Server score still flapping severely due to timeouts from Newark

Server operators
#1

Hello folks, as many others have stated, there seems to be a network error in either the Newark monitoring node or a transit provider. IPv6 behaves normally while IPv4 does not.

Here’s data from my server: https://www.ntppool.org/scores/195.128.100.150 https://web.beta.grundclock.com/scores/195.128.100.150

I can see that this issue has been discussed since last year. Will an administrator from the Pool Project fix it? Can the misbehaving node be replaced?

The project is awesome but this error is not cool at all.

#2

The NTP pool project is in the mercy of the transit ISPs. Some are maintaining NTP packet rate limit filters to control the effect of some “supposedly real” DDoS.

#4

No chance until the monitoring system will be migrated to the beta system. Even after that you will not see the 20 all the time, but it will be better.

The reason is that the Newark monitoring system uses a network which has Zayo as one of the strongest BGP peers and Zayo is known for dropping NTP packets.
The route from Netcup to EWR goes over Zayo at one point. You should see this by running traceroute 139.178.64.41

If you check the forum there are several posts about this issue.

#5

I just recently joined the pool and I have seen these issues firsthand. from nothing happening to -100 in just a couple hours, but the new one just reports steady 20 (I think I deserve it since It’s always online andcorrect time:
main: https://www.ntppool.org/scores/50.35.68.36
test: https://web.beta.grundclock.com/scores/50.35.185.215

#6

Hi @tennisbowling, welcome to the forum!
What do you mean by “new one”? Isn’t your configuration using static IP?

#7

Sorry, I mean the beta site