NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
RAX120 Router Random Disconnects
Hey everyone, I know that I'm not the only one with problems on the Nighthawk WiFi 6 devices, but thought I'd post my most recent headache. My hardware consists of a Motorola MB8600 DOCS 3.1 modem w/ 1Gbps service from my ISP, the modem links a RAX120 router via a single 3' CAT6 (i know, overkill), and I have a single 100' CAT5E connecting the RAX120 to a RAX80 in AP mode on the other side of my property. I have a dynamic IP address provided to my modem from my ISP, the modem provides a single address to the RAX120 WAN connection via DHCP, DHCP is configured on the RAX120 to provide a separate NAT for all clients, and the RAX80 has a static IP address set within that NAT subnet. When the problem began today the RAX120 was running FW 1.0.1.114 and the RAX80 is running V1.0.1.70_1.0.30. Nothing in my environment has changed for a month and this issue arose out of nowhere.
Anyway, this morning I noticed that none of my TV streaming devices were able to connect to their streaming services. The streaming device in one side of the house is connected via WiFi (802.11ac) and the device on the other side is connected directly to the RAX80. I found that I was unable to ping any web sites (www.google.com) from my laptop when connected via 5Ghz or 2.4GHz on the router or AP, but I was able to ping the IP addresses for these sites (172.217.0.36). I could ping DNS server IP addresses for my ISP and other public DNS server addresses, yet I couldn't query DNS running on the RAX120 to get a non-authoritative response. However, I did get an authoritative response when I ran an nslookup and specified an external DNS server. After logging into the RAX120 I could see that it was pulling the DNS server IP addresses from my modem as expected, which were some of the same that I was able to ping without issue. I checked the logs and nothing abnormal popped up. Just the standard logins of MAC addresses and some BS DoS from known services. So after kicking the RAX120, everything began functioning again and I was able to query the router get DNS answers.
Issue solved with a simple kick of the router right? Wrong… 20 minutes later the issue reoccurred. This time I took it a step further and sequentially powered down and up the hardware. Modem, router, and AP. Again, the issue was resolved for a short time, but then randomly came back. I dug into the RAX80 logs to see if it was causing the pain and saw the following.
[Time synchronized with NTP server] Sunday, Mar 08,2020 19:50:31
[Internet connected] IP address: X.X.X.X, Sunday, Mar 08,2020 19:50:20
[Initialized, firmware version: V1.0.1.70_1.0.30] Sunday, Mar 08,2020 19:50:20
[Time synchronized with NTP server] Sunday, Mar 08,2020 19:47:26
[Internet connected] IP address: X.X.X.X, Sunday, Mar 08,2020 19:45:57
[Initialized, firmware version: V1.0.1.70_1.0.30] Sunday, Mar 08,2020 19:45:57
[Time synchronized with NTP server] Sunday, Mar 08,2020 18:54:28
[Internet connected] IP address: X.X.X.X, Sunday, Mar 08,2020 18:54:08
[Initialized, firmware version: V1.0.1.70_1.0.30] Sunday, Mar 08,2020 18:54:08
[Time synchronized with NTP server] Sunday, Mar 08,2020 14:27:44
[Internet connected] IP address: X.X.X.X, Sunday, Mar 08,2020 14:27:04
[Initialized, firmware version: V1.0.1.70_1.0.30] Sunday, Mar 08,2020 14:27:04
The FW reinitializing without me intervening with a reboot is a bad sign, so I isolated it out for a factory reset and rebuild. As an aside, the RAX80 has been problematic in the past with it basically corrupting it’s ARP table and not forwarding L2 traffic, so I thought it definitely could be horked once again. Nevertheless, after a rebuild and a few hours later the problem resurfaced.
The next troubleshooting step I took was to upgrade the RAX120 FW to the latest security patch 1.0.1.118, which still didn’t resolve the issue permanently. Now I’ve downgraded the firmware back to 1.0.1.108, and stepped back to .114 and .118 hoping it won’t resurface. It’s been about an hour, so my fingers are crossed. I don’t want to go down the path of a factory reset of the RAX120, cuz it’s a major PITA to redo all my settings with ACLs and port forwarding. I could just restore from the backed-up firmware after a reset though, which is the next step I’ll take if this crops up again.
Thoughts? Let me know if you guys have seen oddities like this and if Netgear is doing anything to actually solve customer problems. I liked the RAX120 cuz I have a system connected via the 5GbE port and NAS via LACP and the performance is great. But this is only one of many problems that I’ve had with these devices and I’m on the verge of giving Netgear the finger to go with a 10GbE switch and some Ubiquiti APs… and thrown psfense back in while I’m at it.
232 Replies
The router stayed up all night for about 8 hours and then the same symptoms occured once I logged into my laptop and started checking emails for about 5 minutes. I attempted to login to the RAX120 and it would not accept my username and password... I attemtped a number of times and I was not fat fingering, because I was pasting the password from my PW database. I logged into the RAX80 just fine without any issue. Nonetheless, I just reset my router back to factory and the issue still occured after it came online with a default config. WTH?!?!? Is my modem hosed and causing me the problem? The logs in the modem have nothing telling and only a couple status messages. Possibly it's a coincidence and I wasn't being patient. Eitherway, I reapplied my RAX120 config from a backup and the connections are up again.
Anyone out there have any ideas? Are these Nighthawks trash like the community has been evangalizing over the past year? Or am I just a fool? Even though I've managed numerous other network devices from Cisco, Brocade, Linksys, TP-Link, Apple, and even Extreme Networks without ever seeing an issue like this...
Hello rax120_rax80,
Welcome to the community! Thank you for taking the time to write a detailed report of your experience. We are currently running a beta firmware for the RAX120 that may help resolve your issues. Please let me know if you'd like access to the beta firmware.
Best,
Christian
Sure. I have nothing to loose. My network stability is already lacking. :-D
Please provide access to beta firmware for both the RAX120 and the RAX80, since I operate both devices.
Just an update. After re-applying my configuration this morning the network has been stable since. However, I did disable all device MAC addresses, except for my laptop, via the security ACLs as well. About an hour back I enabled 3x IP cameras, a security NVR, and my alarm system gateway. So far so good. Possibly something on my network is being a bad neighbor after an automagic FW update the night before. I'll find out sooner or later.
The issue is back! The modem ran well all day yesterday, so at about 10pm I decided to allow a few more devices. I allowed an Amazon cloud cam, a thermostat, 5 light bulbs, and 4 smoke detectors. Everything ran until about 1am and the issue resurfuced. Here's the interesting part. If I connect my phone via LTE, I am able to communicate with my alarm system that is physically connected via a CAT5e cable and to the light bulbs that are on the 2.4GHz guest network. However, the thermostat and smoke detectors that are on the non-guest WiFi network are not online. I connected my laptop once again to the non-guest WiFi and I can ping external IP addresses, but I cant resolve hostnames/URLs. The same occurs on my phone or other devices connected to that network WiFi network. I connected my laptop to the guest 2.4GHz network, but I still couldn't resolve anything to the routers DNS. The next time it happens I'll have to connect my laptop via a physical cable to see if I can get DNS resolution, but I'm not optimistic since my streaming TV device didn't work. So here's a few questions.
Why are the light bulbs accessible from outside, inbound to the network when they are on the internal 2.4GHz guest network?
Why is the security system accessible from outside, inbound to the network when it's physically plugged with a CAT5e cable network?
This is getting very tiresome though... Everything functioned wthout an issue for a few months and no changes were made. Then on Sunday morning, it all went to hell in a handbasket.
Christian_R What is Netgear doing for customers? After 90 days you are stuck with no support and a paperwieght cuz of poor FW code.
I am having the EXACT same issue with my new RAX120.
I went through all of the same troubleshooting as you, but below are my findings in a nutshell.
The issue is with the RAX120 providing name resolution....period. I thought it could be bad actors on the network as well, but have rulled that out as you did by only allowing 1 device at a time. (I generally have 50+ devices of all sorts like yourself).
I have Spectrum cable hooked up to a netgear nighthawk cable modem CM1200 (working fine with a ASUS RT-AC3200 running asus-merlin for 8+ months).
My RAX120 receives my dynamic IP from spectrum, but I always specify my DNS after running DNSBench. When my "internet" went down on my laptop, I noticed that web pages wouldn't load, my exchange server connection failed, etc, but services which use IP addresses were still connected (PuTTY, pidgen, etc). So I pinged my old DNS server 4.2.2.2...no issues. Pinged google.com...unable to resolve.
I logged into the RAX120, and changed my DNS servers (I was using a new set based on my DNSBench run so wasn't confident in them). Applied them and was off and running again....for 20 minutes.
Same thing happened...changed the DNS to new ones....applied....20 min later, no name resolution.
As you noted, if you manually add the SAME DNS servers to your IPv4 adapter in your laptop/pc/etc, everything works fine.
I did this for a while and tried other things to get the RAX120 to work with no avail, and then called support (I have never done this in my life (I'm 41) given my line of work, but figured WTH). Needless to say we tried nothing I hadn't already tried, but he did have me change my DNS servers to 1.1.1.1 and 1.0.0.1.
I generally do like these servers for the non-logging aspect, but since they aren't the fastest I don't use them. Well low and behold, I went 16 hours straight with no issues.
I went back into the RAX120 and changed my DNS from 1.1.1.1 and 1.0.0.1 to some of my other ones, and AGAIN within 20 min, I was "down". Changed them back to 1.1.1.1 and 1.0.0.1, and here I am typing this 3 hours after the fact.
I don't know why these seem to be working and others do not. I wish I could just have the RAX120 pass the DNS servers to my devices who request it and not do the resolution itself.
Its hard to come from managed devices and devices that run Merlin to something as "consumer grade" as this. It may be going back to Amazon if Netgear can't fix this asap...
Let me know what else you find, and I'll do the same.
Thanks,
Hanz
Dallas TX
Bahahahaha....I should have known I'd jink myself. After I posed my reply, I noticed an email from netgear and when I clicked on it, none of the images loaded. Sure enough...no name resolution again.
Here is one funny thing though. I logged into the device, went to Internet under the basic tab, and click the bottom Test button. It went to the "Updating" page (not sure why as it should be testing...), but when it finished and popped up the product registration box (I assume this is the test). It's working again. No change to the DNS needed.
So all you have to do is log into the admin interface every 20 min, and "Test" the internet setup....mark this one as resolved...you're welcome!
Good info Hanz. Maybe we can right an API call to test every 10 minutes. :-) I actually did figure out how to make a call to reboot either the RAX120 or RAX80 (I dont remember which one) once a day when I had some crashing issues last year, but that's a pretty crappy solution when they should have a reboot schedule built into the OS. *Hint, Hint netgear give the users more command and control*
My network has been up for almost 5 hours now, but I know it's only a matter of time. I've been given access to the .115 beta so I'll give that a after work tonight and report back.
Ha...I may have to. I work from home and this could end up being a major problem for me.
Ah, scheduled reboot...yet another thing missing from this console. The list of things is growing that I'm accustomed to having. Jeez...this thing may be going back. The Advanced section isn't very advanced. I sent a PM to the person in this thread to try and get the beta as well. Hopefully they respond soon...and hopefully it works.
I can deal with software short-comings as long as the device is reliable.
- I too started having DNS issues a few days ago. Seems to just die and have to reboot each time. Hope they come out with a fix if the beta firmware hasn’t resolved the issue.
DNS drops out after some period of time between 15 minutes and a day.
I found changing the DNS or sometimes just re apply the same one brings it back for a while.
The Google DNS 8.8.8.8 and 8.8.4.4 worked for a while then the same thing.
It dropped again as I was writting this. Hit the Apply button on Internet Setup and its back. Need this fixed. Just installed this Sunday. Buggy from the start. Help please Netgear.
Firstly why have you used a RAX80 which is very buggy instead of an access point, also you have connected a 12 stream router to a 8 stream router which I imagine will cause cause conflicts as well, tbh you would better off having two 12 stream routers, same chipset etc to avoid these conflicts I'm not sure why you have downgraded your Wi-Fi to 8 streams anyway. I have had no DNS issues whatsoever using just the RAX120, neither has a friend using two EX8000 extenders, one of which is in AP mode the other is in wireless mode.
I think a lot of issues are down to different chipsets and different wifi 6 specs and bugs, I would suggest using a RAX80 as a AP is a giant waste of time, an access point like the EAX80 would have been better or two 120s. You basically are running two different protocols and drafts of Wi-Fi 6 and Im not sure they really are compatible with what you trying to do, better off using two R7800's and sticking with WI-FI-5. Also could you please not talk about a piece of firmware that if you were given official access to, means you agreed to the terms of a NDA which you are currently breaking. Please talk about betas in the appropriate forum. Note always factory reset your router when using bleeding edge tech it will save a lot of headaches down the line when features and drivers are being updated.First, I don’t need to explain to you why I am using a RAX80 as an AP. However, since you type with such authority as an ‘expert’ in networking equipment, I can tell you that I received it for FFFFFFRRRRRRRRRRREEEEEEEEEEEEEEEEEEE from a friend who went with another vendor.
Second, I’ve had my RAX80 UNGPLUGGED for all of these tests since Sunday and the issue has repeated a dozen times. Tell me again how the different chipsets are the cause of the issue again?
Third, no one on this forum is discussing the details of the .115 beta firmware, only that it is experiencing the same problem as EVERY OTHER CODE TESTED. Stop being a shill for Netgear.
Fourth, if you read the entirety of the posts you will know that myself, and others, have done factory resets and still experience the problem. We are all happy for you and that your RAX120 has no issues whatsoever. Congratulations.
This thread is for people that are having disconnects due to a DNS resolution failure with the RAX120, or those that are willing to work towards a solution. If anyone wants to post something in a lecturing tone like a university professor, don’t waste ours’ or your time. It’s counterproductive.
Code .108 stopped resolving DNS requests at 1am this morning. It made it about 7 hours. Below is a summary of what we’ve seen.
- After 3/1 many users experienced their network devices being disconnected
- It was determined that internal and external hosts could be reached via IP, but DNS in the RAX120 was not functioning
- The problem occurs when DNS servers are set both statically or dynamically in the router.
- The problem occurs regardless of the DNS servers set statically in the router.
- The problem occurs in code levels .108 through .118
- The problem occurs again after a factory reset
- The time between failures varies and the temporary fix is to restart the router (or dnsmasq service)
Since users were hit with this bug without upgrading or making change to their configuration prior to the impact, this feels like a time bomb bug in the code. netgear Are you going to assist your user base? We're hearing a lot of crickets.
- Not sure it helps. But I noticed I had a DNS issue. So I went through the app to reboot the router from my phone. A few days ago when I used the app it wiped my traffic history.
Today when I reboot the router using the app it factory reset my router. I’m wondering if this is all connected? Up until this morning is been using the web interface to reboot my router. I got lazy today and used the app.- I haven't had an incident in over 24 hours, this may be fixed for me.
I updated to the 118 hot fix firmware and factory resetted the router.
I did not restore from backup, I simply took note of my configuration and manually re-applied it.
So far no issues.I truly hope you guys have found a way to keep this thing running. Mine was a new purchase/install on Monday (3/9). The router came with .114 and had a DNS issue before I upgraded to .118. It had an issue on .118 before and after factory resets. It had an issue with .115 provided through this thread. I wasn't able to have any success, I tried mostly default settings with just a couple wired devices, I tried it setup the way I want with 50+ devices. I set it up the way netgear support wanted....everything I tried gave no joy. Believe, me, I tried everything I could and spend HOURS on this. I wanted it to work badly.
- I have the exact same issue. Running fine since December until this pops up in early March with 0 changes from me. This is bizarre. What happened in March to set this off for everyone? Is it related to daylight savings somehow?
I'm betting there is a code issue that was triggered after the DST change. I've thought about disabling it to see if it makes a difference. Not a big deal if the router clock is an hour off.
Regarding the issue, I'm running for over 24 hours without a failure after a factory reset and manually changing my config to match my preexisting setup. It was a big PITA since I have 75 clients on my ACL, some port forwarding, and reserved IPs. If this becomes a thing that needs to be done every few months, it's going to be a major hassle. Anyway, if I make it through a week with no issues, great. If not, I'll try a cronjob running every 30 minutes with ./etc/init.d/dnsmasq restart. Hopefully that will fix me for the long term until I can deploy a more enterprise class solution.
"What happened in March to set this off for everyone?" (FiatNox)
Just an observation: “Mozilla has started rolling out encrypted DNS-over-HTTPS (DoH) by default for US users of the nonprofit's Firefox browser.” (25 FEB 2020; Firefox now defaults to DNS-over-HTTPS for US netizens)
Also: “In the US, Firefox by default directs DoH queries to DNS servers that are operated by CloudFlare” (Firefox DNS-over-HTTPS)
Is this observation a piece of the puzzle, in a way that has yet to be identified?
- My ax12 has worked flawlessly since October 2019. Yesterday, it started going offline. Why is this? I’ve spent too much money for this to happen. A reboot of the device temporarily fixes this. However, I need a permanent fix?
Hello I am chiming in to see what everyone thinks is the best fix for them. I have the rax120 and determined it is a dns issue. I would have to reboot at least once a day and if i rebooted from the app it would wipe my settings. I changed the dns on the router to 8.8.8.8 and it worked through the night but in the morning it was down again. I installed the 118 fw and didn't seem to help. I will try and factory reset it when I get home...I have done this before but not after installing fw 118. I read someone has a beta fw that they are testing that is not the public beta 118. Has this fixed the issue? This is super frustrating I have over 40 devices and everything I have to reboot or something it causes more issues.
I use my ISP's DNS server, it is a small niche provider called IDNet which use DNSSEC on their dns servers. I have mine set to pick up DNS automatically. I really suggest a factory reset via the pinhole in the router, and since 1.0.1.118 is beta hotfix and there may possibly be issues, although I am not them seeing myself just use your ISP's DNS servers untill the next update, compromise sometimes is the best way with beta firmware. Also contact Netgear if you are within the support period to report this issue bearing in mind Covid-19 will mean delays. Always factory reset after every update and never use a config file to restore your device which will bring back low level code changes into the new firmware update which is is known to cause issues, always lean on the side of caution even if it takles longer to set up.
I just posted this in another thread, but thought I'd throw it up here.
For what it's worth, I think I've found the cause of this on my router. I've had my RAX120 a few months now and haven't had a problem. Like most other posts on the subject my issues started happening early March. I saw another post about firefox switching the default protocol to DNS-over-HTTPS (https://www.computerworld.com/article/3529424/firefox-starts-switching-on-dns-over-https-to-encrypt-lookups-stymie-tracking.html) and tested it using the routers Default DNS server settings. Sure enough, soon after having a couple pages open in Firefox all DNS was broken and thus the 'internets' were unavailable.
I left the setting enabled in Firefox (and stopped using it for now). I switched my default to Chromium Edge and the network has been solid for 24+ hours vs. the 1-2 hours I was getting.
Currently my firmware is rolled back to 108 and it's been solid. After another 24 hours I'll go back to 114 with 118 hotfix. My DNS server is set to cloudflare right now.
Curious on anyone else's testing, have you checked for DNS over HTTPS causing your problems? I've not seen a solid list of who uses this or doesn't use it, but I believe I saw a reference that Netflix in Windows will use DoH as well as Firefox. For the time being I'm avoiding both of those as I have 2 people working from home thanks to C-19. Will try to 'break' DNS resolving again this weekend by doing everything in Firefox.
FYI: Readers of this thread may be interested in OpenDNS Prosumer, a product that uses an IPsec tunnel to encrypt DNS traffic for privacy, and blocks malware domains for security. I have used it since installing a RAX120 a few months ago, and have not observed any DNS problems to-date.
Yeah OpenDNS is nice but it doesn't help with this problem at all since the problem is in the Router Proxying DNS request. If you happen to be manually setting DNS on your downstream devices you could bypass the issue or if your using DHCP server not on the Netgear you may also bypass this particular problem and configure DNS resolver on your endpoints that is something other than the Router address.... If you don't use the router as your DNS resolver this particular issues would not affect you.
