12 ms·
List of 8000 security vulnerabilities in 1200 Wi-Fi routers
- politelemon 5y agoI couldn't find the ever popular Asus RT Ac86u in that list, or the X version. Would it be converted under "asuswrt"?
- clb92 5y agoSure, 1200 routers. Except someone listed single Synology applications as router models for some reason. Synology only has a couple routers, not 32 different models. If the quality of the rest of the data is similar, this list isn't very useful.
- daneel_w 5y ago8000 reasons to go for an OpenWRT or DD-WRT compatible router.
- DangitBobby 5y agoMy experience with DD-WRT is that it took tons of tweaks and fiddling to get the same router to perform not quite as well as it did pre-flash, including manually tweaking the power outputs of the antenna to get a signal one room away. Recently I bought a new router with failover support and it's so nice that the basic (and advanced) things I want to use just work. Clean UI, sensible organization of settings, actually reliable documentation... all wonderful things. So unless you are okay with lots of fiddling and restarting your router 10-30 times over a weekend, maybe don't use DD-WRT. I have no experience with the others.
- daneel_w 5y agoPerhaps your router just wasn't quite the fit for it. "Lots of fiddling and restarting your router 10-30 times over a weekend" sounds nothing at all like my experience with OpenWRT and DD-WRT.
- lostmsu 5y agoI don't see this to be too useful. Many vulnerabilities listed on this web site are for outdated firmware versions.
- jvolkman 5y agoHow many people actually update their router firmware?
- lostmsu 5y agoHow many people who would go to that web site do not update their router firmware?
- jvolkman 5y agoFair. But if the site doesn't list vulnerabilities in older firmware, then someone stumbling upon it that hasn't kept their router up to date won't see their actual vulnerabilities listed.
- RealStickman_ 5y agoThis list is still useless in that case, as it doesn't list the affected firmware versions as far as I can see.
- ndata 5y agoOP here: There is a misunderstanding. The affected firmware versions are always shown in the description text. Its just that in some cases we are not able to show them at the "header" level due to parsing failures
- dtaht 5y agofiled with the fcc, 6 years back: http://www.taht.net/~d/fcc_saner_software_practices.pdf http://www.taht.net/~d/fcc_saner_software_practices.pdf
- Kikawala 5y agoReminds me of routerpwn.com
- SavantIdiot 5y agoInteresting. a) How does someone compile this and keep it current? FTWP: "17,000 routers per month" ... ? That's ... daunting. b) Was Ubiquiti UniFi (or brand ___) excluded because their routers have no vulnerabilities or because they weren't tested?
- capableweb 5y agoA) Shortly: Automation. Long: "Every month, We evaluate 17000 routers for security Vulnerabilities using the national vulnerability database and publish the list with the remediation steps" from the website
- pixl97 5y agoNot sure, but if you go to the main site, they do list instructions for Ubiquiti equipment so they seem to know about them.
- mbesto 5y agoHow has someone not made a commercially available open source hardware router and just load it up with DD-WRT or Tomato?
- mikeyschaefer 5y agoThese come pre installed with Openwrt. I haven’t tried any of their products though. https://www.gl-inet.com/products/ https://www.gl-inet.com/products/
- zikduruqe 5y agoThese are my next to investigate if my current Eero network gets replaced. The ability to put Wireguard on the router and not behind it, is the thing I need.
- genewitch 5y agoWhy would it matter where wireguard is? The gl.inet does have the nifty internet kill switch which forces everything through VPN, which is useful I guess.
- funnyflamigo 5y agoLinksys does! They have a series of routers designed to support OpenWRT (which IMO is better then DD-WRT but preferences of course). If it supports OpenWRT then others shouldn't be difficult to load on it either. https://openwrt.org/toh/linksys/wrt_ac_series https://openwrt.org/toh/linksys/wrt_ac_series I've had a decent experience with OpenWRT on a WRT1200AC EDIT: I haven't used it for actual wifi (just routing/switching) in a few years so I don't know how good they are nowadays. EDIT 2: OP asked for open source hardware, not hardware that runs open source firmware - my bad!
- silasdavis 5y agoAlso very happy with openwrt on this device. Really quite a decent gui tui and config. Setting up always on open vpn and wireguard was reasonably painless and works well.
- gennarro 5y agoThis is pretty cool. What dataset is this based on?
- Jon_Lowtek 5y agoThe vulnerability list seems to be CVE data filtered for routers, grouped by vendors.
- gennarro 5y agoAh yes. Makes sense. Thank you
- aliswe 5y agoI am more amazed that there are so many routers than the number of vulnerabilities!
- jaclaz 5y agoWell, some are not even routers: https://modemly.com/m1/brother/model-nc-3100h https://modemly.com/m1/brother/model-nc-3100h
- hk1337 5y agoInteresting, I don't see arris (ATT gateway) on the list. Not sure if that's because there's no vulnerabilities or not as well known?
- willis936 5y agoHoly smokes. The router I used for many years has 79 listed vulnerabilities. I got a used Ruckus earlier this year and it's been great.
- fairytale 5y agoJust wait and see till all those thirsty script kiddies start abusing these even more now. On a side note, those who write router software like this need to step up their security and stop being lazy. Seriously.
- fencepost 5y agoInteresting to not see Mikrotik on the list, though I'm not sure how far back you'd need to go to find hardware that's not still receiving firmware updates - certainly well over 10 years.
- paulmd 5y agoI also don't see Buffalo. The question is whether they're not vulnerable, or whether they were simply not tested (being smaller brands).
- chasil 5y agoI have a Ubee cable modem with integrated wireless, and this manufacturer is not on the list either. It would also be helpful to see how many vulnerabilities are in the latest release of Gargoyle. I have heard that the best countermeasure for router vendor abandonware is to avoid the 192.168 network entirely, so I configured mine on a random 10. subnet.
- jve 5y agoYeah, incomplete list. Ubiquiti is missing too. Not that Mikrotik haven't had CVE https://www.cvedetails.com/product/23641/Mikrotik-Routeros.html?vendor_id=12508 https://www.cvedetails.com/product/23641/Mikrotik-Routeros.h...
- PaulKeeble 5y agoA lot of these routers have DD-WRT, OpenWRT, FreshTomato or maybe pfsense support. Since the manufacturers long ago abandoned security updates and feature upgrades an open source firmware will vastly improve the security and the devices functionality. Not all routers can run one of these firmwares but many can and I wouldn't choose a device that didn't in the future. Its relatively easy to setup a basic secure home router using a Raspberry pi 4 and USB Ethernet and then attach one to a hub and the other to the modem and you have a 1 gbit/s capable routing device that can do SQM and remove bufferbloat and not a lot of consumer routers can remotely achieve that level of performance. It is more hassle than the manufacturers firmware, but its also a surprisingly good way to extend a routers usable life and functionality as well. VPNs, Virtual LAN, File and web servers or just better QoS you can do just about anything you might want.
- paulmd 5y agoThe lack of ongoing support from device manufacturers is really awful. There were some major UPnP vulnerabilities (last year, as well as some previous ones iirc) and a parade of attacks against WPA of various levels and very few devices ever get patched for them - including high-spec devices. Running open-source firmware is basically necessary to have any chance against all these attacks, because manufacturers simply won't do the work. There really really needs to be some regulation on this, internet of things devices as well. Give a defined minimum software update lifespan on the box at time of purchase and require that it be at least 3 years from the date of sale, for example.
- throw0101a 5y ago> The lack of ongoing support from device manufacturers is really awful. One of the reasons why I specifically when with Asus: they've got pretty good long-term support. There's also third-party firmware that uses the open source nature of most/all of the code (e.g., Asuswrt-Merlin).
- 3np 5y agoMaybe I just had bad luck but my two most recent Asus Wi-Fi routers are dead, unsupported, and no Linux/OpenWRT.
- lend000 5y agoIs there a way to filter these by remotely exploitable? Things that can be compromised locally just seem like the cost of doing business at this point (for non-business use, anyway).
- netizen-936824 5y agoFiltering by "not bullshit" or "patched years ago" would be a better start
- unfocused 5y agoI had no idea Huawei produced so many routers.
- dreamcompiler 5y agoMe neither, and the apparent fact that so many are out there being used chills me to the bone.
- ghostly_s 5y agoEvery exploit listed for my router (tplink Archer A7) has no "affected version" listed nor indications if it has been patched, but clicking through to the CVEs indicates all have been remediated. As near as I can tell this website is just scraping CVEs, poorly, in order to sell their security services (which consist, at least in part, of an email reminder to rotate your router password...seriously?).
- wnevets 5y agoso spam?
- ndata 5y agoOP here, Yes, I agree, we (modemly) could do better 1. Show the vulnerability status (patched / open) 2. Show affected firmware versions 3. Display manufacturer's last patch release date Though lots of the dataset is clean, still we do lots of parsing and regexing to extract insights out of a massive haystack. The intention of this tool is for everyone to realize and keep their firmware updated And No, we don't sell any services. The security reminder emails are free
- ndata 5y agoOP here. To clarify. The affected firmware versions are always available in the detailed description text. Its just that we couldn't show them in the header due to parsing failures in few cases
- chasil 5y agoA few questions. I am guessing that very few of these can be exploited on the side that faces the public internet. True? What percentage require access to the router via administrative login on the private network? How many would need nothing but a hostile .exe on the private network? How many could achieve an exploit with nothing but hostile JavaScript on the private network? Will changing the private network away from 192.168.* to something on 10.* protect a vulnerable router from some of these exploits? What else can be done to protect a vulnerable router?
- xedrac 5y agoSo 80% of these vulnerabilities are on Netgear routers, and nearly all of them are rated as High severity. That's really impressive. I don't think I'll buy a Netgear router ever again.
- chasil 5y agoIf the router allows 3rd-party firmware and well-maintained ROMs are available, why avoid the hardware (unless you don't want to apply the upgrades)? Cable modems typically do not allow this; only the cable provider is able to apply oem firmware updates.
- proactivesvcs 5y agoSkirting a discussion on how relatively good or bad Netgear are, the results seem to be vague as to whether they're resolved, how bad each vulnerability is, and it seems to list a device for each firmware version. I don't think the front-page numbers are necessarily particularly helpful.
- outsomnia 5y agoYou're right that the bulk of them will have been resolved... but... given the sheer continuous volume of them for Netgear, does that actually help? It seems there will always be enough undiscovered or unresolved vulns "in flight" to compromise it on demand.
- emkoemko 5y agodevice makers should be forced to support their devices and if they don't they must have something like 6 month period where if they don't push a security check flag to their devices they initiate code to nag the user telling them this devices is not secure anymore because manufacture is not supporting it anymore, in this case they should also be forced to release way to load 3rd party code etc to allow others to fix their crap. This is a serious issue because many people use old devices without knowing anything is wrong.
- Flameancer 5y agoShouldn’t that be up to the user to update their device or replace as they see fit? Also the manufacturer has no obligation to open up a device for 3rd party software installs. At the most they should let their users know of the expected life cycle of the devices and warn them even the life cycle it’s end or has ended but in no way nag the user. It would be nice for the manufacturer to allow 3rd party software but they don’t have too and shouldn’t be forced to do so. You as a consumer have the choice to choose a manufacturer that allows you to install 3rd party software.
- Namidairo 5y agoI already have little hope for consumer networking equipment, this just seems like a big old list of scraped CVE's. One has to remember that the majority of the development ends up being by the SoC vendor, usually a horribly out of date fork of OpenWrt with weird looking proprietary kernel modules to support wifi, accelerated nat, etc. Quite a few of the older devices lack some pretty basic mitigations as well; ASLR, Position Independent Executables, Stack Canaries, etc. Either they get forgotten or they're off because of they can't be bothered getting the drivers up to scratch. (Assuming they haven't just been handed a binary)
- tim333 5y agoI'm surprised my very cheap and crappy consumer router doesn't seem listed anywhere (Technicolor TG588 v2)
- david_draco 5y agoWhen will BrickerBot be reborn? The world needs you!
- zwieback 5y agoBack in the day we did an industry project with a customer with lots of early Symbol Wi-Fi access points and it had some weird non-IEEE 802.11-standard behaviors, some of the FW was probably written before 802.11 was done. WPA was added at some point during the project but kept crashing. I didn't see them on the list but of course for entirely different reasons, the business was bought by Motorola and then petered out, I think.
- harha 5y agoIt should be much simpler to make changes to the manufacturers software, since they all come with the GPL notice. It’s so silly things that don’t work, e.g. the guest network not really being separated. I’ve found it very hard to get a decent solution, especially given that I can’t really change much on the canceling, now considering getting a pfsense/opnsense router in front of a consumer mesh Wi-Fi, though that still isn’t ideal, won’t be able to really do much for the Wi-Fi devices. I think I’ve posted this once, what’s also missing out there is a guide for the home user to set up networking with typical scenarios, along with hardware recommendations (apart from companies producing better products, would consider paying the premium for ubiquity but here again it seems to require too mutant cables to be laid).
- 41b696ef1113 5y agoThis just further emphasizes to me that we need a microkernel OS for internet facing things. Followed with a memory safe or at least security audited network layer.
- zinekeller 5y agoHuh, some of the Huawei and Nokia vulnerabilities are not related to their routers at all (a Symbian bug was even included). Maybe for other brands it's easy because of their networking equipment specialty, but Huawei's and Nokia's list needs to be manually filtered to see only the router vulnerabilities rather than an Android or Symbian bug.
- dtaht 5y agoWe could use an 'upgrade in place': https://docs.google.com/document/d/1T21on7g1MqQZoK91epUdxLYFGdtyLRgBat0VXoC9e3I/edit https://docs.google.com/document/d/1T21on7g1MqQZoK91epUdxLYF...
- chmod775 5y agoThat list is missing AVM/Fritz, which has at least some old CVEs: https://www.cvedetails.com/vulnerability-list/vendor_id-435/AVM.html https://www.cvedetails.com/vulnerability-list/vendor_id-435/....