This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
Hi, I'm have an RBR50 running Firmware V2.7.2.104 that has been working fine for a couple years now. Very recently, devices connected to the Orbi have started getting failed DNS queries, causing them to claim they're offline. When this happens, I can check the router, and it says it's online, can perform speed-checks, etc. Devices are actually online as well, and can ping/traceroute load pages via direct connection to IP address. If I manually set the clients to point to my ISP (or Google, etc.) DNS servers, they work fine as well.
Running manual nslookup and dig requests to the Orbi gateway address (the one it sets as DNS server, the one that the router uses as DNS cache/pass-through) the lookups always fail.
The problem is 100% with whatever DNS app the router is using. Does anyone have a fix?
Thanks
PS If any Netgear folks read this, please just give us a way to have the Orbi DHCP Server pass the ISP provided DNS instead of defaulting/forcing clients to use the router's IP! This is a bad oversight.
Ran into this issue as well. This Reddit thread has the best workaround I've found for the bug(s) that Netgear pushed out with its most recent release for this model.
There are several options. I think the best one is to use the debug.htm page to enable telnet, log in, and modify the DHCP settings to tell clients to contact external DNS servers directly thus bypassing the failing proxy on the router. This change sticks until the router is rebooted which is rare for me. All devices will renew theri DHCP lease within 24 hours (or less) to get the new setting. I recommend forcing devices to renew their DHCP lease to get the new settings sooner. You can unplug the LAN connection, toggle wifi on/off, reboot a device, or issue the command to renew specific to your device/OS. However if you have one computer plugged directly into a LAN port, one simpler way to reset all the wifi attached device's DNS config is to change the wifi name configured on the Orbi, wait a few minutes, and then change it back. I judged "a few minutes" by monitoring my phone to see when it decided the wifi connection was lost. Once back to the original name, my (Android) phone took a couple of minutes to accept the wifi was stable.
There are also reports that the bug is an interaction issue between the DNS process (dnsmasq) and the traffic measuring/metering feature, and that disabling the traffic measuring feature will avoid the bug. Since I don't *need* the measuring feature, I disabled it just in case a thunderstorm glitches the power, there's a chance the Orbi will remain reliable afterwards without the effort/distraction of the reconfiguration workaround.
Ran into this issue as well. This Reddit thread has the best workaround I've found for the bug(s) that Netgear pushed out with its most recent release for this model.
There are several options. I think the best one is to use the debug.htm page to enable telnet, log in, and modify the DHCP settings to tell clients to contact external DNS servers directly thus bypassing the failing proxy on the router. This change sticks until the router is rebooted which is rare for me. All devices will renew theri DHCP lease within 24 hours (or less) to get the new setting. I recommend forcing devices to renew their DHCP lease to get the new settings sooner. You can unplug the LAN connection, toggle wifi on/off, reboot a device, or issue the command to renew specific to your device/OS. However if you have one computer plugged directly into a LAN port, one simpler way to reset all the wifi attached device's DNS config is to change the wifi name configured on the Orbi, wait a few minutes, and then change it back. I judged "a few minutes" by monitoring my phone to see when it decided the wifi connection was lost. Once back to the original name, my (Android) phone took a couple of minutes to accept the wifi was stable.
There are also reports that the bug is an interaction issue between the DNS process (dnsmasq) and the traffic measuring/metering feature, and that disabling the traffic measuring feature will avoid the bug. Since I don't *need* the measuring feature, I disabled it just in case a thunderstorm glitches the power, there's a chance the Orbi will remain reliable afterwards without the effort/distraction of the reconfiguration workaround.
I'm having the same issues. Every day thereabouts for a week or so my RBR50 has been disconnecting from the Internet with a DNS server error. I've tried contacting Netgear, but, of course, I do not have pay for support service and so no dice. Strikes me that this is their issue to fix since everything was working fine until a week ago. I am the user of their product, not their technician nor someone who desires to be. From my perspective they produce the product, and it should work reliably as advertised.
Since I use this router to service multiple computers and NAS drives, I cannot have it malfunctioning with this frequency, in fact, not at all. If I cannot find a solution soon, I will have to find another router system to cover. Given the Netgear response to this problem, I will likely find another piece of equipment, one with a more user-friendly service policy.
In the meantime, if anyone knows of a fix for this issue, please let me know.
simplest solution i think is to download the previous release of firmware for now. that seems to work for me. sure - not the best idea long-term as the previous release probably lacks some bug fixes and security updates - but at least for now it seems to keep things working - so i'll keep an eye out and try the next firmware release once this issue is resolved.
I had this identical problem recently, also on the 2.7.2.104 firmware, multiple times per day, for about the last week or so, which seems to have been resolved since yesterday by doing a factory reset and reconfiguration. To add to what you've observed, the problem seems specific to DNS queries passing through an Orbi satellite (RBS50). Unplugging the satellite appeared to solve the issue for as long as the satellite remained unplugged (all devices then being forced to connect directly to the router), though, admittedly, I never left it unplugged longer than a few hours to see for sure whether it truly addressed the problem.
It may be that merely doing a factory reset on the satellite is sufficient to fix it. But doing a full factory reset on the router (RBR50) appears to have fixed it for me for going on 24 hours now, though it's a much bigger pain, and, unfortunately, I'd already done a full factory reset on the router before it occured to me to just try it on the satellite. So I don't know that only factory resetting the satellite will work, but it's low-cost and high-reward, so I'd give it a shot before factory resetting the router.
As an aside, the Orbi configuring connected devices to use it as the DNS server, rather than passing on the configured DNS servers, serves multiple purposes. One is it allows orbilogin.com to work for accessing the router's web interface, as someone else pointed out. Another is that it allows DNS recursion, which generally improves performance. If multiple users/devices on your network want to use the same services (FB, YouTube, Netflix, whatever), instead of each device independently querying your DNS server (say, Cloudflare's 1.1.1.1), the first one queries the Orbi, the Orbi queries Clouflare, gets the response, passes it back to the requesting device, and, this is the important part, stores the result in the cache, so when another device makes the same request, instead of it going all the way to Cloudflare a second time, the Orbi already has the answer in the cache, and just immediately responds to the DNS request without needing to query outside your own LAN. In addition to resolving the DNS request to an IP address sooner for subsequent queries for your devices, it also reduces traffic to and queries of the DNS server, improving performance for everyone else. This is also what your DNS server (whether you use your ISP's, or Google's, Cloudflare's, etc.) does, caching results for faster turnaround, rather than having to pass each request up the chain to an authoritative DNS server.
There is next to zero benefit from running your own local DNS caching server. Your ISPs caching name server speeds are sufficiently fast, as are any of the major player's caching resolvers. Additionally, you're not relying on a device that isn't as well maintained as the major player's resolvers. If your Orbi is compromised, your local cache could be poisened by an attacker, opening up all kinds of security issues.
It's fine if you don't know, or simply don't mind using the local resolver, but I prefer systems that at least give me the option to fully configure my local services.
As for using "orbilogin.com" to logon, I've never done that, nor was I even aware it was an option. Simply browsing to the gateway address and/or using the ORBI app is typically how I access the router.
I had tried a factory reset already, and the problem reoccured not long after. (Same or the next day, if I recall.)
For what it's worth, since I have manually set the DNS servers in the udhcp settings, I have had no issues on the network. I'll deal with re-configuring it should the Orbi ever need to be reset.
Thanks. I actually rolled back the firmware on the router, not the satellites, yesterday. The problem hasn't recurred, but if it does, I will try your suggestion. I'm not clear, however, how to find an external DNS server directly. Where does one find these?
Your local cache could be poisoned, yes. But the cache automatically clears itself, either by regarding old entries as stale, and/or by newer cache entries pushing the oldest ones out. You should probably run your own DNS service locally, point the Orbi to it, and have it use DoT/DoH (neither of which the Orbi currently supports) to resolve queries securely from only authoritative DNS resolvers. This is the logical endpoint of your concern about DNS poisoning.
And I discovered after posting my previous reply that the issue was with the traffic meter and daylight saving time settings. The factory reset disabled the traffic meter, which I hadn't gotten around to re-enabling. Subsequently, I restored my saved configuration, disabled DST, changed time zones so the displayed time would be correct, and enabled the traffic meter, and no problems since.
I do not understand why this post was tagged as solved... the proposal of downgrading or turning off the NPT or the traffic meter, access by telnet and reboot the DHCP and the DNS service, define the router as AP, and soo on, is not a solution.
The real solution is to having Netgear releasing a new firmware with a final solution for the DNS issue.
I do not understand why this post was tagged as solved... the proposal of downgrading or turning off the NPT or the traffic meter, access by telnet and reboot the DHCP and the DNS service, define the router as AP, and so on, is not a solution.
This appears to be a philosophical issue. Orbi users were frantically dealing with their network use being interrupted every couple of hours and the only thing they could do was reboot the router. The solutions of disabling automatic Daylight Saving Time or disabling Traffic Meter made the problem stop. (Certainly did for me.) We no longer experience the problem. That seems to be the definition of a "solved."
As for getting Netgear to focus attention on the underlying issue, keeping a post "unsolved" on the community forum is not likely to have any effect. My guess is that letters to Netgear's President or Board of Directors would do more to "goose" people into action. Outraged comments on Amazon or wherever the product was purchased, etc.
Unfortunately it is not an option. I bought this router a couple of years ago and the Netgear only gives me the option to use the forum to have support.
Unfortunately it is not an option. I bought this router a couple of years ago and the Netgear only gives me the option to use the forum to have support.
