[Cerowrt-devel] using an alternate tftp server
Hi cerowrt-devel, I needed to tell dnsmasq to use an alternate tftp server and the ability to do so isn't exposed through the web UI, so I figured out how to do it. Here is a description, if it looks OK maybe it can go in the FAQ? === Q: How do I use an alternate tftp boot server? A: The web configuration interface lets you enable a local TFTP server and set its root and boot file name, but if you just want the DHCP response to supply a different path/file/server do the following: * edit /etc/config/dhcp and in the "config dnsmasq" section add something like "option dhcp_boot 'tag:se00,pxelinux.0,,172.30.42.5'" where 'se00' is the interface, 'pxelinux.0' the path/file, an empty server hostname, and IP address of the next-server. * run '/etc/init.d/dnsmasq restart' * if you are debugging and want to check that the line you added resulted in the changes you want, look at /var/etc/dnsmasq.conf ======= Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] FAQ fixes
Hi, The cerowrt FAQ is kind of stale, here are some things that need to be fixed http://www.bufferbloat.net/projects/cerowrt/wiki/FAQ#What-about-QoSAQM Link to 'Setting up AQM', page says it's deprecated, so FAQ Q/A probably needs to be adjusted http://www.bufferbloat.net/projects/cerowrt/wiki/FAQ#Why-so-many-Interfaces- 148 'guest' and 'babel' links are 403 maybe babel could point to http://www.bufferbloat.net/projects/cerowrt/wiki/ Babel_SSID 'internal QOS' is 403, the answer probably needs to be adjusted http://www.bufferbloat.net/projects/cerowrt/wiki/FAQ#Why-so-many-SSIDs-on-th e-wireless-interfaces 'CeroWrt router configuration' points to http://www.bufferbloat.net/projects/cerowrt/wiki/CeroWrt_router_configurat ion which has a note which refers to 'Setting up AQM' which is deprecate (listed above) -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] Glossary wiki page
Hi cerowrt-devel, The Glossary at http://www.bufferbloat.net/projects/bloat/wiki/Glossary is pretty sparse and needs some editing, here are some suggestions: Probably thanks to the effort of people on this list, Wikipedia (now) has a bunch of good pages, lists of terminology, queuing disciplines, bufferbloat, etc I think it's still useful to have some definitions in the Glossary, but maybe it can now refer to wikipedia pages for a lot of things (and if needed more things could be added to wikipedia). Good general jumping off point https://en.wikipedia.org/wiki/Network_scheduler The wikipedia 'network performance' category, lists lots of things that might be worth describing https://en.wikipedia.org/wiki/Category:Network_performance In particular, here are links from things that the glossary already mentions or might want to https://en.wikipedia.org/wiki/Throughput https://en.wikipedia.org/wiki/Throughput#Channel_utilization_and_efficiency https://en.wikipedia.org/wiki/Quality_of_service https://en.wikipedia.org/wiki/Queueing_theory https://en.wikipedia.org/wiki/Rate_limiting https://en.wikipedia.org/wiki/Goodput https://en.wikipedia.org/wiki/Explicit_Congestion_Notification https://en.wikipedia.org/wiki/CoDel https://en.wikipedia.org/wiki/Bufferbloat Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] using an alternate tftp server
Matt Taggart writes: > Hi cerowrt-devel, > > I needed to tell dnsmasq to use an alternate tftp server and the ability to > do so isn't exposed through the web UI, so I figured out how to do it. Here > is a description, if it looks OK maybe it can go in the FAQ? > > === > Q: How do I use an alternate tftp boot server? > > A: The web configuration interface lets you enable a local TFTP server > and set its root and boot file name, but if you just want the DHCP > response to supply a different path/file/server do the following: > * edit /etc/config/dhcp and in the "config dnsmasq" section add > something like > "option dhcp_boot 'tag:se00,pxelinux.0,,172.30.42.5'" > where 'se00' is the interface, 'pxelinux.0' the path/file, an empty > server hostname, and IP address of the next-server. > * run '/etc/init.d/dnsmasq restart' > * if you are debugging and want to check that the line you added > resulted in the changes you want, look at /var/etc/dnsmasq.conf > > === I discovered that making changes via the web ui ends up overwriting any by hand changes to this option :( I haven't investigated why yet (or how I might prevent that or expose this functionality via the ui). -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] FAQ fixes
Jim Gettys writes: > There are good reasons it's a wiki. Please go ahead and fix problems you > see. Ask on the list if you aren't sure of the answers... I don't appear to have the right redmine permissions and I don't see an edit link (and adding "/edit" to the url doesn't work). My redmine account is 'taggart'. Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] default zones including interfaces and babel
Hi cerowrt-devel, In a default 3.10.50-1 install on the Network->Firewall->General settings page, there are some default zones for wan, lan, guest. They don't appear to have any interfaces assigned to them, I am guessing the intent is: ge00: wired wan port, should be 'wan' zone gw00, gw10: guest 2.4/5 wireless, should be 'guest' zone se00: wired switch ports, should be 'lan' zone sw00, sw10: secure 2.4/5 wireless, ? zone gw01, gw11: babel 2.4/5 wireless, ? zone 0) shouldn't the interfaces be assigned to zones? 1) If the intent is that se00, sw00, sw10 can all communicate freely, maybe the zone name should be 'private' or 'secure' (rather than 'lan') and they should all be part of that? 2) What zone should the babel devices be in, what do they need to be able to do? This is maybe a good segway into some other questions I have: * is there a good description of how the babel stuff works? I found this http://www.bufferbloat.net/projects/cerowrt/wiki/Mesh which explains configuring, but I guess I would like something like a walk through of how a wireless client connects to an interior router and how things make it to the internet and back. * if I need to secure both my guest wireless and secure wireless networks, what does that mean for security of the babel networks and what (if anything) stops someone from using them? given that I haven't set credentials anywhere on the routers to make it work, I am guessing nothing. I think the last time I wirelessly connected two routers it was using the old Linksys WDS and it used credentials somehow... Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] more FAQ/wiki additions
Hi cerowrt-devel, The following are very helpful, they should have an entries in the FAQ or links on the main wiki page! I can add them if I get permissions, but also I list some questions and people should review them to make sure they are good to point people at. 'Setting up an interior gateway router' http://www.bufferbloat.net/projects/cerowrt/wiki/Setting_up_an_interior_gate way_router mostly up to date since jg edited recently 'Tuning your CeroWrt default gateway' http://www.bufferbloat.net/projects/cerowrt/wiki/Tuning_your_CeroWrt_default _gateway but * broken link to 'Setting up CeroWrt to bridge', I can't find that with the wiki search * The /etc/config/babel (sic) change is done by default in 3.10 I think, so maybe that can go away or be minimized? 'Enable ECN' http://www.bufferbloat.net/projects/cerowrt/wiki/Enable_ECN 'CeroWrt and BCP38' http://www.bufferbloat.net/projects/cerowrt/wiki/CeroWrt_and_BCP38 'Building Cerowrt on your own Linux machine' http://www.bufferbloat.net/projects/cerowrt/wiki/Building_Cerowrt_on_your_ow n_machine getting old, probably needs review 'Mesh Networking' http://www.bufferbloat.net/projects/cerowrt/wiki/Mesh from 2012, looks mostly correct but could use some updates/additions Thanks Dave, jg, etc for writing them! -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] fq_codel tuning on distros
Hi, http://www.bufferbloat.net/projects/codel/wiki/Best_practices_for_benchmarki ng_Codel_and_FQ_Codel#Tuning-fq_codel explains that the default packet limit of 1 is designed for 10GigE speeds and that for slower links it should be turned down. Is that still true? Looking at 3.16 source in fq_codel_init I see: sch->limit = 10*1024; q->flows_cnt = 1024; But I don't know what those correspond to. Are they sysfs tunable or only at compile time? If Linux distros are going to turn on fq_codel by default, are these reasonable values for the installed base (which I am assuming is mostly 1GigE)? What recommendations should the distro documentation make for tuning on various speeds? I'm excited for this to go into distros, what needs to be done to make that easier? Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] ping loss "considered harmful"
Dave Taht writes: > wow. It never registered to me that users might make a value judgement > based on the amount of ping loss, and in looking back in time, I can > think of multiple people that have said things based on their > perception that losing pings was bad, and that sqm-scripts was "worse > than something else because of it." This thread makes me realize that my standard method of measuring latency over time might have issues. I use smokeping http://oss.oetiker.ch/smokeping/ which is a really nice way of measuring and visualizing packet loss and variations in latency. I am using the default probe type which uses fping (ICMP http://www.fping.org/ ). It has been working well, I set it up for a site in advance of setting up SQM and then afterwards I can see the changes and determine if more tuning is needed. But if ICMP is having it's priority adjusted (up or down), then the results might not reflect the latency of other services. Fortunately the nice thing is that many other probe types exist http://oss.oetiker.ch/smokeping/probe/index.en.html So which probe types would be good to use for bufferbloat measurement? I guess the answer is "whatever is important to you", but I also suspect there is a set of things that ISPs are known to mess with. HTTP? But also maybe HTTPS in case they are doing some sort of transparent proxy? DNS? SIP? I suppose you could even do explicit checks for things like Netflix (but then it's easy to go off on a tangent of building a net neutrality observatory). On a somewhat related note, I was once using smokeping to measure a fiber link to a bandwidth provider and had it configured to ping the router IP on the other side of the link. In talking to one of their engineers, I learned that they deprioritize ICMP when talking _with_ their routers, so my measurement weren't valid. (I don't know if they deprioritize ICMP traffic going _through_ their routers) -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] DOCSIS 3+ recommendation?
Hi cerowrt-devel, My cable internet provider (Comcast) has been pestering me (monthly email and robocalls) to upgrade my cable modem to something newer. But I _like_ my current one (no wifi, battery backup) and it's been very stable and can handle the data rates I am paying for. But they are starting to roll out faster service plans and I guess it would be good to have that option (and eventually they will probably boost the speed of the plan I'm paying for). So... Any recommendations for cable modems that are known to be solid and less bufferbloated? I (like probably everyone on this list) will have router doing SQM/etc connected to the device, so that reduces the damage large buffers in it can do, but it would still be good to have something that designed well and to reward a vendor that's paying attention. My personal ideal is a simple device, cable-in gig ethernet out, and does not have wifi, usb, do NAT, etc. (that's what cerowrt on the router/AP is for). Are there DOCSIS 3.1 devices available yet? Or if those aren't available/affordable, maybe an inexpensive but good 3.0? Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] new Android feature
Android 5.1 apparently has a feature where it will remember wifi networks that didn't work very well and avoid them http://www.wired.com/2015/03/google-android-broken-wifi/ -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] failing to find the "declared victory" in a current wifi router
Rich Brown writes: > With that framework in mind, let me respond to your questions. > > TL;DR - if you just want to fix your home network today and get on with your > life, I recommend: > - OpenWrt Barrier Breaker (BB) release. As of July 2015, it's the stabl > e version. Stay away from CC or trunk, as they're still evolving. > - Install OpenWrt using the instructions at: http://wiki.openwrt.org/do > c/howto/installopenwrt > - Install SQM/fq_codel to solve bufferbloat using the instructions at: > http://wiki.openwrt.org/doc/howto/sqm > - What router to choose? I bought the TP-Link Archer C7 v2 for ~$90 (US > ). http://wiki.openwrt.org/toh/tp-link/tl-wdr7500 In a one-out-of-one test, i > t seems to work well with BB, SQM works fine, and I'm happy. This message made me realize I hadn't posted the CC+SQM HOWTO I wrote, maybe it will be useful, https://we.riseup.net/lackof/openwrt Feedback welcome. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] ar71xx CC builds and tc
Hi, I have installed a bunch of wndr3800's in the last few days with CC and noticed a couple things: 1) sometime around Aug 11 the ar71xx-generic builds jumped from having a date of mid july and using a 3.18 kernel, to having a Aug 11 date and 4.1.4 kernel. I guess builds weren't working and then got fixed? Anyway maybe a good time for people to test newer builds. 2) CC doesn't seem to include tc by default now, so when I installed luci-app-sqm (which pulls in sqm-scripts) things weren't working correctly. After I installed tc then it started working. Toke, maybe you need to have sqm-scripts start depending on tc? Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] google wifi
Google is working with TP-LINK (and soon ASUS) on wifi (is there a make-wifi-fast list this should have gone to?) Google Blog: Meet OnHub: a new router for a new way to Wi-Fi https://tinyurl.com/nloy3jm product page https://on.google.com/hub/ -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] google wifi
Matt Taggart writes: > Google is working with TP-LINK (and soon ASUS) on wifi (is there a > make-wifi-fast list this should have gone to?) > > Google Blog: Meet OnHub: a new router for a new way to Wi-Fi > https://tinyurl.com/nloy3jm > > product page > https://on.google.com/hub/ I talked to a friend that worked on it: the kernel for the Onhub router is under "whirlwind" project name in the chromium.org source tree. The firmware is coreboot and is also public. Openwrt has all the support for the Qualcom chipset but not this board. Openwrt also require fastboot and won't work with Coreboot. Key bits are the Device Tree description of the HW in this directory: https://chromium.googlesource.com/chromiumos/third_party/kernel/+/chromeos-3.14/arch/arm/boot/dts/ qcom-apq8084-mtp.dts qcom-apq8084.dtsi qcom-ipq8064-ap148.dts qcom-ipq8064-arkham.dts qcom-ipq8064-storm.dts qcom-ipq8064-thermal.dtsi qcom-ipq8064-v1.0.dtsi qcom-ipq8064-whirlwind-sp3.dts qcom-ipq8064-whirlwind-sp5.dts qcom-ipq8064.dtsi whirlwind-sp5 is what shipped. (AFAIK) btw, all of this was reviewed on a public chromium.org gerrit server. openwrt does support AP148 and at some point chromeos was booting on AP148 though I don't expect it to work "out of the box" -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] google wifi
David Lang writes: > how open is the wifi driver? Is it something that we can dive into and modify > for make-wifi-fast? or is it a typical vendor blob? more info.. ath10k ChromeOS uses tools to share/merge config files. The config parts specific to OnHub are here: https://chromium.googlesource.com/chromiumos/third_party/kernel/+/chromeos-3.14/chromeos/config/armel/chromiumos-ipq806x.flavour.config it's using a v3.18 kernel mac80211 code base. (USE="wireless318") -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] USB-JTAG pod?
Mike O'Dell writes: > anyone have any recommendations for a USB-JTAG goober? > a friend is try to un-brick some router boxes that > someone tried to upgrade and failed. he was going to > do a bit-boffer out a gen-u-ine parallel printer interface > but I offered to check out USB-JTAG things first. Someone pointed me at this page documenting a fully compatible cheap knock-off that works well http://www.gniibe.org/FST-01/q_and_a/swd-debugger.html The second link has bitrotted some, it should be http://blog.nodonogard.net/2015/11/compiling-and-flashing-gnuk-binaries-to.h tml (and requires turning on a bunch of javascript/referrers) There is also this list on openocd.org http://openocd.org/doc/html/Debug-Adapter-Hardware.html -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] archer c7v2 gets third party unupgradable firmware
dpr...@reed.com writes: > I'm giving a talk in a couple months at a very high level, about "what's > at stake" as we move into the era of "5G" (for lack of a better word, > this is what the media all think is happening, and what has the ear of > the FCC). > > I'd love to have a list of brands and models that have "gone dark" to > security improvements, bug fixing, and innovation - mainly just to > point at, implicitly shame the industry and its captured regulators, > etc. This will be a modest part of the talk, which has some other > well-docmented bombshells in it (like CG-NAT, for example, and the > predictable failure of "white spaces" and the CEO-driven, rather than > science-driven PCAST "spectrum sharing" that we are now experiencing). In addition to the "gone dark" concerns you mention, I think there are a couple larger issues in the "what's at stake" discussion that you might want to think about and include in your talk: 1) Just as we've seen cell phones all but replace "land lines"; smartphones, phablets, and tablets replace laptops and desktops; I think we are starting and will continue to see cell data replace home broadband. For the non-geek market, 4g/5g is more than fast enough (faster than a lot of DSL), already built-in and working, doesn't require setting up additional equipment(that sucks by default unless you are a geek that can fix it), already paid for, etc. For the average consumer, it's increasingly making more financial sense to just buy a capable phone/tablet with good data plan rather than deal with broadband, a laptop, etc. Most of my relatives seem to be going this route. This will result in further centralization of control of the internet. We're starting to see carriers in North America offering non-net-neural features like free data to certain sites (youtube, pandora, etc) but data caps for everything else. Many consumers will still be paying for Cable TV and have a need for their SmartTVs, gaming consoles, and IoT crap to connect to the internet. So the provided proprietary cable/DSL/fiber gateway will still have (crappy) wireless. And maybe they use that wireless on their phone/phablet/tablet, but the cell data will probably replace that soon. But they will no longer have a traditional "wifi router" as we have in the past. So it's not just _some_ models going dark, but _most_ will and the wifi router will become a geek niche market? Maybe the google and amazon premium wireless voice activated things will take over part of that niche? But most consumers may be giving up control of the network in their own house and won't be able to run something that properly solves bufferbloat/bad wifi/security problems/etc Maybe the OpenHardware SBC on Kickstarter world will be what geeks turn to to replace the commodity wifi router for running openwrt etc? Or the nuc offerings? But mainstream people are unlikely to do so. 2) Unrelated to wifi/routing, but it's getting increasingly difficult to run your own internet services. Running an SMTP server successfully now requires many hours a month of staying up on spamfighting tech, dealing with spammer attacks/phishing/poisoning/etc. Same with a web CMS. I think the freedombox and openwireless (cerowrt inspired) folks were insightful in realizing these problems were coming. But I don't know if we're any closer to solving them. dpreed, Who is your audience for this talk and what do you hope to achieve with your talk? -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] sourceforge speedtest
I don't remember seeing this SF.net speedtest reported here yet https://sourceforge.net/speedtest/ In my browsers the graphs didn't work and the details button was a little hard to click(aim for the bottom edge). I learned of it in this reddit thread about SF.net being acquired http://tinyurl.com/go6ag8j -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] Intel latency issue
http://www.theregister.co.uk/2016/12/03/intel_puma_chipset_firmware_fix/ "Modems powered by Intel's Puma 6 chipset that suffer from bursts of game-killing latency include the Arris Surfboard SB6190, the Hitron CGNV4, and the Compal CH7465-LG, and Puma 6-based modems rebadged by ISPs, such as Virgin Media's Superhub 3 and Comcast's top-end Xfinity boxes. There are other brands, such as Linksys and Cisco, that use the system-on-chip that may also be affected." -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] netperf.bufferbloat.net
Hi Cerowrt-devel (and mostly Rich), I was running some betterspeedtest.sh today using the default netperf.bufferbloat.net server (aka atl.richb-hanover.com) and only getting about 2mbit down and 5mbit up (at home, Comcast in Seattle). The broadband reports speedtest gives me 30mbit down/6mbit up (which is the level of comcast I have). Am I bottlenecked somewhere? Is there another server closer to Seattle I should be using? Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] WNDR3800 improvements?
Hi, I love the WNDR3800 platform, it's been great over the years first with cerowrt and then openwrt. Of the many I've deployed I have only had hardware problems with 2 of them, and usually uptimes go over 100 days. You can also still buy them used for $20 w/free shipping on amazon! With the recent improvement for cake, make-wifi-fast, driver improvements, etc is there any chance in seeing some of these things land for the WNDR3800 specifically? It would be really nice if this hardware could continue to do SQM, etc for some of the faster broadband speeds the cable providers are offering (comcast xfinity has 100, 200, 250 plans now) and see some of the wifi improvements too. Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] netperf.bufferbloat.net
Noah Causin writes: > There is flent-freemont.bufferbloat.net > > (California). Much better here in seattle, was able to saturate my connection. Thanks! -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] WNDR3800 improvements?
Aaron Wood writes: > I thought it did better than that with Cake? Yes, I was referring to newer cake, possible BQL improvements, possible ath9k improvements, newer kernel, the stuff listed at https://www.bufferbloat.net/projects/make-wifi-fast/wiki/Wifi_Stack_Rework/ etc I've also seen discussion of "policing" rather than full SQM, etc. Basically, what is the state of the art we should be running on the 3800? > On Tue, Dec 13, 2016 at 8:41 PM, Dave T=C3=A4ht wrote: > > > > > > > On 12/13/16 4:58 PM, Matt Taggart wrote: > > > Hi, > > > > > > I love the WNDR3800 platform, it's been great over the years first with > > > cerowrt and then openwrt. Of the many I've deployed I have only had > > > hardware problems with 2 of them, and usually uptimes go over 100 days. > > You > > > can also still buy them used for $20 w/free shipping on amazon! > > > > > > With the recent improvement for cake, make-wifi-fast, driver > > improvements, > > > etc is there any chance in seeing some of these things land for the > > > WNDR3800 specifically? It would be really nice if this hardware could > > > continue to do SQM, etc for some of the faster broadband speeds the cab= > le > > > providers are offering (comcast xfinity has 100, 200, 250 plans now) an= > d > > > see some of the wifi improvements too. > > > > We already support the 3800, although it peaks at 60 mbits of inbound > > rate shaping. If your primary use case is wifi, with the latest fq_codel > > code, you can live without inbound shaping and probably get 150mbits > > well managed. Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] bcp38 and the caida "spoofer" tool
Dave Taht writes: > I am curious as to how many here are using the lede/openwrt bcp38 package? I always install it everywhere since I consider it part of being a good netizen, even if I think the odds of it getting used are low. So far it's always just worked, with one exception where I was doing something weird with rfc1918 ranges, and then I just had to use the luci interface to adjust. I've had similar ideas for ways to use openwrt/lede to help protect against IoT devices participating in botnets. Ideally each time you added an IoT device to your network, you'd have to go in to luci and approve the device and what types of things it was allowed to do. Possibly separate ESSIDs for them? Maybe dedicate one wired port to be sort of an IoT DMZ? -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] quad core arm
On 12/03/2017 09:44 AM, Dave Taht wrote: Anyway the nanopi folk are now producing a wide range of boards I haven't tried... til tomorrow: https://www.amazon.com/gp/product/B0728LPB2R/ref=oh_aui_detailpage_o00_s01?ie=UTF8&psc=1 Is this the same thing for cheaper? http://www.friendlyarm.com/index.php?route=product/product&product_id=180 (but slow and non-free shipping) It appears to be the H5. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] quad core arm
On 12/03/2017 11:49 AM, Dave Taht wrote: https://www.solid-run.com/marvell-armada-family/armada-8040-community-board/ looks rather promising. (recommendation courtesy koen koi) I also picked up two 30 dollar 10GigE interfaces for spaceheater and whatever I end up calling the second box https://www.amazon.com/gp/product/B016OYD0D4/ref=oh_aui_detailpage_o02_s00?ie=UTF8&psc=1 Interesting! These use the mlx4 driver? Does it have support for all the features the vendor driver supports? Does it require loading any firmware? Has any bufferbloat/mwf specific tuning been done on it? Thanks, -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] quad core arm
On 12/03/2017 09:44 AM, Dave Taht wrote: [snip] Another goal was a largely fruitless quest to find the ideal next gen replacement for the wndr3800. These days I'm using a AC2600 as my main device and waiting for the ath10k support to catch up. I used to use an olimex something or other for my NAS, I upgraded it to a pine64, which was better but crashed hard a few months ago and I've not had a chance to go fix it. RE: replacing the wndr3800... Lately I've been thinking about a different model, switching to one of these newer multi core devices to act as a central router running SQM/etc and _no_radios_ and then deploying lightweight ath9k APs around the building. Maybe something like ubiquiti unifi AP (non-AC) running LEDE? Any other good and cheap ath9k AP options? (hard to compete with the wndr3800 now costing $25 shipped, but an AP might have better antennas and PoE) If/when ath10k catches up then just switch the APs out. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] spacebee
On 03/13/2018 11:06 AM, Dave Taht wrote: I am painfully aware of this. On of my big fears in the SDI 80s was that someone would deploy pebbles in a reverse or polar GEO orbit, rigged to explode in a war extending to space. Continuing a tangent... The Sci-Fi TV show The Expanse recently had a similar plotline https://www.imdb.com/title/tt3230854/ Also the Neal Stephenson book Seveneves https://en.wikipedia.org/wiki/Seveneves People interested in such things would love both. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] spacebee
On 03/13/2018 10:47 AM, Christopher Robin wrote: With all the noise around this launch, I haven’t been able to find info on expected operational lifespan vs expected orbit decay. LEO’s can still last for decades. The only thing I’m finding is an expected use for 6mo to 2yr, but not sure how long after that the Spaceebee will stay in orbit and/or be responsive with positional data. While just 4 of these things in space isn’t a major concern, rogue launching objects into space isn’t a scalable solution. This is especially true as the cost of launching comes down into the “cheap” startup range. These types of companies aren’t usually concerned 25yr impact plans, and most wont last long enough to be around to assist if any problems occur past that 2-3yr window. To possibly bring this tangent back to the topic for this list... A space start-up launching inexpensive devices into orbit with no plans for support, upgrades, or disposal is not totally unlike the situation we're in with consumer routers and other IoT things. When you think of it on that scale it gets quite a bit more scary It's the standard business tricks of shifting profit forward at the expense of the future and externalization of costs. (The nuclear industry is another good example, but that's a whole other tangent). -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] I so love seeing stuff like this
On 2/1/19 8:28 AM, Dave Taht wrote: > I haven't had to touch sqm personally for years now, and I'd like to > thank everybody for keeping the package updated and relevant. Yes it's great! I was pleasantly surprised the other day when switching a WNDR3800 from using fq_codel+simple to CAKE+piece_of_cake and getting a 30mbit improvement (70 to 100, comcast). New life for old hardware! I'm still on the lookout for a good replacement for the dozen WNDR3800s I have deployed as Friends-and-Family-IT(tm). Starting to experiment with espressobin + Linksys PoE switch + Ubiquity APs. Anyone else have inexpensive, better cpu, and 802.11ac capable replacements for WNDR3800? -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] Will transport innovation collapse the Internet?
This is from Jan 12th but I hadn't seen it yet. https://huitema.wordpress.com/2019/01/12/will-transport-innovation-collapse-the-internet/ -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] raspi4 (was Re: WebOS goes more open source (In collaboration with the South Korean government))
On 6/26/19 7:35 PM, Dave Taht wrote: > and the raspi4 : > http://linuxgizmos.com/quad-a72-raspberry-pi-4-finally-gets-its-ram/ Only one gig port, but 2 USB3 ports, so maybe using USB3 gig adapters would work. Anyone have recommendations for the the best ones? BQL support, mature drivers, no binary blobs, etc. The ones I have so far: ASIX Elec. Corp. ASIX AX88179 USB 3.0 Gigabit Ethernet idVendor=0b95, idProduct=1790 linux driver: ax88179_178a $13.59 from https://www.amazon.com/gp/product/B00MYTSN18 Realtek USB 10/100/1000 LAN idVendor=0bda, idProduct=8153 linux driver: r8152 Came with my ASUS laptop Also the raspi4 supports being powered by PoE, which fits nicely with the way I have started deploying things: separate router, PoE switch, multiple PoE APs in strategic locations, PoE VOIP phones, etc. and a UPS to power it all. The built-in gig port could be on the internal side of the network plugged into the PoE switch, and the USB3 adapter could be used for WAN. (I suppose this model works with other USB WAN options like LTE, DSL, wireless, etc). Nothing listed yet at https://openwrt.org/toh/raspberry_pi_foundation/raspberry_pi -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] Ubiquiti Launches a Speed Test Network
https://blog.ui.com/2019/08/13/ubiquiti-launches-a-speed-test-network/ The web app (which requires a lot of js) at http://speed.ui.com/ mostly just gives a graph of bandwidth over time and only mentions a single ping number (which was 10ms for me, so maybe just a bucket?) The android app gives a few single ping times to major service providers (google, facebook, twitter) but no further latency results when running the test. The press release says the UniFi Network Controller can run automated speed tests and in the updated version (5.11.39) I can see where to enable it, but it also says it requires the UniFi Security Gateway product to enable. To setup your own test server at https://speed-admin.ui.com/ requires an account (and also goes into an infinite redirect loop on my browser at least) So an interesting idea but they have some things they could improve. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] GL-MV1000
Another little 3 gig port SBC https://www.gl-inet.com/products/gl-mv1000/ https://store.gl-inet.com/pages/brume-gl-mv1000-edge-computing-vpn-router Marvell Armada 88F3720, Dual-Core ARM Cortex-A53 @1.0GHz DDR4 1GB/ FLASH 16MB + EMMC 8GB 3 x Gigabit ports, 1 x USB 2.0, 1 x MicroSD slot, 1 x USB Type-C power port, 1 x reset button, and 1 x mode switch MSRP $129 Similar to espressobin (which I am still playing with, anyone else have experience with it and SQM?) found here https://forum.armbian.com/topic/11742-another-3720-box-glinet-mv1000/ -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] Looking for MORE SQM Router Recommendations !
On 3/16/21 11:30 AM, Charles Rothschild via Cerowrt-devel wrote: > I want a 1GB capable SQM router. What options are good nowadays hw wise? I've been using the Qotom x86 boxes and running x86 OpenWRT on them. They have no problem doing CAKE on 1gig symmetrical connection. http://qotom.net/ and then Products -> Mini PC -> Multiple NIC But the Qotom vendor site is confusing and also doesn't list all the products you can find available. So a while ago I made this decoder ring https://we.riseup.net/lackof/x86-router-candidates#qotom The prices listed there are old now, but maybe still help to compare them. The two models I have ordered are Q355G4, for $215.00 + $33.64 shipping + tax (seems to have gone up a little) https://www.amazon.com/gp/product/B077ZWR8Q9 Q330G4, for $169 + $42 shipping + tax (not available) https://www.amazon.com/dp/B07117QWFH OpenWRT doesn't need much RAM (unless you plan to do more than just routing with SQM) so I just used some orphaned 2gb DDR3 SODIMMs I had laying around. I boot it from a small USB drive, but you could use a proper hard drive if you wanted it to do more. Here's another thing I learned about lately, you can buy the ROCK64 1gb V2 board on ebay for $20 including power supply ($5 shipping, decent cases for $10-20). https://www.ebay.com/itm/274678920361 It's got a built-in gigabit NIC and also has a usb3 port. Anyone know about the SQM potential of that device? https://wiki.pine64.org/wiki/Rock64 Could it work for just a gateway and then use a separate switch and APs? (I'm much more likely to deploy things that way these days rather than an all-in-one device). It could maybe make a nice AP-only device too with a suitable usb3 wireless adapter. (and what usb adapters do people like these days? Maybe this could be the replacement for the WNDR3800 that were such a good deal? (and if not that one, maybe another PINE device) -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] usb3 gigabit adapters
Has anyone reviewed the various available usb3 gigabit adapters for features, linux driver support, bufferbloat, BQL(can usb NICs do BQL?), etc? With some single board computers having reasonable usb3 now, I can see people making routers out of them. But it would also be good to know if that's reasonable and in general which to recommend. Some searching in drivers/net/usb/ for gigabit capable things I found: aqc111: Aquantia AQtion USB to 5GbE ax88179_178a: ASIX AX88179, ASIX AX88178A, Sitcomm LN-032 lan78xx: 7800/7801 usb3 devices r8152: r8153 usb3 devices smsc75xx: only usb2? Maybe there are others? In real life I have: 1) an r8152 device (came with a Asus laptop). It's been reliable. 2) an ax88179_178a device (https://www.amazon.com/gp/product/B00MYTSN18). Worked at first, but started acting up after a recent kernel upgrade and I haven't tracked down what changed yet. I've attached some ethtool output for the features of each. Ebay has tons of $10 options, but little details. Some also have integrated USB ports on them, some are "docking station" like and have hdmi as well. If you search for "usb 3.2" you can find some that are 2.5gbit. Also some interesting (but from 2019) info in this post https://raspberrypi.stackexchange.com/questions/100040/what-sort-of-throughput-is-achievabe-over-the-usb-3-0-port-on-the-pi-4 -- Matt Taggart m...@lackof.org# ethtool -i eth3 driver: r8152 version: v1.11.11 firmware-version: rtl8153a-2 v1 10/23/19 expansion-rom-version: bus-info: usb-:00:1d.7-1 supports-statistics: yes supports-test: no supports-eeprom-access: no supports-register-dump: no supports-priv-flags: no # ethtool -k eth3 Features for eth3: rx-checksumming: on tx-checksumming: on tx-checksum-ipv4: on tx-checksum-ip-generic: off [fixed] tx-checksum-ipv6: on tx-checksum-fcoe-crc: off [fixed] tx-checksum-sctp: off [fixed] scatter-gather: on tx-scatter-gather: on tx-scatter-gather-fraglist: on tcp-segmentation-offload: on tx-tcp-segmentation: on tx-tcp-ecn-segmentation: off [fixed] tx-tcp-mangleid-segmentation: off tx-tcp6-segmentation: on generic-segmentation-offload: on generic-receive-offload: on large-receive-offload: off [fixed] rx-vlan-offload: on tx-vlan-offload: on ntuple-filters: off [fixed] receive-hashing: off [fixed] highdma: off [fixed] rx-vlan-filter: off [fixed] vlan-challenged: off [fixed] tx-lockless: off [fixed] netns-local: off [fixed] tx-gso-robust: off [fixed] tx-fcoe-segmentation: off [fixed] tx-gre-segmentation: off [fixed] tx-gre-csum-segmentation: off [fixed] tx-ipxip4-segmentation: off [fixed] tx-ipxip6-segmentation: off [fixed] tx-udp_tnl-segmentation: off [fixed] tx-udp_tnl-csum-segmentation: off [fixed] tx-gso-partial: off [fixed] tx-tunnel-remcsum-segmentation: off [fixed] tx-sctp-segmentation: off [fixed] tx-esp-segmentation: off [fixed] tx-udp-segmentation: off [fixed] tx-gso-list: off [fixed] fcoe-mtu: off [fixed] tx-nocache-copy: off loopback: off [fixed] rx-fcs: off [fixed] rx-all: off [fixed] tx-vlan-stag-hw-insert: off [fixed] rx-vlan-stag-hw-parse: off [fixed] rx-vlan-stag-filter: off [fixed] l2-fwd-offload: off [fixed] hw-tc-offload: off [fixed] esp-hw-offload: off [fixed] esp-tx-csum-hw-offload: off [fixed] rx-udp_tunnel-port-offload: off [fixed] tls-hw-tx-offload: off [fixed] tls-hw-rx-offload: off [fixed] rx-gro-hw: off [fixed] tls-hw-record: off [fixed] rx-gro-list: off macsec-hw-offload: off [fixed] # ethtool -i enx000ec6bd480a driver: ax88179_178a version: 5.10.0-13-amd64 firmware-version: expansion-rom-version: bus-info: 2-2:1.0 supports-statistics: no supports-test: no supports-eeprom-access: yes supports-register-dump: no supports-priv-flags: no # ethtool -k enx000ec6bd480a Features for enx000ec6bd480a: rx-checksumming: on tx-checksumming: on tx-checksum-ipv4: on tx-checksum-ip-generic: off [fixed] tx-checksum-ipv6: on tx-checksum-fcoe-crc: off [fixed] tx-checksum-sctp: off [fixed] scatter-gather: off tx-scatter-gather: off [fixed] tx-scatter-gather-fraglist: off [fixed] tcp-segmentation-offload: off tx-tcp-segmentation: off [fixed] tx-tcp-ecn-segmentation: off [fixed] tx-tcp-mangleid-segmentation: off [fixed] tx-tcp6-segmentation: off [fixed] generic-segmentation-offload: off [requested on] generic-receive-offload: on large-receive-offload: off [fixed] rx-vlan-offload: off [fixed] tx-vlan-offload: off [fixed] ntuple-filters: off [fixed] receive-hashing: off [fixed] highdma: off [fixed] rx-vlan-filter: off [fixed] vlan-challenged: off [fixed] tx-lockless: off [fixed] netns-local: off [fixed] tx-gso-robust: off [fixed] tx-fcoe-segmentation: off [fixed] tx-gre-segmentation: off [fixed] tx-gre-csum-segmentation: off [fixed] tx-ipxip4-segmentation: off [fixed] tx-ipxip6-segmentation: off [fixed] tx-udp_tnl-segmentation:
Re: [Cerowrt-devel] Fwd: Realtek RTL8156 devices defaulting to CDC-NCM instead of vendor mode, resulting in reduced performance
On 5/2/22 15:53, Dave Taht wrote: .6ms considered good. Well they don't say to _where_ the round trip was, so I read it as the difference between 3ms to 0.6ms is 2.4ms that the non-vendor mode was adding. For the baseline of 0.6ms, we don't know where that's getting introduced. (probably still in the building, could be to their router or another host running who knows what). -- Forwarded message - From: Forest Crossman Date: Mon, May 2, 2022 at 3:49 PM Subject: Realtek RTL8156 devices defaulting to CDC-NCM instead of vendor mode, resulting in reduced performance To: , , Cc: , Hi, all, I recently purchased a pair of USB to 2.5G Ethernet dongles based on the RTL8156, and have so far been very happy with them, but only after adding some udev rules[0] to to take advantage of the r8152 driver by switching the devices from their default CDC-NCM mode to the vendor mode. I was prompted to use those rules to switch the driver because one of the adapters (based on the RTL8156A) would get very hot, up to 120 F (49 C) even while idle, and the round-trip latency directly between the pair of adapters was about 3 ms, and I couldn't help but wonder if maybe the vendor mode might be more efficient. After performing some tests of latency and power consumption, testing first with both adapters in NCM mode and then again with both in vendor mode, I proved my hunch correct. I discovered that, in a disconnected state, the RTL8156A adapter used about half as much power (0.64 W -> 0.30 W) while the RTL8156B adapter saw a 21% reduction in power (0.34 W -> 0.27 W). Similarly, in a connected-but-idle state the RTL8156A again saw about a 55% savings in power consumption (2.17 W -> 0.97 W) and a 40% savings in the RTL8156B adapter (0.94 W -> 0.56 W). It was only under full load that the fewest power savings were seen, with a reduction of only 15% in the RTL8156A (2.23 W -> 1.90 W) and no savings for the RTL8156B (0.96 W). Similarly, round-trip latency while idle went from 3 ms to 0.6 ms. I also tested under load and saw much larger latency savings and reduced packet loss, but forgot to write down the numbers (I can run the tests again if someone really wants me too). Also, jumbo frames drastically reduced performance under NCM mode, while vendor mode handled it like a champ (again, I forgot to write down the numbers but can test again if asked). So, with all the benefits I've seen from using these adapters in their vendor mode, is there still a reason to let the kernel prefer their NCM mode? It'd be nice to be able to get the maximum performance from these adapters on any Linux system I plug them into, without having to install a udev rule on every one of those systems. If anyone would like to try replicating the results I listed here, or to perform new tests, the specific RTL8156A adapter I used is the Ugreen CM275[1] and the RTL8156B adapter is the Inateck ET1001[2]. Curious to hear your thoughts on this, Forest [0]: https://github.com/bb-qq/r8152/blob/160fb96d2319cdf64ae7597e8739972934ac83b2/50-usb-realtek-net.rules [1]: https://www.amazon.com/gp/product/B081TY1WQX/ [2]: https://www.amazon.com/gp/product/B08VN3DGK6/ -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] Minirouter with pi compute module 4
This looks like an interesting router candidate https://www.seeedstudio.com/Dual-GbE-Carrier-Board-with-4GB-RAM-32GB-eMMC-RPi-CM4-Case-p-5029.html Description says: * one NIC is Broadcom BCM54210PE (from the CM4) * the other is "Microchip's LAN7800" behind usb3 * 2 additional usb3 ports * the usb3 uses the CM4's PCIe 2.0 x1 (500MB/s) * wifi/BLE is the CM4's onboard, I think "Cypress CYW43455"? It sort of reminds me of the Espressobin device from a few years back, but much faster and the pi has a much larger installed base, better support, etc. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] 2.5gbit for $59
On 6/2/22 10:11, David P. Reed wrote: There are small, low-TDP Intel systems for up to ~$250 or so (including case) that use current generation Celerons with 4 2.5 GigE ports, and with the I/O bandwidth to easily support a full-on router at wirespeed on those ports. I'm thinking of upgrading my entry-router (which is based on Fedora Server 36 now, not Cerowrt, just because that's my general go-to distro on x86_64 and Aarch64) from an old Celeron system with two full speed 1 GigE ports to 2.5 GigE, in advance of my expectation that 2.5 GigE DOCSIS 3.1 will become cheap enough soon at my home. The problem with the low-end boards is that you need enough PCIe lanes to move packets at 10 Gb/sec bidirectionally. The contained ARM chips may be fast enough in principle, but the board and the PCIe are a bottleneck. AliExpress sells such boards and also barebones, but prices and specs vary. The ones I see there seem to be using Celeron N5090 or N5105. Both have "PCI Express 3.0 controller supporting 8 lanes (multiplexed); 4 lanes available externally" They all seem to be using "4x Intel i225-V" Apparently earlier revs of that had problems but the "B3" stepping is supposed to be fixed. Each uses pci-e 3.1 x1. So depending on how the board is laid out, they should have the bandwidth to actually do 2.5Gbit. All the usb ports, wifi, graphics, etc should all be using the internal lanes I think. Here is a comparison of those celerons, the nanopi, and the pi4 https://www.cpubenchmark.net/compare/ROC-RK3568-PC-HDMI-(Android)-vs-BCM2711-vs-Intel-Celeron-N5105-vs-Intel-Celeron-N5095/4752vs4297vs4412vs4472 more details on the specific pages. The nanopi seems mostly better than the pi4, except some floating point and matrix. The Celerons are much better CPUs, but are in a different power consumption and price class. -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
Re: [Cerowrt-devel] binary blobs struck again
On 8/17/22 14:07, Dave Taht via Cerowrt-devel wrote: lack of trust in turtles all the way own. https://www.bleepingcomputer.com/news/security/exploit-out-for-critical-realtek-flaw-affecting-many-networking-devices/ More justification for your mass-router-trade-in-refurb program :) Seriously, there should be a press release. eCos seems to just be the embedded O/S in these router devices, so presumably if you had linux/openwrt/etc on these devices you wouldn't be affected? Realtek's former website http://www.realtek.com.tw/ doesn't resolve, but they seem to have realtek.com too. Here are a couple related product pages https://www.realtek.com/en/products/communications-network-ics/item/rtl8196e https://www.realtek.com/en/products/communications-network-ics/item/rtl8197f Here's a good wiki page https://wikidevi.wi-cat.ru/Realtek/SoC http://en.techinfodepot.shoutwiki.com/wiki/Realtek/SoC (same page?) Seems to be AKA Lexra and never really got full OpenWRT support https://openwrt.org/docs/techref/hardware/soc/soc.realtek most of the people attempting things seemed to be working on it back in the Barrier Breaker days and there hasn't been anything since then. So we can't just advocate people install openwrt on them. Probably all the devices are 4mb flash and 32mb ram or worse, so at this point should just be recycled anyway https://openwrt.org/supported_devices/432_warning Searching on the openwrt table of hardware I found a few popular devices that received hardware revs to use it and never got support: D-Link DIR-615 Revs J1, M1, T1 https://openwrt.org/toh/d-link/dir-615#unsupported_versions NETGEAR WNR612 Rev v3 https://openwrt.org/toh/netgear/wnr612v2 Maybe someone will write a worm that just bricks them... (NOT ADVOCATING FOR SUCH A THING, THAT WOULD BE ILLEGAL) rtl819x seems to be the general name of the SoC but it's really just rtl8196/rtl8197 and there are other devices with rtl819* names, mostly wireless https://wireless.wiki.kernel.org/en/users/drivers/rtl819x https://wiki.debian.org/rtl819x https://openwrt.org/docs/techref/driver.wlan/rtl819x Also common rtl81* things: * RTL8111/8168/8411 pci-e gigabit NICs (r8169 driver) * RTL8153 usb gigabit NIC (r8152 driver) Many of these realtek devices can load firmware binary blobs and those are found at https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git (see rt*) and are available on Debian in the non-free firmware-realtek package. If anyone finds exploits in those then we're _really_ in trouble... -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] OTA exploitable wifi bugs
https://www.openwall.com/lists/oss-security/2022/10/13/2 Presumably openwrt and other router firmwares (FOSS and proprietary) will be effected? Also android and maybe TVs, etc? That's a whole lot of devices. Lots of updating in our futures... maybe this will help get newer SQM rolled out more (but maybe not enabled by default). Sorry, this is probably my fault, I just updated a bunch of stuff last night (after I washed my car causing it to rain). -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
[Cerowrt-devel] x86 router boxes with 2.5G ports
Serve the Home https://www.servethehome.com/ have been doing a bunch of reviews of some fanless router mini-PCs similar to the Qotom that I researched before and got a couple of(which have been great). They have been buying them from Aliexpress and they are mostly the ‘Topton’ brand, but also some similar no-name units. What is particularly cool is they are now coming with 4-6 Intel 2.5Gbit I225 NICs. Here’s a bunch of links * A 6x 2.5GbE Intel Pentium N6005 Fanless OPNsense pfSense Firewall Option https://www.servethehome.com/a-6x-2-5gbe-intel-pentium-n6005-fanless-opnsense-pfsense-firewall-option/ * Topton Intel N5095 4x 2.5GbE Firewall Review https://www.servethehome.com/topton-intel-n5095-4x-2-5gbe-firewall-review/ * Cheap Intel Pentium N6005 4x 2.5GbE Fanless Firewall and Router Review https://www.servethehome.com/cheap-intel-pentium-n6005-4x-2-5gbe-fanless-firewall-and-router-review/ * Topton Jasper Lake Quad i225V Mini PC Report forum https://forums.servethehome.com/index.php?threads/topton-jasper-lake-quad-i225v-mini-pc-report.36699/ * Two Fanless Intel Celeron N5105 4x 2.5GbE Options Reviewed https://www.servethehome.com/two-fanless-intel-celeron-n5105-4x-2-5gbe-options-reviewed/ * video review https://www.youtube.com/watch?v=tZK1l9bXDgs Here are the CPUs they seem to come with (in StH’s preferrred order by price/perf/wattage) model, family name, release, core/thread, base/turbo, cache, TDP --- i7-1165G7, TigerLake, 2020Q3, 4/8, 2.8/4.7, 12mb, 28W N5105, JasperLake, 2021Q1, 4/4, 2.0/2.9, 6.5mb, 10W J4125, GeminiLake, 2019Q4, 4/4, 2.0/2.7, 4mb?, 10W N6005, JasperLake, 2021Q1, 4/4, 2.0/3.3, 4mb, 10W N5095, JasperLake, 2021Q1, 4/4, 2.0/2.9, 4mb, 15W (search for "Intel" and model name to get the Intel ark page for each) The N5105 seems the best if you care more about power usage than being able to host a bunch of other cpu intensive stuff on it. Also has more capable pci-e, faster ddr4, newer, etc. The are also some based on this one: J6413, ElkhartLake, 2021Q1, 4/4, 1.8/3.0, 1.5mb, 10W, LPDDR4 3733 (faster ram but less cache). Where to buy: * Aliexpress Topton store https://www.aliexpress.com/store/1101224793 * N5105 4x 2.5Gbit $166, no serial https://www.aliexpress.us/item/3256804153987410.html * N5105 6x 2.5Gbit $218, has serial https://www.aliexpress.us/item/3256804173890686.html It's interesting that most of the units are marketed specifically for pfSense(but often say "soft router" or "network appliance" too). All seem to be using Intel i-225. Some specs on this product sheet: https://cdrdv2.intel.com/v1/dl/getContent/639476 Apparently it's important to have the B3 stepping or later of that, earlier revs had issues, and many of the units on aliexpress explicitly mention B3. There is an i-226 now too, which some units have. Some people are speculating that it's just a rebrand of the i-225v B3 since the i-225 name is tainted now. Anyone know about the state of the linux driver for this device? Does it have all the nice things one would want for SQM? Any hardware features that are nice (or should be turned off)? I ordered a couple of the 4-port models to experiment with. (FYI: StH also have some reviews of inexpensive 2.5G unmanaged switches. some with PoE too, but I haven't looked into them much yet.) -- Matt Taggart m...@lackof.org ___ Cerowrt-devel mailing list Cerowrt-devel@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cerowrt-devel
