Askey RAC2V1K / RT4230W REV6 Support

@StephenR0 -
your step by step is complete - and I agree, it's not difficult. for some, it might be even easier; there are some no-serial-access workarounds for those who have a rac1v1k if you look earlier in this long thread.

regarding access point / radio performance - these devices are excellent. i have had 3 of them in my home (large, old with chicken wire or thicker steel/iron mesh in every wall) for more than 6 months. they have performed extrmeely well an have eliminated nearly every wireless issue i previously had. i've been using @lmore377 builds or a local build adapted from his git. see https://github.com/lmore377/openwrt-rt4230w

i'll be interested to update my builds soon. thanks to all who have contributed here

Since you're using these devices in a similar manner to the way I am, I'd like to compare notes on a couple of things. I'm currently using the non-ct version of the ath10k firmware, board, and driver. Things seem to be somewhat more stable that way. The other thing that I've noticed is that I need to dial back the transmit power. Currently I'm only using 39mW on all the radios. If I try to use more, some devices try to find a more distant access point. But the signal strength on everything is good and everything works well. How is yours configured? Thanks.

@StephenR0 - excuse my ignorance - what do you mean by the non-ct version?
im building from the @lmore377 git:
|---|---|
|Model|Askey RT4230W REV6|
|Architecture|ARMv7 Processor rev 0 (v7l)|
|Firmware Version|OpenWrt SNAPSHOT r16204-ed5c94c09c / LuCI Master git-21.080.52349-023b632|
|Kernel Version|5.4.105|
|Local Time|2021-05-15 06:51:06|
default signal strength

how can i help?

upstream/ mainline ath10k (and corresponding firmware) versus CandelaTech ath10k-ct (and corresponding firmware), which is selected as default by OpenWrt.

Sorry to be late. :slight_smile: What @slh said. It's easily installed from @lmore377's packages. I'm not suggesting that you change anything. Actually, you're answering my questions. Thanks. Do I understand correctly that you aren't running the latest version that @lmore377 hosts? If you're not and you decide to update, I'll be interested in your experience.

Just a heads up, I'm gonna build an updated image so it includes the patches for the Frag Attack vulnerability. I'll make another post when it's done.

@lmore377 - great news - I'll look for it and test
thanks

Will your image include support for the Ath10k non-ct driver? That's what I'm currently using on my access points.

1 Like

@lmore377 -
since your builds require a different mtd layout already, do you think a process like this
https://github.com/dangowrt/linksys-e8450-openwrt-installer#steps
might make it easier to get your builds installed for those who dont have serial access or a cracked config?
i'm not complaining - and i dont have th expertise to do this; seems like you might!

The firmware upgrades on this router are des3 encrypted so something like that wouldn't be possible without encrypting the update file first. It's really easy to get the encryption password but you need serial or SSH access for that so you might as well just flash the firmware manually. I would post the password but I don't want to somehow get into legal trouble or something like that.

The new firmware is up. The image is available here and I'll also put it on GitHub in a bit. If you still need to use the old firmware and packages for whatever reason, just edit distfeeds.conf and add .old to the end of snapshots so it's like /snapshots.old/. I haven't had a chance to test it so if anything is broken just send a message or post it here.

no, you dont want any legal trouble!
im fine with serial. i just sysupgraded your recent build over a few images i built from your last repo and aside from my customizations, all good. - will be testing more
.thank you!

5g wireess connectiivty seems good. but a problem that i have experience on old builds and new: the 5ghz

wirelss channel analysis does not work - only shows the local radio.

2.4ghz wireless channel anaysis does work -

i was hoping this would be fixed in with developmens in kernel, kmods, and packages, i'm not sure if this is restricted to your builds. i've noiticed devlopment in this feature over the past few months; may not be quite ready for prime time yet!
thanks

I've noticed that as well, but I thought it was an artifact of using the DFS channels. That's consistent with my experience, but I haven't tested it thoroughly.

To illustrate, here's what I get if I set the 5GHz radio to a non-DFS channel, in this case channel 40.

interesting. i had my channel set to auto, was on 52. when set manually to 40, i got the scan as you did.

Yeah, you could also set the channel to 157 and it would work. There's only two basic areas of the band that are non-DFS. I use this wikipedia article to decide where I'm going to put my access points. That's one of the advantages of OpenWrt. Many access points won't even let you use the DFS channels

It looks like other people are having this issue on a completely different OS.

There's also this
https://ath10k.infradead.narkive.com/W5ZsOP6C/ath10k-firmware-sends-probes-on-dfs-channels-without-radar-detection

Doing iw dev wlan0 scan just returns command failed: Resource busy (-16)

Edit: Yeah this seels to be something with ath10k. I tested on a Luma Mesh Rotuer which uses a ipq4018 for the wireless radio that I flashed and it shows the same behavior.

Actually, I had not looked at this as a bug, more of a limitation. But I suppose it could be a bug of some kind.

Hey guys just a heads up, if you have something that back feeds power into the USB port (some poorly made powered USB hubs do this), unplug it before powering up the router. If you don't, there's a chance you'll have some very unhappy electronics. The router seems fine tho.