That’s a good idea in theory, but it would be pretty rough for some people if their traffic goes way up as the system adjusts…
Could perhaps be an “opt-in” model?
You didnt even make a dent in the server by adding .br, .cn and .tr. - it only took a hit of 5-20Mbit 
It is one of Leo’s NTP servers and it has a dedicated 100mbit fiber connection. There is still capacity left, so if there is a need you can add it to other countries as well. What about eg. India? They seem to have only 5 NTP servers.
Hi @ask
Is there other countries in need of additional servers? I was thinking about India.
If so, please feel free to add to this one.
http://www.pool.ntp.org/scores/5.103.139.163
Br,
I would like to mention that the China zone would still benefit from new servers. One of my servers is in Singapore and also serves the China zone. Singapore itself seems to be a fairly busy zone as well. The server used to have a 2TB/month transfer limit, but seeing that the demand is much higher than that, I upgraded to a 4TB/month plan a few weeks ago. I switched my netspeed setting in the pool to 1000Mbit, and for the last week I have been sending 200+GB of NTP traffic per day. That would mean 6+TB/month, so I’ll need to adjust my netspeed setting a bit lower to stay below 4TB/month.
I would not mind if others took a slice of this traffic.
1 Like
Done! I added it to in, too.
Any suggestions for what to name the “I volunteer my server to extra busy underserved zones”-zone? “overflow”? “overflow-optin”? (It will show up on the website next to the country/continent ones).
It was 5.103.139.163 I volunteered to serve in the .in pool 
, 77.213.68.1821 should only be in .cn - it is an ordinary consumer ADSL router that cannot cope with this amount of traffic, so it is thrown out of the pool every time is has gained “10+”.
I am not sure overflow-optin is descriptive, what about “Assist underserved regions”.
Re naming. If you are looking for a short term to complement the current europe, north-america, oceania etc zones, perhaps just plain “extra”? overflow would also work. relief-valve? superhero? reserve? boost? servers-without-borders? reinforcement?
Zones: @ asia cn sg fill-in-the-word-here
I know this is difficult, sorry I can’t help you here.
1 Like
What about “manually-added”?
It seems like the .cn zone is slowly failing again. If we look at the statistics from the last year there is been a significant drop in servers over the year.
Not only cn pool affected. Even jp pool shows obvious drop in ipv4 server counts over the last year (not counting the spike). Right now tw pool has only 3 ipv4 servers published; mine got kicked out at 768kbps… Several months before I could maintain publishing at 25mbps. Sigh.
In India zone is also a disaster. Even at the 384k setting in (IP 4 and 6) my Bangalore server is going to bust it’s 3TB traffic cap this month. With no way to keep it in budget I just scheduled it’s deletion from the IP4 pool.
In India there is even fewer servers than in China and a similar sized population, so traffic must be very high.
Right now, there is 7 servers operational in China and http://www.pool.ntp.org/scores/5.103.139.163 is receiving about 70Mbit/s of NTP traffic, so it will be very difficult for someone to join and maintain a position.
Two new China servers in Shanghai
1.networktime.org (120.132.6.211)
2.networktime.org (120.132.6.225)
ref GPS
1 Like
We have massive increase in traffic on two of our Chinese zone servers since last few days.
We have reached full wirespeed on the interfaces (over 100ksps)
http://leobodnar.com/LeoNTP/all-LeoNTP-servers.php#2d
Is anything significant going on with NTP in China?
Leo
I think it is because several servers has either left or has been kicked out due to the high demands.
We really need some mechanism to restrict times of a particular server entry being returned by pool dns to clients. Burdening those not-so-strong servers will result in eventual pool collapse.
We saw LeoBodnar’s tweet yesterday and so added a stratum-2 node (preferring our LeoNTP GPS receiver on the same network) to the China pool, with a notional bandwidth setting of 1Mbit/sec — I understand that this is a “weight” and I didn’t want to pick a setting which would pull loads of traffic straight to our new node. That’s also why we run this NTP server on a separate IP address — in case we needed to nullroute it with upstreams due to DDoS.
With 1Mbit/sec set, we’re currently receiving around 10k requests per second, peaking at 40k requests per second. That’s a bandwidth usage of around 25Mbit/sec at 95th percentile (we are an ISP, we pay for transit links with CDRs and 95%iles) — which equates to about 8Tbytes/month for people billed that way by VPS providers.