubious-looking links. *Oddly, FFmpeg and youtube-dl aren’t in the list.

2 5

Tip For Repairing Ancient Linux Installations
by Lawrence D'Oliveiro 02 Apr '24

02 Apr '24

Here’s the scenario: there is a malfunctioning server running a really ancient Linux installation (e.g. 2.6.x-series kernel). You boot up with a reasonably recent copy of SystemRescue, mount the machine’s filesystem, try to chroot into it and run a shell ... and it segfaults. Seems there has been a subtle incompatible change in Linux kernel behaviour since those times, to do with the disabling of something called the “vsyscall” mechanism (which is specific to 64-bit x86-based machines, I understand). This was supposed to be a way to speed up certain commonly-executed system calls (I think getting the system time was one of them) by allowing direct transfers to certain entry points with fixed memory addresses for invoking them. It was later decided this was a bad idea (I think for security reasons). The mechanism has not been completely removed (yet), but nowadays it is disabled by default. To get around this, you need to boot SystemRescue with a change to the kernel command-line parameters, adding this item vsyscall=emulate After you get to a shell, you can confirm that this setting is in effect by checking the /proc/self/maps file (the memory mappings for your own process). At the bottom, there should be an entry with “[vdso]” in it, and below that another line with “[vsyscall]”. The presence of the latter line indicates that vsyscall emulation is in effect. This is not present when I check a process on my own machine, for example. Linus is legendary for being a stickler for never, ever breaking backward compatibility for userland. So the reasons for turning this off must have been pretty compelling, for him to agree to it. But the compromise is, it’s still there to be enabled if you really, really need it. There’s a bit more technical info here <https://0xax.gitbooks.io/linux-insides/content/SysCall/linux-syscall-3.html>, if you are interested. By the way, the above page is part of a book beginning here <https://0xax.gitbooks.io/linux-insides/content/>, about the insides of the Linux kernel. The whole book seems to be accessible from that link, but if you step up a level, the site doesn’t seem to have any idea about the existence of that book.

3 4

Enabling Case-Insensitive Filenames On Linux
by Lawrence D'Oliveiro 06 Feb '24

06 Feb '24

The ext4 filesystem has had the “casefold” option for a while now <https://lwn.net/Articles/784041/>. This allows you to specify, when you initialize a new filesystem, that you want file/directory names to be treated as case-insensitive. At the same time, you have to specify that the filenames are to be interpreted as UTF-8 encoding; normally the kernel doesn’t care, it only singles out ASCII slash and NUL bytes for special interpretation, and simply passes through any other byte values in the names. I was able to create such a filesystem, but I wasn’t able to mount it in my last-upgraded-just-a-few-days-ago Debian Unstable setup. When I try, it fails, and in the dmesg log I see Filesystem with casefold feature cannot be mounted without CONFIG_UNICODE Turns out the Debian 5.10 kernel has not been built with this CONFIG_UNICODE option enabled. However, this situation looks set to change shortly, according to <https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=985689>.

2 3

'Microsoft Stole My Chrome Tabs, and It Wants Yours, Too'
by Peter Reutemann 31 Jan '24

31 Jan '24

'Tom Warren, writing for The Verge: Last week, I turned on my PC, installed a Windows update, and rebooted to find Microsoft Edge automatically open with the Chrome tabs I was working on before the update. I don't use Microsoft Edge regularly, and I have Google Chrome set as my default browser. Bleary-eyed at 9AM, it took me a moment to realize that Microsoft Edge had simply taken over where I'd left off in Chrome. I never imported my data into Microsoft Edge, nor did I confirm whether I wanted to import my tabs. But here was Edge automatically opening after a Windows update with all the Chrome tabs I'd been working on. I didn't even realize I was using Edge at first, and I was confused why all my tabs were suddenly logged out. After the shock wore off, I looked to make sure I hadn't accidentally allowed this behavior. I found a setting in Microsoft Edge that imports data from Google Chrome on each launch. "Always have access to your recent browsing data each time you browse on Microsoft Edge," reads Microsoft's description of the feature in Edge. This setting was disabled, and I had never been asked to turn it on. So I went to install the same Windows update on a laptop, which actually resulted in it failing and my having to do a system restore. Once the system restore was complete, the same thing happened. Edge opened automatically with all of my Chrome tabs. I haven't been able to replicate the behavior on other PCs, but a number of X users replied to my post about this saying they have experienced the same thing in the past.' -- source: https://it.slashdot.org/story/24/01/30/1824233 Cheers, peter -- Peter Reutemann Dept. of Computer Science University of Waikato, Hamilton, NZ Mobile +64 22 190 2375 https://www.cs.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0

Raspberry Pi Is Planning a London IPO, But Its CEO Expects 'No Change' In Focus
by Peter Reutemann 31 Jan '24

31 Jan '24

'Ars Technica: The business arm of Raspberry Pi is preparing to make an initial public offering (IPO) in London. CEO Eben Upton tells Ars that should the IPO happen, it will let Raspberry Pi's not-for-profit side expand by "at least a factor of 2X." And while it's "an understandable thing" that Raspberry Pi enthusiasts could be concerned, "while I'm involved in running the thing, I don't expect people to see any change in how we do things." CEO Eben Upton confirmed in an interview with Bloomberg News that Raspberry Pi had appointed bankers at London firms Peel Hunt and Jefferies to prepare for "when the IPO market reopens." Raspberry previously raised money from Sony and semiconductor and software design firm ARM, and it sought public investment. Upton denied or didn't quite deny IPO rumors in 2021, and Bloomberg reported Raspberry Pi was considering an IPO in early 2022. After ARM took a minority stake in the company in November 2023, Raspberry Pi was valued at roughly 400 million pounds, or just over $500 million. Given the company's gradual recovery from pandemic supply chain shortages, and the success of the Raspberry Pi 5 launch, the company's IPO will likely jump above that level, even with a listing in the UK rather than the more typical US IPO. Upton told The Register that "the business is in a much better place than it was last time we looked at it [an IPO]. We partly stopped because the markets got bad. And we partly stopped because our business became unpredictable." "It's a good thing, in that people care about us," Upton said in response to concerned hobbyists and tech enthusiasts. "What Raspberry Pi [builds] are the products we want to buy, and then we sell them to people like us," Upton said. "Certainly, while I'm involved in it, I can't imagine an environment in which the hobbyists are not going to be incredibly important." The IPO is "about the foundation," Upton said, with that charitable arm selling some of its majority stake in the business entity to raise funds and expand. "We've not cooked up some new way for a not-for-profit to do an IPO, no," he noted. [He told Ars that Raspberry Pi's business arm has had both strategic and private investors in its history, along with a majority shareholder in its Foundation (which in 2016 owned 75 percent of shares), and that he doesn't see changes to what Pi has built. He also noted that the foundation was previously funded by dividends from the business side.] "We do this transaction, and the proceeds of that transaction allow the foundation to train teachers, run clubs, expand programs, and ... do those things at, at least, a factor of 2X. That's what I'm most excited about." Upton said there would be "no change" to the kinds of products Pi makes, and that makers are "culturally important to us." [...] "If people think that an IPO means we're going to ... push prices up, push the margins up, push down the feature sets, the only answer we can give is, watch us. Keep watching," he said. "Let's look at it in 15, 20 years' time."' -- source: https://hardware.slashdot.org/story/24/01/30/2212232 Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, Hamilton, NZ Mobile +64 22 190 2375 https://www.cs.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0

Linux On Mars: The Final Chapter
by Lawrence D'Oliveiro 30 Jan '24

30 Jan '24

You may have heard that the “Ingenuity” helicopter, which has been flying on Mars for the last three years, came a cropper and broke one of its rotor blades a couple of weeks ago. Not bad for a project that was really put together as an afterthought, a proof of concept that was planned to maybe last 30 days and make five flights at most. Here <https://arstechnica.com/space/2024/01/now-that-weve-flown-on-mars-what-come…> is a description of some more details of how it was put together, within stringent limits on both mass and money. Apparently, a radiation-hardened RAD750 computer, which is normally used for such space missions, costs a quarter of a million USD. And would have weighed too much anyway. So they used a cheap, lightweight Qualcomm chip instead, state of the art for smartphones of just a few years ago. And massively more powerful. And of course they ran off-the-shelf open-source software on it <https://www.jpl.nasa.gov/news/meet-the-open-source-software-powering-nasas-…>, <https://www.theverge.com/2021/2/19/22291324/linux-perseverance-mars-curiosi…>, <https://spectrum.ieee.org/nasa-designed-perseverance-helicopter-rover-fly-a…>. Flying a chopper on Mars isn’t easy: the rotor tips on Ingenuity were moving at Mach 0.65 (on Mars, Mach 1.0, the speed of sound, is about 240m/s). A future chopper is planned to push this to Mach 0.95. You might say they were cutting corners on this design, which made it inherently more risky. But if the lower costs and simpler development can lead to shorter lead times, and hence more missions, maybe a higher failure rate would be worth it.

1 0

Remembering Unix Desktops - and What We Can Learn From Them
by Peter Reutemann 29 Jan '24

29 Jan '24

'"As important as its historically underhanded business dealings were for its success, Microsoft didn't have to cheat to win," argues a new article in the Register. "The Unix companies were doing a great job of killing themselves off." You see, while there were many attempts to create software development standards for Unix, they were too general to do much good — for example Portable Operating System Interface (POSIX) — or they became mired in the business consortium fights between the Open Systems Foundation and Unix International, which became known as the Unix wars. While the Unix companies were busy ripping each other to shreds, Microsoft was smiling all the way to the bank. The core problem was that the Unix companies couldn't settle on software standards. Independent Software Vendors (ISV) had to write applications for each Unix platform. Each of these had only a minute desktop market share. It simply made no business sense for programmers to write one version of an application for SCO OpenDesktop (also known as OpenDeathtrap), another for NeXTStep, and still another one for SunOS. Does that sound familiar? That kind of thing is still a problem for the Linux desktop, and it's why I'm a big fan of Linux containerized desktop applications, such as Red Hat's Flatpak and Canonical's Snap. By the time the two sides finally made peace by joining forces in The Open Group in 1996, it was too late. Unix was crowded out on the conventional desktop, and the workstation became pretty much a Sun Microsystems-only play. Linux's GPL license created an "enforced" consortia that allowed it to take over, according to the article — and with Linus Torvalds as Linux's single leader, "it avoided the old Unix trap of in-fighting... I've been to many Linux Plumbers meetings. There, I've seen him and the top Linux kernel developers work with each other without any drama. Today's Linux is a group effort... The Linux distributors and developers have learned their Unix history lessons. They've realized that it takes more than open source; it takes open standards and consensus to make a successful desktop operating system. And the article also points out that one of those early Unix desktops "is still alive, well, and running in about one in four desktops." That operating system, of course, is macOS X, the direct descendent of NeXT's NeXTSTEP. You could argue that macOS, based on the multi-threaded, multi-processing microkernel operating system Mach, BSD Unix, and the open source Darwin, is the most successful of all Unix operating systems. ' -- source: https://tech.slashdot.org/story/24/01/28/0241228 Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, Hamilton, NZ Mobile +64 22 190 2375 https://www.cs.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

2 1

I abandoned OpenLiteSpeed and went back to good ol’ Nginx
by Peter Reutemann 29 Jan '24

29 Jan '24

'Since 2017, in what spare time I have (ha!), I help my colleague Eric Berger host his Houston-area weather forecasting site, Space City Weather. It’s an interesting hosting challenge—on a typical day, SCW does maybe 20,000–30,000 page views to 10,000–15,000 unique visitors, which is a relatively easy load to handle with minimal work. But when severe weather events happen—especially in the summer, when hurricanes lurk in the Gulf of Mexico—the site’s traffic can spike to more than a million page views in 12 hours. That level of traffic requires a bit more prep to handle. For a very long time, I ran SCW on a backend stack made up of HAProxy for SSL termination, Varnish Cache for on-box caching, and Nginx for the actual web server application—all fronted by Cloudflare to absorb the majority of the load. (I wrote about this setup at length on Ars a few years ago for folks who want some more in-depth details.) This stack was fully battle-tested and ready to devour whatever traffic we threw at it, but it was also annoyingly complex, with multiple cache layers to contend with, and that complexity made troubleshooting issues more difficult than I would have liked. So during some winter downtime two years ago, I took the opportunity to jettison some complexity and reduce the hosting stack down to a single monolithic web server application: OpenLiteSpeed.' -- source: https://arstechnica.com/gadgets/2024/01/i-abandoned-openlitespeed-and-went-… Cheers, Peter -- Peter Reutemann Dept. of Computer Science University of Waikato, Hamilton, NZ Mobile +64 22 190 2375 https://www.cs.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/

1 0