My MediaTek access point refused to start due to "daemon.err hostapd: could not get valid channel". Apparently it doesn't want to start on DFS channels anymore, and my selected one was marked DFS for country USA.
It's now in Russia since they didn't mark it as DFS and the access point starts as normal. Could be the same in your situation.
Netgear WNDR4300 V2, anyone tried on this device?
Success on Archer C7 v2. Waiting on the EA8300 build as well, thanks all for your hard work!
I'm going to use this snapshot with my packages until there is a stable one
23.05-Snapshot is a preview of the next 23.05 release.
Today the difference of 23.05-Snapshot to 23.05.0 is only one commit: https://git.openwrt.org/?p=openwrt/openwrt.git;a=shortlog;h=refs/heads/openwrt-23.05
Just want to show my appreciation and send big thanks to all developers and contributors of the project - your great! Can also confirm successful upgrade on a Netgear R7800, well maybe obvious with such an amazing software 
I updated my two R7800 APs, everything seems to be running fine.
Tomorrow I will try to update my NanoPi R4S router.
Thank you very much devs!
P.S.
The same behavior was also there on the previous stable 22.03.05 (I tried before upgrading).
First of all, thanks to everybody involved for the effort. I upgraded my old and trusted TP Link Archer C7 v2 from 22.03 to 23.05 with a customized image (AP-only use) and it seems to work fine.
Two nitpicks. First:
That is already half a year old 
I would have hoped that the final release would have come with a more recent backport from the 6.1.x branch.
And second:
zram was initially broken after the upgrade because kmod-zram doesn't depend on kmod-lib-lzo anymore (it did in 22.03), despite being the default compressor for zram. So, you need to manually install that now. While I get that not everybody might want to use lzo for zram, shipping a package that doesn't work by default due to missing dependencies, doesn't really make sense to me.
In my case, it was simply a matter of needing to replace wpad-basic-mbedtls with the full wpad-mbedtls
Thanks to all the core Devs and all the package Devs. I'd also like to thank all the people that answer questions in the forums. I learn a lot just reading through the posts.
Used factory.bin for the A8300 from here:
https://downloads.openwrt.org/releases/23.05-SNAPSHOT/targets/ipq40xx/generic/
Adjusted kernsize, did not preserve settings on flash, but looks successful.
Sorry, I did not notice that the ipq40xx/generic and bmips/bcm6362 build failed, I triggered the build again and the build is queued now. The binaries should be available in the next few hours.
Build status ipq40xx/generic: https://buildbot.staging.openwrt.org/images/#/builders/84, bmips/bcm6362: https://buildbot.staging.openwrt.org/images/#/builders/102
Thank you so much
Hello, is it already supporting comfast ew72-v2?
Tried to attendedsysupgrade on RPI4 and got this error:
Collected errors:
* opkg_conf_parse_file: Duplicate src declaration (openwrt_core https://downloads.openwrt.org/releases/23.05.0/targets/bcm27xx/bcm2711/packages). Skipping.
* opkg_conf_parse_file: Duplicate src declaration (openwrt_base https://downloads.openwrt.org/releases/23.05.0/packages/aarch64_cortex-a72/base). Skipping.
* opkg_conf_parse_file: Duplicate src declaration (openwrt_luci https://downloads.openwrt.org/releases/23.05.0/packages/aarch64_cortex-a72/luci). Skipping.
* opkg_conf_parse_file: Duplicate src declaration (openwrt_packages https://downloads.openwrt.org/releases/23.05.0/packages/aarch64_cortex-a72/packages). Skipping.
* opkg_conf_parse_file: Duplicate src declaration (openwrt_routing https://downloads.openwrt.org/releases/23.05.0/packages/aarch64_cortex-a72/routing). Skipping.
* opkg_conf_parse_file: Duplicate src declaration (openwrt_telephony https://downloads.openwrt.org/releases/23.05.0/packages/aarch64_cortex-a72/telephony). Skipping.
Any solution? Thanks
Kudos on the 23.05 release! Much appreciated to everybody who contributed, in any way.
Archer C6 v3.2 upgraded from 23.05 rc3 to 23.05.0. I tried to retain settings but there was a problem connecting to the internet even after several reboots. I dont know what the problem was. Troubleshooting was taking too long, so I simply used the reset button on the router and manually re-did the configuration.
Seems to be working pretty good now. I will let you know in a few days how its all going
Had been running 23.05 snapshots and RCs on a pair of:
Upgraded both to the final release, via local imagebuilder.
So far so good, no changes from the snapshots for my use case. Many thanks to the devs!
The Netgear still has a peculiar problem that only seems to surface, very occasionally, with just one client (Fedora 38 laptop using an Intel AX210). I described the original problem here:
The network topology has changed a bit since then, but the TL;DR is the same: On either DFS or non-DFS channel, and only apparently with this one client, the client (or the AP) will sometimes just "refuse" to pass any traffic. Forcing a disassociation and re-connect (by just selecting the WLAN again from the GNOME UI) will work instantly, every single time.
Even more interestingly, a second laptop (same type, same OS, same Intel AX210) does not seem to have this problem.
I'm waiting for a different WiFi adapter (Mediatek based) for the laptop to do some more testing, thought interesting to share.
Thank you very much for the great work!!!
I tested on the Dlink DIR-2640 A1 (I reinstalled from scratch, using the Dlink recovery tool to install from zero).
I noticed that the wifi 5 GHz shows error "wireless is not associated" when I manually set a channel different from 36, or when I change the power, the country or when I use 160 MHz for the wifi.
With the previous firmware I did not have these specific issues, not sure if this is due to new regulation with Wifi.
Thanks for this. The build (https://buildbot.staging.openwrt.org/images/#/builders/84/builds/108) appears to have passed, but it's still not appearing in the downloads section