They are working on an OpenWRT Two at the moment which will be Wifi 7.
OpenWRT runs on a lot of hardware and its a great way to extend the life of a router past the manufacturers patches as well as gain a lot of capabilities. I wouldn't buy a commercial router that wasn't supported by OpenWRT now.
I've been hunting for a good switch with at least two 10Gbps ports too, surprisingly hard to find today still. I've ended up with a XGS1250-12 (3 10Gbps ports) for now, but OpenWrt support isn't great, the ports end up 1Gbps (or some other similar problem, not sure I remember the details 100%), so still running Zyxel firmware which, well, isn't OpenWrt... Other people here might have recommendations for suitable switches with good OpenWrt support?
The other devices based on the same filogic chip do have dual 2.5Gbps at least.
You can get a Wifi 7 device and 2x2.5GBps with Wifi 7 support already with the Asus BT8 and a few other devices. Asus's bootloader firmware flasher will take the initial OpenWRT image so its really quite simple to get going.
How will it handle PPPoE at gigabit speeds? I've been wanting to replace by terrible router from my ISP, but the options that can handle gigabit+ PPPoE are limited.
Very well, IIRC I have measured its capability to route at about 16gbps IIRC although that isn't PPPoE just the usual iperf test, it handles my 1.2gbps without any drama.
Gl.Inet ships with their openwrt version. I have the last version and it can be flashed to vanilla openwrt or one a high speed branch. It’s been good and fast. I don’t need wifi 7 yet so I have time.
Why do they always have to look like some unholy blend of a cybernetic spider and a Knight Rider? What happened to a plain unassuming looking piece of industrial hardware...
if your rep depends so much on visual aesthetics then I'd say you don't have much rep to begin with. if someone trusts me they'll buy whatever I say regardless of how it looks, and likewise if I trust somebody to recommend a piece of hardware I know the aesthetics are irrelevant and they know more than I do about the specs compared to the competition.
Still have one, still works. Would happily use it if it were still practical. I think you can still load OpenWRT on them, but there's no software route around the hardware being outdated and slow.
As a counter opinion to "too slow" in regards to 54Mbps, as it likely depends on your use cases. My WRT54GL was the primary family router until 2018 and even worked just fine for live video broadcasting. Even today it sees good use for video calls with no issues at all. Lovely little piece of hardware that refuses to die. Just a shame that OpenWRT has dropped support since 2013, which feels a bit ironic given their name.
I bought it in anticipation of the Nintendo DS having WiFi capabilities, which I had never heard of before (I was like 13 or 14 then). Had to convince my parents to get broadband internet too.
Wifi 5-7 happened, now operating at 3 different frequency ranges (2.4, 5 and 6Ghz) and using techniques like beam forming and MIMO. All those antennas need to go somewhere.
If you want plain unassuming looking hardware get dedicated wifi access points and place them all over the building. There are plenty of those shaped liked big smoke detectors.
If you want single device there are also quite a few trash can shaped home routers.
A shit ton of beam forming and phased arrays. Why do you think all of a sudden there's a bunch of "WiFi Radar Imaging" projects popping up on HN? It's not just because of advances in ML. Boost the output power by a few more magnitudes and you can probably ship them to Ukraine.
Because the gamers and dads who make all the purchasing decisions about household routers love F-117s, and stealth bombers, and consider fighter jets and black “Nighthawk” branding to be way sexier than a pink box of candy.
Because a modern wifi router requires a minimum of 6 antennas. 9 is even better.
This lends itself to a spider like design with just a ton of antennas sticking out of a box, or a trash can with the antennas hidden inside around the outside edge.
But usually it is a phased-array setup, or it seems to be, with a row of antennas.
Another setup is circular or semicircular. I suppose it allows for a more uniform directional diagram across the entire 360°, because a straight phased array has harder time emitting sideways.
Because a shocking amount of consumers buy things based purely on how they appear and the gamer adjacent aesthetic looks surprising and advanced to consumers. Unassuming business boxes are much harder to sell via the visual marketing
I have been extremely happy with cisco 3802 access points purchased on ebay for $25 each. Sure, it's only wifi5, but they're pretty solid and you can just deploy a swarm of them.
And they don't look fugly.
It is a tremendous shame that cisco hasn't opensourced / unlocked this generation of kit.
Unfortunately you need to use the cisco software / firmware. The access points run linux but they're locked down like crazy with signed firmware blobs and such.
That said, the cisco firmware for this specific generation of access points is actually free and trivial to get -- create yourself a cisco account and go to downloads and download the 3802 "mobility express" firmware. The last ME firmware came out in 2024 and all this equipment and software is now totally unsupported by cisco so don't run PCI transactions at home... I'd also avoid running their captive portal or some of their other weird features...
Actually setting it up is a bit of a chore but it is a full featured "enterprise" (cough) AP management system with all the knobs and twiddles you could ask for.
It's really only a good idea if you don't value your time (like me) or if you have a sprawling plaster house where you want to have lots of cheap access points instead of a couple super fast ones.
Lastly, for better or worse, I haven't been able to make my kid's switch 2 work on the network.
It is probably a combination of hitting a (low) cost and mimo; cisco makes pretty reasonable looking APs with lots of radios and decent coverage and they look like UFOs not alien spiders.
But it's probably easier / cheaper to get maximum coverage at larger distances from a single AP using a big array of sticking out antennae, and that's what a normal home user is going to want.
In my opinion, get the Flint 2, the Flint 3 doesn't work with vanilla OpenWRT (but it does work with GL.iNet's OpenWrt fork). Then again, I don't need the 5x2.5G ports or Wifi 7 since my internet only goes to 1G.
I got the Flint 3 because I wanted 6E for my Quest 3. It’s not bad, but still doesn’t reach enough through the walls. What I like about GL.Inet UI is that it’s very easy to set up WireGuard /OpenVPN profiles per MAC, and you can drop into LuCi for more advanced stuff.
Source? A cursory check on their website shows a Hong Kong and a US address. Some people seem to be claiming mainline China associations as well, which could be true, can't find anything on that in either direction.
But Israeli, no can't find that. Sure you didn't confuse it with some other company?
It looks like the person you replied to deleted their comment, but I assume it was about GL-iNet being a Chinese company.
They have a tiny Hong Kong office that handles marketing as well as a US office for technical support, but the entirety of engineering and manufacturing is in Shenzhen and Chengdu.
Because they provide a hosted site-to-site VPN service they are obligated to hold a B13 license. One of the conditions of which is the ability for the Chinese government to request access to devices worldwide.
That is interesting (and I recommend you flash stock openwrt instead, I did on mine). But no, the parent comment I replied to claimed Israeli / IDF connections.
I have really liked my GL.iNet travel router, also with OpenWRT. I didn't think I would need a travel router but they're pretty handy.
I didn't realise routers like theirs existed, and had been paying through the nose for your standard brands like TPLink and hoping it didn't get popped.
As some not deep into networking, just home(lab) stuff, learning OpenWRT took some time. But now cannot think of buying a router not for OpenWRT, once I learn what is possible I want to use it.
Fot example from my ISP I can have two options for PPPoE connection, first is legacy IPv4 only but lacks IPv6, second is IPv4 but behind CGNAT and with modern IPv6.
With OpenWRT, I am able to make two PPPoE connection over the same wire and have the best of both worlds.
Off topic, but what amuses me about the "Wrt" name is that it was originally alternate firmware for the Linksys WRT54G router from 25 years ago. The name has stuck for whatever reason; I guess since only geeks use it and know what it is.
The best model of the WRT54G line. I would snag them at thrift stores for cheap to use for silly utility functions. I always referred to that particular model as "The highly-coveted WRT54GL."
I used a pair to provide Internet access at a Customer's construction site back in 2010. Cell phone hotspot wasn't a thing for me yet. We took a pair of WRT54Gs, configured one as a WiFi client, the other as a bog-standard router/AP, connected the LAN from the client to the WAN on the router/AP, pur a directional antenna onto the "client", and pointed it down the road toward a big business who offered free WiFi for Customers. We leeched off that until the real Internet service got installed. (It was a restaurant and we ate there at least once so we were Customers, right? >smile<)
It seems crazy to me that Linksys didn't look at the success of the WRT54GL and the higher prices they commanded and decide to just keep doing that. Why every company feels the need to roll their own firmware that is buggy, slow, crashy, and doesn't implement half of the promised functionality properly is still baffling to me.
Companies roll their own, I think, because of a combination of Not Invented Here and secret-sauce binary blobs. They work within the script that the chipset/radio maker gives them to follow.
---
They don't often offer inexpensive, deliberately-hackable units like the WRT54GL, I think, because of support costs.
And by "support costs," I don't mean that it was expensive to hold users' hands while they installed custom firmware -- that's never been a service that has been provided.
Instead, I mean that there are people who start goofing with this stuff and run out of skill when hacking close-ish to the metal on this kind of hardware. They don't know how to get themselves out of a jam and unbrick their device.
So they find a way to lie their way into getting an RMA and get the device replaced under warranty, and that's expensive for companies to deal with.
I'm pretty sure the software side of the project is a direct descendent from the WRT54G stack.
LinkSys got sued to release the firmware as it was GPL linked. This dump got modified to make the WRT54G way more powerful than LinkSys ever planned but they got to sell the hardware for years more than would have been expected at the time.
Yeah, I loved it because it allowed me to boost the signal above FCC-approved power requirements and saturate my house with that sweet 2.4GHz connection everywhere.
It is basically always better to run more APs at lower power in the areas where you need coverage, than to boost the power. Especially today with the radio spectrum being so congested.
Despite this, I could expect 3-5 people to hunt me down at PyCon when I was running the wireless to tell me that I had misconfigured the wifi because it was set to low power. More reports of that than reports of wifi not working, IIRC. ;-)
(I was running the wireless because the people we paid do to the wifi would just set up one or two APs and crank the power)
Yeah it was so popular they even released a specific WRT54GL model (where the L stands for Linux) in order to keep supporting third-party firmware after the main hardware series moved on to a more optimized VxWorks-based OS that let them ship less RAM and Flash.
A mainstream hardware company releasing a specific product SKU to support third party firmware really sounds crazy from the perspective of the current market where a substantial portion of the value in selling hardware is supposed to come from subscriptions and surveillance.
I used OPNSense briefly when I got 1gig synchronous fiber ~2015 at home on an old i5 desktop, which I think was shortly after the pfSense fork, and then found Mikrotik and RouterOS. Used RouterOS at home since and have been replacing aged out Cisco switches in the datacenter (100gig) and closets (20-40gig) cheaply. I'm looking to dump a handful of ASA's for OPNSense and the messing around I've done so far has been positive. It's aged nicely.
> it was originally alternate firmware for the Linksys WRT54G router from 25 years ago
There's a couple of fun examples like that. xda-developers is named after the O2 xda, a smartphone from 25 years ago that not many people ended up developing software for.
Well we still use Roman months and weekdays, and carry over the 30/31 days months even though we could’ve had a way better system by now (base 10 and equal divisions)
I just received my OpenWrt One because I’m tired of dealing with the questionable quality of most routers.
And I don’t feel like resurrecting my old PC that I used as a router for a while. I stopped doing that because it’s loud. Pretty sure the power supply fan is about to fly off.
But Qualcomm WiFi pci card with giant antenna in a dirt cheap PC running ancient Ubuntu and a simple hostapd setup is so far the most reliable WiFi router I’ve ever had. I hope openwrt one is even better :-)
In case it is not, an old PC with a dual port Intel NIC running OPNSense is so far the best router I have ever used. I mean rock solid performance with near zero maintenance beyond adjusting VLANs and setting up a 6in4 tunnel over the past 5 years solid. My home network is larger, more diverse, and more complex than what I suspect most people have, with several hundred devices and yet I log into the OPNSense UI maybe twice a year and usually just out of curiosity.
The learning curve is a little steeper than more consumer stuff but it is by no means beyond a person who is capable of using OpenWRT and the docs and forum support are better than 99.9% of open source projects I have seen over the past 25 years.
I don't disagree, however using an old PC as a router almost certainly wastes an enormous among of power. An old non-gaming PC could use 70kWh of power a month if running continuously (as a router would), which is around 11 a month and almost 140 a year. At that price you could just buy a nice router, or an OpenWrt One which will possibly also have newer, faster WiFi standards (WiFi 6)
I currently use an old boring HP tower. Nothing fancy, I think it’s a quad core AMD APU. But if building it from scratch I would get a $35 thin client off eBay and stick a NIC in it. The CPU load is minimal as the network card does all the processing. I do have 8 TP-Link access points and a hardware controller as well as three unmanaged PoE switches for the Wi-Fi but that would be the case regardless of what my router is.
>however using an old PC as a router almost certainly wastes an enormous among of power
I don't think you're quite right on this, or at least you're imagining using something inappropriate when the comparison here involves buying something new right? So it's not "OpenWrt One" vs "whatever you happen to have in your closet" but "OpenWrt One (~$110-130)" vs "whatever can be bought used for $110-130, if you have nothing appropriate". And while they won't go to near-zero like some ARM stuff might, idle power for PCs improved a ton after around the 2013 era. There are lots and lots of small systems available for equivalent prices on Ebay or the like made since then (like Intel NUCs or various other mini PCs) that will idle around 4-10W. Like to take something in the same price range as this OpenWrt One, I regularly see 7th gen era NUCs going for <$140. An i5-7260U will have single threaded performance about the same as the MediaTek in this unit and multi-thread close, but will also generally have 8-16 GB of RAM and often a 250-500GB NVMe drive as well. It'll probably have only one native ethernet, but USB or TB adapters work fine with Linux & FreeBSD at this point.
There's definitely a question of values and exactly what you're trying to focus on, but there are a lot of niceties in having lots of RAM on tap and extremely standard fallbacks to interface with a system, back it up, etc.
>or an OpenWrt One which will possibly also have newer, faster WiFi standards (WiFi 6)
If you want an AIO style device that's definitely a consideration, though again USB WiFi dongles are a thing too. But regardless of router choice, for someone considering going beyond what their ISP offers at all I think it's usually well worth spending the $50-80 to get a dedicated wireless access point. It'll make a major difference in real world performance in most spaces I've seen to just physically have a unit in an ideal spot (on a ceiling or high up on a wall, away from metal). Aesthetically the clean disks or rectangles those tend to have also blend well and mean that various boxes can be tucked away. And of course you get to upgrade networking bits separately from your router.
Anyway, definitely good there are multiple approaches, this is an area of life where people can have very different needs driven by very different physical environments and "stakeholders" (like significant others). But I think OPNsense (or other bog-standard-PC FOSS alternatives like VyOS) can be competitive even in TCO, depending on how much value you place on pushing your networking stack and what else you have going on (like solar power).
For some real numbers, I picked up a 2018 generic office PC (Core i5-7500) used for $50 as a backup machine to run linux on when my laptop was in for repairs, and it idles at 14 W, vs the OpenWrt One which appears to run at 5.5 W.
So that's 10 kWh/mo for the PC vs 4 kWh/mo for the OpenWrt One.
Yep, and that's anywhere from about $1.30 to $3 something in the really expensive states for electric rates in the US. Half that if you only count the delta between that and a low power device.
Spending hundreds on new hw to save $20 a year in power cost.
Hi I tried with OPNSense, but I use a screen reader they made a big song and dance about fixing 11y on there web interface. In the end they did fuck all. OpenWRT has bin good with a screen reader since the start and the few times I have pointed out things to be fixed they have bin fixed with in days. So yeah go OpenWRT.
That is a very solid argument against OPNSense! I do wonder if the advent of AI can be used to fix issues like this, but in the meantime I totally see why you would choose OpenWRT.
"Enterprise class" wifi routers from ten years ago sell on eBay for about 1/5th as much, and work just as well for most home or small business applications.
Aren't they also missing security patches? I'm not sure about you, but I'd rather have SOHO routers with up to date firmware, than enterprise routers with out of date firmware.
It's worth being careful here: a lot of the affordable enterprise-class routers from 10+ years ago aren't as fast as cheap consumer hardware - like the OP or just a decent mini PC. The primary arguments for the enterprise-class gear are around feature availability and certain aspects of reliability (redundant power/fans, better heat tolerance, higher quality components). It's also worth remembering that this kind of gear tends to be built for dedicated environments: loud fans, higher power draw/lack of power saving features, etc.
Beside the potential performance and environmental issues the other big downsides tend to include firmware availability - either because download from the site requires a login on the vendor's site or, increasingly commonly, the gear has hit LDOS and images just aren't posted. Obviously there are other "unofficial" places for such images, but the risk/legality are a whole other (potentially serious) question.
There's an additional issue mapping the requirements of home networking to enterprise gear: Ethernet switches are lousy firewalls (little or no NAT, primitive built-in security, DLNA/mDNS and friends aren't really sane options, etc). Finally, even at "1/5" the price the gear may still be quite a bit more expensive than other options. And if it's not expensive, it's usually because nobody wants it any more because of the issues mentioned above (being near- or beyond- LDOS).
FWIW this is from someone who literally built a commercial-class machine room in his house with dedicated AC, subpanels, commercial UPS, etc for data center class Ethernet, Fibre Channel and Infiniband switching as well as carrier-grade routers and still runs "enterprise" grade WLAN and switching and can lay hands on as much as I could reasonably want without too much drama or cost... Going down this road can absolutely be amazing if you either have a.) the background to properly source and run the hardware/software or b.) have a driving desire to learn how to do so or c.) have some very atypical requirements for home networking. Otherwise it tends to not be something to be done lightly.
>I just received my OpenWrt One because I’m tired of dealing with the questionable quality of most routers.
My latest 3 home routers have been mikrotik, mikrotik and juniper.
And the Juniper is still in there, just running as a switch. The old juniper srx sucks at uPNP but is otherwise still a fantastic router if I wanted to swap back to it.
Middle tier mikrotik took a power surge and lost 50% of its ports. It still routed just enough for me to get a replacement.
How about OPNSense on open hardware of your choice, and passing messy wireless to separate AP?
OpenWRT is very good, but the installation and upgrades are not easy. There is a zoo of images for different hardware, installation options and tools. It has to run on small devices, so there are limitations. The documentation on Wiki is scattered and could be improved.
I had to search forums for weeks for a custom package installation for my router. Right now I have been trying to upgrade to the latest version via LUCI for a while, and it stucks. Probably have to wait for few weeks, go through CLI and maybe search forums again.
I just thought I am paying a hefty time price for a bit more expensive x86 mini pc and AP.
I agree on the upgrade story, though supposedly the recent move to apk will help in that regard.
I moved from pfSense to OpenWRT due to the really poor IPv6 support in pfSense. I don't use the AP capability either. How are things in OPNSense these days?
Particular pain points from pfSense was that it published global IP as DNS address to LAN clients and no way around it, so connectivity broke every time prefix changed, and no real support for specifying prefix-less firewall rules or similar, so couldn't really expose anything via IPv6 without pain.
You're right, it's interesting that this device isn't the most technically superior in hardware or software, and isn't the most casual user friendly. It seems to be targeting a segment I can't bucket other than loyalists. Maybe good hardware & software for the cost?
Maybe so. The documentation seems to be all over the map, and the GUI suggests using "attended sysupgrade" for upgrades.
...which I tried doing, a week or so ago, for a minor point release update within the 25.12.x series. And then the router went out to lunch and didn't come back.
Getting it going again wasn't so bad as such things go. My router has a huge advantage here in that it's a Raspberry Pi 4, so it's easy to remove/replace/re-do the flash device and start over.
(Except: I get all out of sorts when I need to do Internet stuff to fix my Internet connection while that Internet connection is absent.)
Yeah, for non-X86 devices, getting to U-boot with pressing a combination of PINs in particular order and conditions and releasing at right time is a pain.
I think I wasted $100,000 in salary for $100 more in device cost, in setting up an OpenWRT router.
Apart from installation and upgrades, the OS itself is nice, very flexible and capable.
I've got other options for routing hardware and software (of course I do), but I generally keeping using OpenWRT. Looking back, it seems like I've had it around in some form or other in active use for about 20 years so far.
Part of what keeps it around is the flexibility and the home-network-centric hack-value. I mean, this whole thing grew out of a shell injection exploit on a Linksys WRT54G. :)
Anyway, it can keep whatever counts as a slow WAN connection today feeling responsive and quick with cake SQM, even while loaded heavy with traffic and users. It's nice in that way, even though enterprise types don't seem to be interested in that kind of thing at all.
I could take a nice Juniper router home from work to use instead and it would absolutely trounce the packet-forwarding performance of my cheap OpenWRT box...while also doing nothing at all to make my home-gamer WAN limitations more tolerable.
So OpenWRT is still my answer, with the warts and upgrade woes and all.
All of that is a nope. The solution is that a router should have a standard unattended upgrade system built into it that is on by default and pulls from the stable release stream, preserves your configuration automatically with a 100% guarantee of it working, automatically falls back to the last known working image if the update fails, and has a way to notify you of what’s going on with it. This must work out of the box with the first install without you having to do anything at all or even be aware of it except perhaps setting the time of day and day of week/month when the router is allowed to reboot itself for the upgrades (but the default should be set automatically by the system). Anything less than that is simply broken for anything that is considered production quality. Words like “image builder” and “config baked into the image” are for those developing the system, not end users.
> The solution is that a router should have a standard unattended upgrade system built into it that is on by default...
Mmm, no. Unexpected downtime for infrastructure is godawful... just ask Windows Home users.
OpenWRT has a "Click a button to upgrade" thing, just like just about every consumer/prosumer-grade equipment does. [0] It also has a command-line tool that one can use to automate upgrades, for environments where the phrase "production grade" is actually an important thing to think about. [1]
[2] Those documents mention that you need to install some things to get operator-initiated upgrades. As of March, the button to click is installed by default, and the CLI tool is installed on systems that have enough disk space for it. [3]
With the 25.12 release, the luci app to use ASU for upgrades became installed by default in OpenWrt's "vanilla" images the project builds and provides for supported hardware and devices.
Previous OpenWrt releases at least as far back as 21.02 could be equipped with the same degree of ASU support by installing a single package (luci-app-attendedsysupgrade) and its dependencies.
> How about OPNSense on open hardware of your choice
Yes, it's a possibility, but if you want to tinker, I think a plain Linux distro like Debian is better. Turning it into a router is literally a couple of kernel parameters and a few iptables rules to set up NAT. Nowadays that's less than fives minutes of work with Claude.
This buys you much better performance and hardware compatibility relative to a BSD system, as well as lower resource usage and attack surface (no GUI or other unnecessary additions). WiFi support on BSD is bad, but on Linux you can use hostapd and almost immediately get an access point for free. And of course Linux is also better if you intend to run other stuff on the same hardware.
Anecdotal, but I was never able to get FreeBSD (TrueNAS CORE, OpnSense) to get anywhere near 10 Gbps. Using Mellanox or Intel NICs. The Linux equivalents (TrueNAS SCALE, OpenWrt) handled it fine on the same hardware.
But what if you don't want to tinker? I switched to OPNsense as a direct replacement for our Asus "WiFi routers", and it has been phenomenal, reliable, and does everything needed - when you just want it to work, it really just does. But when you want more advanced functions, there are tons of plugins and stuff that you can run natively, while still having a true CLI.
I suppose it comes down to what you said - "if you intend to run other stuff on the same hardware." Is it a good idea to run all sorts of extra stuff on your literal firewall/router? And if you did, I'd assume using a hypervisor is safer anyway? That way you can have the GUI and reliability of OPNsense but have a Linux distro beside it.
You also said that Linux has much better performance vs BSD, which seems rather far fetched. Got any data for that?
One other thing: OPNsense comes with a ton of helpful rules to eliminate bot traffic, allow IPv6, different NATs, VLANS, etc which you'd have to add manually. Not the end of the world, but worth considering.
> Is it a good idea to run all sorts of extra stuff on your literal firewall/router?
I don't see any reason not to. I run dozens on services, both bare metal and containerized (Podman) on my router/firewall. It doubles as an all-purpose home server with plenty of headroom to spare. It's just a computer that sits at the edge of my network, and running services meant to be exposed to the Internet on it is natural.
> You also said that Linux has much better performance vs BSD, which seems rather far fetched. Got any data for that?
I should have worded this more carefully. What tends to happen is that BSD has worse (or no) drivers, that's when BSD's performance can noticeably degrade vs. Linux. From memory, people online were reporting issues with Realtek chips. With Intel NICs, the routing performance should be broadly equivalent .
Thats more or less what I did, and nix just made sense for the job. For 99% of people I'd say no its not worth it to tinker, just go with opnsense virtualized so you get at least some the benefits of the better linux drivers. By that I mean NBASE-T on various intel chips and while intel's igb is fairly solid on unixes many other vendors' drivers are less so.
However if you're willing to figure out configuring per your needs you definitely can get a lower latency router with all the same capabilities and more, with it's components more sanely isolated via containers.
i don't understand the use case for 1x2.5gbs + 1x1gbs for a router. Why not both 2.5gbs it's not like you'll be running lots of stuff on the router itself so it would be more useful to have a wan AND a lan at 2.5 (outside of load balancing for a lot of wifi devices of course)
I have similar board, from same producer - BananaPi BPI-R3 Mini.
Got cheaper version without case - designed simple box with cutouts and 3d printed it out.
I can say it works flawlessly. I'm very happy about it. Previously I was using openwrt on different consumer-grade routers and I always had some issues, even when selecting supported devices.
Without performing any work at all to optimize RAM usage: My all-singing, all-dancing OpenWRT router projects have always used less than 100MB of RAM. These days, they usually occupy less than 64MB.
1GB is a ton of RAM for this kind of application. :) What do you anticipate needing more for?
But pihole asks for a pi with 512gb ram. Add openwrt’s 100mb, now your ram budget for running something else (file server, pairdrop, irc, tailscale, etc) is <400mb. One nodejs app could use all of that.
I have and love my OpenWrt One for my main router. I have two, so that I have a backup one I can switch to if the first one ever dies. It is the best device to run OpenWrt on as it is fully supported hardware that has great images/packages for it. Routing speeds/buffer/latency are great, everything just works, price is very reasonable.
I don't use it for my APs, but that is mostly because I already had 3 TP-Link routers setup as dumb APs using OpenWrt that have been working great. If I did it again, I'd buy OpenWrt Ones though. Although Deco mesh kits I've used have worked exceptionally well, and have become my recommendation for friends/family that don't want to do things like run arbitrary packages on their router/APs.
another happy user here too. having openwrt with all features working and no tinkering out of the box (since it's their reference target) is a dream. this, plus the warm fuzzies of buying open-source, makes it worth it to me even with the 1GBps limitation and outdated WiFi (i use a separate AP anyway, like you). i swapped my ports in software to have 1GB WAN and 2.5GB LAN (which also lets me power the router with PoE, which i have coming in on the LAN port).
If you just want a good WiFi router or access point, unless you need something cannot do (e.g. WiFi 7 or 10 Gbit/s Ethernet), and if want to spend minimal time messing around with routers today and in the future, just get this one.
After getting this, I see no reason to ever buy any closed-source router again.
No need to learn/remember any other Router config either. It's just all OpenWRT, always looks the same, always works the same. Setting up a new one takes me 2 minutes max.
The recent OpenWRT update also brought the one feature the project was most sorely missing: A simple "Download and install latest firmware" button in the device UI.
Now they just need to add an unattended-upgrades option and I never have to log in again after initial setup.
I have been using them for years and I'm really happy. I recently bought the WiFi 6 upgrade kit for both of my turris. They "recently" released their latest version which is expensive but comes with WiFi 7 and 2.5 Gbps RJ45 and 10 Gbps SPF.
I've recently been test driving SPR[1] which is a security-oriented distro for Wi-Fi routers. The team behind it are serious about Wi-Fi security and have a research lab[2] that has been credited with several CVEs in the likes of Apple's network stack. The headline feature is strong device isolation for semi-trusted guest and home automation devices, and the software stack is based around containerized and audited Go daemons.
It ran pretty well for me as a travel router I cobbled together from a Raspberry Pi and Netgear A7500 USB dongle for a stay in a short-term rental where the infrastructure network was shared with other units. More recently I have been trialing their CM5-based model with Wi-Fi 7 and 2.5GbE PoE for use as primary home Wi-Fi.
I switched from a Google Wifi to this and found it to be just as stable, but with better range/signal strength, and easier to apply the parental controls I want.
I do it the opposite way, disable my kids' devices at night, but I suspect your desired method would also be supported using native features. I have found LLMs to be very helpful in providing the right settings.
There is a plugin marketplace that provides more features, like ad-blocking. I haven't played with those yet, so I cannot vouch for them.
What I want is a cheap, brainless wifi 6 or 7 device, that's easy to mesh and extend. I currently have Orbis that are Wifi-5, and I'm sick of the general untrustworthiness. The thing is, like my WRT54G from back in the day, they just work in all situations. It's amazing how stable this kit has been for 7 years.
I'm not in the US. I can't Amazon. I don't want to spend the equivalent of 1k USD just to get 3 devices in a brainless mesh that covers my ~125sqm place made of insane amounts of radio blocking concrete.
I no longer want to maintain my network, to have network. I want things that just work. The Orbi did that for me - but the costing makes me think abut the future, and not finding a painless solution. I guess that's the tradeoff.
To wit, I also want the mesh comms on another channel (i.e 6Ghz, rather then the 2.4/5.0) and the computers/IOTs/etc isolated from that. Perhaps cheap tri-band is insanely wishful thinking. It sure seems that way.
I use opnsense with an aliexpress n100 router. It works very well and I enjoy it. But upgrades scare the crap out of me. I've only had 1 upgrade where things went bad. I have zfs snapshots and everything, but just because its a headless unit, I get super anxiety upgrading the system waiting for the beeps for it to come back online.
I virtualize mine with proxmox and pass the necessary NICs through to the VM. Snapshots make rollbacks a breeze. You can also run HA pairs if you want to do updates without downtime.
Bpi does a nice job on open source hardware and openwrt integration, but only 2 ports in this model is really good?
Just update my bpi-r3 purchased years ago from 24.x to 25.12, it need a bit of extra work because the sfp interface is renamed(though I never used that), and i finally just do plain sysupgrade only and add some required packages because i run a customized fd.io vpp build on the bpi-r3, connected via vhost-net backed tuntap, it works well, I never regret bought this machine, with current agent stuffs, i think bpi r3 model may be more fun to play with
It does raise the question that if it is for developers, what exactly is being developed? Especially if its not representative of hardware that is available or desired; is there some advantage targeting a very particular chipset? This seems to be the only device using it (from what i could find briefly)
A 5 port 2.5GbE switch would upgrade this to 5 total ports (4x 2.5GbE), and costs less than $100. If you only need 1GbE then it's even cheaper.
Outside of home-labs, it's rare for me to see any devices connected to the LAN side of a wireless router these days, and more than 1 (i.e. the non-portable device that is closest to the router) is exceedingly rare.
Neither my parents nor my wife's parents have their desktop connected to their router. The cable modem isn't even in the same room as the desktop.
[edit]
If it matters, my mom no longer has a desktop (she uses a docked laptop now), but it is true of the docking station and was true of her previous desktop.
Ok, so your wan is 1gig, and your lan 2.5...handy but not much of a perk. Lets just call this an AP, which would clear up many issues people seem to have with it
It'd be handy for me: The fastest WAN pipe I can get is less than a gigabit while my LAN is still gigabit. I can't be the only person with this situation, wherein: If anything, then the 2.5-gig port is overkill.
I don't have any direct interest in the wifi radio that the box includes (I already have Mikrotik APs that I like just fine), except to configure it as a failover station-mode interface to use with a phone hotspot when the DOCSIS connection is on the fritz.
...which doesn't happen often at all, but it's annoying when it does happen. It's nice to be able to work around problems like that with OpenWRT.
Chaining a switch off the gateway is the best way to do it anyway. If you do that, then when you reboot your gateway, your lan devices do not lose their physical link and can continue talking to each other.
That's enough connectivity for a gigabit WAN pipe and a LAN full of stuff (including one or more better/faster APs), if a person wants to slice it up that way.
I'd like something bigger than 1gb for the LAN port...yeah, most of my stuff is quick from a wifi standpoint, but I don't want the hardwired stuff to be second class citizens.
Does it have hardware PPPoE offloading? Because it's a huge issue for those of us stuck with old-school telecoms for our fibre connections. Doing PPPoE at gigabit speeds needs something that can handle it.
Have you looked at the BPI-R4? It's a pricier option than the OpenWRT One, but it has excellent hw acceleration for networking tasks. I am 90% sure I recall someone reporting using it for a 2.5Gbps PPPoE connection and it handled it. It's also supported by the OpenMPTCProuter project if you need network aggregation support.
While they're okay on paper, they managed to burn a lot of customers in regards to the add-on wifi 7 module.
Quite a few of the modules went out without their eeprom programmed correctly, and all of them appear to be plagued by quite mediocre performance from (alleged) inadequate RF shielding on the card.
It looks like they've revised the design, but it's peeved off quite a few of the Banana Pi/Sinovoip customers who have bought them with the intention of using it as a router/ap with their R4. (It's dual-PCIe fingers with unique spacing and requires out of spec PCIe voltages, so it's only practically usable with a BPi R4.)
At the very least, the customers using them as wired-only routers are likely to be having a slightly better experience.
Been using Openwrt for years, router and access points. Any new AP I buy has to has to be supported by Openwrt. I have probably bought 7 or 8 old AP's at yard sales over the years for $5/$10 and turned them into decent wifi access points back to an Openwrt router running vlans, bandwidth monitoring, QoS, tcpdump for ssh dump with wireshark.
Just excellent, some pretty smart people in the Openwrt forum as well if you have problems.
As someone who knows very little about WiFi, I always thought it sucked that if you wanted to go from 802.11this to 802.11that, it always requires brand new hardware with a different WiFi chip that implemented the new standard. Is there a good reason that software-defined 802.11 doesn't exist and that every new standard requires a different radio+SoC?
One example is the introduction of MIMO, a technique to send multiple data streams in the same frequency band in parallel. This requires multiple antennas, i.e. hardware which wasn't there in the previous wifi version. Note this was 2009.
I have one of these and love it, especially after I once bricked it during a manual software update and got to use the dip switch reset to reflash it using the ROM.
I wish it had more ethernet ports but I've managed to live with that. I'd be up for buying an OpenWrt Two as a backup or to replace this if it has even one more LAN jack.
I started down my “custom” home network journey with OpenWrt and some aftermarket hardware routers. Enjoyed my time using / playing with it.
After some time though, I eventually moved over to using OpenBSD directly. My small brain has a much better understanding of all the moving parts compared to that of OpenWrt :P
I’ve been running one for a while and love it. Have also built a Nix -> OpenWRT config language transpiler so that I can keep my router state in Nix files and have nice deterministic rollbacks etc. It’s been great!
I became interested in OpenWrt when I noticed that the cloud portal for my ISP reported to me the names and types of devices that were associated to my home access point/router.
Suddenly I want to put every IPS device into dumb bridge mode, and run my own damn router.
I once tried to flash OpenWRT onto a very old router that was EOL, and apparently wherever I had downloaded the firmware it was pre-pwned. There was literally an interface numbered “eth666” and the CLI was playing crazy mind games with me as I tried to configure it.
So I thoroughly destroyed the router and e-wasted it. I honestly never want an open-source router OS ever again.
There is definitely beauty in having a separate router device that chugs on just fine regardless what happens to the rest of your network. But I got bored with the constantly-churning embedded culture, bespoke OS's (sorry, OpenWRT), and VPNs generally want more CPU than what purpose-built "routers" have. So I just went back to the old way of using a plain Linux machine as the gateway (now virtualized, with NixOS and nftables) and couldn't be happier. WiFi AP is done by that same physical machine (not virtualized) and by two other amd64 machines that double as Kodi boxes. When you learn netfilter/iproute2, that experience carries to anything else you might switch to.
It's not obsolete, it's basically the contemporary baseline. Remember, this is a cheap device. And unlike most Chinese garbage, you can be reasonably certain that it isn't backdoored.
That's such an assumption of needs. For someone to be using a 10g capable NAT would be some sort of super nerd with such a small portion of the population. More and more households have no compute device other than their devices. With WiFi routers now being sold with OpenWRT installed, it no longer means you're in the nerd category for installing it.
They are working on an OpenWRT Two at the moment which will be Wifi 7.
OpenWRT runs on a lot of hardware and its a great way to extend the life of a router past the manufacturers patches as well as gain a lot of capabilities. I wouldn't buy a commercial router that wasn't supported by OpenWRT now.
Hopefully, dual 2.5gbe too?
I'm hoping for one with at least two 10Gbps ethernet ports (one for upstream, one for downstream). Ideally more, but two would be great.
You might be interested in the Turris Omnia NG which has two 10g SFP+ ports:
https://www.turris.com/en/products/omnia-NG-wired/
I've been hunting for a good switch with at least two 10Gbps ports too, surprisingly hard to find today still. I've ended up with a XGS1250-12 (3 10Gbps ports) for now, but OpenWrt support isn't great, the ports end up 1Gbps (or some other similar problem, not sure I remember the details 100%), so still running Zyxel firmware which, well, isn't OpenWrt... Other people here might have recommendations for suitable switches with good OpenWrt support?
https://mono.si/
The other devices based on the same filogic chip do have dual 2.5Gbps at least.
You can get a Wifi 7 device and 2x2.5GBps with Wifi 7 support already with the Asus BT8 and a few other devices. Asus's bootloader firmware flasher will take the initial OpenWRT image so its really quite simple to get going.
How will it handle PPPoE at gigabit speeds? I've been wanting to replace by terrible router from my ISP, but the options that can handle gigabit+ PPPoE are limited.
Very well, IIRC I have measured its capability to route at about 16gbps IIRC although that isn't PPPoE just the usual iperf test, it handles my 1.2gbps without any drama.
The planned specs are here, they say it will be made by GL.iNet:
https://openwrt.org/voting/2025-02-12-openwrt-two
Otherwise this router from GL.iNet has OpenWRT preinstalled, Wifi 7, 5x2.5G:
http://www.gl-inet.com/en-gb/products/gl-be9300
> expected availability is late '25.
T-T. Any update on the timeframe (and presumably also I would expect the expected price to be solidly in the mid to high 300s at this point)?
Latest speculation would be that they don't have a manufacturer anymore https://www.reddit.com/r/openwrt/comments/1rnr0sv/what_happe...
Gl.Inet ships with their openwrt version. I have the last version and it can be flashed to vanilla openwrt or one a high speed branch. It’s been good and fast. I don’t need wifi 7 yet so I have time.
Would you be able to set this up as a simple mesh with two units?
I have two old Amplifi HD units in wireless backhaul mesh that I’d like to upgrade
If the routers use OpenWrt then you can use mesh.
Why do they always have to look like some unholy blend of a cybernetic spider and a Knight Rider? What happened to a plain unassuming looking piece of industrial hardware...
I couldn't agree more. It makes it difficult to attach my reputation to when making suggestions for hardware purchases.
if your rep depends so much on visual aesthetics then I'd say you don't have much rep to begin with. if someone trusts me they'll buy whatever I say regardless of how it looks, and likewise if I trust somebody to recommend a piece of hardware I know the aesthetics are irrelevant and they know more than I do about the specs compared to the competition.
The WRT54 is still one of my favorite pieces of industrial design. Small case, ~~purple~~blue and grey, two antennas.
https://en.wikipedia.org/wiki/Linksys_WRT54G_series
Still have one, still works. Would happily use it if it were still practical. I think you can still load OpenWRT on them, but there's no software route around the hardware being outdated and slow.
Yeah me to. I hade a WRT54G V2.2 for ages. Loved that thing. Just toslow now tho.
As a counter opinion to "too slow" in regards to 54Mbps, as it likely depends on your use cases. My WRT54GL was the primary family router until 2018 and even worked just fine for live video broadcasting. Even today it sees good use for video calls with no issues at all. Lovely little piece of hardware that refuses to die. Just a shame that OpenWRT has dropped support since 2013, which feels a bit ironic given their name.
My first WiFi router :)
I bought it in anticipation of the Nintendo DS having WiFi capabilities, which I had never heard of before (I was like 13 or 14 then). Had to convince my parents to get broadband internet too.
One could stack them.
Wifi 5-7 happened, now operating at 3 different frequency ranges (2.4, 5 and 6Ghz) and using techniques like beam forming and MIMO. All those antennas need to go somewhere.
If you want plain unassuming looking hardware get dedicated wifi access points and place them all over the building. There are plenty of those shaped liked big smoke detectors.
If you want single device there are also quite a few trash can shaped home routers.
That complaint was about its styling, not number of antennas
Why does Wifi 5-7 require designing the case so it is shaped like an F-117 fighter jet instead of a box of candy?
A shit ton of beam forming and phased arrays. Why do you think all of a sudden there's a bunch of "WiFi Radar Imaging" projects popping up on HN? It's not just because of advances in ML. Boost the output power by a few more magnitudes and you can probably ship them to Ukraine.
Because the gamers and dads who make all the purchasing decisions about household routers love F-117s, and stealth bombers, and consider fighter jets and black “Nighthawk” branding to be way sexier than a pink box of candy.
Because a modern wifi router requires a minimum of 6 antennas. 9 is even better.
This lends itself to a spider like design with just a ton of antennas sticking out of a box, or a trash can with the antennas hidden inside around the outside edge.
Do you have other ideas for how to lay it out?
My radio knowledge is not up to par, but couldn't a phased array antenna setup allow for friendlier form factors?
But usually it is a phased-array setup, or it seems to be, with a row of antennas.
Another setup is circular or semicircular. I suppose it allows for a more uniform directional diagram across the entire 360°, because a straight phased array has harder time emitting sideways.
Some 3-band wifi 7 AIO routers that resemble neither spiders, nor trash cans:
https://mikrotik.com/product/hap_be3_media
https://eero.com/shop/eero-pro-7
https://www.asus.com/us/networking-iot-servers/whole-home-me...
The Mikrotik HAP have terrible coverage, and the other two are the hallmark trashcan design I mentioned.
Right?
I guess they think consumers need them to look like this crap.
Because a shocking amount of consumers buy things based purely on how they appear and the gamer adjacent aesthetic looks surprising and advanced to consumers. Unassuming business boxes are much harder to sell via the visual marketing
I have been extremely happy with cisco 3802 access points purchased on ebay for $25 each. Sure, it's only wifi5, but they're pretty solid and you can just deploy a swarm of them.
And they don't look fugly.
It is a tremendous shame that cisco hasn't opensourced / unlocked this generation of kit.
Documentation for how to set these up without the cisco control platform being present is hard to come by.
You have any docs on how to set these up? I believe a firmware change is required.
Unfortunately you need to use the cisco software / firmware. The access points run linux but they're locked down like crazy with signed firmware blobs and such.
That said, the cisco firmware for this specific generation of access points is actually free and trivial to get -- create yourself a cisco account and go to downloads and download the 3802 "mobility express" firmware. The last ME firmware came out in 2024 and all this equipment and software is now totally unsupported by cisco so don't run PCI transactions at home... I'd also avoid running their captive portal or some of their other weird features...
Actually setting it up is a bit of a chore but it is a full featured "enterprise" (cough) AP management system with all the knobs and twiddles you could ask for.
It's really only a good idea if you don't value your time (like me) or if you have a sprawling plaster house where you want to have lots of cheap access points instead of a couple super fast ones.
Lastly, for better or worse, I haven't been able to make my kid's switch 2 work on the network.
The spider look comes from multiple antennas. Multiple antennas are needed for beamforming [1]; they represent a minimal phased array.
[1]: https://www.networkworld.com/article/967954/beamforming-expl...
It is probably a combination of hitting a (low) cost and mimo; cisco makes pretty reasonable looking APs with lots of radios and decent coverage and they look like UFOs not alien spiders.
But it's probably easier / cheaper to get maximum coverage at larger distances from a single AP using a big array of sticking out antennae, and that's what a normal home user is going to want.
In my opinion, get the Flint 2, the Flint 3 doesn't work with vanilla OpenWRT (but it does work with GL.iNet's OpenWrt fork). Then again, I don't need the 5x2.5G ports or Wifi 7 since my internet only goes to 1G.
I got the Flint 3 because I wanted 6E for my Quest 3. It’s not bad, but still doesn’t reach enough through the walls. What I like about GL.Inet UI is that it’s very easy to set up WireGuard /OpenVPN profiles per MAC, and you can drop into LuCi for more advanced stuff.
Doesn't the Flint 3 require a binary blob to run the broadcom chip? If so you'll always have a locked down black box inside your "open" system.
I got the Flint 2 to avoid that, and I'm really happy with it for the reasons you mentioned.
Very, very happy with my Flint 2. Rock solid.
deleted.
based
Source? A cursory check on their website shows a Hong Kong and a US address. Some people seem to be claiming mainline China associations as well, which could be true, can't find anything on that in either direction.
But Israeli, no can't find that. Sure you didn't confuse it with some other company?
Apologies, you are right.
It looks like the person you replied to deleted their comment, but I assume it was about GL-iNet being a Chinese company.
They have a tiny Hong Kong office that handles marketing as well as a US office for technical support, but the entirety of engineering and manufacturing is in Shenzhen and Chengdu.
Because they provide a hosted site-to-site VPN service they are obligated to hold a B13 license. One of the conditions of which is the ability for the Chinese government to request access to devices worldwide.
That is interesting (and I recommend you flash stock openwrt instead, I did on mine). But no, the parent comment I replied to claimed Israeli / IDF connections.
Why spread disinformation?
https://www.gl-inet.com/en-us/pages/about-us
My apologies. I must have confused them with someone else.
I have really liked my GL.iNet travel router, also with OpenWRT. I didn't think I would need a travel router but they're pretty handy.
I didn't realise routers like theirs existed, and had been paying through the nose for your standard brands like TPLink and hoping it didn't get popped.
As some not deep into networking, just home(lab) stuff, learning OpenWRT took some time. But now cannot think of buying a router not for OpenWRT, once I learn what is possible I want to use it.
Fot example from my ISP I can have two options for PPPoE connection, first is legacy IPv4 only but lacks IPv6, second is IPv4 but behind CGNAT and with modern IPv6. With OpenWRT, I am able to make two PPPoE connection over the same wire and have the best of both worlds.
The Two at least seems to have some features to it. Albeit catering to home ethernet enthusiasts.
I would really love to see something like this with just 10 sfp cages, no switching all routed interfaces. A real open source alternative to mikrotik.
> I wouldn't buy a commercial router that wasn't supported by OpenWRT now.
Is there any closed-sourced firmware that exceeds OpenWRT performance on the same level of hardware?
I think that some proprietary firmware may have hardware optimizations that aren’t possible in a community-developed environment.
Off topic, but what amuses me about the "Wrt" name is that it was originally alternate firmware for the Linksys WRT54G router from 25 years ago. The name has stuck for whatever reason; I guess since only geeks use it and know what it is.
Similar to XBMC at least for a long time.
I still have a WRT54GL sitting in a box somewhere.
The best model of the WRT54G line. I would snag them at thrift stores for cheap to use for silly utility functions. I always referred to that particular model as "The highly-coveted WRT54GL."
I used a pair to provide Internet access at a Customer's construction site back in 2010. Cell phone hotspot wasn't a thing for me yet. We took a pair of WRT54Gs, configured one as a WiFi client, the other as a bog-standard router/AP, connected the LAN from the client to the WAN on the router/AP, pur a directional antenna onto the "client", and pointed it down the road toward a big business who offered free WiFi for Customers. We leeched off that until the real Internet service got installed. (It was a restaurant and we ate there at least once so we were Customers, right? >smile<)
It seems crazy to me that Linksys didn't look at the success of the WRT54GL and the higher prices they commanded and decide to just keep doing that. Why every company feels the need to roll their own firmware that is buggy, slow, crashy, and doesn't implement half of the promised functionality properly is still baffling to me.
Companies roll their own, I think, because of a combination of Not Invented Here and secret-sauce binary blobs. They work within the script that the chipset/radio maker gives them to follow.
---
They don't often offer inexpensive, deliberately-hackable units like the WRT54GL, I think, because of support costs.
And by "support costs," I don't mean that it was expensive to hold users' hands while they installed custom firmware -- that's never been a service that has been provided.
Instead, I mean that there are people who start goofing with this stuff and run out of skill when hacking close-ish to the metal on this kind of hardware. They don't know how to get themselves out of a jam and unbrick their device.
So they find a way to lie their way into getting an RMA and get the device replaced under warranty, and that's expensive for companies to deal with.
(Those people fucking suck.)
Used to work for that model. Great device for it’s time.
I'm pretty sure the software side of the project is a direct descendent from the WRT54G stack.
LinkSys got sued to release the firmware as it was GPL linked. This dump got modified to make the WRT54G way more powerful than LinkSys ever planned but they got to sell the hardware for years more than would have been expected at the time.
Yeah, I loved it because it allowed me to boost the signal above FCC-approved power requirements and saturate my house with that sweet 2.4GHz connection everywhere.
It is basically always better to run more APs at lower power in the areas where you need coverage, than to boost the power. Especially today with the radio spectrum being so congested.
Despite this, I could expect 3-5 people to hunt me down at PyCon when I was running the wireless to tell me that I had misconfigured the wifi because it was set to low power. More reports of that than reports of wifi not working, IIRC. ;-)
(I was running the wireless because the people we paid do to the wifi would just set up one or two APs and crank the power)
Please travel back to 2003 and talk to old me when I could only afford one AP and had no idea how to make them work in concert together. :)
Good to know
> It is basically always better to run more APs at lower power in the areas where you need coverage, than to boost the power.
Only if your clients are competent at roaming.
Yeah, Linksys made a killing from this and the WRT54GS 2.0 because of OpenWRT.
Yeah it was so popular they even released a specific WRT54GL model (where the L stands for Linux) in order to keep supporting third-party firmware after the main hardware series moved on to a more optimized VxWorks-based OS that let them ship less RAM and Flash.
A mainstream hardware company releasing a specific product SKU to support third party firmware really sounds crazy from the perspective of the current market where a substantial portion of the value in selling hardware is supposed to come from subscriptions and surveillance.
Recycled 4 or so WRT54G variants a couple years ago I ran Tomato on for friend's small businesses and my home in early 2000's.
I miss the Tomato UI/UX... I don't care for LUCU or OpnSense's UI by comparison...
Been using OpnSense for about 8 years now though... it's just been the best option for me, I use separate commercial AP.
FreshTomato is still active, and they are doing x86 builds now. I'm running it on an aging Netgear R7000 and it has been stable.
I used OPNSense briefly when I got 1gig synchronous fiber ~2015 at home on an old i5 desktop, which I think was shortly after the pfSense fork, and then found Mikrotik and RouterOS. Used RouterOS at home since and have been replacing aged out Cisco switches in the datacenter (100gig) and closets (20-40gig) cheaply. I'm looking to dump a handful of ASA's for OPNSense and the messing around I've done so far has been positive. It's aged nicely.
Did WRT mean anything? Wireless RouTer?
> it was originally alternate firmware for the Linksys WRT54G router from 25 years ago
There's a couple of fun examples like that. xda-developers is named after the O2 xda, a smartphone from 25 years ago that not many people ended up developing software for.
Well we still use Roman months and weekdays, and carry over the 30/31 days months even though we could’ve had a way better system by now (base 10 and equal divisions)
What a coincidence to see this on the front page!
I just received my OpenWrt One because I’m tired of dealing with the questionable quality of most routers.
And I don’t feel like resurrecting my old PC that I used as a router for a while. I stopped doing that because it’s loud. Pretty sure the power supply fan is about to fly off.
But Qualcomm WiFi pci card with giant antenna in a dirt cheap PC running ancient Ubuntu and a simple hostapd setup is so far the most reliable WiFi router I’ve ever had. I hope openwrt one is even better :-)
In case it is not, an old PC with a dual port Intel NIC running OPNSense is so far the best router I have ever used. I mean rock solid performance with near zero maintenance beyond adjusting VLANs and setting up a 6in4 tunnel over the past 5 years solid. My home network is larger, more diverse, and more complex than what I suspect most people have, with several hundred devices and yet I log into the OPNSense UI maybe twice a year and usually just out of curiosity.
The learning curve is a little steeper than more consumer stuff but it is by no means beyond a person who is capable of using OpenWRT and the docs and forum support are better than 99.9% of open source projects I have seen over the past 25 years.
I don't disagree, however using an old PC as a router almost certainly wastes an enormous among of power. An old non-gaming PC could use 70kWh of power a month if running continuously (as a router would), which is around 11 a month and almost 140 a year. At that price you could just buy a nice router, or an OpenWrt One which will possibly also have newer, faster WiFi standards (WiFi 6)
I use an m920q as my router and it idles around 5W.
I currently use an old boring HP tower. Nothing fancy, I think it’s a quad core AMD APU. But if building it from scratch I would get a $35 thin client off eBay and stick a NIC in it. The CPU load is minimal as the network card does all the processing. I do have 8 TP-Link access points and a hardware controller as well as three unmanaged PoE switches for the Wi-Fi but that would be the case regardless of what my router is.
>however using an old PC as a router almost certainly wastes an enormous among of power
I don't think you're quite right on this, or at least you're imagining using something inappropriate when the comparison here involves buying something new right? So it's not "OpenWrt One" vs "whatever you happen to have in your closet" but "OpenWrt One (~$110-130)" vs "whatever can be bought used for $110-130, if you have nothing appropriate". And while they won't go to near-zero like some ARM stuff might, idle power for PCs improved a ton after around the 2013 era. There are lots and lots of small systems available for equivalent prices on Ebay or the like made since then (like Intel NUCs or various other mini PCs) that will idle around 4-10W. Like to take something in the same price range as this OpenWrt One, I regularly see 7th gen era NUCs going for <$140. An i5-7260U will have single threaded performance about the same as the MediaTek in this unit and multi-thread close, but will also generally have 8-16 GB of RAM and often a 250-500GB NVMe drive as well. It'll probably have only one native ethernet, but USB or TB adapters work fine with Linux & FreeBSD at this point.
There's definitely a question of values and exactly what you're trying to focus on, but there are a lot of niceties in having lots of RAM on tap and extremely standard fallbacks to interface with a system, back it up, etc.
>or an OpenWrt One which will possibly also have newer, faster WiFi standards (WiFi 6)
If you want an AIO style device that's definitely a consideration, though again USB WiFi dongles are a thing too. But regardless of router choice, for someone considering going beyond what their ISP offers at all I think it's usually well worth spending the $50-80 to get a dedicated wireless access point. It'll make a major difference in real world performance in most spaces I've seen to just physically have a unit in an ideal spot (on a ceiling or high up on a wall, away from metal). Aesthetically the clean disks or rectangles those tend to have also blend well and mean that various boxes can be tucked away. And of course you get to upgrade networking bits separately from your router.
Anyway, definitely good there are multiple approaches, this is an area of life where people can have very different needs driven by very different physical environments and "stakeholders" (like significant others). But I think OPNsense (or other bog-standard-PC FOSS alternatives like VyOS) can be competitive even in TCO, depending on how much value you place on pushing your networking stack and what else you have going on (like solar power).
That makes sense, but I really do have to emphasize just how gloriously stable my old PC router setup was.
It was the most incredibly hassle free router I've ever had.
Hopefully, the OpenWrt One is like that. But if not, I'm going to go back to killing the planet with my loud ass beater PC.
For some real numbers, I picked up a 2018 generic office PC (Core i5-7500) used for $50 as a backup machine to run linux on when my laptop was in for repairs, and it idles at 14 W, vs the OpenWrt One which appears to run at 5.5 W.
So that's 10 kWh/mo for the PC vs 4 kWh/mo for the OpenWrt One.
Yep, and that's anywhere from about $1.30 to $3 something in the really expensive states for electric rates in the US. Half that if you only count the delta between that and a low power device.
Spending hundreds on new hw to save $20 a year in power cost.
Hi I tried with OPNSense, but I use a screen reader they made a big song and dance about fixing 11y on there web interface. In the end they did fuck all. OpenWRT has bin good with a screen reader since the start and the few times I have pointed out things to be fixed they have bin fixed with in days. So yeah go OpenWRT.
That is a very solid argument against OPNSense! I do wonder if the advent of AI can be used to fix issues like this, but in the meantime I totally see why you would choose OpenWRT.
"Enterprise class" wifi routers from ten years ago sell on eBay for about 1/5th as much, and work just as well for most home or small business applications.
Aren't they also missing security patches? I'm not sure about you, but I'd rather have SOHO routers with up to date firmware, than enterprise routers with out of date firmware.
No I’m saying load openwrt on them.
It's worth being careful here: a lot of the affordable enterprise-class routers from 10+ years ago aren't as fast as cheap consumer hardware - like the OP or just a decent mini PC. The primary arguments for the enterprise-class gear are around feature availability and certain aspects of reliability (redundant power/fans, better heat tolerance, higher quality components). It's also worth remembering that this kind of gear tends to be built for dedicated environments: loud fans, higher power draw/lack of power saving features, etc.
Beside the potential performance and environmental issues the other big downsides tend to include firmware availability - either because download from the site requires a login on the vendor's site or, increasingly commonly, the gear has hit LDOS and images just aren't posted. Obviously there are other "unofficial" places for such images, but the risk/legality are a whole other (potentially serious) question.
There's an additional issue mapping the requirements of home networking to enterprise gear: Ethernet switches are lousy firewalls (little or no NAT, primitive built-in security, DLNA/mDNS and friends aren't really sane options, etc). Finally, even at "1/5" the price the gear may still be quite a bit more expensive than other options. And if it's not expensive, it's usually because nobody wants it any more because of the issues mentioned above (being near- or beyond- LDOS).
FWIW this is from someone who literally built a commercial-class machine room in his house with dedicated AC, subpanels, commercial UPS, etc for data center class Ethernet, Fibre Channel and Infiniband switching as well as carrier-grade routers and still runs "enterprise" grade WLAN and switching and can lay hands on as much as I could reasonably want without too much drama or cost... Going down this road can absolutely be amazing if you either have a.) the background to properly source and run the hardware/software or b.) have a driving desire to learn how to do so or c.) have some very atypical requirements for home networking. Otherwise it tends to not be something to be done lightly.
>I just received my OpenWrt One because I’m tired of dealing with the questionable quality of most routers.
My latest 3 home routers have been mikrotik, mikrotik and juniper.
And the Juniper is still in there, just running as a switch. The old juniper srx sucks at uPNP but is otherwise still a fantastic router if I wanted to swap back to it.
Middle tier mikrotik took a power surge and lost 50% of its ports. It still routed just enough for me to get a replacement.
Latest tik is doing great.
I dont see what the One offers me at all tbh.
How about OPNSense on open hardware of your choice, and passing messy wireless to separate AP?
OpenWRT is very good, but the installation and upgrades are not easy. There is a zoo of images for different hardware, installation options and tools. It has to run on small devices, so there are limitations. The documentation on Wiki is scattered and could be improved.
I had to search forums for weeks for a custom package installation for my router. Right now I have been trying to upgrade to the latest version via LUCI for a while, and it stucks. Probably have to wait for few weeks, go through CLI and maybe search forums again.
I just thought I am paying a hefty time price for a bit more expensive x86 mini pc and AP.
I agree on the upgrade story, though supposedly the recent move to apk will help in that regard.
I moved from pfSense to OpenWRT due to the really poor IPv6 support in pfSense. I don't use the AP capability either. How are things in OPNSense these days?
Particular pain points from pfSense was that it published global IP as DNS address to LAN clients and no way around it, so connectivity broke every time prefix changed, and no real support for specifying prefix-less firewall rules or similar, so couldn't really expose anything via IPv6 without pain.
You're right, it's interesting that this device isn't the most technically superior in hardware or software, and isn't the most casual user friendly. It seems to be targeting a segment I can't bucket other than loyalists. Maybe good hardware & software for the cost?
Upgrades are “owut upgrade” these days. Pretty straightforward.
Maybe so. The documentation seems to be all over the map, and the GUI suggests using "attended sysupgrade" for upgrades.
...which I tried doing, a week or so ago, for a minor point release update within the 25.12.x series. And then the router went out to lunch and didn't come back.
Getting it going again wasn't so bad as such things go. My router has a huge advantage here in that it's a Raspberry Pi 4, so it's easy to remove/replace/re-do the flash device and start over.
(Except: I get all out of sorts when I need to do Internet stuff to fix my Internet connection while that Internet connection is absent.)
Yeah, for non-X86 devices, getting to U-boot with pressing a combination of PINs in particular order and conditions and releasing at right time is a pain.
I think I wasted $100,000 in salary for $100 more in device cost, in setting up an OpenWRT router.
Apart from installation and upgrades, the OS itself is nice, very flexible and capable.
Well-said.
I've got other options for routing hardware and software (of course I do), but I generally keeping using OpenWRT. Looking back, it seems like I've had it around in some form or other in active use for about 20 years so far.
Part of what keeps it around is the flexibility and the home-network-centric hack-value. I mean, this whole thing grew out of a shell injection exploit on a Linksys WRT54G. :)
Anyway, it can keep whatever counts as a slow WAN connection today feeling responsive and quick with cake SQM, even while loaded heavy with traffic and users. It's nice in that way, even though enterprise types don't seem to be interested in that kind of thing at all.
I could take a nice Juniper router home from work to use instead and it would absolutely trounce the packet-forwarding performance of my cheap OpenWRT box...while also doing nothing at all to make my home-gamer WAN limitations more tolerable.
So OpenWRT is still my answer, with the warts and upgrade woes and all.
>OpenWRT is very good, but the installation and upgrades are not easy.
The solution is to use image-builder and bake your config into the image.
All of that is a nope. The solution is that a router should have a standard unattended upgrade system built into it that is on by default and pulls from the stable release stream, preserves your configuration automatically with a 100% guarantee of it working, automatically falls back to the last known working image if the update fails, and has a way to notify you of what’s going on with it. This must work out of the box with the first install without you having to do anything at all or even be aware of it except perhaps setting the time of day and day of week/month when the router is allowed to reboot itself for the upgrades (but the default should be set automatically by the system). Anything less than that is simply broken for anything that is considered production quality. Words like “image builder” and “config baked into the image” are for those developing the system, not end users.
hahaha You try building that all in to a small flash chip mate. Good luck!
That’s why I prefer things like Debian, OPNSense, etc. It IS hard. But that doesn’t mean that not doing it should be considered a done deal.
A multitude of router products manage to do this.
> The solution is that a router should have a standard unattended upgrade system built into it that is on by default...
Mmm, no. Unexpected downtime for infrastructure is godawful... just ask Windows Home users.
OpenWRT has a "Click a button to upgrade" thing, just like just about every consumer/prosumer-grade equipment does. [0] It also has a command-line tool that one can use to automate upgrades, for environments where the phrase "production grade" is actually an important thing to think about. [1]
[0] <https://openwrt.org/docs/guide-user/installation/attended.sy...> [2]
[1] <https://openwrt.org/docs/guide-user/installation/sysupgrade....> [2]
[2] Those documents mention that you need to install some things to get operator-initiated upgrades. As of March, the button to click is installed by default, and the CLI tool is installed on systems that have enough disk space for it. [3]
[3] <https://openwrt.org/releases/25.12/notes-25.12.0#integration...>
Your OpenWrt ecosystem knowledge seems oudated; upgrades are a solved problem since the advent of "Attended Sysupgrade": https://openwrt.org/docs/guide-user/installation/attended.sy...
It's been included in all suitable default image configurations starting with OpenWrt release 25.12.
I do run OpenWrt on my x86-based router, on my AP, and even on my managed switches, and have no regrets.
Indeed, I was referring to Attended Sysupgrade in luci or CLI.
So an image released in March? I’m not sure I’d proclaim it completely solved when it’s been ga for all of three months.
With the 25.12 release, the luci app to use ASU for upgrades became installed by default in OpenWrt's "vanilla" images the project builds and provides for supported hardware and devices.
Previous OpenWrt releases at least as far back as 21.02 could be equipped with the same degree of ASU support by installing a single package (luci-app-attendedsysupgrade) and its dependencies.
What switches are you running it on?
Netgear GS308T, Netgear GS108T v3, and HPE JG925A.
> How about OPNSense on open hardware of your choice
Yes, it's a possibility, but if you want to tinker, I think a plain Linux distro like Debian is better. Turning it into a router is literally a couple of kernel parameters and a few iptables rules to set up NAT. Nowadays that's less than fives minutes of work with Claude.
This buys you much better performance and hardware compatibility relative to a BSD system, as well as lower resource usage and attack surface (no GUI or other unnecessary additions). WiFi support on BSD is bad, but on Linux you can use hostapd and almost immediately get an access point for free. And of course Linux is also better if you intend to run other stuff on the same hardware.
Please support your claim about (networking) performance of BSD-based systems and Linux with some source(s). It surprised me. Thank you.
Anecdotal, but I was never able to get FreeBSD (TrueNAS CORE, OpnSense) to get anywhere near 10 Gbps. Using Mellanox or Intel NICs. The Linux equivalents (TrueNAS SCALE, OpenWrt) handled it fine on the same hardware.
But what if you don't want to tinker? I switched to OPNsense as a direct replacement for our Asus "WiFi routers", and it has been phenomenal, reliable, and does everything needed - when you just want it to work, it really just does. But when you want more advanced functions, there are tons of plugins and stuff that you can run natively, while still having a true CLI.
I suppose it comes down to what you said - "if you intend to run other stuff on the same hardware." Is it a good idea to run all sorts of extra stuff on your literal firewall/router? And if you did, I'd assume using a hypervisor is safer anyway? That way you can have the GUI and reliability of OPNsense but have a Linux distro beside it.
You also said that Linux has much better performance vs BSD, which seems rather far fetched. Got any data for that?
One other thing: OPNsense comes with a ton of helpful rules to eliminate bot traffic, allow IPv6, different NATs, VLANS, etc which you'd have to add manually. Not the end of the world, but worth considering.
> Is it a good idea to run all sorts of extra stuff on your literal firewall/router?
I don't see any reason not to. I run dozens on services, both bare metal and containerized (Podman) on my router/firewall. It doubles as an all-purpose home server with plenty of headroom to spare. It's just a computer that sits at the edge of my network, and running services meant to be exposed to the Internet on it is natural.
> You also said that Linux has much better performance vs BSD, which seems rather far fetched. Got any data for that?
I should have worded this more carefully. What tends to happen is that BSD has worse (or no) drivers, that's when BSD's performance can noticeably degrade vs. Linux. From memory, people online were reporting issues with Realtek chips. With Intel NICs, the routing performance should be broadly equivalent .
Some people want a webinterface tho.
Thats more or less what I did, and nix just made sense for the job. For 99% of people I'd say no its not worth it to tinker, just go with opnsense virtualized so you get at least some the benefits of the better linux drivers. By that I mean NBASE-T on various intel chips and while intel's igb is fairly solid on unixes many other vendors' drivers are less so. However if you're willing to figure out configuring per your needs you definitely can get a lower latency router with all the same capabilities and more, with it's components more sanely isolated via containers.
i don't understand the use case for 1x2.5gbs + 1x1gbs for a router. Why not both 2.5gbs it's not like you'll be running lots of stuff on the router itself so it would be more useful to have a wan AND a lan at 2.5 (outside of load balancing for a lot of wifi devices of course)
I have similar board, from same producer - BananaPi BPI-R3 Mini.
Got cheaper version without case - designed simple box with cutouts and 3d printed it out.
I can say it works flawlessly. I'm very happy about it. Previously I was using openwrt on different consumer-grade routers and I always had some issues, even when selecting supported devices.
$106usd or $84usd without a case and antennas. That’s a solid price. Wish it had more than 1gb ram - goddamn datacenters.
Without performing any work at all to optimize RAM usage: My all-singing, all-dancing OpenWRT router projects have always used less than 100MB of RAM. These days, they usually occupy less than 64MB.
1GB is a ton of RAM for this kind of application. :) What do you anticipate needing more for?
Blocky/pihole type stuff is what comes to front of mind, which probably fits but add anything else and now you’re hitting the limit.
How many blocklists does one need to have in order for them to use >900 megabytes of RAM?
I mean, there is a lot of spam on the internet.
But pihole asks for a pi with 512gb ram. Add openwrt’s 100mb, now your ram budget for running something else (file server, pairdrop, irc, tailscale, etc) is <400mb. One nodejs app could use all of that.
I mean:
On one hand, 400MB is still a ton of RAM to do useful work with -- and unused RAM is wasted RAM.
On the other hand: How many non-routing tasks do you really expect or require a ~$100 home-router-device to perform? :)
Require? Just OpenWRT + blocky. Expect? As much as I can get! I run a full home server, but if I could offload some things to a router, i would.
Not even for the OS itself, but for the freaking wifi send / receive windows...
I have and love my OpenWrt One for my main router. I have two, so that I have a backup one I can switch to if the first one ever dies. It is the best device to run OpenWrt on as it is fully supported hardware that has great images/packages for it. Routing speeds/buffer/latency are great, everything just works, price is very reasonable.
I don't use it for my APs, but that is mostly because I already had 3 TP-Link routers setup as dumb APs using OpenWrt that have been working great. If I did it again, I'd buy OpenWrt Ones though. Although Deco mesh kits I've used have worked exceptionally well, and have become my recommendation for friends/family that don't want to do things like run arbitrary packages on their router/APs.
another happy user here too. having openwrt with all features working and no tinkering out of the box (since it's their reference target) is a dream. this, plus the warm fuzzies of buying open-source, makes it worth it to me even with the 1GBps limitation and outdated WiFi (i use a separate AP anyway, like you). i swapped my ports in software to have 1GB WAN and 2.5GB LAN (which also lets me power the router with PoE, which i have coming in on the LAN port).
It's got only one LAN ethernet port. Do you use only wifi or did you add a switch between the router and your devices?
I have it going into a 16-port gigabit switch, which my OpenWrt-based APs and wired devices are hooked into.
I have 4 OpenWRT Ones and they are good.
If you just want a good WiFi router or access point, unless you need something cannot do (e.g. WiFi 7 or 10 Gbit/s Ethernet), and if want to spend minimal time messing around with routers today and in the future, just get this one.
After getting this, I see no reason to ever buy any closed-source router again.
No need to learn/remember any other Router config either. It's just all OpenWRT, always looks the same, always works the same. Setting up a new one takes me 2 minutes max.
The recent OpenWRT update also brought the one feature the project was most sorely missing: A simple "Download and install latest firmware" button in the device UI.
Now they just need to add an unattended-upgrades option and I never have to log in again after initial setup.
If someone is interested in open hardware running openwrt, also check out Turris.
https://www.turris.com/en/
I have been using them for years and I'm really happy. I recently bought the WiFi 6 upgrade kit for both of my turris. They "recently" released their latest version which is expensive but comes with WiFi 7 and 2.5 Gbps RJ45 and 10 Gbps SPF.
Since we're talking WiFi, I'll mention
https://www.wiisfi.com/
The single best wifi reference I've found to date.
This is awesome, thank you for sharing.
This answers so many questions that I had but never had the true desire to research every niche WiFi term.
I've recently been test driving SPR[1] which is a security-oriented distro for Wi-Fi routers. The team behind it are serious about Wi-Fi security and have a research lab[2] that has been credited with several CVEs in the likes of Apple's network stack. The headline feature is strong device isolation for semi-trusted guest and home automation devices, and the software stack is based around containerized and audited Go daemons.
It ran pretty well for me as a travel router I cobbled together from a Raspberry Pi and Netgear A7500 USB dongle for a stay in a short-term rental where the infrastructure network was shared with other units. More recently I have been trialing their CM5-based model with Wi-Fi 7 and 2.5GbE PoE for use as primary home Wi-Fi.
1: https://www.supernetworks.org 2: https://www.supernetworks.org/security-labs.html
I switched from a Google Wifi to this and found it to be just as stable, but with better range/signal strength, and easier to apply the parental controls I want.
Does it have parental controls natively or did you have to install something extra?
I would love to be able to whitelist which devices are allowed to access the internet during night time hours.
I do it the opposite way, disable my kids' devices at night, but I suspect your desired method would also be supported using native features. I have found LLMs to be very helpful in providing the right settings.
There is a plugin marketplace that provides more features, like ad-blocking. I haven't played with those yet, so I cannot vouch for them.
What I want is a cheap, brainless wifi 6 or 7 device, that's easy to mesh and extend. I currently have Orbis that are Wifi-5, and I'm sick of the general untrustworthiness. The thing is, like my WRT54G from back in the day, they just work in all situations. It's amazing how stable this kit has been for 7 years.
I'm not in the US. I can't Amazon. I don't want to spend the equivalent of 1k USD just to get 3 devices in a brainless mesh that covers my ~125sqm place made of insane amounts of radio blocking concrete.
I no longer want to maintain my network, to have network. I want things that just work. The Orbi did that for me - but the costing makes me think abut the future, and not finding a painless solution. I guess that's the tradeoff.
To wit, I also want the mesh comms on another channel (i.e 6Ghz, rather then the 2.4/5.0) and the computers/IOTs/etc isolated from that. Perhaps cheap tri-band is insanely wishful thinking. It sure seems that way.
I use opnsense with an aliexpress n100 router. It works very well and I enjoy it. But upgrades scare the crap out of me. I've only had 1 upgrade where things went bad. I have zfs snapshots and everything, but just because its a headless unit, I get super anxiety upgrading the system waiting for the beeps for it to come back online.
How much did you pay for it?
$200.93 shipped in April 2024 with a wall bracket. No storage or ram though, but it was cheap at the time.
I virtualize mine with proxmox and pass the necessary NICs through to the VM. Snapshots make rollbacks a breeze. You can also run HA pairs if you want to do updates without downtime.
Bpi does a nice job on open source hardware and openwrt integration, but only 2 ports in this model is really good?
Just update my bpi-r3 purchased years ago from 24.x to 25.12, it need a bit of extra work because the sfp interface is renamed(though I never used that), and i finally just do plain sysupgrade only and add some required packages because i run a customized fd.io vpp build on the bpi-r3, connected via vhost-net backed tuntap, it works well, I never regret bought this machine, with current agent stuffs, i think bpi r3 model may be more fun to play with
Just two Ethernet ports (1+2.5GbE), and it’s dual-band (no 6GHz)… I’m not sure who’s the target audience or what’s the use case.
It’s for developers as far as I understand, it’s not meant to buy as a consumer router. There is far better hardware you can get that runs OpenWRT.
It does raise the question that if it is for developers, what exactly is being developed? Especially if its not representative of hardware that is available or desired; is there some advantage targeting a very particular chipset? This seems to be the only device using it (from what i could find briefly)
A 5 port 2.5GbE switch would upgrade this to 5 total ports (4x 2.5GbE), and costs less than $100. If you only need 1GbE then it's even cheaper.
Outside of home-labs, it's rare for me to see any devices connected to the LAN side of a wireless router these days, and more than 1 (i.e. the non-portable device that is closest to the router) is exceedingly rare.
>Outside of home-labs, it's rare for me to see any devices connected to the LAN side of a wireless router these days
I would assume every gaming desktop computer would be? I actually assumed every desktop would be...
Neither my parents nor my wife's parents have their desktop connected to their router. The cable modem isn't even in the same room as the desktop.
[edit]
If it matters, my mom no longer has a desktop (she uses a docked laptop now), but it is true of the docking station and was true of her previous desktop.
2.5 is the WAN, so your lan is only getting 1g anyway
With OpenWRT, the network interfaces are whatever you want them to be. That's one of the perks. :)
Ok, so your wan is 1gig, and your lan 2.5...handy but not much of a perk. Lets just call this an AP, which would clear up many issues people seem to have with it
It depends.
It'd be handy for me: The fastest WAN pipe I can get is less than a gigabit while my LAN is still gigabit. I can't be the only person with this situation, wherein: If anything, then the 2.5-gig port is overkill.
I don't have any direct interest in the wifi radio that the box includes (I already have Mikrotik APs that I like just fine), except to configure it as a failover station-mode interface to use with a phone hotspot when the DOCSIS connection is on the fritz.
...which doesn't happen often at all, but it's annoying when it does happen. It's nice to be able to work around problems like that with OpenWRT.
Chaining a switch off the gateway is the best way to do it anyway. If you do that, then when you reboot your gateway, your lan devices do not lose their physical link and can continue talking to each other.
That's enough connectivity for a gigabit WAN pipe and a LAN full of stuff (including one or more better/faster APs), if a person wants to slice it up that way.
I'd like something bigger than 1gb for the LAN port...yeah, most of my stuff is quick from a wifi standpoint, but I don't want the hardwired stuff to be second class citizens.
Looking at one of these when TP-link stops patching my Wifi. https://www.toptonpc.com/product/2x10g-sfp-solid-firewall-mi....
Does it have hardware PPPoE offloading? Because it's a huge issue for those of us stuck with old-school telecoms for our fibre connections. Doing PPPoE at gigabit speeds needs something that can handle it.
Have you looked at the BPI-R4? It's a pricier option than the OpenWRT One, but it has excellent hw acceleration for networking tasks. I am 90% sure I recall someone reporting using it for a 2.5Gbps PPPoE connection and it handled it. It's also supported by the OpenMPTCProuter project if you need network aggregation support.
While they're okay on paper, they managed to burn a lot of customers in regards to the add-on wifi 7 module.
Quite a few of the modules went out without their eeprom programmed correctly, and all of them appear to be plagued by quite mediocre performance from (alleged) inadequate RF shielding on the card.
It looks like they've revised the design, but it's peeved off quite a few of the Banana Pi/Sinovoip customers who have bought them with the intention of using it as a router/ap with their R4. (It's dual-PCIe fingers with unique spacing and requires out of spec PCIe voltages, so it's only practically usable with a BPi R4.)
At the very least, the customers using them as wired-only routers are likely to be having a slightly better experience.
Yes. Mediatek SoCs have hardware acceleration support for that.
https://github.com/torvalds/linux/blob/master/drivers/net/et...
This is the official shop page afaict: https://www.bpi-shop.com/products/banana-pi-openwrt-one-rout...
Been using Openwrt for years, router and access points. Any new AP I buy has to has to be supported by Openwrt. I have probably bought 7 or 8 old AP's at yard sales over the years for $5/$10 and turned them into decent wifi access points back to an Openwrt router running vlans, bandwidth monitoring, QoS, tcpdump for ssh dump with wireshark.
Just excellent, some pretty smart people in the Openwrt forum as well if you have problems.
If you're remotely interested in this stuff you should go ahead and set all this up yourself on a debian or such. Great learning experience.
As someone who knows very little about WiFi, I always thought it sucked that if you wanted to go from 802.11this to 802.11that, it always requires brand new hardware with a different WiFi chip that implemented the new standard. Is there a good reason that software-defined 802.11 doesn't exist and that every new standard requires a different radio+SoC?
One example is the introduction of MIMO, a technique to send multiple data streams in the same frequency band in parallel. This requires multiple antennas, i.e. hardware which wasn't there in the previous wifi version. Note this was 2009.
Radio modulation / coding (at least for 802.11) benefits greatly from paralelism (lots of matrix multiplication etc).
I imagine that using an ASIC is way more cost efficient vs using a CPU.
I have one of these and love it, especially after I once bricked it during a manual software update and got to use the dip switch reset to reflash it using the ROM.
I wish it had more ethernet ports but I've managed to live with that. I'd be up for buying an OpenWrt Two as a backup or to replace this if it has even one more LAN jack.
Flashed my TP-LINK router last week with openwrt bin file, to get wifi to ethernet bridge. Working awesome.
There are too many settings. Feels like Gimp (vs paintbrush). But once figured out, it works well.
I started down my “custom” home network journey with OpenWrt and some aftermarket hardware routers. Enjoyed my time using / playing with it.
After some time though, I eventually moved over to using OpenBSD directly. My small brain has a much better understanding of all the moving parts compared to that of OpenWrt :P
I’ve been running one for a while and love it. Have also built a Nix -> OpenWRT config language transpiler so that I can keep my router state in Nix files and have nice deterministic rollbacks etc. It’s been great!
I became interested in OpenWrt when I noticed that the cloud portal for my ISP reported to me the names and types of devices that were associated to my home access point/router.
Suddenly I want to put every IPS device into dumb bridge mode, and run my own damn router.
I have an ASUS RT-AX53U running OpenWRT for a few years now.
I bought it, turned it on and flashed it out of the box lmao
I like the idea of having their own hardware but it is not easy to get, so buying normal ones and flashing them still the best option.
I once tried to flash OpenWRT onto a very old router that was EOL, and apparently wherever I had downloaded the firmware it was pre-pwned. There was literally an interface numbered “eth666” and the CLI was playing crazy mind games with me as I tried to configure it.
So I thoroughly destroyed the router and e-wasted it. I honestly never want an open-source router OS ever again.
What does it take to add support for a router? Support for tp-link ax5400 seems missing for a long time.
Where do you buy the open wrt one? EU
I have one of these for a few months now. Works like a champ. Firmware updates are very easy now through the web interface.
I purchased one of these for a family member and it has been great.
Are these for sale in the US?
Some previous discussion around the launch in 2024:
https://news.ycombinator.com/item?id=42285689
Next step: open source hardware ASIC for the open router ?
gl·iNet also runs openWRT. You can even ssh into these routers.
There is definitely beauty in having a separate router device that chugs on just fine regardless what happens to the rest of your network. But I got bored with the constantly-churning embedded culture, bespoke OS's (sorry, OpenWRT), and VPNs generally want more CPU than what purpose-built "routers" have. So I just went back to the old way of using a plain Linux machine as the gateway (now virtualized, with NixOS and nftables) and couldn't be happier. WiFi AP is done by that same physical machine (not virtualized) and by two other amd64 machines that double as Kodi boxes. When you learn netfilter/iproute2, that experience carries to anything else you might switch to.
Did anyone, ever do a security evaluation of the most common network chips on the market? How much would it cost, are there even companies doing this?
madwifi lives on...
where to buy?
Gigabit / 2.5 Gbit connectivity is already obsolete. Any modern product must have 10gbe WAN with the hardware to back up NAT at that throughput.
It's not obsolete, it's basically the contemporary baseline. Remember, this is a cheap device. And unlike most Chinese garbage, you can be reasonably certain that it isn't backdoored.
For a home network 2.5 is plenty.
https://frontier.com/shop/internet/fiber-internet/7-gig
7gig WAN is widely available residentially at a reasonable cost.
it most certainly is NOT available at a reasonable cost in many places.
200 horsepower vehicles are also obsolete. Any modern car should have 500 horsepower with the braking and safety systems to back it up.
That's such an assumption of needs. For someone to be using a 10g capable NAT would be some sort of super nerd with such a small portion of the population. More and more households have no compute device other than their devices. With WiFi routers now being sold with OpenWRT installed, it no longer means you're in the nerd category for installing it.