RT-BE92U Persistent Thermal Zone Faults

[RicoSuave]

First post here so bear with me if I miss anything. Since purchase, my RT-BE92U has been plagued with the same issues others have posted here. Up until the new firmware version dropped, I was having constant problems with network stability. I am currently running the latest firmware version (3.0.0.6.102_37435) and a lot of the issues seem to have been resolved except for one. Before the firmware update, I was seeing intermittent thermal zone errors in the system log. Since applying the update, I am consistently getting the following errors that appear every 3 seconds exactly:

Jan 4 18:29:52 kernel: read_bpcm_reg_direct_keyhole: Wait for completion failed, devAddr 0x3, offset 24, idx 2
Jan 4 18:29:52 kernel: thermal thermal_zone0: Failed to get RAIL 0 temperature, ret=256
Jan 4 18:29:52 kernel: thermal thermal_zone0: Failed to get temperature, ret=-1
Jan 4 18:29:52 kernel: thermal thermal_zone0: failed to read out thermal zone (-1)

On occasion, the errors coincide with a drastic decrease in wireless network performance. Under normal conditions, when connected to the 6GHz network, my wireless network download speed averages around 1100Mbps. This will drop to a nearly unusable 2-3Mbps. The 2.5GHz and 5GHz networks also see a decrease in speed to around the same 2-3Mbps. Wired clients seem unaffected. Once I perform a hard reset to the router, it seems to correct the speed issue for a while, but the thermal zone errors continue to persist. This seems to happen anywhere from once a week, to every day. Performing a hard reset isn't something I want to nor should have to keep doing to correct the issue every time it happens. Please let me know if there is anything else I can do to stop this from happening. I am getting very close to returning this router and exchanging it for a different brand. Thanks.

[changkai]

I'm seeing the same problem! My router reboot seems to happen a few times a day.  And here are some logs I gathered right before one of the reboots:

Jan 11 22:16:56 kernel: thermal thermal_zone0: Failed to get RAIL 0 temperature, ret=256
Jan 11 22:16:56 kernel: thermal thermal_zone0: Failed to get temperature, ret=-1
Jan 11 22:16:56 kernel: thermal thermal_zone0: failed to read out thermal zone (-1)
Jan 11 22:16:58 wlceventd: wlceventd_proc_event(656): wl0.3: Deauth_ind 10:5A:17:0B:83:88, status: 0, reason: Unspecified reason (1), rssi:0
Jan 11 22:16:58 wlceventd: wlceventd_proc_event(696): wl0.3: Auth 10:5A:17:0B:83:88, status: Successful (0), rssi:0
Jan 11 22:16:58 wlceventd: wlceventd_proc_event(656): wl0.3: Deauth_ind 10:5A:17:0B:83:88, status: 0, reason: Unspecified reason (1), rssi:0
Jan 11 22:16:58 wlceventd: wlceventd_proc_event(696): wl0: Auth 10:5A:17:0B:83:88, status: Successful (0), rssi:0
Jan 11 22:16:59 wlceventd: wlceventd_proc_event(656): wl0: Deauth_ind 10:5A:17:0B:83:88, status: 0, reason: Unspecified reason (1), rssi:0
Jan 11 22:16:59 wlceventd: wlceventd_proc_event(696): wl0.3: Auth 10:5A:17:0B:83:88, status: Successful (0), rssi:0
Jan 11 22:16:59 kernel: read_bpcm_reg_direct_keyhole: Wait for completion failed, devAddr 0x3, offset 24, idx 2
Jan 11 22:16:59 kernel: thermal thermal_zone0: Failed to get RAIL 0 temperature, ret=256
Jan 11 22:16:59 kernel: thermal thermal_zone0: Failed to get temperature, ret=-1
Jan 11 22:16:59 kernel: thermal thermal_zone0: failed to read out thermal zone (-1)
Jan 11 22:17:02 kernel: read_bpcm_reg_direct_keyhole: Wait for completion failed, devAddr 0x3, offset 24, idx 2
Jan 11 22:17:02 kernel: thermal thermal_zone0: Failed to get RAIL 0 temperature, ret=256
Jan 11 22:17:02 kernel: thermal thermal_zone0: Failed to get temperature, ret=-1
Jan 11 22:17:02 kernel: thermal thermal_zone0: failed to read out thermal zone (-1)
Jan 11 22:17:02 wlceventd: wlceventd_proc_event(656): wl0.3: Deauth_ind 10:5A:17:0B:83:88, status: 0, reason: Unspecified reason (1), rssi:0
Jan 11 22:17:02 wlceventd: wlceventd_proc_event(696): wl0.1: Auth 10:5A:17:0B:83:88, status: Successful (0), rssi:0
Jan 1 08:00:37 kernel: klogd started: BusyBox v1.24.1 (2024-12-08 22:32:46 CST)
Jan 1 08:00:37 kernel: Linux version 4.19.294 (root@asus) (gcc version 10.3.0 (Buildroot 2021.02.4)) #1 SMP PREEMPT Sun Dec 8 22:36:47 CST 2024
Jan 1 08:00:37 kernel: ARM_SMCCC_ARCH_WORKAROUND_1 missing from firmware
Jan 1 08:00:37 kernel: Kernel command line: coherent_pool=4M cpuidle_sysfs_switch pci=pcie_bus_safe console=ttyAMA0,115200 rootwait rng_core.default_quality=1024 mtdparts=brcmnand.0:2097152(loader),265289728@2097152(image) root=/dev/ubiblock0_4 ubi.mtd=image ubi.block=0,4 rootfstype=squashfs
Jan 1 08:00:37 kernel: bootloader version U-Boot2019.07(12/08/2024-22:34:48+0800)50404p3@523100
Jan 1 08:00:37 kernel: BoardId [RT-BE92U], NVRAM_ULBOARDSTUFFOPTION=0
Jan 1 08:00:37 kernel: size of sk_buff 576
Jan 1 08:00:37 kernel: offset of cb in sk_buff 400
Jan 1 08:00:37 kernel: OTP driver initcall
Jan 1 08:00:37 kernel: OTP driver scanning DT
Jan 1 08:00:37 kernel: Strap Register: 0x7fffebc3

[Chobra]

I was about to give up as well but my issue has been fixed by doing the following:

Once you confirm you have the latest (3.0.0.6.102_37435) firmware installed 

Power down the BE92U router, unplug from the wall outlet, let it sit for a few minutes for the cache to clear

Plug it back into the wall outlet, power back up, thermal issue no longer appears in the log, CPU core doesn't bounce from 50-100% anymore every 1-2 seconds

[RicoSuave]

I'm pretty sure I've already done this at some point, but I'll try it again when I get home and report back in 24hrs to see if the system logs report anything. 

[changkai]

That's great to hear! Thanks for sharing! Does anyone know if there is a way to clear cache without powering down?

[RicoSuave]

Update:

Powering down and letting it sit for over 45 min last night did not work. I checked the system log again this morning, and the faults are still asserting every 3 seconds exactly. RAM usage is at a constant 68% with little network activity. One thing I did notice just now that I didn't see before is what you said about the CPU bouncing from 50-100% 2 seconds. Core 2 on my router is doing that right now. Maybe I was so distracted with everything else wrong with the router and finding how to fix it that I didn't notice it before. @Aureliannn_ASUS Here are some more data points to add to this issue if you need them.