I just bought a GL-MT6000 and is very happy for the coverage and Speed.
I have a 1000/50 MBit connection via Cable from Vodafone in Germany.
The Vodafone router had really bad coverage. So I connected a cable from Vodafone LAN to MT6000 WAN. The speed on the 5GHz WLAN is impressive. I measure 600/50 MBit on iPhone and computers, which must be the maximum throughput.
I split the SSID of the 2.4GHz and 5GHz to two different, so I have control over which network devices connect to. Before that, my Android phone would connect to the 2.4GHz channel, and I could not find any way to control that.
Anyway, I upgraded the firmware to 4.6.2 since the UI promted for this.
The router is placed on ground floor, and our office is on second floor.
Unfortunately, on the "first work day" with MT6000 , we experienced the router simply rebooted during work hours 3 times. The reboot takes around 30s, but we both have video calls to customers and multiple online systems.
I have tried downgrading to 4.5.8, but we have also experienced the reboot 1 times today.
After upgrading to op24, uptime shows 7 days. So that is progress. I still need to see if it's stable for work. Upgrading to op24 though did have a remarkable bad influence on the range of the wifi. It seems halfed of what it used to be.
After upgrade, pluggin in with LAN, I configured Wifi. Enabling iBF (MediaTek’s proprietary beamforming) had a noticable and positive effect on range for both 2.4 and 5GHz.
I also enabled VHT QAM-256 in 2.4GHz 802.11n, following the screenshot of the guide. Not sure what i does?
Recently there was a new mt76 commit merged which might cause regressions, therefor i don't know if this version from pesa is affected.
See the pieces of info here see what eginnc mentoins.
The symptoms you describe about range sound like you are using the usb connection?, the flint 2 interference with the usb it can create these exact symptoms the port likely isn't shielded.
802.11N is very slow wifi even before AC generation it is expected to be slower, however 256-QAM allows only clients who support it, to reach ax speeds on 802.11N, its kinda like a sort of hack, but OEM SDK also include these type of things, clean OpenWrt as of now still doesn't allow it, pesas build is patched to do so.
My suspicioun points to a usb, but do you use usb?
Edit: it can also be a power issue, especially since you mentoin a crash without a log, that is weird.
The flint2 router is positioned in staircase at ceiling, in the cellar. There is metal nearby. Wired from WAN to upstream provider router on LAN, no USB.
Measurements in living room on ground floor and office 2 floor.
On stock gl inet firmware, both range and speed was impressive. But random reboot happened every few hours. Changing to op24, fixes reboot issue but range was hit severely.
Changing to Custom build seems to improve on range, when iBF enabled.
Hmm im not a scientist but i suspect there might be a few things which cause it:
I see the poe connector hanging on the antenna, these antennas are really vulnerable to close objects from what i have observed to my own Flint, it might also cause unwanted magnetic interference.
As for the placement and antenna orientation.
I think it is omni orientation but that is going to the ground.... And then circle around, but you want it preferely the other side with the antenna up either on a wall.
I wish i had some images to show how that looked.
But if you had a antenna straight up, it should be like a mushroom cap shape as signal and then going down, however if you point the orientation down it hits the ground.
Hi. New here too, having purchased the MT6000 a month ago. Its started to do the same thing (just now while I was using the network, and a few days ago while my other half was home. There may have been others...). Simply seems to lose connection. It resets, then internet is restored around a minute later. I do note I dont have access to any apps anymore, and I did setup Adguard Home.
Crash log empty.
System log contains a LOT of entries like this:
Sat Oct 19 15:05:43 2024 user.notice AdGuardHome[25887]: 2024/10/19 15:05:43.324363 [error] dnsproxy: upstream 8.8.8.8:53 failed to exchange ;developer-api.govee.com.
IN
A in 68.073µs: dialing 8.8.8.8:53 over udp: dial udp 8.8.8.8:53: connect: network is unreachable
Sat Oct 19 15:05:43 2024 user.notice AdGuardHome[25887]: 2024/10/19 15:05:43.384405 [error] dnsproxy: upstream 8.8.8.8:53 failed to exchange ;api-nl.wiserair.com.
IN
A in 173.493µs: dialing 8.8.8.8:53 over udp: dial udp 8.8.8.8:53: connect: network is unreachable
Happy to help debug; I have not reset the device yet... I did take a dump of the log if anyone's interested; Haven't the slightest idea where to look tbh..
oops, forgot the most important part!
Model GL.iNet GL-MT6000
Architecture ARMv8 Processor rev 4
OpenWrt Version OpenWrt 21.02-SNAPSHOT r15812+1075-46b6ee7ffc
Nope. Back to normal.. No issues so far. I know from experience though it will probably happen again at some point. I might dive into the logs to try and make sense of it. Fingers crossed.
Its not a massive thing tbh, although it did sever my wife's important zoom call for a short while.
Umm, perhaps. I wonder though if its a similar problem I had with the Glinet Beryl. That device consistently attempted to renegotiate link speed (2.5<->1G) on the WAN port - It drove me nuts tbh, as it was the main IoT AP and dropped clients. I changed cables etc, ll to no avail.
I also see this device doing the same 'I think'..
Forgive me for diving into detail on this thread, please feel free to tell me off and I will re-route this if its important: Issue: WAN port (eth1) intermittently renegotiates link speed on MediaTek-based device.
Device: MediaTek SoC with a 2.5Gbps WAN port.
Problem: The WAN port (eth1) sometimes renegotiates its link speed, causing brief disconnections. I've noticed similar behaviour on another MediaTek-based device. The link is configured for 2.5Gbps but periodically drops and renegotiates.
Kernel Log Extract (relevant portion):
[ 1.317804] mtk_soc_eth 15100000.ethernet eth1: mediatek frame engine at 0xffffffc011f00000, irq 119
[ 3.451029] mediatek_soc_hnat 15100000.hnat: wan = eth1
[ 13.189436] mtk_soc_eth 15100000.ethernet eth1: PHY [mdio-bus:01] driver [RTL8221B-VB-CG 2.5Gbps PHY]
[ 13.199073] mtk_soc_eth 15100000.ethernet eth1: configuring for phy/2500base-x link mode
[ 17.153410] mtk_soc_eth 15100000.ethernet eth1: Link is Up - 2.5Gbps/Full - flow control rx/tx
[ 23.169050] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
The log shows eth1 initially coming up at 2.5Gbps, but this is intermittent, and I've seen the link drop and renegotiate on occasions. Additionally, eth0 (LAN port) exhibits similar behavior, dropping and then re-establishing the link at 2.5Gbps:
[ 12.785164] mtk_soc_eth 15100000.ethernet eth0: Link is Down
[ 12.805352] mtk_soc_eth 15100000.ethernet eth0: Link is Up - 2.5Gbps/Full - flow control rx/tx
I'd upgrade the firmware and do a reset of configuration
Never deploy (normal unshielded Ethernet cables) parallel to electricity cables, if they have to cross they cross, but never parallel.
If you have light tubes in rooms to give light, those use mercury gas to give light produce a lot of EM. Change them for normal bulbs or leds .
Never put a charger or any source of electricity next to the the antennas, like in the picture, whatever it is , move it away at least 0.5 meter.
Whatever that thing is it will degrade the signal and may even do hardware damage.
Needless to say. A microwave oven is the worst thing to have in the same room as a router, they leak EM and they use exactly 2.4ghz to heat the food, so there is that
