The Uncast Show
Owning an Unraid server gives you the ultimate control over your data and reduces your dependence on The Cloud. Join us to learn how to get the most out of your Unraid server, stay up to date on relevant news and topics, and get to know members of the Unraid community!
The Uncast Show
Unraid 8 Explained: New Distro, Docker Compose, Backups and more | Uncast (Part 2 of 2)
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Last time you met the CEO. This time, meet the man who started it all and the person building what comes next.
In Part 2, Ed sits down with Unraid creator Tom and product director Eli. Tom walks through 20 years of Unraid — from a 2005 hobby project born out of a pile of cheap hard drives, to parity-protected storage, a web GUI built to hide Linux, and the arrival of plugins, Docker, VMs and ZFS. Then Eli lays out what's coming in Unraid 8.
What's next: a move to a Fedora base (built on Universal Blue's uCore) for faster security updates and far better hardware support out of the box; native Docker Compose; a built-in backup solution — to the cloud, an S3 bucket, or another Unraid server, with snapshots and full restores; a plugin permission system so you can see exactly what each plugin can touch before you install it; and a cleaner, more consistent interface across the OS, your account, and Community Apps — all without taking anything away from the power users who go deep.
Same idea that started it all: your hardware, your data, your choice. Here's to the next 20 years.
— CHAPTERS —
0:00 – Welcome back: meet the founder (Part 2 of 2)
1:05 – 2005: the problem Tom set out to solve
2:24 – Parity & redundancy: the first Unraid array
4:41 – The web GUI, "Tower" & the birth of plugins
7:23 – 64-bit, VMs & Docker: the road to Unraid 6
9:37 – ZFS support & mirrored USB boot (Unraid 7)
11:49 – Why Unraid 8 moves to a Fedora base
13:29 – Immutable Linux & Universal Blue (uCore)
15:58 – Kernel flexibility & better hardware support
20:32 – One cohesive interface & own-your-data
22:38 – Built-in backups & native Docker Compose
26:20 – Plugins you can trust: the new permission system
29:35 – Community Apps, reimagined (ca.unraid.net)
31:23 – When can you try it? Beta, RC & getting involved
33:05 – Early days: building servers by hand
35:34 – From flash drives to digital downloads
37:17 – 20 years, still independent: Tom's reflection
39:00 – Closing remarks & how to get involved
— LINKS —
Watch Part 1 (meet CEO Tiffany Jones).
Read Tiffany's letter (companion blog).
Sign up for the Unraid 8 beta.
Submit feature requests & bug reports.
What will you build with Unraid?
Get Started with Unraid in 15 minutes or less.
i7-14700, dual 10GbE, Arctic fans, Lifetime Unraid license included. Starting at $2,999.
Reserve your spot with a $99 refundable deposit now
Other Ways to Connect with the Uncast Show
Meet Unraid’S Founder And Team
SPEAKER_02Hey everyone and welcome back. This is the second part of our two-part podcast with the people behind Unraid. Last time you met Tiffany, our CEO, and today you get to meet her dad, the man who started this whole thing off 20 years ago, and the man you can probably blame for a fair few of your home lab addictions, mine very much included. So Tom, it's brilliant to have you here. Thank you very much for making the time.
SPEAKER_01Hey, thanks, Ed. And uh the real brilliance is over here with uh Eli. Yep.
SPEAKER_02And joining you, Tom. Um, one of the people who's building Where Unraid's Heading Next is our product director, Eli. Great to have you on the show too, as well, Eli.
SPEAKER_00Thanks, Ed. Yeah, it's great to be here. And um, yeah, uh, Tom, you're wild for that. Thanks.
SPEAKER_02Anyway, so today we're gonna go back to where Unraid began, how it's grown over 20 years and where it's heading next. Tom's gonna take us through the story, and then Eli is going to bring us up to date and what's next. Unraid's always brought its community along as it's grown, so that's exactly what we're doing here. So, Tom, let's go back to the start. Um, before there was a company or a product, what was the problem you were actually trying to solve for yourself back way in 2005?
SPEAKER_01Okay, well, hey, and um, I've actually told the story a few times, but uh sorry I won't go into too many gory details, but it actually really started in the early 2000s, and in those days, uh, a few things sort of came together at the same time. One was big plasma TVs, 16 by 9, high definition, right? Another was you're able to rip your DVD collection to hard drives in those days. And and then thirdly, there was uh the beginnings of um home theater PC software. So, for example, my HTPC. Then the last thing that happened for me in that time is I was in between jobs, so I had a lot of time on my hands. So, what I decided to do is in my background in those days was uh in enterprise storage and disk arrays, and so I knew a lot about working closely with hardware, also a little bit of Linux, although I hadn't really done a lot of programming in Linux. But anyway, so I had some time and um I immediately
The 2005 Problem That Sparked Unraid
SPEAKER_01discovered that ripping DVDs was very time consuming, and and I was using really cheap white label hard drives, and it suddenly dawned on me, you know, if one of these drives fails, I've just lost like you know, a hundred hours of work. And so I decided that you know what we need is some kind of redundancy. So I looked at what was available in Linux and what have you, and decided that you know I didn't really want to take and build a uh a RAID 5 out of the Linux MD driver at the time and move all the files again. And so um I undertook to to realize that, hey, if all you really need is each individual drive, and then you need to have it parity protected, is all you really need because the bandwidth of drives in those days was good enough to keep up with uh any kind of media stream, either writing to the drive or reading from it. So that's really where the idea came from. And so the unrayed array was born where I just took a set of drives and added a parity drive. And what I actually did is put the um the Linux MD driver and figured out how it worked and studied it and made some changes and came up with what is today I call MD Unrayed driver. Anyway, so that's where that started. The software itself to control all of it was really uh a hobby project using the Linux command line and you know just using it. And I had a Windows PC that was driving the display, and and so it was connected over uh, I think even that in those days, a fast internet, not even gigabit. But very soon after that, we had gigabit. So anyway, I decided that now the next step would be to add some kind of a user interface. And my son actually, you know, to this day when you install Enray, the default name of your server is called Tower, and and that came from my son at the time going, Hey, hey dad, what do you have in that tower over there? And that's where that name came from. Is he he said, okay, that's a good name for it. And um, he said, you know, hey, this is pretty cool. You ought to think about, you know, providing this, you know, selling it or providing it some way to other users. And so I thought, okay, great, and undertook a project to create a web GUI. And the the idea there at the very beginning, and even to this day, is that I wanted to hide Linux. I I didn't want a user to have to go in and use the command line to do things, to move files, any of those things. And and so um unrayed versions one through three were sort of my the hobbyist side of me, you know, figuring out how things work. Okay, Unraid version four was really the first version that was really available to other people to use. And uh I created an HTTP server, very basic Web GUI. Uh it just let you uh manage the individual devices of an Unraid array. Later, add a cache drive that came along with gig Ethernet and realizing that you can now copy computer to computer faster than what that you could write a single drive at an unrate array because of the re-modify write required. And so the idea came along of a cache drive. Okay, let's write to that first, and then in the background, when everybody's asleep, just write that, you know, flush it out to the to the main array storage. So that's that's where uh Unray version 4 had the cache drive, and then it's sort of make a consistent view of the storage. It also included user shares, but we're called user shares. Um and and that simply was you could look at one share, what's called now a union file system, it would
From Parity Array To Web GUI
SPEAKER_01look at that same directory on all the storage, including the cache drive, and and you wouldn't even know necessarily what device it was stored on, you could just access it. Um and so Unray version 4 was NAS only, and that's pretty much what that was. Uh then later on, okay, in those days I was living in um in Sunnyvale on uh Lime Drive, and that's where uh Lime technology came from. Anyway, we ended up moving to Colorado, and while I was out there, people were you know really liked Unraid, and they were writing different add-ons. They were plugins, but they're not officially called plugins. Um but they're you know, they wanted to extend the functionality of the server. So um I undertook version five, and the big change in Unray version five was moving to PHP instead of sort of this homegrown system that was very rudimentary to present web pages and and very basic conditional testing and what have you, integrated PHP. And along with that came uh plugins, official method for adding plugins. And then the community really took off of that. And and one of the big things that came out of it was the so-called dynamics web GUI, which we have today. And so the beginnings of that was back in Unrate version 5. And so an independent community developer developed that, and he and I became friends and and worked at it over the next few years. So I was in Colorado four years and then finally decided to move back to California, and I ended up in San Diego. That was around 2012 now. We got there, um, decided to start embarking on Unraid 6, which is just the next evolution. And what mainly with that was several things. So, first of all, moving to 64-bit. So up until that time, Unraid was a 32-bit application. But now most motherboard processors, everything was 64-bit. And so that was uh an easy transition to to get the code base running on X64, x8664. And the next thing we wanted to do was really increase the functionality that you could do on the on the server. I wanted to implement the ability to have apps. And the first go-around of that was implementing a KVM, implementing virtual machines. And the idea there was that each virtual machine would be one app, and we're gonna design a system where you'd have all these icons, and you click it, and it would fire up the VM and auto-start the app. And uh we've got down that path. Okay, but then the other technology that happened right around the same time was Docker. So everybody said, Wow, this is pretty cool. And and so we shifted, added Docker, and now Docker provides that exact functionality. Um, and again, the community ran with that. Next thing you know, there's an awesome plugin called Community Apps, where someone took upon themselves to organize everything that people did and create a nice little plugin that had categories and let you automatically download it. And that was done by a fellow that's with us now named Squid. Everybody knows who Squid is, and he is the curator of the app feed to this day, and that that has just been an awesome, uh just awesome. Uh technically, we added uh, I think in Unraid version six, we added multiple pools as well, butterfS pools and for version six. And uh and so yeah, the community involvement just really grew in those days. So next thing to happen is we decide, okay, let's move to version seven. And now what we're doing with version seven is we're finding that there are more business type applications, more business opportunities, and advanced home users. And one of the big things that's always it's been a two-edges sword is the USB boot and USB licensing. Okay, has always been a two-edged sword. On the one side, it's super easy. Okay, super easy to prepare to flash, you boot from it. All the config data is there, you know, it's just updates are are easy. There's a lot of advantages, but the downside is USB Flash is flaky and tends to fail, especially lately. It seems like the quality in general of Flash has just gone down. And people want redundancy. If your flash fails, you can't boot. Okay, so we embarked on the main thing in Unray 7 was two things. Well, one was ZFS. We needed to add ZFS support because the community wanted ZFS support. And again, we wanted to open the door for developers to come
Unraid 7 And Leaving USB Boot Behind
SPEAKER_01in and just continue uh doing things that they want to do with the platform. And so we integrated ZFS, but that also let us integrate ZFS uh mirrored boot. We call that internal boot right now in uh version 7. And what that does is you still initially configure your server, you can boot your server up with a flash, but now we have a way of transferring the boot to a mirrored ZFS pool. And on top of that, too, what we wanted the user to be able to do is once that's accomplished, you know, take that flash out and throw it back in your junk drawer. That's that's what we wanted. So in order to do that, we implemented TPM licensing as well. So it's totally off the USB at this point. So that that was the main thrust of version seven. Of course, all along this road, there's been a lot of refinements in the management interface, the web viewing, in the distro, etc. That's you know, I've glossed over all that, but but but that's sort of the rough roadmap that brings us to where we are today with version seven. Um we're still working, of course, producing that version seven releases with refinements. But coming up on the future is unrate version eight. And uh we want to talk a little bit about that. Uh, I'm gonna hand it off to Eli in a minute, but what I will say about that is what we're facing today is not unique to us. A lot of companies have this issue where in this age of AI, AI is really good at finding security vulnerabilities. And we are seeing, I everybody knows this, we are seeing massive numbers of vulnerabilities in every Linux package and even the kernel. Okay. And it's getting to be difficult to keep up with that. Okay. Unray traditionally has been uh Slackware-based. People know that. Uh, we use a lot of Slackware packages, but we also use a lot of custom packages, and we use packages from other sources, from like, for example, GitHub. There's some packages that we use from there. And keeping up with the avalanche of updates that are happening almost daily is getting to be distracting from our core mission of working on the core UnRate OS product itself, keeping keeping the distro
Security Pressure And The Move To Fedora
SPEAKER_01secure is is distracting from that. So Unrate 8, one of the big things we're moving to in Unraid 8 is going to be to a Fedora-based distro. Now, Eli will speak more about that, but uh one thing about Unraid that has also been nice is that we've always been uh a so-called immutable OS, meaning the system boots, it loads the root file system from a squash fs file system. And and so no matter what you do, however you modify your server, if you go in there, you type the wrong command, you delete all the files from your root, you still could just reboot and you're back to the virgin release. Okay, it's immutable. And and that's a big reason why we've chosen Fedora.
SPEAKER_00It's uh Universal Blues maintenance package. UCore is the Fedora sub-variant we're looking at. We uh did a lot of research internally on what distro would allow us to maintain our ability to like let people get updates really easily, um, not shoot themselves in the foot if, so to say, if like a install a package and it's not compatible with some other package and you can't like get back to the base state. That's a really big problem with like a fully mutable Linux install, which is like a traditional Linux install would be a mutable Linux install, and you end up with like just a bunch of packages that are all dependent on each other, and if one of them changes, now you have to somehow get back to some baseline that maybe you snapshotted a month ago, maybe you don't have a snapshot, and how are you ever gonna get back? So we looked through the whole industry of like what people are doing. Tom is a brilliant engineer, and he years ago built this immutable system in Slackware that in my opinion is very ahead of its time. He uh he managed to basically, before anybody thought about immutable Linux as a thing, was doing immutable Linux. So I'm like super impressed that we've been able to use this system in Slackware with immutable Linux that people started doing immutability like five years ago about, and Tom was doing it. He's like, Oh, I'll I'll just do this like 15, 20, 25 years ago. That's freaking awesome that he was able to do that. Uh yeah, so we we looked at Fedora as uh as an option. We looked at a bunch of these other options that I've seen the community talking about. A lot of people were talking about us going to Debian. And the reason we landed on Fedora and specifically Universal Blue, which is what Bazite, I'm sure people are familiar with Bazite, is built on this Universal Blue U-Core system. The reason we landed on this is that there's a lot of really cool things about it. One of it, one of the things is that it's fully buildable inside of this like crazy Docker setup. So we can do like we can basically build our OS in uh Docker, and then there's this other really cool component of it, which is just that it's this fully immutable updates and it has a layering system. So the benefit of all that is that eventually I would love to see us providing our community with like the ability to install whatever package you need to make your uh software work without having to do a bunch of workarounds. So, for example, like I have some weird Wi-Fi card on one of my servers, and I had to build my own driver package and ship a plugin file, and it's in the CA store now. But it would be nice if instead of having to do that, I could have just gone and pulled it down from Fedora and just had it available to me. And so I think Universal Palladable would give us a lot of ability to do that kind of thing. So you can install driver packages and stack them on top of the base OS much more easily.
SPEAKER_02With with Fedora, will users expect to have to see maybe a larger storage area for boot, or will it be a similar one?
SPEAKER_00That's a really good question. We're we're still in the process of figuring out what that storage area is going to look like. My hope it's gonna be a little bigger for sure. One of the reasons it's gonna be bigger is that one of the goals we have is actually that we start running, we start moving towards a more standardized kernel so that our issues that our users are having are the same issues that other people on other Fedora distros are having. And the benefit of that is that we benefit from the whole community who's running Fedora working and helping to fix these issues. A lot of the times we'll see, you know, random uh niche issues come in, and they could be because of our custom packages, they could be because of our custom OS build pipeline, whatever it is, it would really be helpful to be able to say, oh, this is upstream and it's been fixed already. And so that's the real benefit is that we'll probably have a larger distro base, but that larger distro base will have a lot better hardware support out of the box for like random things that people want to run in their own rate servers. It's a trade-off, but I'm hoping that we can still keep that under. I definitely hope the installed size will stay under like four gigs. And honestly, with storage being as cheap as it as it is these days, I would imagine fully installed size under four gigs would be probably totally fine for people.
SPEAKER_01Yeah, another another thing we want to try to do is have a uh cleaner separation between the distro itself and all of the unrayed magic above it. Yes. So that uh you know, down the road, hoping it's possible to just update your distro and that's it. That will help with uh with people who are concerned about stability.
SPEAKER_00Yeah, and and we're also hoping to be able to very quickly get people security updates with this new workflow. So the the main thing is we don't want anybody to have to worry about being behind if like another one of these major Linux CVEs comes out. There have been quite a few kernel vulnerabilities discovered in the last few months, and uh as we've seen these come out, we've realized that the best thing we can do for our community is to give you guys the ability to self-you know, help yourselves be secure and give you more flexibility so that you're not just waiting on line tech is the the goal. We want our distro to be decoupled as much as we can from the management interface because at the end of the day, the management interface of unraid is what makes Unraid really, really, really special.
SPEAKER_02So, so guys, is it because the decoupling of the operating system from the Unraid parts that allows the users to actually upgrade the base themselves without having to wait for you guys to maybe do certain fixes for things? It it gives the freedom to do that. Am I understanding that correctly?
SPEAKER_00We're not entirely sure how it's going to pan out for our first release, but the goal longer term is that we do allow more flexibility with like running a base distro that as long as you're in line with some set of packages you can run on Raid on that base distro. And our hope is also that we'll just and I think this will come to fruition. We're trying to make a lot of our systems more capable of doing more fast releases. So we'll probably see more security patches coming out that are specifically security only versus the future security patches that we've been doing historically.
SPEAKER_01Something else that decoupling opens the door for us later down the road is the ability to have multiple kernels. So right now, Unraid, we've made the a decision that our stable releases are going to be used in a long-term supported kernel. But uh, as you know, those only come out once a year. At the end of the year is typically when the next LTS kernel is declared. However, Linux is very fast moving, hardware is very fast moving, and often there's features, especially hardware support, that's only available in the newest kernels. But in the interest of stability and testing, these days we don't just arbitrarily update the kernel, especially in a minor uh release. However, by decoupling the distro and the kernel from the main software that runs UnRate itself, we have the ability to let users choose if they want to run the latest Linux kernel. Okay, they can. If something doesn't work, well, they decide if they want to use it or not. But it gives again, it gives our community more flexibility, is really what we want to do.
SPEAKER_02So that's great because again, it's giving freedom to people because by making it that you can choose to upgrade the kernel, it means that everyone doesn't have to do it just to get the latest hardware. You can stay very stable, but the people who have got bleeding edge can give it a go. And um, I think that sounds really, really awesome, yeah.
SPEAKER_00Yeah, and with Slackware, we've we've actually been able to stay very on top of kernel updates. We're on a very, very modern Linux kernel, especially on 7-3. And so I think it'll be really nice to be able to give people the option of like, I know there's like, you know, sometimes people have issues with an updated kernel on a new OS release and being able to stay on the old one while updating the OS for security fixes will be really helpful, I think. That's like one of the core things that we're trying to make sure is possible in the new distro. I think the other thing that we shouldn't talk about while we're here is probably the uh work we're doing on the unraid interface as well. And so, as part of Unrate 8, we're trying to improve consistency across the Unray platform. And so what that basically means is we have a lot of different parts of the OS that were built through many, many, many, many years of development time and energy. And those different parts of the OS don't all work the same. They might be a little bit different. They're all really powerful, and we really want to make it so that people can come in and use those components of Unray without having to learn a new or different interface for each subcomponent. We kind of want the whole thing to feel like a cohesive user interface. So if you know how Docker works, you know how VMs work, if you know how storage works, you know how to do file manager stuff. So the goal with all of that is to basically, and then one of the big things that's going to be happening as part of UnRate 8 is this interface and API overhaul that will allow us to uh make a lot of progress towards that goal. And the really cool thing about this is that basically we really want anybody, no matter
Decoupling OS And Multiple Kernel Options
SPEAKER_00who you are, to be able to boot up Unrate and get onboarded into how. Works and then run your own cloud and have data independence. So one of the biggest things that I see my friends and family and coworkers and random people I talk to seem really interested now in owning their own data and getting their data out of the cloud. And that is one of the core goals of this work we're doing for UnRate 8 is to make it really easy to help those people get off of the cloud and own their own data and own their own storage. And our partnerships with companies like 45 Drives are like starting to let people buy off-the-shelf systems that come with Unraid. So you don't necessarily have to be as technical to be able to run an Unraid server. You don't have to build one from scratch. You can just like buy one. And the the goal with the work we're doing on Unraid 8 is to make that even easier and to make it even easier also for the other side of the business. Customers coming in from small businesses that want to run Unraid in their small business, but they don't have a backup tool or something like that. Unraid 8 is going to have a built-in backup solution that will let you back up to cloud providers. It will let you back up to your other Unraid servers. It'll let you back up to storage wherever you want, pretty much. And uh that backup solution. It's like support snapshotting and all sorts of really advanced things and full restores in the product as well. So you'll be able to do backups and restores from there. Um so that's really exciting. I think.
SPEAKER_02I think people are going to be really, really happy to see if you learn one part of Unraid, like the interface there. Uh that interface is going to be very, very similar for other sections. I think that's going to be really, really good for people new to the space. But I'm going to ask you with with these things making it easier for people, I'm sure it's not going to cut anything off for the more advanced users who want to go a bit deeper.
SPEAKER_00I was going to add, as part of this work, we're being really, really careful to make sure we're not losing any advanced functionality. That's one of the coolest things. Um, I started working in Unraid almost four years ago. Actually, four years ago, probably on this date. I think I started like July 15th. When I started, I had no idea how much chest there was to our product. I just used it like a NASA to spin up some Docker containers. Now I'm like going down VLANs and I'm like, oh my gosh, we have all these networking features that are like built out. You can do things like if you have multiple NICs on your server, you can bridge them all together and like use your server as like a an essentially a switch for other servers. It's super cool. And so yeah, I really think it's it's super important as we go through this work to make the UI consistent that we're not losing anything that's advanced. We're we're actually doing a lot of work uh to make sure that all of those advanced features are shown when they're needed and and we're trying to make it easier to like actually surface those when you might want them. So I think it's actually going to end up being easier as an advanced user to navigate um the options and the sort of settings that we have uh than it was before.
SPEAKER_02Yeah, because advanced users, it doesn't necessarily mean you want things to be difficult. You know, being an advanced user, we still like things we still like things to be you know easy and we don't want to spend ages doing things.
SPEAKER_00So yeah, I'm I would say I'm an advanced user at this point, and uh there's a lot of times where I'm like, man, I want to click two buttons and have something go for me. So you know it's like when you want the option, you've got to go in and dive down in and find it, and that that can be harder. So um yeah, we're trying to make that a little bit easier for people.
SPEAKER_02So talking about Unraid 8, what are the main features users can look out for?
SPEAKER_00That's a great question. So the main features that I would say are are like really gonna be pretty different from a day-to-day is like having this built-in backup feature, the ability to just backup your server to the cloud, uh, an S3 bucket if you want to another person's server. We historically have a really a lot of people have been relying on tools like various Docker plugins to do backups, and this will be a lot easier to set up and it'll be more built
A Cohesive UI For Newcomers
SPEAKER_00directly into UnRay. That's a really cool one for me. The other ones are Docker Compose support. We're trying really hard to make Docker Compose a part of uh the existing Unray uh templating system. So the goal with Docker Compose is that we end up with it being like very much a part of the existing Docker community app system. Um, community apps is gonna have support for compose templates. Um, you're gonna be able to use compose templates alongside Unraid XML, like normal Unraid templates. So it'll be very flexible. You can keep using single apps as XML, and I say XML, that's probably a little bit confusing for users. Basically, our current system is like these templates that are written in XML that came out again before Docker Compose even existed, really. I think we had this templated Docker system. And so Docker Compose comes out, and we've had a lot of community members build amazing Compose plugins. There's a really good one, Compose Manager Plus, that I use on my server every day. And we really wanted to make that a core part of the OS and make it feel like part of the rest of the Docker system. So you're not having to like jump between a Compose view and a Docker view.
SPEAKER_02So got it. So with Docker, it's basically going to be a unified system, still based off templates, but it will be the existing templates like what we've got, but also Docker Compose as well. So it's gonna be very easy for people to actually understand how the system works. So talking about um community developers, what about plugins?
SPEAKER_00Yeah, that's a that's a great question, too. So we talked a lot about security in this call. And I think that one of the things that we really wanted to help our users to understand and help our users to keep track of is just making sure people know when they're installing a plugin what access it has to your system, and also just help our community developers, because they've done so many amazing things for the product, to basically have an easier time building integrations into Unraid. And so in that vein, we're building a system that allows a developer to, first of all, we're building basically a plugin like developer SDK. So you'll be able to develop and build on Unraid by following a very, very well structured guide on how plugins work. And then on top of that, we're building a whole plugin uh capability system so that a plugin can say, Hey, yeah, I need access to execute shell commands on the system because I'm running user script, something like that. And so if it's a plugin like that, you'll be able to see when you install it, this plugin is going to be able to execute these things. Because the current plugin system that exists in Unread is extremely powerful, but right now every single plugin has access to do every single thing on the system. And so there's some amount of like security uh that we want to try to build into that.
SPEAKER_02Excellent. So basically, then installing a plugin, you can see exactly what it's doing. If you're not comfortable with various permissions it's got, you've got the opportunity and the freedom to choose whether you want to install that one. You can think, yeah, I'm happy with that level of risk or not
Backups And Docker Compose Built In
SPEAKER_02not so much risk. I'm happy with that level of um depth that plugin can go. Or if you're not happy with that, you you don't have to, as of before. It was very hard to see that.
SPEAKER_00Yeah, that's right. And um, and now you can also like you know be a little more skeptical. Like, why does this plugin need access to run shell scripts on my machine if all it does is adds a page to the GUI or something like that? So it'll be really, really good.
SPEAKER_02And is is there going to be any change to community applications in respect to this?
SPEAKER_00That's a that's a fantastic question. I'm I'm glad you asked it. So yeah, we um we actually just went through a pretty substantial overhaul of our whole community applications flow. We've been really spending a lot of time over the last year reinvesting in our community's contributions and trying to make sure that like the systems that people work with in Unraid, but outside of the core Unraid product are really cohesive. So we did a lot of work on our account application. We redid our purchasing flow. We've also done a ton of work to make it so that managing your keys is easier inside the account application. And we also just launched this new ca.unrate.net website, which is actually, if you go and look at that, our goal and our hope is that that becomes the face of community apps in Unrate 8. So the CA website allows sign-in. You can uh sign in with your Unrate account if you want, you can save apps. So Ed, you had this great idea when we were talking uh the other day, and you said, hey, what uh how cool it'd be if I could go on my phone and uh YouTube video about some Docker or some Unraid plugin has a link to community apps. And so now you'll be able to go on your phone, click that link to community apps, ca.unray.net, and then save it. And then when you go on your Unrate server, you can go to your saved applications in community apps, and there's your app right there, and click install, and you're you're off to the races. So I think that's gonna be really nice to have that sort of cohesion where you can kind of use CA on the web, you can use it on your Unrate server, and it's sort of the same interface no matter where you are.
SPEAKER_02Cool. So, guys, you're saying earlier that the interfaces within Unraid 8 are going to be very similar. So it's going to be a very cohesive experience for users. I'm assuming then that's going to be that same user interface, it's going to spread across things like account app and community applications, not only on the actual server locally, it's going to be on the other services as well.
SPEAKER_00Yeah, we're really trying. One of the things that that drives me nuts is when I go to somebody's website and you have the sexiest website ever, and you're like, wow, this is amazing. Like this product's got to be so good. And then you boot up the product and you're like, this is not at all like what I was thinking I was getting into here. And so, yeah, the goal is that like what you see is is what you get from our websites to the Unrate OS. And I'm really excited. It's happening. Like, we are doing a lot of work to make it so that what you're experiencing on the web and on the account app is is very similar to like what you experience at Unrate OS.
SPEAKER_02Right. Well, I'm gonna have to ask you a question because I'm sure our um listeners are gonna kill me if I don't actually ask you those. When can they start getting involved? When's it gonna be an RC? When are we gonna start seeing some of this work that we can tinker with?
SPEAKER_00For sure. That's a great question. So um we're gonna be reaching out probably right after this this podcast goes out to community developers to get our community developers on board and and help them know what's coming and how to work on this new system and what they need to know for their plugins and Docker apps and all of that stuff. And then after that, what we should have a beta sign-up period. So I think this video probably will may have a beta list sign up on it. And so you'll be able to sign up for a sort of closed beta. Um, and that closed beta will lead into the standard open beta, which we'll publish on our normal release channels. So you'll be able to go in your on rate server and update to that beta when it's available, and then we'll have an RC and all those things. So yeah.
SPEAKER_02Awesome. So exciting
Safer Plugins And Better Developer Tools
SPEAKER_02times going forward. I'm really looking forward to it myself. I can't wait to start experimenting. And I'm sure you listeners, you know, are going to be very, very happy when you get your hands on it as well. So, Tom, in the last podcast when Tiffany and I were talking, she actually told us that you forgot to put a power lead in one of the servers that you first shipped. Now, what I'd like to know is how did you actually realize that the power cable was missing?
SPEAKER_01Well, the server was a full tower. And if I recall, the the weight of the packages was over 75 pounds, it was like 80 pounds. And my sister-in-law actually helped me put it in the box, tape up the box, and loaded it in the back of our, I think I had a Durango, and we drove that thing to the post office and shipped it through USPS because it was so heavy. And we get back home and I look on the bench and there's the power cables, and I go, um Those I sent through FedEx.
SPEAKER_00Yeah. You wanted them to get there first or something.
SPEAKER_01And there's there were two power cables because the the tower had redundant had two power supplies, redundant power. Wow. Yeah. It was a piece.
SPEAKER_02So what do you think got to the chap first? The power cable or the server?
SPEAKER_01Power cables, and I let them know. I let them know uh hey, you're getting your cables first.
SPEAKER_02Awesome. That's an interesting sorry. Isn't it crazy that um the last the last server you shipped was like 10 years from the one shipping now? 2025.
SPEAKER_00Yeah, that's probably crazy to think.
SPEAKER_02Pretty much pretty much ten years, anyway. A little bit more than ten.
SPEAKER_01I have sitting next to me over here is the um the last server. It was the last server I uh designed, I is a semi-custom design. And this is the prototype
Community Apps On The Web
SPEAKER_01here that I I just kept. You know, I use it as a backup target and um personal server, the old ABS 10 for yeah.
SPEAKER_02How long how long does it take to put together a server when you were building them?
SPEAKER_01Well, I I had a production line. So first thing you do is you you get the motherboards and populate them with the processor, the fan and memory, and then hook them up to a power supply. Okay, and so I would have I'd have several benches and I would have them all lined up, maybe 10 at a time, 10, 12 at a time, and just bear like that. You know, they're they're on a foam pad, but not in the chassis or anything, just hook up to a power supply. And I had you know a flash device that I would plug into each one and um run memory tests, okay, and let them just burn in for 72 hours typically, three days, and just let them run. Okay, and then each one that was still running, I would uh install into a chassis and build up the chassis. And uh yeah, it was quite time consuming. Uh and and the worst part is is not only you know building them, but also sourcing all the parts because motherboards would have a run, you know. You couldn't get after a certain time, you couldn't get the same motherboard anymore. And so I was constantly always trying to evaluate motherboards and qualify them and make sure it was gonna do the job. And um same with disk controllers, add-on disk controllers. Yeah.
SPEAKER_02Did you ship them with disks or without disks?
SPEAKER_01I shipped them without disks, but if they ordered disks, they can't I had to put them in a separate box. So I I would assemble a server and completely populate it with all the devices that they bought, okay, and then run tests, you know, just run it.
SPEAKER_02So, Tom, in the early days of Unraid, when was it that you were able to actually download Unraid as a digital product as opposed to just having flash drives sent out through the post or complete servers sent out?
SPEAKER_01Yeah, right. That's a good question. I'm trying to remember now. At the very beginning, the software itself was only shipped on a USB flash device that we uh sent out, and we had that product for for a number of years. And there's been various server products, there's starter sets, and it was just the motherboard and controllers. Um
Betas, RCs, And How To Join
SPEAKER_01when it first was a download, it was probably when when we switched to x8664, is also when we um provided digital download. Every single license up until part way through version six, I hand generated and and attached a file to someone's email and sent it back to them. And I remember you know on vacation sitting uh in the Bahamas one time on their flaky internet processing orders. And um, but during Unrate six, we created the key server and uh automated a lot of that. Uh so that that was a big help.
SPEAKER_02Wow. So so the original ones then were only just um shipped out on a flash drive. I tell you, Tom, I would love to actually get hold of one of those old flash drives.
SPEAKER_01Um Yeah, I I have stuff floating around.
SPEAKER_02Really?
SPEAKER_01Oh yeah.
SPEAKER_02That does it have the older OS's on there?
SPEAKER_01Yeah, you know, I I'm forgetting now when we had to do this, uh I was trying to find I have version four, but I don't have any earlier releases. And I've got probably, you know, 40 hard drives floating around in boxes and storage, and I know it's on there, so soon I'll be able to uh retrieve those and I'll I'll find I'll find all the source code. And and in fact, there's you can
The Missing Power Cables Story
SPEAKER_01boot it still, it'll it'll work. Amazing.
SPEAKER_00You should get edible that does to do a video on RAID for it.
SPEAKER_02Yeah, I I would love one. Okay, so Tom, if you um were sat in your garage when you were putting servers together in 2005 and you could see now what's happened to Unraid and where it is now, what would you have made of it back then?
SPEAKER_01Well, you know, uh I had no idea. I mean, it it started out as I've said this before, and it is actually true, is just beer money. But now we're up to something like 30 employees, I want to say, 30 people that are that are working on a product in the company. And it's it's been super rewarding, is the right word. It's been very rewarding to see how the platform has evolved and and how people have stepped in and added their own ideas. You know, the features that in the in the timeline I sort of recited at the beginning, you know, those were added because those are things I wanted to do, basically. And the community did and I I integrate it. But but now we have a real team of uh brilliant people who are saying, hey, you know, we need to do this instead. And I'm like, yeah, you know what? That's a great idea.
Looking Back On 20 Years
SPEAKER_01I hadn't thought of that. And and so that's been really rewarding that um people, you know, the the culture in our company is anybody here can come up with any idea and it's worth consideration. Um so so yeah, that's that's the best word, is rewarding. It's been great. And um, I just think we should go another 20 years.
SPEAKER_02Yeah. Awesome. And I just want to say on behalf of all of the people watching here, thank you very much, Tom. Thank you for actually back in 2005 having the problem that you had to fix. We're very glad you had that problem, because without that problem, we wouldn't be here doing what we're doing. And I must say, 20 years is a remarkable run for any software product, and it's rarer still to have one that stays independent the whole way. And I'd just say that, you know, it must be awesome to actually see that as the founder that it's still independent and it's it's made it this far, and here we are. And here's to the next 20 years. Very excited to see what happens.
SPEAKER_01Yeah, thank you, Ed, and uh great job, man. Yeah, thank you, Ed. This is really good. Appreciate it.
SPEAKER_02Well, that's it for this two-part series. Exciting stuff. Now, a big thank you to Tiffany, Tom, and Eli for coming on and sharing all of that with us. And a big thank you to you guys as well for watching. Now, if you haven't seen part one with Tiffany, our CEO, go and check that out. I'll leave a link for you below in the description. And like Eli was saying, keep an eye out for the beta, because we'd really love to have you guys getting involved and come along with us. Now, if you've got any questions at all, just drop them in the comments of either video and we'll do our best to answer them. So thanks a lot for watching everyone, and I'll catch you in the next one.
.jpg)