Administration Panel fails to load Rapps and NTP differential at 19K minutes

[ketheriel]

Hi,

I have the R2 in bridge mode with my operator equipment; The router works fine after a cold reboot for 1 or 2 days, which after I get very insane latencies, the backend administration panel fails to load Rapps with a message 'unable to bind'; I can't access it only after a cold reboot, and I'm afraid I loose a large chunk of the log with possible good information. Another common message I see is a time diferential of 19K minutes on NTP... The configuration is pretty much standard.

Any word on how to work around this? Does the next firmware update provides any solution for any of this problems? Is there a way to access the full logs? Is there a way to do crash dumps to a network storage so I can send you?

Thanks...

[Netduma Fraser]

Could you give us an example of the latency you experience when that occurs? Can you provide a screenshot of the unable to bind message? Can you give us an example regarding NTP?

No way to transfer to network storage. You can download the log from system information. Once we get some more information we can work towards fixing the issues if they're not already resolved.

[ketheriel]

3 hours ago, Netduma Fraser said:

Could you give us an example of the latency you experience when that occurs? Can you provide a screenshot of the unable to bind message? Can you give us an example regarding NTP?

I have in-game footage if you want, but there's no metrics on it. I can't provide screenshots because I did factory reset this afternoon (all physical ports stopped responding) and only my cell could connect to wifi (not even laptop). The current log has information from the reset onwards. I wasn't aware the log could be downloaded, so when it happens again I will provide it to you. I did used your tools now, and got what is in the attached file (after factory reset, nothing special there as expected). If you want latency tests, I have a few servers here and there and while iperf3 isn't designed to test latency I can try a few things. If you have any suggestions that help for testing, feel free to share. Router tools are probably not of much help as they get unnavailable.

3 hours ago, Netduma Fraser said:

No way to transfer to network storage. You can download the log from system information. Once we get some more information we can work towards fixing the issues if they're not already resolved.

As I said above, factory reset worked against us. I will download log when it happens. I did changed one thing though, I am no longer on bridge mode with the fiber modem, it's just now a client with only the playstations attached to it.

Also regarding the attachment, it's not happening now, but in the past I did actually got once a lot of packet loss in the ping test. I don't use any special features of the router, I only use geo filtering for a few hours during the week for a special gaming session I do with some friends, other than that it's always disabled.

[Netduma Fraser]

The fact that you're not using QoS and that the upstream router devices are pulling away bandwidth from the gaming device means you can't control it so that's probably why you're seeing some latency - no clips needed. If you can get us the logs, some stats like when you're having packet loss etc when it next happens that would be great. No need for iperf.

[ketheriel]

On 9/17/2020 at 8:58 PM, Netduma Fraser said:

The fact that you're not using QoS and that the upstream router devices are pulling away bandwidth from the gaming device means you can't control it so that's probably why you're seeing some latency - no clips needed. If you can get us the logs, some stats like when you're having packet loss etc when it next happens that would be great. No need for iperf.

I've contacted my ISP regarding this also. I followed their recommendations and switched the bridge port from LAN1 to LAN4. I've done further testing and I found a cute stuff, in previous modem/routers from operator I had, when bridge mode was enabled all the IP functionality would go to the bridged device, with this one, I get a public IP on the bridged device but the isp router/modem keeps it's own public IP and functionality. So in reality I have two public IP's... 

I've gone through more trouble here and I found a consistent line when the service drops in the netduma logs:

> quantum of class 60001 is big. Consider r2q change.

When this happens, my connection goes away; 

I did some research on this, and found interesting stuff on netgear side which seemed to be related to NTP (I did not see the NTP error line again):

https://community.netgear.com/t5/Hardware-VPN-Firewalls-and/HTB-quantum-of-class-11024-is-big-Losing-internet/td-p/1538066

https://community.netgear.com/t5/Hardware-VPN-Firewalls-and/SRX5308-Bug-HTB-quantum-of-class/td-p/1143086

I wonder if it's related  Anyway I can change the NTP servers? I would like to test using my company's NTP servers or CentOS/Red Hat NTP servers (as both options have been proven working perfectly).

NMM

[Netduma Fraser]

Okay that's quite interesting, thanks for that. There is no way to change the NTP servers currently but I'll add that to our roadmap. Perhaps your time zone is incorrect, in the top right of the interface click the 3 dots icon and then click Advanced > Timezone. Pick your correct timezone and toggle daylight savings on as well and see if you notice a difference then.