I noticed that queries from Tokyo BETA monitor server to my 2 servers are frequently dropped recently, as shown in the attachments. Queries from the LA monitor are fine.
Is there a problem with the Tokyo monitor?
Thanks, I didn’t know there is a new Tokyo monitor. It seems to have no problems with my NTP-servers, except it doesn’t seem to check them via IPv6 currently (just IPv4).
It’s a test instance for the new monitoring system (hence only active on the beta system)! I’ve been working on streamlining how the monitoring API keys, etc, are managed.
When that’s done and there are more test monitoring systems I’ll work on revising the scoring algorithm to filter out this sort of thing. (In other words, this is a good test case for the new system, stay tuned 
)
1 Like
Good to know, hope the new monitoring system can be pushed to PROD soon.
Regarding the poor score seen by the Tokyo beta monitor, I range some network tests from this beta machine. The path from Tokyo to the two servers listed above seems solid.
You might try running a traceroute to monnrt1.ntppool.net using target UDP port 123.
[The traceroute syntax is OS dependent]. That may show where the loss occurs.
Even if the NTP filtering location is known, little can be done about it.
