LINUX Unplugged - 663: The 99.8% Rescue

Episode Date: April 20, 2026

We all have data to rescue, you just don't realize it yet. This week we build our own custom live rescue distros, recover real data, and show you how to make your own.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:💥 Gets Sats Quick and Easy with Strike📻 LINUX Unplugged on Fountain.FMLinuxFest Northwest 2026 - Back to Root — April 24-26, 2026 - Bellingham, Washington6 ultimate boot CDs (system rescue) for troubleshooting Windows PCsSystemRescue — A Linux system rescue toolkit available as a bootable medium for administrating or repairing your system and data after a crash.Finnix — Debian-based, CLI-focused with a comprehensive set of repair tools. Recent updates improved modern hardware support, boot speed, zram compression, and SSH remote access.Changes/ModernizeLiveMedia - Fedora Project Wiki — Modernize the live media by switching to the "new" live environment setup scripts provided by livesys-scripts and leverage new functionality in dracut to enable support for automatically enabling persistent overlays when flashed to USB sticks.Fedora 41: Modernize the live media Overlay Features?aegis-boot — A signed UEFI Secure Boot rescue environment that lets operators pick any ISO from a USB stick's data partition and kexec into it — without leaving the chain of trust.netboot.xyz — A network-based bootable operating system installer based on iPXE.Greg's nixos-configSelf-Hosted 129: Forged Alliance — The battle for code forges is heating up. We chat about HexOS' big promises and get excited about Meshtastic.pocket-id — A simple OIDC provider that allows users to authenticate with their passkeys to your services.What Is Claude Mythos—And Why Anthropic Won’t Let Anyone Use Itv4call — v4call is a decentralised video, voice and text platform built on the Hive blockchain. Users set their own rates for receiving calls and messages.Pick: Metadata Cleaner — View and clean metadata in filesMetadata Cleaner on FlatHubPick: mat2 — mat2 is a metadata removal tool, supporting a wide range of commonly used file formats, written in python3Pick: Little Snitch for Linux — Every time an application on your computer opens a network connection, it does so quietly, without asking. Little Snitch for Linux makes that activity visible and gives you the option to do something about it.littlesnitch-linux-flake — A Nix Flake for Little Snitch for LinuxPick: PCAPdroid — No-root network monitor, firewall and PCAP dumper for Android

Hello, friends, and welcome back to your weekly Linux doc show. My name is Chris. My name is Wes. And my name is Brent. Hello, gentlemen. Well, coming up on the show this week, it's easier than ever to build your own custom live rescue environment. So we're going to take a look at some of our favorite tools and some new ones we came across, then recover some real data and also go over how to build your own. They'll round it out with some great boosts and picks and a lot more.

So before we get to all of that, let's say time appropriate greetings to our virtual lug. Hello, Mammaloo. Hey, Chris, hey, Russell. Hello, Brent. And good morning. And good morning to all of you. And shout out to everybody in our Matrix chat room, too, that everybody joins us. It's a Tuesday on a Sunday, and we have a lot of fun.

We've been going now for about 45 minutes or so, just to kind of warm up, you know. That's the secret to a good show. You don't just come in cold. Certainly not. And we do our warming up live. We also make it available for our members. Also, good morning to our friends over at Defined Networking. Go to defined.net slash unplug.

If you're building network infrastructure, you've got to think about the long game here. Managed Nebula is a decentralized VPN built on the open source Nebula platform for people who want speed, resilience, and most importantly, real ownership of the stack that you depend on. Nebula was originally built to securely connect Slack's global infrastructure, and it still reflects all kinds of that scale first design today, with optional self-hosted lighthouse nodes for extra control so you manage the resilience and the source of truth. I use it, you know, just a host backups between two hosts, and companies like Rivian use it for securing their entire global fleet of cars. The range there is incredible. Get started 100 hosts for free, no credit card required.

Go to Defined.net slash unplug. Check it out. Support the premier sponsor of the show. We appreciate them very much to find.net slash unplugged. Well, next weekend is Linux Fest Northwest. Starts, as you're listening to this, just five days away, which is probably physically less time than Brent has to get here. How about that? I think I was doing the math, and it puts you roughly at eight-hour days. No, my math says 10-hour days.

Really? Yeah, yeah. I did a little look up here before the show, because, like, someone should calculate this, and I figured I should probably do that. 3,745 kilometers, which in miles is, well, the same. 36 hours, Goog's claims. 36 hours Google time, which means probably about 42, 43 hours van time. Yeah, yeah, yeah.

Still possible. Yeah, just a little time. I hope everything's in great shape. I would go over everything. Oh, yeah, I've been charging the battery for the last day because I left it, you know, I left my little tracking cell phone plugged in and it totally killed the battery all the way down to zero. So starting off. So road ready is what you're saying.

Road ready. Road ready. I mean, the opportunity here for a betting pool, you know, does he make it on Saturday? Does he make it after the live show on Sunday? That's good odds, you know, good odds. We'll see. We'll see. But we do have ourselves on the schedule.

At least Wes and I'll be there. Probably, Brent. If you want to boost in between now and then, we can kind of see how old folks. did in that episode. Yeah, you can make your prediction and then the sats will also go to his gas to get him here. So, you know, it's a win-win. You should also boost in which part of the van is going to fall apart halfway throughout the continent and see if we win. What? No, you don't like that. That stress you out. Oh, sorry. Why are you putting that out there? What are you thinking? What are you doing? Why? I'd like a heads up before I get there. You've got like 45 hours of solid, you have more than the average work week of driving ahead of you and you're putting that out there. You realize. You realize. I'd like a heads up before I get there. You realize. You realize. You're you're going to do more than a week's worth of work of driving. It's going to be fun. You want to join me? You can fly out here.

We'll do it together. If I didn't have other shows, I would. That's why I love you. Also, want to just keep putting it out there. The BSD Challenge for episode 666. We're still collecting your ideas on how to make that great. I think we'll technically probably kick it off in 665 and do the completion in 666.

Don't you think that's the way to do it? We need a little time to... Yeah, and then that's the canonical episode of the actual challenge and the results will be 666. That's our plan. We'd love your ideas, and we'd love you to do it with us, too. We covered it more in last week's episode in detail. And I think the mission will be

to try to have all the details pretty solid for you in probably next episode or so. The tricky thing about that is it's Linux Fest. So that might not happen, but by 665 for sure. I'll be talking PSD at LinuxFet? What's that? Just kidding. It's a welcoming Unix-like environment.

Yeah, that's true. That's true. Well, we were taking a look recently because we have had some system outages, different reasons, different stories, which we'll share later in the show. And so it was a great opportunity to just poke our heads up and see what new tools are out there to build a rescue session on Linux. And there are new tools, some that are coming just around the corner that aren't out yet that we'll tell you about.

And they are extremely handy because it makes a podcast. to extend these things in ways that wasn't originally possible with these. So before we get there, I wanted to talk about how the Windows side does this for a second. Because this time around, I spent some time trying out what is considered the best rescue live session for Windows. It's Heron's Boot CD, and it uses that Windows PE tool to customize it. And, guys, it's great. If you are a Windows user listening to this, or if you are maybe at work in a Windows universe and you need a rescue environment that is Windows-based,

Heron's Boot City is really good. It's a very customized version of Windows 11. It's the best implementation of Windows 11 I've ever seen because it looks like Windows 2000. It's like all classically themed and everything. Basic standard start. But it's real Windows 11. Yeah.

That's so great. Yeah. Yeah, I know. I know. It is. It has a plethora of tools. anti-malware and antivirus, which is something you need on these. Lots of different backup and data recovery tools, hard drive management tools,

partitioning tools of various kinds, hardware analyzers to try to test your hardware or just give you a report. Password recovery, specifically stuff that's designed for earlier and now newer versions of Windows NT, both. I mean, I was telling the boys, it's like it's a kitchen sink. It's really a kitchen sink of utilities, but in the Windows world, that's kind of what you want, because you don't have a go-to package manager. Maybe you don't have a driver for the network card. So in the Windows world, I think it is really nice to just have all this stuff pre-installed.

It makes it one of the bigger live CDs we're going to talk about because it comes in at like 3.2 gigabytes. Still, that's not bad considering Windows. And just about every freaking recovery tool that runs on Windows is pre-installed on this thing. And it's well organized. This is what I've heard about. I just never quite intersected with it in my Windows adminning days. Same. And it's, you know, snarky Linux user here.

I'm like, oh, Windows is going to do a live CD. Oh, that's going to be rich. That's the attitude I went in. So I'm sitting, I got my notes up, and I'm like, oh, I'm going to roast this. And I'm like, the first couple of times I couldn't get it to boot. And then I swapped a couple of my VM options came right up after that. I mean, it takes a little bit because it has to do some sort of pre-boot hardware detection thing. I don't know quite how they're doing it.

And then once Windows loads, it has to run some pre-loaded modification. stuff to set up the networking and all that, but it does it all very quickly, relatively. And it was immediately a great experience. So if I were in this ecosystem and I was more comfortable in Windows to try to recover data, this is absolutely a way to go. And I'll put a link to this in the show notes. I ended up spending a lot more time with this than I expected just because I liked it so much. I wonder if it could be in a nice environment just to have as a quick way to get a Windows

thing running? Could you just boot it up in a VM or something as a... if you just needed to do something quick to test on Windows? Yeah, I wonder. I think the limiting factor, and I'm not positive. Is it like a PE environment? Yes. Ah, okay.

Yes. And I don't know, I don't think you could, I don't know if you could sign into a Microsoft account. Like, so I don't know if you could install things from the store. So you might be, yeah, limited to what you could actually get running on there. But just straight up trying Windows. Yeah, it works for that. It was, it was good.

Or maybe you needed to run a quick random Windows binaries, to run a line or something. Not a giant. I'm not saying install office on there or anything, but just. Or, you know, maybe you need just need a quick Windows VM, right? Yeah. You don't want to, because you don't have to go through the whole install,

which is so much fun. Yeah. All right, now, we have to give honorable mention to the classic. One of the OGs in this space, formerly known as System Rescue CD, now just goes by System Rescue. It's kind of the definitive Linux rescue toolkit out there, bootable.

You go in there, it's got just about every tool you could possibly imagine a minimal GUI environment if you want it. data recovery tools now clocking in at 1.2 gigabytes. Not bad. And I'm going to try to make my case later in the show, but I'm just going to present this now as we get into more.

I don't even know if you can find a smaller than 2 gig USB drive out there. Well, that's a fair point. That's a fair point. The next one we're going to talk about would still fit on a CD. But before I got off System Rescue CD, they have so much stuff pre-installed.

I think these distributions should consider also adding open code to these environments. And I know that's going to sound weird, but I'll try to make my case later in the show. I think they should consider adding open code as a recovery tool. And I think it matters a lot. We'll come back to that because first I want to talk about Phoenix. I will say, are you going to mention that, you know,

System Rescue recently added support for our favorite file system. ZFS? Well, I think it does have ZFS. already. But no, B-Cash-FS. Whoa. Oh, that's great. Just recently, end of last month, they got the kernel up to long-term supported Linux 61820 and the B-Cash-FS

tools and kernel module to version 1-373. So, yeah, there's also a bunch of stuff like there's IQ now. Okay, you got me beat there. But Phoenix is great if A, you're Davian first and B, you want something that could actually still fit on a

CD-ROM. So Phoenix clocks in at 577 megabytes. It's based on Debian, I forget, whichever one runs Linux 616, with some updated packages from that. And this also ships now, Phoenix as of 251, is the first version to release a OCI container image. Oh, nice.

And I'm wondering what you could use that for. It feels like there's something you could use that for. I'm not quite sure what, but could be useful. And it's Debian-based, C-L-I-focused. So that's really where they go out with a number. of comprehensive repair tools, and they've done things like recently they've turned on Z-Ram compression, they've enabled SSH remote access and other things to kind of make it useful. I suppose one version might be if you just booted any old random ISO that had internet access,

you could pull down that container and then mount in, you know, whatever actual disks you were trying to do recovery on and you have all of its tools. Yeah, that would be very useful. Sort of on-demand rescue environment. Mm-hmm. Hmm. So that's Phenix F-N-N-N-I-X. link to that in the show notes.

Before we get into what we ran into recently, I want to talk about how this is going to get so easy to create your own in just probably about a year, maybe less. This has been in the works for a while, but Fedora 45 plans to enable persistent overlays

when you flash Fedora to a USB stick. Now, our buddy Neil has been working on this in Fedora since 2024, and his original code, he started on this since 2020. So that's how long feature has been in the works. And the idea is, is that when you flash Fedora to a thumb drive, you'll leave some unallocated

space on that sucker. And then, by default, when Fedora 45 boots, there'll be a boot menu entry selected by default, by default. Hey. They will just initialize that new partition and mount a writable overlay to it. Now you have a persistent storage to install packages, tools like open code, or DD Rescue, whatever it might be that you need. Or away even just for some of the stuff,

those tools write to dot files in your homest or whatever you could have. And you can, and you take that thumb drive with you and that stuff persists. And the next machine you go to, it has that. It has your SSH key setup perhaps. Maybe it has your, you know,

whatever you need to be logged into, logged in, your password manager, whatever it is. It's got it. Super neat. And all you have to do is just leave a little unallocated space on the disc and Fedora 45 will just take care of the rest. And that is going to be such an unlock

for building your own custom desk. It's not there yet because it requires some upstream work in Drag-Cut and some other things. But it does make sense, though, right? Talking about sizes of USB drives, like, if all you're doing is DD and this thing on there, there probably is some space at the end of that that isn't actually being used. Yeah, when do you not? Right, right? Like, even a 16 gig, really.

I mean, these images are, what, four or five gigs sometimes? Yeah. You touched a bit on the, I don't know, the complexity, the things needed to get across the line. and I just like the text here in the change. Since we introduced live media in Fedora Linux 7, the actual mechanism in which the live environment sets itself up has been complex and intricately tied to the method in which we produced the media using Kikstarts.

The nature of the implementation of those scripts means that they are hard to understand and debug, which has caused problems in the past whenever we've needed to update them. So hopefully this kind of change would also set to modern, you know, not only would you get some better features, but like better, more maintainable plumbing too. I think we could talk about this when it lands, but it's also a great example of, I guess, on the surface, it seems like a seemingly simple feature. And yet it's been in the works for years and years because of just the upstream cooperation it takes.

And it's a great example of a small feature that just takes persistence to get actually shipping. But when 45 lands, 44 is next. So the next release after 44, when that lands, this will just, in theory, be a default feature and you just create your own live environments just by using it. I mean, that is going to be massive for just everyone out there. So that will probably be one of the ways you could just make your own with very little fuss. What we're going to talk about is not that. First, I want to thank our members for making this episode possible. I mean, quite truly making it possible.

You can support the show by going to Linuxonplug.com slash membership or support the whole network and get access to all the features for all the shows at jupiter. When you become a member, you get access to the bootleg version, clocking in about an hour right now. or you can get yourself the ad-free version, which still has all of Editor Drew's nice touches and is a little bit tighter. So if you got a little shorter schedule, it's perfect for that,

and you don't have to hear these segments over and over again. Or you can just support the show on autopilot and just do the right thing, knowing that this show is made for the community and possible because of the community. And we really do support and appreciate your support. And really try to do right by you because of it, because you are our number one client.

You are the number one customer. Thank you so much. And if you haven't done it yet, Could be a great time with Linux Fest coming up. You could show up and say, hey, I just became a member. Shake a hand. We'll give you a virtual beer at least because we can't afford to buy everybody a beer. But the sentiment will be like we got you a beer, right?

Yeah, let's go with that. Yeah, sure, sure, sure. Well, I decided to do it not that way. Why would you do it the standard way when you could do it the non-standard way? Really, this whole rescue USB drive has been a progression for me in, supporting family members mostly. Years ago when I set up my parents and brothers and other family members on Linux, I thought some stuff occasionally goes wrong.

Sometimes hardware, sometimes, you know, they're traveling and you get run over, you know, your laptop gets run over by a dump truck or something like that. And I always wanted some way to be able to help family members from a distance if their computer goes down because I felt responsible. If I'm going to switch you to a new operating system that I, tout as, you know, better, then I want to have some way to help you if it's not better. So I have always convinced everyone I moved to Linux to always, always, always, always carry a live USB drive that's dedicated with Linux on it so that something goes wrong, you can boot

into that thing and I can help you. Now, that has been very useful. And I have to say people have followed that advice quite nicely because they typically forget the USB drive in their laptop bag and then like a year and a half later I'm like, hey, do you have that USB drive? Like, whoa, do I look for that somewhere? Yeah, it's in your bag somewhere. You just got to find it. And sure enough, it saves the day. But the hardest part I always had was that these were just live drives. And so they were always vanilla. So whenever I was booting it up to try to help someone, I was like, okay, I need to now spend the next 15, 20 minutes

guiding you through connecting it to the internet, getting me SSHSS or some kind of remote access, maybe through something like a RustS. So that process I wanted to eliminate. And I started working on that recently where I iterated this concept. And instead of having a live drive on a USB, just installed Linux to the USB. Yes, that is a terrible idea most of the time. But with a rescue you drive, you're only booting this thing up. Well, hopefully you're never booting it up. But if you need to, you're going to boot it up once or twice, maybe a year, right? So drive longevity with having a whole operating system running off of just the flash drive is a bad idea generally. But in this particular case, I thought that's maybe an okay compromise. Talk to me a little bit about before

you go further, because I actually just did the same thing for my process. I thought, why am I using these custom, why don't, why don't, why don't I just install to a thumb drive and then have something I can install tools to and whatnot? Did you, like, download a distro installer and run through the installer and point it at the thumb drive? Or did you, like, do an image and flash it? What was your process for that bit right there? Well, in this iteration, I just installed directly to the flash drive. So just treated that like any old install. Nothing special, just said, hey, please install. you know, please install Linux to this flash drive.

And then I did a couple things like connected it to the Wi-Fi network that they use most of the time and logged into my mesh VPN and things like that. So that at least I had a couple default tools available. And so then it was really just another Linux system on a flash drive that you could plug into any computer. And it would boot up and have a couple, you know, pre-set up configurations just to make the process easier. How was the performance? Well, you know, fine in a rescue situation, but, you know, you're running off a flash drive, so that's not great. And what I ran into was it works great until you feel like you need to update that drive.

Yeah, yeah, yeah. So I made one, you know, for my mother, which she carried and forgot about, which is perfect. And then I was like, hey, I'm around. like maybe I can update your flash drive to something more modern considering it's a year and a half old and we should have some modern Linux on there. What a good son. You know, I know. I'm thinking ahead. And then the update process was painful because it's way behind.

Right, because it's been a year and a half. So you're replacing every single package on that thing. And that is the exact thing flash drives are terrible at is the random, you know, rights to all over the blight. It took forever. to the point where I thought, I need to iterate this idea again because this is not working for me. So while I was waiting for it to update,

I was like, there's got to be a better way. So it turns out, of course there is. So my next iteration, which I'm always looking for advice, you help me make this better, but the next iteration just worked on that idea and did a bit of a hybrid. So I used NixOS

because NixOS is really great for building your own custom versions of something. So I already have a NixOS configuration for my mother's laptop. But then I was like, well, I just want to give her exactly what she's used to, but trim that down to just the essentials. And then let's do some other crazy stuff. Like maybe we could just boot the entire thing to RAM because I don't think we need to be using the flash drive at all, right? Just use it to hold the information and transfer that somewhere more useful when you're booting.

up the system. So sure enough, that is extremely easy to do. And the other way that helps is her current laptop only has one, it's like a really super slim, low budget laptop. She freaking loves that thing because it doesn't weigh anything, but it only has one USBA port. Oh, brother. So if you're plugging in a live USB drive, perfectly fine. But if you want to do anything like plug in a backup hard drive or something like that, you can't. So moving the entire system to RAM means you can pull that live drive out and a system still runs perfectly fine. So that is a

big plus as well. And other nice things like not installing to the drive, but creating an image instead using the NixOS installer ISO. So you can use

installation cdbase.nix module, which is exactly what the NixOS installer uses. And there's just pile on top a few of your custom configurations. So I just somehow managed to craft my own NixOS, not the installer, but a specialized NixOS live boot environment with my own customizations that automatically logs into my Mesh VPN. It has recovery tools on there that I like.

It even boots up fish on the terminal by default for me. And it has custom extensions in Firefox. and takes out a bunch of plasma stuff that I don't need by default on a rescue disk. And it's fantastic. Nicely done. This is very, very similar to what I did based off my desktop for my recovery process. I could just take most of what is my desktop, strip out some of the ancillary desktop applications and flat packs, and then I have all the tooling I need.

I have all my keys. Let's go. Like, it actually worked really, really well. but I did not do your copy to RAM idea, which I'm thinking... Oh, you've got to take that one. Yeah. Yeah, was that easy? That's one thing I haven't done either.

I mean, I've done it in the past with other ISOs, but not in the combo with this particular setup. Yeah, I've used this trick for years, actually. Ubuntu certainly has a kernel option, just called 2 RAM that you can use. So at boot, even if you're just on any live ISO, at boot, you can just change the kernel parameters before you boot and add 2 RAM, and it'll do that for you. So that I've used specific, I learned that specifically for my mother's laptop because it had this one USB problem I had to get around. So that was, that was a really nice trick. In NixOS, it's slightly different called copy to RAM, but it is a trick that the Linux kernel has been able to do for a very long time. So

it's a, it's a nice little thing to keep in, in your hat. It is really nice in a rescue environment to have things like your preferred desktop environment, to have like the ButterFS tools, or to have Firefox, or you know, like all these things that you kind of need to really get a system up and going. It's nice to just have that.

Yeah, so you don't have to spend a bunch of time every reboot, reinstall it is. Yeah, yeah, exactly. So now you have something you can keep. But how do you keep it up to date, right? That seems to be, the next level of this would be something that would auto update somehow, right?

Like my thinking has been, what if instead of a thumb drive, next time I install a system, the first thing I do is I install a rescue system in a little partition. Yeah, that's something I've done before and liked a lot. And then I install the rest of the system as a separate install. And then maybe there's a once a week reboot into the rescue environment that updates the rescue environment. And if the update is successful, then reboots back into the main environment. That's the bit I'm... You could probably set up a setup where you, like, flashed it instead.

if you wanted to, if you kept, if you isolated. That's a great idea. You build it, updated on the host system, and then you just flash the update. Yeah, which could also maybe work if you were still doing a USB drive too. Again, you'd want to make sure you'd separate it out with partitions or some mechanism for the data you wanted to keep. And just leave the drive plugged in. And then you could just update it that way, too. I believe PopOS implemented this kind of feature when they first came out and users loved it. One thing I would argue, although I, most of the reason I was trying to, you know,

to trim down this image as much as possible is that flashing to any USB drive is just slow and painful and you should only ever do if you really need to. But it also saves for the use case where if you're on the move

and you lose a laptop, it gets stolen, you drop it in the water or something like that. Maybe you still have a USB drive and you can install on a new machine you get at Best Buy or something just on the go. If you're you're just on the go, if

You have a partition. Obviously, you can't do that if you, you know, your entire laptop gets stolen or you left at a coffee shop or something. But you can easily have both solutions, I think. Yeah. True. Yeah. Well, that's, it's funny that you and I kind of both went down the same path, but you had success. You did recover pretty much all the data that you set out to recover with when you were working on a separate issue.

So tell me a little bit about that because I think you told the members, but I don't know if you told the audience that you were doing like two weeks of drive recovery splunking too. Just give us a, like, the quick recap on that. Yeah, when you love your family, you do crazy things for them. The things you do for your brother. So when your friend brother says, hey, I got some old, you know, the hard USB hard drives I've been backing up to for years. Now all of a sudden today, they don't seem to, you know, be behaving like they used to. So I thought, okay, I'll, I'll spend, you know, an hour helping him recover some date off,

some old hard drives that he's been backing up to that I think I probably suggested that he do that years ago. And, well, I was able to successfully use open code to help me just, like, test my assumptions on how I was going to approach recovering data. Luckily, I was able to do it from a distance, and I had a server already built there that I should have moved all those backups to years ago. Okay. Okay, so you, uh, okay, so you, you had a remote box. I did have a remote box with plenty of storage in it, just sitting there waiting. Okay, I got you.

That's handy. It is super handy. And then like you had made, you had made some assumptions about how you're going to recover the data and then you're like, hey, I'm going to run it past the machine, see what the machine says.

Yeah, so he was getting a couple read errors. So I thought, okay, one way to do is just, you know, use our sync to pull a bunch of data off. But I'm going to run into read errors, obviously. I could log those and then investigate later. But then, you know,

there's smarter people on the internet who've done it probably better than I have. So it turns out DD Rescue is a fantastic tool that's perfectly built for this use case. I'd heard of it forever. Never used it before. I always planned, hey, next time I want to do this, I'm going to use DD Rescue. But I'm not familiar with it. So having a little, I have to say, having a little open code wrapper around some of these, like,

bulletproof tested tools for data recovery is so nice. Because I could just think about the big picture. and then the implementation detail about, you know, which flags you need to use for a tool that you haven't used before. I didn't need to think about. So I could just worry about are we keeping the data as safe as possible and where to put it, how to treat it, what is an acceptable level of risk while doing the recovery? And then the implementation was just taking care of by the computer, which is a really nice way to do it. So it was sort of pulling the strings on DD Rescue for you?

Exactly. So I could say, okay, let's use DD Rescue, but treat the drive really kindly. It's a USB drive. It heats up. So let's reduce the bandwidth that the rescue is going to happen so that we don't heat this thing up. And then, of course, when I told my brother I was making good progress, he was like, oh, well, I found four other USB hard drives in my closet. Would you pull the date off those too? So I said yes, because I love my brother. And, but that meant I could get open code to help me just write some scripts to, so that I could just check in every couple hours and say, okay, well, pull all the data off all of the drives, but do it sequentially. And there's different file systems on here. So use check tools for the different file systems in line before you do the recoveries. And while you're doing the recovery, reduce the bandwidth for all the drives because they're all USB drives. But, you know, so don't saturate. the USB either because I don't want the USB bus to like fail for some reason four hours into the recovery of all these things. So it was nice to be able to put like safe parameters around both the hardware and the data on disk and the software to just do this as a gentle recovery from a distance. Like I'm literally thousands of miles away. And also to get a bunch of really sweet-looking reports about how the rescue went and all that stuff. So it's a great experience.

And you got almost all the data, right? You got like with with acceptance. Yeah, I got 99.889% of the data. Nice. You know? That's not so bad. Not so bad. And a happy brother, I'm sure.

Yeah, I'll still have to break to him that one of his files isn't working. Well, when you do that, what you do is you go over to like drive savers and you just get them the cost to do data rescue from drive savers. and then be like, but I did it for free, I just didn't get this file. You know, like, that's not so bad. Buy me dinner sometime. Yeah. Now, Wes, I know you've been messing around with a few things kind of adjacent to this area as well recently. And we have some cool tools linked in the show notes.

What from that batch do you want to talk about? Yeah, well, I will just say that it is kind of delightful how easy NixOS makes it to make your own, where you could just, you can build an actual ISO or have just like a system that you can flash on with an IMG file or just be able to, to mount or whatever. I was just looking, because I need to update mine based on your two's brilliant ideas, because I tried this, I don't know, a year and a half ago, two years ago. Yeah. But it's under 100 lines of code because there are just these modules, as Brent was talking about, that you can import. And then the rest is just as much, like, you can go all the way to having, like, most of a full desktop, or you can have a minimal system that just, you know,

presents a CLI that shows up on your mesh or whatever. But I was kind of musing on, like, you know, what is, what would you want in terms of like an installed system on your host or on a USB drive? There's also now, you know, we live mostly in like an EFI world. Maybe you want to have legacy compatibility. Maybe you don't kind of depends on the fleet you're trying to support and all that. And we live in like a world of UKIs and other things where maybe you could have just a single file that you drop on your EFI partition that you boot into. And I was kind of looking around to see what people were doing. And we've used Vento in the past. But, you know, it has its quirks and issues and skepticism, rightfully so.

And there's just, there's like a lot of baggage in history with Ventoy. And I was interested to see that there was sort of a new project I hadn't heard of called Aegis Boot that I think plays a similar role that you guys might find interesting. Okay. It's a tool that creates USB rescue sticks capable of booting any Linux ISO while keeping UEFI secure boot fully enforcing. Unlike alternatives like Ventoy, UME, multi-boot USB. they're required disabling it or enrolling

custom keys. Aegis boot leverages the same signed boot chain that distributions already use. So it uses the Microsoft's signed shim that a lot of distros are using and then it's using a canonical signed grub and their kernel.

And then it boots a ratatooey rust, rescue toey. And it has some custom code to go hunt around. So it's a two partition setup, kind of like Haventoy does, right?

It's got its system code and like the EFI stuff that it boots into. And then it's got like an ISO area you can dump these files onto. And it mounts and scans the ISO files to find the Linux kernels and the parameters for them. And? And then, because this is a project I'm highlighting,

it presents them to you. It checks them out to see if they're signed and you can actually boot them or not. Yes. And then when it goes to boot them because you're already running Linux, it uses KXAC. Oh yeah.

There it is. Age is boot. So this is potentially a vent toy replacement, maybe a little more modern. Yeah, it's a single Rust EXC you can download. They have some different profiles if you want it to pre-download like a bunch of ISOs onto the drive. It's an all-in-one sort of flashing tool that's available is part of it. It has NixOS, Windows, Arch Linux, Fedora, all kinds of support. Very cool.

Well, leave it to West to find the most modern tool using the most, I don't know, Interesting and fun tooling under the hood, Chris. I feel like we got to catch up again this week on the tooling that Wetz is using. I know. I know. Well, I want like a hybrid of yours and Wes's setup. Well, I think you could get that because you could make the USB with this Aegis tool, but then put your custom NixOS or whatever ISO on there, and then it would just boot it. The only thing you boys aren't considering, and it's a project we talked about a long time ago,

so fair enough, that it's easy to forget these things. We've talked about so many of these. But do you remember, I think it was net boot. I'm going to it right now, XYZ. Oh, yes. Netboating was another thing

I wanted to mention as part of this, although it's a different whether or not you want to rely on the network. Right. But if you've got a machine that supports network boating,

which is like all of them, you can run Netboot. And I actually do have Netboot. XYZ running in a container on my network. You do have to have DHCP and all of that's giving out. It's a little bit to set up

because you have to have DHCP that can give out like the TFTP server information and the information to boot off of the IPixie server. But it's a little. If you have the ability to set that, then you can have a network boot system for all your systems on your land and just choose the network option. And they will discover this and boot the ISO image you have queued up.

Also another one where if you didn't, depending on your firmware, you can also drop like a UA5 thing that does the pixie as well. Yeah. So I was wondering, I mean, could you have Aegis in there too? Like there's a whole, like I'm just thinking with net boot, there's a whole angle to this that could take this to the next level where you wouldn't even need a custom partition or a. thumb drive if you're just working on your land. And that, to me, that could be a real unlock. If everybody out there has that setup, send it in. You can almost imagine you, you know, you pixie to Aegis, but then it's Linux so it can sort of mount the samba share of your ISOs and then it kegsax into one of them. Oh, that's glorious. That's glorious. I was just wondering

how, uh, how modern Aegis actually was, since I called it modern. And it looks like the project is very active with the last commit, uh, eight minutes ago. So, Wes is right on the most modern activity here. I've tried it all of once last night, so I just found it. A tester beware. Yeah, it looks old. I mean, it looks new. It does look new, but it looks very promising.

It's a good find, Wes. And now it is time for the boost. We have several ballers this week, but at the top of the list, Greg, the lawyer, sent in three boosts with a total of 122,345 sets. Wow, thank you, Greg. I started listening a couple years ago when I got back into Linux after a long break. I was very pleased to discover the world of home lobbying and to see how far Linux had truly come. My first distro back in the day was Slackware.

When I returned, I tried Debian and Fedora and then landed on our. for quite a while. Thanks to you guys. I'm now on that NixOS train. I'm just getting started, but I really like Nix. Everything about it is so intentional. Everything I add is documented because, well, it has to be. That's true. I do love that. It's great. It's great for people that are very, very implicit about how their system must be set up to every detail. And it's also great for people that cannot remember at all how their system is set up. You wouldn't know anyone like that. No, I wouldn't. No, I wouldn't. No, I would. Oh, man.

Thank you, Greg. Appreciate that very much. Yeah, Greg, I have the full... It kind of got cut off. I have the full one, and Greg, links to the NXOS config repo that he has. So we can include that. Nice. Thank you, Greg.

Appreciate that. All right, our next baller is Optic Gray, and he came in with 88,88s. I hoard that which all kind covered. Coming in half with the boost. That's great. 4GEO sounds like a great GitHub alternative, but why not use GitLab locally hosted? that's another good option for sure. GitLab's like a little more

enterprise-y and feel and sort of set up and components and stuff. Totally doable to self-host. But Forgeo is just a really easy sort of like turn-on one NixOS service and it runs like a mostly single sort of system D thing. And the front ends a lot like what you would expect if you were just coming over from GitHub as well, which is nice. I have used GitLab a lot so that part would be fine.

I like GitLab. Well, there's a lot of free software projects that we use that are just on GitLab. Yep. So it's definitely a great option. Yeah, and there's a lot of, you know, there's more forges than that, too, to consider. That was just one that I knew had functionality that I liked and was easy to get going. Also just shipped an update after our episode. That's true.

So a new version is out. Well, the due to buy it comes in with 77,77s. Hey, it's been a while. I wanted to boost so many times, but I didn't have. have the time. We understand. For the stage 4 BSD challenge, I say you test how gaming works. You pick a game

that's supposedly supported in BSD, and you get points for managing to install and play. All right. I'll try to participate. I have a Dell XBS and an X-260 line around. Nice. And then, for what it's worth, I like the AI content. So the dude, I'm writing this down

right now. Level four, level four gaming. Chris frantically writing. Mm-hmm. Taking notes. Thank you, sir. That's good. What did you get the fountain pen? Yeah. It is dipped with the blood of my enemies. Adversaries 20.

Upgraded. Adversary 17 comes in with 20,0001 sets. Fun will now commence. He says, I think the other host name was Alex or something like that. Oh, okay. I'm reading the matter of order, I see. There used to be a podcast about self-hosting that was really done well for many years. There was an episode about a self-hosted code forages. Another one mentioned he roasting us.

like Git-Tee that automatically mirrors to GitHub. I think one of the hosts had a name like yours. There might have been another one in there, something like Alex.

Oh my gosh. Aversary's 20. Roasting me. Coming in hot with the booze. I send in a good link to episode 129 is self-hosted Forged Alliance.

There you go. Talk about some Git-Tee in there. Sounds good. Spooky Sat comes in with 10,000 cents. It's over 9,000. Live boost. Woo-hoo.

Hey, thank you for the live boost. This was from last week. That's awesome. That's always nice getting those live. Thank you, set come. Goofy Ambitions boosted in 1, 2, 3, 4, 5 sets. So the combination is 1, 2, 3, 4, 5. Hello, Chris, Wes, and Brent.

I'm a Jupiter Party member, and I primarily listen to your show to get insight into new open source tools that will be useful for me. I'm currently listening to 661 for the second time, and I just want to, to say, keep up the good work. I really appreciate your content on OpenAI agents and NixOS.

Goofy, thank you for both the membership support and the feedback on that. Thank you very much. Appreciate that. J-Cube 3 came in with 16,482 sats. Boy, they are doing a lot with Mayo these days. I really enjoyed this episode and would love even more in-depth coverage like it. I enjoyed all your AI coverage. I know it can't all be AI all the time, nor should it be.

I think you guys are doing an excellent job. Keep it up. Great. Thanks. You know, and I, we are like the last two episodes. It's come up, but it's not the topic of the show. Thank you for the feedback, everybody. That's good signal.

It'd be interesting to see what the vibe is at Linux Fest Northwest. Mm-hmm. Let that be interesting if people are talking about it. Podbon, boosin with a row of ducks. Maybe the chapters could be more granular and include specific topics such as AI or Nix with the people that complain about too much of either. I don't mind the topics, but some of the nerdyish stuff does go over my head. Shout out to editor Drew's excellent chapter titles and always really great.

You could see a fun project where there's like a GitHub post or maybe a 4Gio of our chapters JSON that could get iterated on by the community. Oh, totally. You could go in there and publish it and update it because it's just a JSON file. That's the beauty of the cloud chapters. I would also say if the nerdyer stuff goes over your head now, it's still sinking in somewhere. And if you pay attention, that nerdier stuff, you will understand more and more and more of it. That's just the beauty of hanging out with a bunch of nerds. Well, Daja boosted in 6,969 Sadoches.

Put some macaroni and cheese on there, too. Oh, does anyone else have that problem of the boost audio clips slipping into your everyday vernacular? Well, I've accidentally dropped. They're doing a lot with mayo these days. And that is a tasty burger in several conversations without even thinking about it. I just called my co-worker a rich lobster. I guess it's just a sign of how integrated J.B. is into my life. I love it.

We definitely do us. Dajah has another boost here. One thing that gets lost in the license conversation is that, as it currently stands, LLMs are just large license launderers. one of the great things about copy left FOSS is that you know that all derivations will continue to benefit humanity by also being Foss. But as it as it is now, we lose all freedoms as soon as someone makes a proprietary derivation using an agent. Roll over and take it because they're going to violate your code and the freedoms of your users anyway. Isn't a very good stance and neither is put up a bunch of useless walls that also violate your user freedoms.

It just feels silly that the two leading arguments on the subject break false ethos in different ways. I think if we can get both sides to come to an agreement on an interface that lets AI access the code and also respect the license, we could get the best of both worlds, even if it's just a setting that you can give to your agent to include or exclude source material that is compatible with the license of your project or something similar. If you're also writing GPL, go ham. If you're writing proprietary, then you can consciously make the choice on whether or not to abide by that license. Another benefit to something like that would be that folks at companies that block copy-left dependencies can actually enjoy all the fun AI stuff we have without worrying of getting in trouble. Of course, that's all really easy to say than to do. True.

It is easy to say. Like, you could see just, it could be a markdown file or a HTML file or a dot file that just says, you know, don't. train. It could be very simple. But I wanted to maybe reframe this a little bit because I think maybe that's where you're getting hung up and having a hard time squaring it, right? And I think when you can't really square it, maybe that's an opportunity to see if you got the math right. And Wes, I'd like to know your input on this. But I think there is a broader question that has to be answered before this particular situation can be sorted out. And that broader question is, what is code laundering? is an open source developer that goes to stack exchange

that gets inspired by a post on stack exchange and writes a similar equivalent version of that in their own code, was that code laundering? And if that is the equivalent of what an LLM, is it code laundering when the LLM does it based on your request? Because it's not Google or Anthropic that's asking the machine to do that.

It's the user prompting the machine to do that, but the user would also be going to stack exchange and just aping what's on stack exchange or an example they find on GitHub. And we don't consider that, code laundering, we just consider that to be standard practice, how you develop software. I'm curious to know what you think on that nuance. Yeah, I mean, there's just a lot of nuance to go around because there's different questions.

There's the, like, there's the training time question and how that works and what code was used there and the statistical nature and how you think that the system works and the actual outcomes of all of that, right? And then there is now this question of agents in practice and what if you have one agent that sort of reads the code and outputs a spec that is an arguable clean room sort of thing? and then you have someone else through the coding, does that work? There's just all kinds, yeah. And so there's just a lot of, like,

I think, I do think Daj is right that, like, we have not really maybe got to that part where a lot of the norms and expectations have adjusted, and we need to keep having these conversations because there are a lot of nuances and sort of things to discuss. So I appreciate the thoughtful response here. It's almost like if we locked in something now

it would probably be too premature. Could be. The thing continues to move forward while we're figuring it out. It sure does. Yeah, that is it right there. Great. boost, thank you, sir. Jane Bean comes in

with a row of ducks. I've had a basic forgeo instance on NICS for a while, and I've had an aothein via Pocket ID, and I'm pretty happy with that. Seems like a nice setup. A pocket ID is a cool little

sort of minimal OADC provider for aoth stuff, but it just uses Paskey. So it's super simple. I'll take Arctic too. Arctic Bender comes in with another row of ducks. For the love of all that is holy. You guys, enough about AI.

I work in IT, and I'm sick and tired of every vendor talking about their AI offering. You can't spell failure without AI. I stop listening about halfway through this one. Please don't give me a safe haven where I, please give me a safe haven where I don't have to hear about the joys of you know what. You know, I was interesting. I was having a conversation with the wife last night. And she finds some of these tools useful for her small business. At the same time, absolutely hates that the vendor software she's using is shoving it into like, she has a field for like filling out patient chart notes. And they've literally put an AI wizard in

every field of the text field, which now accludes all of the text you can fit in there. So she can't read all of the text in there. And none of it's useful. It's extra JavaScript that's just kind of floating around. And there's like tons of HIPAA concerns that they haven't properly explained to her, so she's not comfortable with that aspect either. Yeah, of course, right.

And it's like, so she is both like, boy, it's really useful as a small business owner for some of these things. And boy, I really hate this. So I get it, Artic. I get it. Thank you for that feedback. Appreciate that. Hey, Mr. 20's back.

Yeah, this is a duplicate boost But we still got the SATs So we appreciate 10,250 Thank you adversaries You are great But then Woodland Geeks comes in With a row of ducks And an opportunity for you, Chris

Okay Though I think you've already talked about this on the launch maybe Maybe that was just the pre-show Okay Members You get the good stuff Wanted to share this link Sure you guys are on top of this

And probably will show up in the week's episode But how to share And see many different sides of what is going on here. Chris Rant Opportunity. Talking about mythos. Yes, and it's a link to a Forbes article. What is GLAD Mythos and why Anthropic won't let anyone use it?

Hmm. Mm-hmm. Yeah. Yeah. Yeah. Yeah. Well, there's a dude here trying stuff and succeeded in sending a row of ducks. First agents, now private GitHub replacements.

You all are reading my mind. I'm currently starting my home lab from scratch over again on a new computer. and I see, oh, infrastructure as code and AI are my two primary focuses. It's great to hear y'all's input on that. Please keep it up. A dude, that is such a fun time. You know, when you're rebuilding, like, with everything you've learned, yes, it's a lot of work. And for me, it would be a crushing, overwhelming amount of work.

But it's so fun when you actually, like, it's like how I dream of, I would never do this, but I dream of building my own RV, taking everything I've learned and just building the ultimate RV. That would be it real. Oh, it would be wonderful. Please keep us posted. I'd love to know how that goes. And glad to hear, we're helpful.

We'll keep trying to do so. And if you screw something up, now you know how you can rescue it. Magnolia Mayhem came in with 2,000 sats. B-O-O-S-T. And says boost. Thank you, Maham. Appreciate you. Tomato comes in with 5,000 cents.

Hmm. Danger zone. I love the episode 666, demonic challenge plan. One idea for a challenge. would be to run Linux binaries or even a Linux container. I think Brent in particular might enjoy NetBSD on some dumpster hardware. They can run on remarkably small machines.

He knows your taste, I think. He does know your taste. You do like those dumpster machines. That's a great one. I think I feel like maybe not hard enough because they have the Linux compatibility layer, which gives you a lot. You do have to set that up.

You do, and that could be the challenge. Also, what I find, And this is just me coming in with my preconceptions and bias eyes. But what I find is the BSD guys love to talk about stuff working great. Oh, that's a solved problem. And then you go to use it and you realize it's about a 20 or 30% solution that they're fine with. But if you're used to something on Linux is inadequate. Maybe it'll be like that.

Stay tuned to find out. Boost in now. Complete noobs did boost in with a row of ducks. I was just a link to a website here, v4.call.com. V4Call.com, complete noops project. It says this is a 0.0.1 concept phase. Use at your own risk. Wes, what did you do figure out here?

Well, I found an info page. V4Call is a decentralized video voice and text platform built on the Hive blockchain. Users set their own rates for receiving calls and messages. Callers pay with Hives stable coin or custom tokens. So it's, okay, video and voice. All right. On Hive. Interesting.

Okay, there you go. V4V call. Our v4call.com. Well, I do need to come up with some sort of call system for the launch. You know, because I'm paying for a hosted solution out of my PayPal account, but my PayPal account's been frozen because I used it. How dare you? I know.

Yeah, that's the mistake, really. Yeah, that's what I get. That's what I get. So now I have to go through like a 20-point identification process. I'm not even kidding. I think it's like 17 different documents and like several questionnaires to get access to my account so that way I can pay the bill. to use the thing.

To use the call service. So, yeah, I probably should work on that. But, you know, next week's Linux Fest. Thank you, everybody who boosted in and makes the show possible. It also sends us the signal on the episode. You really have no idea how much it really means to us.

It really is something. And also, thank you everybody who stream sets as you listen. 18 of you did it. Collectively, you stacked 58,000, 35 sets. That ain't too bad, y'all. That ain't too bad. When you combine it together, though,

with our boosters, we really have a good one this week. Very grateful. we stacked 444,740 SATs. Wow. Yes. Fountain FM makes it really easy to boost these days with Fiat or Sats. And of course, there's a lot of options if you go the self-hosted AlbiHub route. New Podcast Apps.com will get you started on that.

And you send in a boost over 2,000 sats and we will read it on a future episode. And it means a lot. Your split goes to myself, to Mr. Payne, to Brentley, to editor Drew, and to the podcast developer and a little bit goes to the index as well. So it supports the whole ecosystem. All of that is in the RSS feed open to you in the public. You can view it, you can audit it, you can see if that ever changes. That's just a little bit of how we do this. So before we go, we got a smattering of picks this week. I found a GUI to clean out metadata from various files. You know, there's so many things that get just embedded in documents, PDFs, images, especially. And metadata cleaner is what

it says right there on the tin. It is a GTK GUI to go through and identify the metadata in a picture file, a text file, a video file, whatever it might be, and then help you strip that metadata out before you share it. And it is available on FlatHub. And it's pretty lean, pretty mean, not much else to it. Very simple UI. Speaking of GitLab, it seems to be primarily hosted over there. Yes.

And it is GPL3. So that is just called simply metadata cleaner. But Wes, you've been messing around with something a little different. Yeah, well, I also kind of had this need. I mean, I have had for a while, but just decided to solve it a new way for whatever reason. So I searched around Nix packages as one does, and I found MAT2, which is just a convenient little CLI metadata removal tool. You just pointed out a file kind of a thing? Yeah, I put a little sort of demo here in the dock.

You can just run it right from Nix packages with Nix Run if you want, but yeah, you just kind of tell it what file you want. You could pass dash S just to show the metadata. Here's like an example from my phone. You can see a bunch of details that I probably don't need other people knowing necessarily. Great for your own archival stuff, perhaps, right? And then all you do is you run it with no arguments. By default, it just makes a new, it adds like baseman. So you get like image.comed.

So you know that one's safe. That's the one you share. Yeah. So it doesn't touch the existing files. It just makes a bunch of new files in there. It's easy to automate with something like XARGs or parallel or, you know, those kinds of tools. So it's clean, simple. easy to use. I do like the idea of just

it's on the command line, run it real quick. Wrap it in a script if you want, so you just do it on regular files, you can take things through. Yeah, yeah. Of course, with these days, you could probably still get a lot of information just from the stuff that's in the photo, depending on where you take it, but it's get at least clean that up, right?

Now, this next one's making some news. So we want to take a quick look at it for you and mention it, because it's one of these quintessential MacOS apps that's actually come to Linux this week. It's called Little Snitch. And you might be familiar with it. Every time an application on your computer opens up a network connection, it just does so in the background without you knowing, which is fine, right? But wouldn't it be

kind of interesting to have insight on what's opening up, what applications are doing what, and kind of how much data are you using? Linux didn't have this with Little Snitch until they released it. They used to be a Mac-only app. Now it's available. You can see exactly what applications are talking to what servers. You can block the ones you don't want talking to those servers. You can keep an eye on historical data and the amount of traffic volume over time. The connection View has a lot of interesting things on their current and past network activity by application, shows you what's being blocked by your rules and block lists and tracks data volumes and traffic history. This will run on any Linux distribution with kernel 612 or above. It's not quite

equivalent to the macOS version. It is missing a few things, mostly because of the limitations for better for worse, I think for better, of what EBPF will allow some of the tooling to get access to. And yes, it is powered by EBPF. That's right. So you actually got it working there on your machine. Yeah. It is like the license situation. There's a bunch of open source parts, but some of it is still proprietary. I think maybe ported over from the...

View and shareable, but proprietary. And so it's like the back end stuff is free. The only thing that's, I think, proprietary is maybe some of the gooey stuff, but it's viewable and shareable. So, there you go. But they provide pre-compiled binaries. You can just download for Linux. So I was able to download those.

And not too much work, a little patch of got them running on NixOS. So I made a, well, had to open code help me make a... basic little slate for that on my GitHub. So it was easy enough to run, so that means it'll definitely run on most standard Linux distributions quite easily. I think you run it. It comes with the system D service already

with like nice, exactly the permissions it actually needs, or you can just run it with pseudo, if you want to test it out. Because it needs some of the like, you know, cis admin, or not full system, cis admin, but need some of the network admin permissions to be able to hook things in and all that. And then it runs a Damon in the background.

So it, you pop it up as a WebUI, up at 3031, and you can can see I put a picture in the dock there, kind of clean-looking modern UI with a nice, feels responsive, drill down, breaks down a lot of the different traffic there. It automatically detected the stuff I was pulling down from Nix as I was building something in the background while I was trying it out. So that was fun to see. Huh. That is kind of nice when you're doing a system update or something and you just kind of want to watch. Like, I'll often use Bottom for this.

I love Bottom, but this is more specifically network level, which was great. Did you try blocking anything? No, not yet. I am curious to try that, though. It is neat to see one of these quintessential Mac apps land on Linux. Even if it's not bully the same, with all of these Mac users switching over, it's great timing. It really is.

And I mean, kind of a long time coming, there'd been other open source sort of like attempts at this kinds of software that we've even talked about on the show. But I do wonder what it says, that now was the time that it moved. Me too. All right, who snuck in P. Cap droid? Who snuck that in here?

Ooh, I snuck in a little bonus pick. on the same topic as Little Snitch that I found recently for Android. So this is a privacy friendly, of course, open source, of course, but it's an Android app that lets you also do similar things, track, analyze, and block the connections made by other apps on your Android device. And you could dump the traffic, inspect the HTTP, decrypt some of the TLS traffic if you want, all sorts of stuff. I use this to just, well, you're going to get curious. What's your cell phone really doing in the back? Especially my cell phone. I'd really like to know what my cell phone's up to.

It is every single time you do this on any device, it is mind-boggling and very enlightening. So I recommend you pick up PCAP droid. Just run it for like an hour and see what the heck's going on on that device of yours. P-CAPDroid, and we will have a link to that in the show notes. That's a nice fine, gentlemen. So now we'll have more info not only on our desktops, but on our mobile devices. We're cleaning up your metadata and protecting your network.

this week while also rescuing your data. It's a very utility-focused episode, I guess. Or we're having a lot of problems behind this. We've got to play it off. No, this one. No, it wasn't that at all. We're getting very, very excited about Linux Fest Northwest. It is next weekend.

What? Yeah. So you will get the full report from the Pacific Northwest's largest community-run Linux event. And one of the OGs going for... something like 25, 26 years now. I don't know. I don't know how long I've been going. It's been a long time.

It doesn't feel real yet, but it's going to be great. I have been going to Linux Fest Northwest since I've been in high school. And I'm a little out of high school now. So that's how long I've been going. Woo! The first time I went, I went with a high school teacher of mine. That's wonderful.

Let's keep it going. I mean, there's a reason I keep going every year, right? You don't go for 26, 27 years, unless maybe 28. I don't know. I don't know what it is. You don't go for that long unless it's really something special. So I hope to see you there. LFNW.org, if you want more details, we'll put a link in the show notes too. And, of course, you can always make it a Tuesday on a Sunday.

Wherever you are, you can join us live 10 a.m. Pacific, 1 p.m. Eastern. We may be, you know, somewhere around there a little bit because we have a live, we've got to do a lot of setup and all of that. But jbblive.fm. or jbblive.tv will be where you can catch it if you can't make it physically to Linux Fest Northwest. Now, Wes, before we go, we should tell people about some of the cool, advanced features we have around the show because it's not just audio. Yeah, that's right. We take advantage of the podcast 2.0 namespace.

In particular, that gets those mighty fine cloud chapters we mentioned earlier on. Dynamic JSON delivered to you just when you need it. And if you want more information, check out our VTT files. Those are also found in the feed under the podcast transcript. It is neat, too, to see more of these, particular these ones we just mentioned, getting picked up by more and more vendors and whatnot, including even Apple Podcasts, which is amazing now. Also, if your podcast client doesn't have support for this,

you should message to the developer and let them know that they've been beaten by Apple. It's all standard stuff, right? VTTSRT files. There's lots of open source libraries folks can bring in to help make use of this stuff. And the other one that hasn't seen as much traction, but it is in apps like Fountain and Podverse

and Castamatic, of course, is live stream support. We are live every single Sunday. There's not a lot of podcasts that do that anymore. JBLIV.com is also where you can listen to it. See you next week. Same bad time. Same bad station. And we have links to just about darn near everything we've talked about.

We try to catch it all, and we put that over at LinuxUplug.com slash 663. We also sneak in alternate enclosure into the feed these days, so just go grip around for that. There might be a video version in there for you. With visuals of what we talk about and stuff. You never know. We are so much looking forward to seeing you, so we hope to see at Linux Fest Northwest. But if we don't, friends, we'll catch you right back here in the feed like we always do next Tuesday, as in Sunday. Thank you.