Det kriminella gänget Loyal to Familia (LTF) är för närvarande inblandat i våldsam gängkonflikt i Köpenhamn med ett annat kriminellt nätverk. Mer specifikt handlar det om LTF Loyal to Familia BGP, där BGP står för Blågårds Plads, och en icke namngiven kriminell grupp på andra sidan.
Svensk och dansk polis har sedan lång tid haft ett nära och gott samarbete för att bekämpa olika typer av gränsöverskridande brott. Detta enligt dem själva. Nu fördjupas samarbetet genom att dansk polis ska stationeras i Malmö och svensk polis i Köpenhamn.
In the previous attempt, we used llama3 but after trying llama 3.1, phi3, qwen2, and mixtral, I settled on gemma 2. It seemed to give the best results with a machine running 32 GB of unified memory. There are three parts to this app.
requirements.txt
The Python equivalent to packages.json is requirements.txt. You can run it from the terminal with pip install -r requirements.txt. Ours is going to look like this:
config.py
Our settings (what model we are using, where our Ollama instance is installed, etc) are all in a separate config.py file so if you want to use this code for your project, it should be pretty easy to modify what I am doing.
app.py
This file has three parts:
for shop_name, url in config.custard_shops.items() loops through the shops defined in config.py and executes get_flavor_of_the_day(shop_name, url) once per shop
get_flavor_of_the_day(shop_name, url) uses SmartScraperGraph(), the prompt, and the page_content variable to get the flavor of the day, its description, the current month, and the current day of the month.
save_to_firestore(shop_name, result) takes the values from get_flavor_of_the_day(shop_name, url) and saves them to a Firebase Cloud Firestore collection.
This is the guts of the app.
How do you run this app, locally?
I covered this a little in the readme for my Python RAG demo. Here are the steps that you need to go through to run this application:
Install Ollama and then run ollama pull nomic-embed-text and ollama pull nomic-embed-text from your terminal application (like iTerm 2)
When you run the script, it should update your Firebase Cloud Firestore collection.
You will notice that it can get the flavors from some shops and not from others. Screen scraping isn’t a great method of pulling in data at the best of times. You might get lucky with some tuning to the prompt or if you can run a 70b model, though.
Have a question, comment, or idea? Please feel free to drop a comment, below.
Som ung arbetare på 70-talet var jag aktiv i Kommunistiska partiet. För oss då var det självklart att solidariskt stödja det lilla landet Vietnams nationella försvarskrig mot den imperialistiska stor-makten USA:s invasion, ockupation och angreppskrig. Symboliskt samlade vi in pengar till Vietnams rättmätiga försvarskrig under parollen ”Luftvärnskanon till Vietnam!”
Men du kan väl ändå inte vara emot fredsförhandlingar. Vietnam fredsförhandlade ju med USA under lång tid. Men de krigade också. Naturligtvis kan Ukraina agera likadant.
I had this idea that I’d love a shirt like that, with the names of the Fediverse platforms I use most often (referring back to my post about moving PeerTube instances, you’ll know I use quite a few). So, I designed one; Heidi typeset and kerned the design and vinyl cut it for me, using the Cricut Maker and heat press in the studio; and, eventually, I’d had enough expressions of interest that we went ahead and put a printed version up on our studio shop
I had this idea that I’d love a shirt like that, with the names of the Fediverse platforms I use most often (referring back to my post about moving PeerTube instances, you’ll know I use quite a few). So, I designed one; Heidi typeset and kerned the design and vinyl cut it for me, using the Cricut Maker and heat press in the studio; and, eventually, I’d had enough expressions of interest that we went ahead and put a printed version up on our studio shop.
I’m really happy with the way this came out. I’ve also designed a few other shirts (albeit, not Fediverse-related), so those may hit the shop in the future.
I’m all-in on the Fediverse as the best way to own my own data and network. It’s based on an open standard (ActivityPub), loosely-coupled and open source services, and everyday people (and, it isn’t subject to the whims of lying, narcissistic billionaires). You can think of it as the next iteration of the open web, with social features baked in.
I’ve got a number of accounts that correspond with content that the different networks are good at – posts on Mastodon, photos on Pixelfed, reading habits on Bookwyrm. If I post something on one that I think my followers on a different network may like, I can boost it directly for them to see; or, folks can follow me on whatever platform they choose. My WordPress blog has federation switched on, too, so you can if you like follow @andypiper@andypiper.co.uk, and read my blog posts directly in your Fediverse platform of choice. (of course, RSS remains another excellent way to follow my blog).
By the way, if you’re curious what the buzz about the Fediverse is all about, I recommend Elena Rossini’s newsletter The Future is Federated – in the edition that was published today, she did a great job of explaining some of the interoperability between federated networks, from the perspective of, you know, just actually, using them, rather than from a deep technical angle. Worth following!
One of the other services I’ve been using is PeerTube, a federated alternative to YouTube. Up until now I’ve been on Diode Zone. However, that instance recently started to run into some storage issues, and also switched off the live broadcast feature that I’d occasionally used to stream some 3D printing and pen plotter content (this is straightforward to configure in OBS, by the way). I really appreciated my time on Diode Zone, but I’ve chosen to move across to MakerTube, a relatively newer instance dedicated to “makers, musicians, artists and DIY content creators”.
One of the core elements of the Fediverse is data ownership, and some form of portability. When I first joined Mastodon I started out at mastodon.social/@andypiper – if you visit that profile page now, you’ll find my posts starting in November 2016 and ending in November 2022, when I moved over to my current home, macaw.social – there’s a large message that points you at my current location if you look at my original profile. When I switched instances, my whole follower network went with me, seamlessly – unlike, for example, Twitter, where I lost everything when I deleted all my accounts; or Facebook, which heavily relies on its lock-in – read Cory Doctorow‘s excellent book The Internet Con to understand what’s happening there.
The current state of portability is absolutely not perfect – in the case of Mastodon, there’s a process which enables to you to migrate from one server to another, and that automatically resubscribes you to your network, and your followers to your new account, but it’s currently not technically possible to take the past posts with you (there is a W3C Social Web Incubator Community Group taskforce that is working on a more complete set of specifications for data portability that may help to improve this in the future).
Here is how I moved from Diode Zone to MakerTube:
Requested an account on MakerTube. They want to know who is part of the community, their content intentions, etc. I explained that I’m a maker and I also may plan to use the streaming feature.
Requested an export of data from Diode Zone.
this was technically a little bit frustrating, as it got stuck the first time (likely because I had requested it when there were earlier storage issues), but the instance owner was really kind and helped to clear the stuck process.
it was also a bit annoying because of the size of my export including the videos, which was a lot of gigabytes; so I ended up having to run a script that kept running wget with a resume flag to get the data in chunks.
Setup the basics of my new account on MakerTube.
Imported the export from the other instance.
Modified a few places that were pointing to Diode Zone, such as my links page, and also updated a few embeds that were loading videos from my previous account, such as some of the Fedidevs.org meeting recordings.
… while I was doing that, I thought I’d take advantage of PeerTube’s multi-channel capability, and created a specific channel for Fedidevs content, which you can now follow @fedidevs from other Fediverse platforms.
There were a couple of slightly rough edges, but nothing very significant:
Unlike Mastodon, PeerTube does not run a process to tell your followers that you have moved, and to resubscribe them to your new account. In my case I didn’t have a huge number of followers, but I will be posting a video there to say that I moved, and I also updated my profile information to point to the new instance. It did re-follow the accounts I had followed, but didn’t do the other side of the process.
The new instance imported my playlists, which was great – but some of them were playlists of my own videos, which I tend to create for curation and organisation, and of course, they still pointed at the videos on the previous instance. This was fairly straightforward to fix, just removed and re-added the videos on the new instance.
So there you are. You can follow my entire MakerTube account, my main channel, or the Fedidevs channel, if you like. I also have it set up to import future content that I may choose to post to YouTube, so that it has a free and open backup that Google can’t delete if I ever lose my account there.
I’m still frustrated that WordPress doesn’t seem to have a good integration for PeerTube content yet – I can paste a YouTube link here and get an embedded video, that’s less easy for PeerTube – but, I’m hopeful that will improve in the future.
Today, I received some fun post from some lovely people in New York City. Those in the know, may recognise these stickers as the logos of Glitch and Fastly.
I’ve been using Glitch to write and host web apps for quite a few years now – it is super helpful when working in a role like developer relations, needing to rapidly spin up demos, examples, or to demonstrate new features. A couple of years ago, Glitch came together with Fastly, and in the past couple of months their
... show more
Today, I received some fun post from some lovely people in New York City. Those in the know, may recognise these stickers as the logos of Glitch and Fastly.
I’ve been using Glitch to write and host web apps for quite a few years now – it is super helpful when working in a role like developer relations, needing to rapidly spin up demos, examples, or to demonstrate new features. A couple of years ago, Glitch came together with Fastly, and in the past couple of months their new developer platform vision really started to come together.
If you haven’t been keeping up with what they have been up to, and were not able to be at their recent special developer event in NYC (don’t worry, I couldn’t get there either), there’s a helpful ~6 minute video that summarises the announcements. I’m particularly interested and excited about this because I know and respect the folks involved – Anil Dash, Jenn Schiffer, Hannah Aubry, many others across their teams – and I know that they get and they care about developer experience, Open Source, and the free and open web. I’m talking about the big stuff, the infrastructure, the stuff that needs to invisibly just work in order for the web to run; and also the smaller things, the quirky indie little pieces, the fun and new experiences, helping people to learn to code and to be creative. It’s no exaggeration to say that Fastly’s Fast Forward program is a massive supporter of Open Source, open standards and the Fediverse. All of these things are reasons why I love Glitch & Fastly.
[ if you’re curious about the sorts of things I’ve been building or working on from a code and web perspective, I’ve also spruced up my GitHub bio, and I have a more general gallery page on GitHub that has links to the source and deployments of different projects – some of which are links to those Glitch apps above ]
Thank you for the stickerage, Glitch friends! And, congratulations on the new Fastly Developer Platform! I’m looking forward to continuing to use your cool technologies 👍🏻
Resources page for Andy Piper's talk on the history of Computer Art, pen plotters, and more. Explore further with links to exhibitions, contemporary artists, tools, and reading materials.
When I poked around with Postmarks (a federated bookmarking service) last year, I briefly mentioned another federated server, Shuttlecraft… but I never followed up on that.
Shuttlecraft is a single-user ActivityPub-compatible application that can run on a platform like Glitch – in fact, it’s more-or-less a single-click deployment to fork and install on Glitch.
I’ve written about being all-in on the Fediverse and I generally try a lot of different things out. I created a Shuttlecraft instance for myself around the same time I set up Postmarks last year – both run on Glitch, both are somewhat experimental / in-progress and unfinished.
... show more
When I poked around with Postmarks (a federated bookmarking service) last year, I briefly mentioned another federated server, Shuttlecraft… but I never followed up on that.
Shuttlecraft is a single-user ActivityPub-compatible application that can run on a platform like Glitch – in fact, it’s more-or-less a single-click deployment to fork and install on Glitch.
I’ve written about being all-in on the Fediverse and I generally try a lot of different things out. I created a Shuttlecraft instance for myself around the same time I set up Postmarks last year – both run on Glitch, both are somewhat experimental / in-progress and unfinished. The fact is that I use my Postmarks instance all the time for logging links of interest, but Shuttlecraft has really just been sitting there without very much interaction. That’s mostly deliberate, it’s a playground more than anything else, and it can be handy for playing with interactions with other ActivityPub servers.
However – yesterday I thought I’d take a quick look at it, only to discover that my Glitch app was down.
After doing some digging in the console and logs, I found that Node was complaining about being unable to load a module. That was surprising to me, as I hadn’t changed anything there for a long while. It turned out that what was actually happening was that the container’s disk allocation had run out, and the log message itself was spurious.
To be fair, the Shuttlecraft README is very clear that a) it’s an unsupported side-project from the author Ben Brown, and b) …
all data is written as PLAIN TEXT FILES to the disk.
Right now, the app builds an IN-MEMORY INDEX of EVERY SINGLE POST. This will work for several thousand posts, but … maybe not for 10,000s of posts. I’m not sure how far it will go. I have ideas about being able to shard the index into multiple files and page through it, etc. But.
It took a bit of very messy and imperfect cleanup inside hidden directories in the Glitch container, but for now, I have my small instance back up again. Realistically, it is probably not worth following / I don’t look at it much / it will fall over again, before much longer – but it’s also a fun thing to tinker with, and the code is quick to tweak in the Glitch editor.
In the past few months, quite a few things have changed in the ecosystem – for example, I’m now on Threads, and my account is federated out from there. Shuttlecraft is missing something in its ability to play nicely with that, which is a bit sad… maybe I’ll have to have a look into it at some stage. I’ll have to look at some more Fediverse-compatible apps soon, hopefully more resilient ones! 🙂 Fediforum is coming up next month, and that’s usually a great place to learn about what developers are working on across the ActivityPub ecosystem.
Today, I received some fun post from some lovely people in New York City. Those in the know, may recognise these stickers as the logos of Glitch and Fastly.
I’ve been using Glitch to write and host web apps for quite a few years now – it is super helpful when working in a role like developer relations, needing to rapidly spin up demos, examples, or to demonstrate new features. A couple of years ago, Glitch came together with Fastly, and in the past couple of months their new developer platform vision really started to come together.
If you haven’t been keeping up with what they have been up to, and were not able to be at their recent special developer event in NYC (don’t worry, I couldn’t get there either), there’s a helpful ~6 minute video that summarises the announcements. I’m particularly interested and excited about this because I know and respect the folks involved – Anil Dash, Jenn Schiffer, Hannah Aubry, many others across their teams – and I know that they get and they care about developer experience, Open Source, and the free and open web. I’m talking about the big stuff, the infrastructure, the stuff that needs to invisibly just work in order for the web to run; and also the smaller things, the quirky indie little pieces, the fun and new experiences, helping people to learn to code and to be creative. It’s no exaggeration to say that Fastly’s Fast Forward program is a massive supporter of Open Source, open standards and the Fediverse. All of these things are reasons why I love Glitch & Fastly.
[ if you’re curious about the sorts of things I’ve been building or working on from a code and web perspective, I’ve also spruced up my GitHub bio, and I have a more general gallery page on GitHub that has links to the source and deployments of different projects – some of which are links to those Glitch apps above ]
Thank you for the stickerage, Glitch friends! And, congratulations on the new Fastly Developer Platform! I’m looking forward to continuing to use your cool technologies 👍🏻
Resources page for Andy Piper's talk on the history of Computer Art, pen plotters, and more. Explore further with links to exhibitions, contemporary artists, tools, and reading materials.
For a while eSIM management on Mobile Linux, such as postmarketOS, was just a dream. Now it’s real!
Background Nowadays many smartphones and other devices have a so-called eSIM - or eUICC - built-in to the device.
Let me start off by saying: If you just want to have a working system to do your thing with minimal effort, Slackware isn't for you (anymore).
Running Slackware today is like being gifted a Ford Model T by a weird, bearded museum curator, and then finding out that after some minor modifications and learning how to drive it, you can keep up with any modern car on the road. Only it has no ABS, AC, power steering, starter motor, crumple zones, airbags or seatbelts.
Most people who still run it (by any realistic estimate, fewer than 10000 people in the world now) have been running it since the 90's and follow the advice not to change a running system to the letter. So why should anyone who hasn't studied CompSci in Berkeley in the 90's try it today?
First of all, the most widely known criticism (it has no dependency resolution) is a bit of a misunderstanding. Slackware is different. The supported installation method is a full installation, which means you install everything in the repository up front. That way, all dependencies are
... show more
Blog post alert
Let me start off by saying: If you just want to have a working system to do your thing with minimal effort, Slackware isn't for you (anymore).
Running Slackware today is like being gifted a Ford Model T by a weird, bearded museum curator, and then finding out that after some minor modifications and learning how to drive it, you can keep up with any modern car on the road. Only it has no ABS, AC, power steering, starter motor, crumple zones, airbags or seatbelts.
Most people who still run it (by any realistic estimate, fewer than 10000 people in the world now) have been running it since the 90's and follow the advice not to change a running system to the letter. So why should anyone who hasn't studied CompSci in Berkeley in the 90's try it today?
First of all, the most widely known criticism (it has no dependency resolution) is a bit of a misunderstanding. Slackware is different. The supported installation method is a full installation, which means you install everything in the repository up front. That way, all dependencies are already resolved. And you have a system you can use equally well on a desktop or server. It uses 20GB but disk space is essentially free now.
What if you need something that isn't in the repo? Well, do whatever the fuck you want. There's Slackbuilds which aren't officially supported but endorsed by Slackware's dev. There's Sbopkg, a helper script with dependency resolution very much like Arch's AUR helpers. There's Flatpaks. There's the old school way of compiling from source and administering your own system yourself. Slackware doesn't get in the way of whatever you want to do, cause there's nothing there to get in the way.
It's the most KISS distro that exists. It's the most stable one, too. Any distro-specific knowledge you acquire will stay valid for decades cause the distro hardly ever changes. It's also the closest to "Vanilla Linux" you can get. Cause there really isn't anything there except for patched, stable upstream software and a couple of bash scripts.
Just be mindful of the fact that Slackware is different (because the Linux ecosystem as a whole has moved on from its roots). One example: Up-to-date Slackware documentation isn't on Google, it's in text files written by the guy who maintained the distro for 31 years, which come preinstalled with your system. Or on linuxquestions.org, where the same guy posts, asks for input from users, and answers questions regularly.
It's still a competent system, if you have the time and inclination to make it work. And it's a blast from the past, where computing was about collaborating with like-minded freaks on a personal level. And I love that.
If you hate bloat you like Slackware. It doesn't assume anything about how you want to use your computer, so it's more painful for a lot of folks. Other distros will try to do things for you and will ultimately end up doing something someone doesn't want. With Slackware you learn a lot and you get a rock-solid system that will do whatever you like, but you have to be willing to manage it.
Late reply, but I suppose you're correct; I think the option to choose what to install, the lack of pre-created desktop stuff ("home/$username/Videos" for example) and the requirement that you handle software that isn't in the base install all make a Slackware installation less bloated than most. Maybe not at install time, but over the life of the install you end up with less garbage IMHO.
Den befängda och oseriösa legaliseringsdebatten. Debatten om avkriminalisering och legalisering av cannabis och narkotika är ofta helt bisarr och befängd. Anhängarna av nuvarande lagstiftning vill ha högre och högre straff även för den minsta mängd narkotika.
I januari 2024 stoppades några polispatruller en kvinna i en bil utanför Västerås. De hittade då cirka 50 kilo hasch i bilen. Spår ledde vidare till ett garage i Sala. När polisen undersökte garaget hittade de över 100 kilo hasch gömt i flyttkartonger och lastbilsdäck. Det har nu lett till att två polisanställda är åtalade för grova brott.
Just a quick note to recognize that the first lines of PieFed code were published on the 28th July 2023, just over a year ago. Since then there have been 1400+ changes made by 9 people, involving adding 88,000 lines of code and removing 28,000 lines. The issue queue has 98 open and 99 closed issues.
While join.piefed.social went live in October 2023, it wasn't until time off work over the christmas holidays enabled a big push to get it ready that piefed.social went live on 4th January 2024.
Since then piefed.social has federated 190k posts, 2.3M comments and 19M votes with 1900 other instances of various types. Besides piefed.social there are 5 other PieFed instances that I know of.
What a year it's been! I've grown significantly as a developer, had a lot of fun and hopefully contributed something meaningful to whatever the fediverse is becoming. Long may it continue!
MNT have been making devices that aim to be “open source, accessible and modular” for a number of years. I didn’t get their original Reform laptop, but I’ve seen those around at events like FOSDEM and I’ve been following the team’s progress with interest. When the Pocket Reform was announced I was immediately intrigued – a smaller form factor, 7 inch full Linux system with open hardware that is easily portable.
MNT have been making devices that aim to be “open source, accessible and modular” for a number of years. I didn’t get their original Reform laptop, but I’ve seen those around at events like FOSDEM and I’ve been following the team’s progress with interest. When the Pocket Reform was announced I was immediately intrigued – a smaller form factor, 7 inch full Linux system with open hardware that is easily portable.
I went for the Hyper edition, which came with a beautiful Piñatex sleeve, SSD, and printed manual. Purple, of course, because I want to be on brand 🤓 and, also because I love it!
I posted a very brief unboxing video. I’ve now had a couple of weeks to occasionally tinker, and I’ve been putting off writing about it all in part because there’s a lot of things to dig into!
One of the things I like, surprisingly, is the chunkiness of the machine. It is really well constructed, solid, and feels great. Fits in a small cross-body bag or satchel. It’s less than half the size of a 14 inch MacBook Pro – you can more than fit two of them side by side on top of the Mac – but it is about 3 times thicker – and that’s OK, because, it is in service of making the innards very accessible and user serviceable. It comes with a complete manual and schematic, which is something I’ve not had in a computer since the 8-bit machines of my youth! The top half contains the mainboard and display (all the ports are in the top half), and the bottom contains the battery cells and mechanical keyboard. The upper panel has a copper layer and acts as a large heatsink for the processor module. The keyboard is ortholinear, which means it is a direct grid layout rather than offset row-by-row. It’s the first time I’ve used this format, which – along with the smaller keycaps – has made it a little challenging to learn, but I’m doing pretty well now. The trackball is nice and responsive. The backlit keys are easy to adjust. The screen is excellent – bright, and sharp. Actually I think the screen is probably the aspect that has impressed me most so far.
In terms of ports and connectivity – I’ve yet to hook up to an additional screen via the micro-HDMI connector, but I’ve used the USB-C connections (one of which is used to charge / power the machine), and the micro SD slot. The industrial iX connector for ethernet is likely a good choice for the target niche, and certainly does give more space on the motherboard than an RJ45 socket would… but I’ve yet to put my hands on a passive adapter to enable me to plug in to a wired connection, so it’s currently not as useful to me.
There’s a lot more to talk about, primarily (but not exclusively) on the software side, and also around hardware enablement. Out-of-the-box the Pocket Reform runs Debian unstable from MMC, with some customisations to provide a nice getting started wizard. It is important to point out that this is a machine for hackers and tinkerers – although it works very nicely, it’s not all fully baked in places – for example, the firmware for the system controller (a Raspberry Pi RP2040 chip) is being tweaked and tested, and I’ve already tested one update to that. The other day I posted about some issues with an NVMe SSD – that, unfortunately ,was on this machine. I actually think there was a physical hardware issue with the drive, as I’ve now replaced it with a higher-performance NVMe and things are moving along nicely (while the problematic SSD continues to report errors when accessed in an adapter over USB). I also managed to temporarily brick the machine by corrupting the uboot in flash, and needed to rig it up with Dupont wires on headers and access the machine from another via USB to get back to where I wanted to be. Very Open Source! 😎 but, I’m comfortable with this, and knew what I was purchasing. The forums and IRC have proven to be useful so far and I’m enjoying learning as well as hopefully (!?) helping the MNT team through my feedback and bug reports. I have a huge amount of respect for what they have built, their ethos, and their commitment to making this as open hardware as they can.
I should be receiving a modem / WWAN card for the second internal expansion slot shortly. I’ve been both learning the Sway desktop environment and also working out how best to organise my setup, so there will be more to cover in future. I particularly want to play more with the onboard I2C, and other hardware opportunities, as well – for example, potentially swapping in a Raspberry Pi CM4 if that becomes a modular option in the future. andypiper.co.uk/2024/08/06/mnt…
Today, I’ve been trying to resolve a strange error that just started showing up on one of my Linux devices – specifically, errors in the dmesg output on boot, during runtime, and at shutdown: [17411.181979] nvme0n1: I/O Cmd(0x2) @ LBA 587247848, 8 blocks, I/O Error (sct 0x2 / sc 0x81)[17411.181991] critical medium error, dev nvme0n1, sector 587247848 op 0x0:(READ) flags 0x800 phys_seg 1 prio class 0[17774.406908] nvme_log_error: 12 callbacks suppressed[17774.406924] nvme0n1: I/O Cmd(0x2) @ LBA 633393408, 8 blocks, I/O Error (sct 0x2 / sc 0x81)[17774.406937] blk_print_req_error: 12 callbacks suppressed[17774.406940] critical medium error, dev nvme0n1, sector 633393408 op 0x0:(READ) flags 0x800 phys_seg 1 prio class 0[17774.590170] nvme0n1: I/O Cmd(0x2) @ LBA 633393440, 8 blocks, I/O Error (sct 0x2 / sc 0x81)[...][21832.814513] nvme0n1: I/O Cmd(0x2) @ LBA 1145110912, 8 blocks, I/O Error (sct 0x2 / sc 0x81)[21832.814544] critical medium error, dev nvme0n1, sector 1145110912 op 0x0:(READ) flags 0x800 phys_seg 1 prio class 0 Needless to say, this doesn’t look happy; but, fsck is not showing errors. This is a Debian install, and I chose to install the nvme-cli and also the smartmontools packages; util-linux and e2fsprogs were installed by default and contain fsck and also the badblocks tool; blktool and hdparm are potentially also useful. The Arch Wiki seems to be particularly helpful on the nvme tools.
I’m currently booted from an alternate source (an SD card) and attempting to run fsck -fcDy /dev/nvme0n1p1 in an effort to look for bad blocks. I’m still seeing these I/O error messages scroll up the dmesg -w output in a second terminal even though fsck is not obviously telling me that there are problems. Based on my research I’m starting to think that this might be a hardware issue with the drive. This is a bit of an unusual setup for me (ARM64, uBoot is in play with eMMC as well) but fortunately the system is recent, so a clean rebuild is not out of the question…
Fairly sure that I’ll end up getting another M.2 NVMe drive (at least they are relatively cheaper than they used to be), but writing this down for now as I continue to investigate. Fun times…
A cattle painkiller introduced in the 1990s led to the unexpected crash of India's vulture populations, which still haven't recovered to their former glory.
È possibile pubblicare post su Lemmy dal tuo blog WordPress? Se questo post apparirà nella community @fediverso@feddit.it allora questa ipotesi sarà confermata!
È possibile pubblicare post su Lemmy dal tuo blog WordPress? Se questo post apparirà nella community @fediverso@feddit.it allora questa ipotesi sarà confermata!
2023 utbröt en konflikt inom Foxtrotnätverket mellan en gruppering kring Ismail Abdo (Jordgubben) och en annan gruppering kring Rawa Majid (Räven). Grupperingen kring Ismail Abdo har av polisen döpts till Rumba. En del av dessa nätverks verksamhet ska ha varit och är postorderhandel med narkotika.
Anders_S
in reply to Hans Norebrink • • •