4.5.8 is based on OpenWRT 21.02 which is EoL. No option for me to use an EoL firmware on a brand new router.
serious? Have you seen Xiaomi and Redmi routers? I had a redmi ax6 based on openwrt 18.06
The best of both worlds is having stability and having an updated version. But if I need one of the two options, I prefer stability than the placebo of having a current version with wifi problems, devices being disconnected out of nowhere and not being able to connect anymore, among other things.
Just check the forum that the stable version is 23.05.3 and the snapshot is far from stable.
I‘m running the snapshot that has been mentioned above and for me it seems pretty stable. No issues so far.
I have been on the snapshot for a week and it is VERY stable. I am using IPv6 (6RD), stubby (DNS over TLS), guest network, main network and have many connected IoT devices. Both wireless bands are very stable. Log files have few entries. If someone is experiencing instability on a snapshot, I would be interested to hear about the issues being encountered. Take care!
Just to also follow up on that post from you @japa182: I can’t follow your reasoning, or did the manufacturers also actively advertise support for OpenWRT 23.05? Please check the previous postings regarding that.
ok ok, but if even the clean version of openwrt presents the same problem that no developer has yet solved, do you think that you demanding an open source version of 23.05 or higher from the gl devs will solve it?
I’m not sure if I follow you, but again: obviously a lot of the problems seem to be solved with the latest firmware from OpenWRT and the patches for the mt76 driver @pesa1234 included in his firmware.
Even if these problems were not yet solved with the OpenSource driver, I would expect that at least the product that Gl-Inet distributes is equipped with a driver that is stable and performant. I get the feeling that we have completely different ideas about manufacturer promises and quality.
I’m also not satisfied with the mt6000, there’s a lot to improve and I’ve already covered a lot here on the forum and also served as a guinea pig for several alpha versions that the devs released. but the only thing we can do at the moment is wait for them to correct it, there’s little point in complaining that they didn’t fulfill what was agreed and blah blah blah, yes they launched a crappy router without further testing, but unfortunately we have to wait and as you repeated me too I’ll repeat, even with the latest mt76, 2.4 is still terrible in openwrt, whether in 23.05 or in versions with kernel 6.1 or 6.6, on smartphones and older devices, those who use IoT based on 2.4 couldn’t even use it…
Strongly agree.
In the end, you only have two options:
-
Accept, wait and see and be happy that things are getting better.
-
Sell the device (or throw it away), label the company as “I’ll never buy there again” and get on with your life.
Please be more specific. I have about 10 IoT devices here that are all connected to the router on 2.4G and the latest snapshot firmware and currently have no problems.
Man, I’m glad they solved it then, how is the latency there? I have a friend who was always testing the opensource, openwrt, immortalwrt, xwrt versions and in all the latency in 2.4 it was 700 800 in both down and up. and this meant that the connection between everything in the residence either took a long time to execute or presented errors.
I havent measured every device, as not always I need low latency, but for example Alexa or starting a power switch reacts much quicker with this Firmware on 2.4G.
I also don’t know what this dismissive tone is all about. If you don’t believe me, please try it out yourself and share your experience with that Firmware. I’m glad that there are people working on improving the software and I’m sure that they are also happy to receive constructive feedback.
Do you use the snapshot firmware with WED enabled?
I noticed on my Asus TUF AX6000 (same chipset as MT6000) a random high WiFi ping and bad rx/tx performance several minutes long.
These ping problems all started after AP-STA-POLL-OK appeared multiple times in the log.
Some google research led me to the suggested solution to disable WED.
After disabling it this problem did not occur yet again.
Yes, it’s enabled (already by default in this firmware), but just confirmed.
It’s not sarcasm.
I really want to use pure openwrt as well, but every time I tried I had problems that unfortunately made it unfeasible for my use.
Wanted to push out a beta before the long holiday but decided not to rush.
Slow and steady wins the race. I appreciate the hard work and dedication of you and the rest of the team at GL.iNet.
Thankfully, 4.5.8 has been rock solid for me since April 16th.
Thanks, I’m using a 4.6.0 snapshot on the MT3000 and it’s been working fine, it has like 3-4 days up, I liked all the optimization that this firmware has,
AP-STA-POLL-OK was not the source of the problem. I also disabled the inactivity timer and “Disassociate on low acknowledgment”. I tested my settings on the TUF-AX6000 and MT6000 with the latest OpenWrt snapshot and 23.05.3 which always ended up with the same results mainly in the 5GHz network:
-
Random ping spikes and terrible WiFi performance
-
Total WiFi lockdown, no more data transfer going through anymore, i can’t even ping the gateway (happened so far only with the MT6000)
In my opinion the mt7986 driver is far away from being perfect because it’s still horrible unstable.