Problems with my RT-AC86U

[Rabelais]

Thanks for your help and advice. On another note I have managed to mac filter out the intruding device and the authentication deauthentication requests have ceased in the log (but just because you can't see something doesn't mean it's gone). A reflash wil take me longer than I have available left in today, so I will leave the thing running overnight to see if it falls down again and if so apply your instructions when I have more time tomorrow evening. A nice tight learning curve today, using SSH reminded me of using PAD in the old days. Thanks for everyone's help and advice and I will report back when and if successfully resolved in the hope that others may be helped if they get something similar.
Cheers everybody and thanks again.
Rab

 

[Rabelais]

Well there's nothing in the output to suggest that there's a problem with memory on the router as a whole, so it's hard to guess where that error message is coming from.

I suggest you try the following to try and narrow down the source of the router's instability.

1. Change all WiFi passwords (including guest networks).
2. Disable the media server.
3. Disable Samba.

Thanks Colin I'll add that to my list of things to try before I reflash stock and report back. Though the passwords are all new but I'll disable all the other bits.
Cheers
Rab

 

[Rabelais]

Just to update all the helpful people here, the router fell over harder than a fatman on a skateboard over night with the merlin .19 build. So hard it infact went back to reset screen. So I followed the advice of reflashing with Asus stock and slowly built things back in over the last 24hrs. I am stable for now and I even put back the MESH and crept back the SAMBA. So far all is well, a couple of out of memory entries but I seem to remember that as being a consequence of samba running. All unidentified mac authorisation deauthorisations have disappeared, though some pop up occasionally from the MESH. I was starting to think it was duff power supply, corrupt memory or component failure but it seems to have stabilised. Thank you all for your suggestions and help, god willing it is sorted.
Cheers to all in the community
Rab

 

[cooloutac]

Ran into same exact issues as you, minus the out of memory errors with merlin 384.19 on my new ac86u. But I did notice the router would creep to 80% memory. Disabling samba and media server does seem to keep it to around 70% usage. Probably because of all the dam wlceventd events in the log just like you. In my case a big culprit was my vizio tv. Devices that use a low power/sleep mode setting just does not work well with these asus routers which I find astonishing, whether wired or ethernet. NO idea about your neighbors router though, Its probably normal stuff. My router crashses just like you after a while which I notice from the up time and May 5th date when scrolling past all the many wleventd's.... I keep trying to make it work because I want to use policy rules with the vpn. But stock is way more stable and I think i'm going to have to flash back too and give up.

 

[cooloutac]

what stopped the crashes for me for over 2 days now on 384.19 merlin, is disabling qos and ai trend micro. don't know if its one or both. will report back if this changes.

 

[FanaticLight7]

Just to update all the helpful people here, the router fell over harder than a fatman on a skateboard over night with the merlin .19 build. So hard it infact went back to reset screen. So I followed the advice of reflashing with Asus stock and slowly built things back in over the last 24hrs. I am stable for now and I even put back the MESH and crept back the SAMBA. So far all is well, a couple of out of memory entries but I seem to remember that as being a consequence of samba running. All unidentified mac authorisation deauthorisations have disappeared, though some pop up occasionally from the MESH. I was starting to think it was duff power supply, corrupt memory or component failure but it seems to have stabilised. Thank you all for your suggestions and help, god willing it is sorted.
Cheers to all in the community
Rab

Hey quick question, that unauthorized MAC you saw in the logs, did it ever connect to your network? Like was it in the online or offline client list at all?
I'm just wondering because I had the same issue on my RT-AC3100 with a ton of similar logs of MAC addresses that I have never seen before. I never got any router crashes though. Stuff like this was happening multiple times a day from MAC addresses that were not mine.

Oct 19 12:05:36 syslog: WLCEVENTD wlceventd_proc_event(500): eth1: Auth EA:06:02:EB:04:5C, status: Successful (0)
Oct 19 12:05:36 syslog: WLCEVENTD wlceventd_proc_event(529): eth1: Assoc EA:06:02:EB:04:5C, status: Successful (0)
Oct 19 12:05:46 syslog: WLCEVENTD wlceventd_proc_event(466): eth1: Deauth_ind EA:06:02:EB:04:5C, status: 0, reason: Deauthenticated because sending station is leaving (or has left) IBSS or ESS (3)

It almost sounds like these devices got authenticated with the network which doesn't make sense. Adding a MAC filter fixed it, but as you said I'm not sure if this means it has gone away. I'm more concerned with unauthorized users using my network somehow.

 

[Matthew Patrick]

what stopped the crashes for me for over 2 days now on 384.19 merlin, is disabling qos and ai trend micro. don't know if its one or both. will report back if this changes.

Any changes??

 

[cooloutac]

Any changes??

ya i got everything running with ai protection and qos on now. not sure what changed it. about 3 days uptime now. no dcd crash. all i can think of is a lucky factory reset and then a formate of the jfss partition. i'm wondering if its the order in which I set my options. no idea...

I basically just flashed merlin, factory reset. put a dummy password and ssid names and passwords. formatted jfss. then switched to https. relogged and changed password and relogged again. then set my ssid and passwords. then added ai mesh node. then i'm not sure what I did first. did I set guest network before turning on ai protection and qos. or afterwards not sure. turned on rebind protection and dnsec. but I did set up the vpn last.

I'm also not sure at what point I changed the to specify dns servers. that might be a factor too. I can't remember if I immediately changed off the isp dns or later on. it was definitely before i turned on dnsec.


The only issues i had. is the first day the guest network started deauthenticating everything after 7 hours. which still has me weary. I'm not sure why it happened since I had it set for unlimited time. I had to remove it and re-enable it again.

I also get blog errors now which i'm ignoring. At first I thought this was related to vpn, but after further research seems to be related to ethernet(with ai protect or qos on). and it could be a new alarm system i just installed at my home on ethernet. or maybe its a magickjack ethernet that is at 100MBs. its like one line a couple times a day. I also had some kernel 0800 protocol is buggy errors, which have now disappeared. and i had some vpn replay errors which have not seen for over day and half now. I think these are all things that can be ignored. as long as I don't get kernel or dcd crashes i think its livable and everything is working. . but i still wanna know what happened to the guest network in those first 7 hours. I couldn't see anything in the log cause all the devices flooded the log with constant wlceventd proc events...i'm surprised the router didn't crash. but it hasn't happened since now over 2 and half days with guest network on.

My guess is i'm just being left alone now. lol

 

[Matthew Patrick]

ya i got everything running with ai protection and qos on now. not sure what changed it. about 3 days uptime now. no dcd crash. all i can think of is a lucky factory reset and then a formate of the jfss partition. i'm wondering if its the order in which I set my options. no idea...

I basically just flashed merlin, factory reset. put a dummy password and ssid names and passwords. formatted jfss. then switched to https. relogged and changed password and relogged again. then set my ssid and passwords. then added ai mesh node. then i'm not sure what I did first. did I set guest network before turning on ai protection and qos. or afterwards not sure. turned on rebind protection and dnsec. but I did set up the vpn last.

I'm also not sure at what point I changed the to specify dns servers. that might be a factor too. I can't remember if I immediately changed off the isp dns or later on. it was definitely before i turned on dnsec.


The only issues i had. is the first day the guest network started deauthenticating everything after 7 hours. which still has me weary. I'm not sure why it happened since I had it set for unlimited time. I had to remove it and re-enable it again.

I also get blog errors now which i'm ignoring. At first I thought this was related to vpn, but after further research seems to be related to ethernet(with ai protect or qos on). and it could be a new alarm system i just installed at my home on ethernet. or maybe its a magickjack ethernet that is at 100MBs. its like one line a couple times a day. I also had some kernel 0800 protocol is buggy errors, which have now disappeared. and i had some vpn replay errors which have not seen for over day and half now. I think these are all things that can be ignored. as long as I don't get kernel or dcd crashes i think its livable and everything is working. . but i still wanna know what happened to the guest network in those first 7 hours. I couldn't see anything in the log cause all the devices flooded the log with constant wlceventd proc events...i'm surprised the router didn't crash. but it hasn't happened since now over 2 and half days with guest network on.

My guess is i'm just being left alone now. lol

Let me know how it goes kk? Since I was fine with 7 days uptime. I was like yes I beat it and then it rebooted. And now it's been rebooting after 2 days ++ . Weird but yeah. If me turning off trend micro stuff doesn't fix the reboots. My last thing to do is rollback to 384.18 and fingers crossing it'll be fine.

 

[Cobra Kai]

Let me try to explain this again... I have the AC86u as my main...I have it in a AIM (Ai Mesh - Wireless) with a AC 3100. I have three full time remote learners. I have been rock solid for over 3 months.
I am using AIP (AiProtect) DaveQoS, Diversion, Skynet, etc - on Merlins .19
What worked for me? I rolled back to .15 and slowly migrated to .19. Every step, I followed all the guides and guidance to start fresh and used the recommended settings for wireless settings, naming conventions, etc.

All of these issues went away - no more ghost power downs, won’t power up, random things to take if offline, random wireless drops, etc.

Plus, I got 20 IoT Devices - smart TVs, cameras, Alex etc...

YMMV... Now, back to finding Daniel riding his bike tonight..

 

[cooloutac]

Let me know how it goes kk? Since I was fine with 7 days uptime. I was like yes I beat it and then it rebooted. And now it's been rebooting after 2 days ++ . Weird but yeah. If me turning off trend micro stuff doesn't fix the reboots. My last thing to do is rollback to 384.18 and fingers crossing it'll be fine.

turning off the trend micro probably will fix it. but my whole thing is we pay alot of money for these routers and thats a major feature adding to the price. plus why should our router be more vulnerable.

just logged in today and i don't have any blog errors now. all i did after i posted last night was mac filter both the main 2.4 and 5ghz networks. i forgot to add i also don't use smart connect i made two seperate ssid's and immediately hid both of them. but only one broadcasting now is my guest network with about 20 iot devices without macfilter. just wanted to see if macfilter would cause problems, but now i'm wondering if that fixed my blog errors and it had nothing do with ethernet after all lol. but i doubt it and it makes no sense. whatever the reason now my log is completely error free for the first time ever. maybe it was just a restart of the wireless or a reinitializing of the settings that shook the blog error off. nothing in log now but some link up downs and wlceventd proc events and romast and dnsmasq messages.

i will let you know if any errors reappear or if i go longer then 7 days uptime. so far i'm at 3 and half days with ai protect and qos on.

 

[cooloutac]

@Matthew Patrick you were right. on the 7th day God rests. i was forced to reboot cause the 2.4ghz band stopped working. and not just my guest network, the whole 2.4 ghz band in general. the light was still on and restarting wireless didn't help, i had to reboot the whole router to get it working again. and once again i have no clue of anything in logs cause its all flooded with wlceventd proc events. what a joke.

I'm over it and returning the router while i still have time. its a garbage device. i'll probably never buy an asus router again, shame on them. i find it suspicious and disgusting that they still sell this thing, its unscrupulous.

 

[OzarkEdge]

@Matthew Patrick you were right. on the 7th day God rests. i was forced to reboot cause the 2.4ghz band stopped working. and not just my guest network, the whole 2.4 ghz band in general. the light was still on and restarting wireless didn't help, i had to reboot the whole router to get it working again. and once again i have no clue of anything in logs cause its all flooded with wlceventd proc events. what a joke.

I'm over it and returning the router while i still have time. its a garbage device. i'll probably never buy an asus router again, shame on them. i find it suspicious and disgusting that they still sell this thing, its unscrupulous.

It was already suggested awhile back that your 86U is defective. It happens. To continue complaining about defective hardware is pointless. Exchange it or move on to another product.

OE

 

[cooloutac]

It was already suggested awhile back that your 86U is defective. It happens. To continue complaining about defective hardware is pointless. Exchange it or move on to another product.

OE

they shouldn't even be selling the model man. doesn't matter when or where it was manufactured. You're in denial. I should of known better from all the complaints across the internet and on these very forums. I'm just pushing it too much with all the settings I use and the amount of devices i connect. Its junk. from what I hear though the ac3100 is even worse lol. I'm gonna avoid the company altogether.