AC66U crashes and reboots itself when Android phone connects to its Wifi

[Marsi4eg]

I've got a new phone (HTC U12 +) and ran into a very strange problem - very often (not always, more likely after wifi module on the phone goes off and then back on) when connecting the phone to the AC66U network, on both ranges, it does not matter 5 or 2.4 GHz, router goes into reboot. Never seen such behavior, although several years a huge number of home devices work with this router, uptime has always been from firmware to firmware, the stability of the work is iron. And now it's like a shoe on the head.

No matter what settings I try to change, I tried the old and new firmwares. I'm using Merlin 380.70 at the moment, I tried to roll back to 374.43 fork, both L and E versions), of course with a full reset - all the same. Even on the default settings AC66U goes into reboot when this phone connects to it. The last thing that happens in the log is how the phone gets the settings from dhcp, it goes through the discovery-offer-request-ack algorithm, after which all the indicators flutter and the router reboots.

Tell me, did any users of this router encounter similar problems? Google gives pretty low information on this topic, but it is noticeable that the people with Pixel phones for example had similar problems with Asus routers...

In general, things are not very good.

As far as I still use the access point-repeater, I had to block my phone by MAC on the on AC66U, so it connects through the repeater, in this mode there is no crash, Ie the problem is definitely not with the dhcp, but with radio.

I even cannot get any consistent pattern and a 100% way to reproduce this, and I don't know how to analyze what actually causes router to reboot in those moments. Please help

 

[ApexRon]

I've got a new phone (HTC U12 +) and ran into a very strange problem - very often (not always, more likely after wifi module on the phone goes off and then back on) when connecting the phone to the AC66U network, on both ranges, it does not matter 5 or 2.4 GHz, router goes into reboot. Never seen such behavior, although several years a huge number of home devices work with this router, uptime has always been from firmware to firmware, the stability of the work is iron. And now it's like a shoe on the head.

No matter what settings I try to change, I tried the old and new firmwares. I'm using Merlin 380.70 at the moment, I tried to roll back to 374.43 fork, both L and E versions), of course with a full reset - all the same. Even on the default settings AC66U goes into reboot when this phone connects to it. The last thing that happens in the log is how the phone gets the settings from dhcp, it goes through the discovery-offer-request-ack algorithm, after which all the indicators flutter and the router reboots.

Tell me, did any users of this router encounter similar problems? Google gives pretty low information on this topic, but it is noticeable that the people with Pixel phones for example had similar problems with Asus routers...

In general, things are not very good.

As far as I still use the access point-repeater, I had to block my phone by MAC on the on AC66U, so it connects through the repeater, in this mode there is no crash, Ie the problem is definitely not with the dhcp, but with radio.

I even cannot get any consistent pattern and a 100% way to reproduce this, and I don't know how to analyze what actually causes router to reboot in those moments. Please help

30,000 foot analysis: No device attempting to connect to the WiFi of a router should cause that router to reboot. So what in the HTC WiFi connection process is different than other WiFi devices but this only needs to be known so that the Merlin coder can find out what needs to be corrected in Merlin. My suggestions would be:

1. Install the latest non-Merlin firmware from ASUS. This will determine if it is a Merlin firmware bug or a hardware bug on the router.
2. Try to find someone in your neighborhood with another model ASUS router and see if they would allow you to connect.
3. On your AC66U, set up a Guest network using a password like 'password' and see if that works. I have seen issues, not on ASUS, where special characters can be the cause.

 

[Marsi4eg]

I tried latest official firmware Version 3.0.0.4.382.50470, reseted it to factory defaults, disconnected anything from the router except the tested phone - and the router still reboots randomly when I manually turn wi-fi switch in the phone to off and then on. Like in 3 of 10 repeats it will happen.

Also I tried to set static IP in saved network profile to make sure that this problem isn't DHCP-related. Router still restarts even with static settings in the phone.

I've also found this (about Pixel phones causing the same):

https://productforums.google.com/forum/#!topic/phone-by-google/JW7fnzOPtDo

This problem exist only on AC66U and N66U, and looks like no other models affected.

 

[ApexRon]

I tried latest official firmware Version 3.0.0.4.382.50470, reseted it to factory defaults, disconnected anything from the router except the tested phone - and the router still reboots randomly when I manually turn wi-fi switch in the phone to off and then on. Like in 3 of 10 repeats it will happen.

Also I tried to set static IP in saved network profile to make sure that this problem isn't DHCP-related. Router still restarts even with static settings in the phone.

I've also found this (about Pixel phones causing the same):

https://productforums.google.com/forum/#!topic/phone-by-google/JW7fnzOPtDo

This problem exist only on AC66U and N66U, and looks like no other models affected.

Excellent find on Google regarding Pixel phones exposing a defect on ASUS routers apparently associated with Pixel ipv6. Can I assume your issue is associated with ipv6 as well? Regardless, you should open a case with ASUS.

 

[Marsi4eg]

I've found some new info on this. HTC U12+ is a very new device, so searching by its name gives almost nothing. But when I started to search for something like "Pixel 2 reboots router" I've discovered a lot of reports and not only on Asus routers, and probably related on global Android Oreo 8.0/8.1 bug with phones doing lookup for Chromecast and other *cast devices in the network on connect and this multicast thing crashes some routers. No IPv6 related.

https://linustechtips.com/main/topic/877211-android-oreo-possibly-knocking-out-wireless-routers/
https://www.zdnet.com/article/google-to-fix-chromecast-wifi-bug-with-update-tomorrow/ (dunno if they released it for all phones)

So I disabled scanning for any cast devices in my phone and will see for router's behavior for a couple of days.

The very odd thing is the problem doesn't show up on cheap router models like tp-links 841, etc. but affects more expensive and powerful ones.

[ADDED LATER]
Unfortunatly that didn't help and my router rebooted twice in the last hour, just because I walk around with my phone at home and sometimes it reconnects.

 

[ApexRon]

Having worked with engineering for two companies, I can say that router companies need to get detailed information from phone companies as to the exact nature of the cause so that they can prevent this from happening.

My AC-86U uses Linux as I suspect other high end routers do and lower end routers use proprietary, non-OS based solutions.

 

[Mikael Kanstrup]

I have the same problem with my AC66 router. First connection with my Android O based phone works fine, disconnecting and reconnecting causes router to restart. I opened the router and found a serial port inside connected to a Linux console. Printing the kernel log and reconnecting my android phone clearly shows the router is rebooting due to a kernel crash. I've contacted Asus support but haven't yet received any answers.

The kernel panic looks like this:

CPU 0 Unable to handle kernel paging request at virtual address 0000000c, epc == c06e93d0, ra == c06f95a8
Oops[#1]:
Cpu 0
$ 0 : 00000000 00000000 00000018 00000000
$ 4 : 00000000 87e56810 86f74138 87778d80
$ 8 : 00000008 86f7413e 00004188 8711e800
$12 : 87923000 87756000 00000000 00000100
$16 : 00000000 00000000 00000000 00000000
$20 : 00000001 00000000 87755000 00000000
$24 : 87778d98 00000000
$28 : 802f2000 802f3cb8 00000030 c06f95a8
Hi : 00000000
Lo : 00000000
epc : c06e93d0 wlc_rxframe_chainable+0x168/0x2e0 [wl] Tainted: P
ra : c06f95a8 wlc_bmac_recv+0x290/0x3b0 [wl]
Status: 1100dc03 KERNEL EXL IE
Cause : 00000008
BadVA : 0000000c
PrId : 00019749
Modules linked in: nf_nat_sip nf_conntrack_sip nf_nat_h323 nf_conntrack_h323 nf_nat_rtsp nf_conntrack_rtsp nf_nat_ftp nf_conntrack_ftp ip6table_mangle usblp tntfs(P) vfat fat ext2 ext3 jbd mbcache usb_storage sg sd_mod scsi_wait_scan scsi_mod cdc_mbim qmi_wwan cdc_wdm cdc_ncm rndis_host cdc_ether asix cdc_acm usbnet ip6t_LOG ip6table_filter ohci_hcd ehci_hcd usbcore jffs2 zlib_deflate zlib_inflate nf_nat_pptp nf_conntrack_pptp nf_nat_proto_gre nf_conntrack_proto_gre wl(P) bcm57xx et(P) igs(P) emf(P) ctf(P)
Process swapper (pid: 0, threadinfo=802f2000, task=802f4188)
Stack : 00008000 87755000 00000000 00000000 00000001 00000000 8711e800 87755000
c06f95a8 c06f95ec c06f97b8 00010000 00000000 00000000 04eb0000 73de6ded
00000000 429c38d4 00008000 87755000 87756000 20000000 c0068000 00000001
802f3d70 8094f038 00000000 c06f9b8c 80300000 802f3d70 04160403 802f3d70
c06f9f00 c063a2c8 87c1c680 80350000 80350000 80350000 8034f9bc 8034ded8
...
Call Trace:
[<c06e93d0>] wlc_rxframe_chainable+0x168/0x2e0 [wl]
[<c06f95a8>] wlc_bmac_recv+0x290/0x3b0 [wl]
[<c06f9b8c>] wlc_dpc+0x1b4/0x520 [wl]
[<c06e74a4>] wl_intrson+0x4a8/0x7a0 [wl]
[<c06e783c>] wl_isr+0xa0/0x4a4 [wl]

Edit: Forgot to mention that I'm running latest stock firmware 3.0.0.4.382.50470

 

[ApexRon]

I have the same problem with my AC66 router. First connection with my Android O based phone works fine, disconnecting and reconnecting causes router to restart. I opened the router and found a serial port inside connected to a Linux console. Printing the kernel log and reconnecting my android phone clearly shows the router is rebooting due to a kernel crash. I've contacted Asus support but haven't yet received any answers.

The kernel panic looks like this:

CPU 0 Unable to handle kernel paging request at virtual address 0000000c, epc == c06e93d0, ra == c06f95a8
Oops[#1]:
Cpu 0
$ 0 : 00000000 00000000 00000018 00000000
$ 4 : 00000000 87e56810 86f74138 87778d80
$ 8 : 00000008 86f7413e 00004188 8711e800
$12 : 87923000 87756000 00000000 00000100
$16 : 00000000 00000000 00000000 00000000
$20 : 00000001 00000000 87755000 00000000
$24 : 87778d98 00000000
$28 : 802f2000 802f3cb8 00000030 c06f95a8
Hi : 00000000
Lo : 00000000
epc : c06e93d0 wlc_rxframe_chainable+0x168/0x2e0 [wl] Tainted: P
ra : c06f95a8 wlc_bmac_recv+0x290/0x3b0 [wl]
Status: 1100dc03 KERNEL EXL IE
Cause : 00000008
BadVA : 0000000c
PrId : 00019749
Modules linked in: nf_nat_sip nf_conntrack_sip nf_nat_h323 nf_conntrack_h323 nf_nat_rtsp nf_conntrack_rtsp nf_nat_ftp nf_conntrack_ftp ip6table_mangle usblp tntfs(P) vfat fat ext2 ext3 jbd mbcache usb_storage sg sd_mod scsi_wait_scan scsi_mod cdc_mbim qmi_wwan cdc_wdm cdc_ncm rndis_host cdc_ether asix cdc_acm usbnet ip6t_LOG ip6table_filter ohci_hcd ehci_hcd usbcore jffs2 zlib_deflate zlib_inflate nf_nat_pptp nf_conntrack_pptp nf_nat_proto_gre nf_conntrack_proto_gre wl(P) bcm57xx et(P) igs(P) emf(P) ctf(P)
Process swapper (pid: 0, threadinfo=802f2000, task=802f4188)
Stack : 00008000 87755000 00000000 00000000 00000001 00000000 8711e800 87755000
c06f95a8 c06f95ec c06f97b8 00010000 00000000 00000000 04eb0000 73de6ded
00000000 429c38d4 00008000 87755000 87756000 20000000 c0068000 00000001
802f3d70 8094f038 00000000 c06f9b8c 80300000 802f3d70 04160403 802f3d70
c06f9f00 c063a2c8 87c1c680 80350000 80350000 80350000 8034f9bc 8034ded8
...
Call Trace:
[<c06e93d0>] wlc_rxframe_chainable+0x168/0x2e0 [wl]
[<c06f95a8>] wlc_bmac_recv+0x290/0x3b0 [wl]
[<c06f9b8c>] wlc_dpc+0x1b4/0x520 [wl]
[<c06e74a4>] wl_intrson+0x4a8/0x7a0 [wl]
[<c06e783c>] wl_isr+0xa0/0x4a4 [wl]

Edit: Forgot to mention that I'm running latest stock firmware 3.0.0.4.382.50470

My personal experience is that ASUS support is sloooooooooooow. It took them several months to find a fix for my issue for which they gave me beta firmware. A month later and still no production level firmware with the fix.

 

[Mikael Kanstrup]

My personal experience is that ASUS support is sloooooooooooow. It took them several months to find a fix for my issue for which they gave me beta firmware. A month later and still no production level firmware with the fix.

Solving this problem (a null pointer dereference in the wifi driver) should be trivial for someone with access to the source code and kernel objects with debug symbols. I don't know of other ways than to have this fixed by Asus.

 

[Marsi4eg]

Oh so great to see someone was able to analyze what causes that crash! I have also reported the problem to Asus (Ukraine), they have answered to me a couple of times with some suggestions, some older firmwares that I should install via recovery and test, and also additional questions. But router crashed on all firmwares.

The last answer I got from them was:

"If such reboots occur only with 1 connected device, the reason may be the driver of the Wi-Fi router module. This is also confirmed by the fact that on different firmwares the router has the same behavior (the drivers of communication modules are provided by the manufacturer, in this case Broadcom, and these drivers remain the same on all versions of the firmware)"

or in original language:

"Спасибо за обращение в поддержку клиентов ASUS
Если такие перезагрузки происходят только с 1 подключаемым устройством, возможно, причиной могут быть драйвера Wi-Fi модуля роутера. Это подтверждает и то, что на разных прошивках одинаковое поведение роутера (драйвера модулей связи предоставляются производителем, в данном случае Broadcom, и эти драйвера остаются одни и те же на всех версиях прошивки)"

Unfortunately I don't know if they forwarded the issue to Broadcom, but now I'll also send them kernel log that
Mikael Kanstrup
posted above. Maybe they will say something on this.

 

[ongame13]

I have the same problem with my OnePlus 6 and Asus RT-N66U. Very frustrating

Edit: This is what Asus responded to my email:

"Thank you for your email.

This issue seems not be caused by the Asus router, but by the two devices you are mentioning. As you describe it, its when your devices connects to your network - its' start draining the network and in the end the router crashes - few months back several router manufacturers including Asus - were the router was drained (that caused network issues with router for other devices) and crashed. That was caused by Chrome Cast in hibernation mode, also a device based on Android - update of the Chrome Cast firmware solved the issue (this is to give you an example). So your problem sound very simular to the example and it's not Asus router that are cause to the problem.

These are the all of the troubleshooting steps you can do on your router.

Check if you the latest firmware installed on the router - login to the router and go to Administration - Firmware Upgrade.

You can do a reset of the router, but since other device has no problem operating in your home network and the network only crashes when you add these two devices you have mentioned - we don't think that will solve the issue. We think only way to fix this issue, is the manufacturers of the devices you mention to update their firmware to their devices. Else it's incompatibility between these to devices and the Asus router - and that we can't do anything for. "

 

[Mikael Kanstrup]

I have the same problem with my OnePlus 6 and Asus RT-N66U. Very frustrating

Edit: This is what Asus responded to my email:

"Thank you for your email.

This issue seems not be caused by the Asus router, but by the two devices you are mentioning. As you describe it, its when your devices connects to your network - its' start draining the network and in the end the router crashes - few months back several router manufacturers including Asus - were the router was drained (that caused network issues with router for other devices) and crashed. That was caused by Chrome Cast in hibernation mode, also a device based on Android - update of the Chrome Cast firmware solved the issue (this is to give you an example). So your problem sound very simular to the example and it's not Asus router that are cause to the problem."

At least the crash I get is not related to that problem. The problem Asus talks about here was caused by a storm (thousands) of MDNS broadcast packets sent out and some routers didn't manage to handle that kind of broadcast packet load. If you run tcpdump/Wireshark or similar on your network you can check if this is the case you have. In any case whatever the devices throw at the router it should of course not crash.

 

[Marsi4eg]

In my case I have reported in my first message to Asus support that I already know about those issues with *cast devices and that I'm sure it's not caused by broadcast storm. If they ask you to check firmware version and do a factory reset - that means it's a first answer and they suggest some basic known things.

 

[indark]

gave away my ac66u to my aunt to replace her wrt54g, but haven't heard her complain about the wifi issue.
she uses HTC ultra or something, i have pixel xl, dont remember seeing that happening.
or maybe it's just not an issue for the phone we have.

I will always give a shot to power cycle and wps method clearing nvram.
it sometimes can help in those awkward situation that persist after factory reset/initialize.
i.e. CPU sit @ 100%, ai-protection signature wont update, bootloop, etc

doesn't hurt to waste another 5 min ;p

power off the router for 30+ sec, hold wps button while powering on
wait till the power led to start flashing/blinking, and then release the wps button
now wait for it finish booting and re-configure everything from scratch.

 

[dsjstc]

I started seeing this behaviour when I upgraded my phone to an LG G7, Android 8.0. It applied to an AC66u and an N66u, on stock, main branch Merlin, and John's fork. It seems to happen in both router and AP modes, although AP is my main mission.

I eventually got the AC66u working acceptably with John's fork, 374.43_2-37LAj9527. I seemed to have to shell in and clear the nvram with `mtd-erase -d nvram` -- the gui's factory reset button didn't do the trick. It also seemed to be necessary to do a jffs format, but I wouldn't swear to it.

I'm having the same trouble with an N66u, and nothing will fix it. Instant crash every time I connect to it . I'm in the process of putting Tomato on it, and would love to hear if anyone has AP mode serving modern phones reliably with a Merlin variant.

 

[L&LD]

I started seeing this behaviour when I upgraded my phone to an LG G7, Android 8.0. It applied to an AC66u and an N66u, on stock, main branch Merlin, and John's fork. It seems to happen in both router and AP modes, although AP is my main mission.

I eventually got the AC66u working acceptably with John's fork, 374.43_2-37LAj9527. I seemed to have to shell in and clear the nvram with `mtd-erase -d nvram` -- the gui's factory reset button didn't do the trick. It also seemed to be necessary to do a jffs format, but I wouldn't swear to it.

I'm having the same trouble with an N66u, and nothing will fix it. Instant crash every time I connect to it . I'm in the process of putting Tomato on it, and would love to hear if anyone has AP mode serving modern phones reliably with a Merlin variant.

I'm sure there are many thousands of users with the exact same configuration as you describe with no issues at all.

Have you tried forgetting the network on the phone and re-associating it after you restart it (the phone)? Is your password using special characters?

After doing the factory resets, are you minimally and manually configuring the router(s) to secure them and connect to the internet?

 

[dsjstc]

are you minimally and manually configuring the router(s) to secure them and connect to the internet?

Hadn't hear that phrase before, but it's a good one. I bypassed the wizard and did a minimal AP config on an L-branch firmware (Firmware:374.43_2-38L4j9527). It's only been an hour, but the N66u now appears to be tolerating connections from the phone.

Thanks for the prod!

 

[giracueca]

I have the same problems here. Using AC66R and a oneplus 6 fone. Never see this before, but I notice that when I turn on wifi on phone and have a notebook near connected by wifi. I need to check this more, but after a "googled" I find this post...

 

[Marsi4eg]

I keep seeing people write about flashing TomatoUSB is the fix. So what can be the actual fix in it? Does Tomato use more stable wireless driver? If so, could it be merged to Merlin firmware (theoretically)?

 

[Marsi4eg]

I have recently tried installing Advanced Tomato and testing it for 2 days (https://advancedtomato.com/downloads/router/rt-ac66u), the latest one: Version 3.5-140
And it really does not have this issue! Router accepts connections of my phone correctly every time and had no kernel panic.

Just wonder if wireless driver from this firmware could be used in Merlin's firmware, because tomato is so unfriendly and poor of some other features that I used to...