Hello, am trying to add new 2-server in the pool . but it showing only Could not check NTP status. My time server is work without issue, my location is Bangladesh. how i fix it?
Use NTP tester like this to make sure your server is currently accessible from Internet first.
Thanks for the pointer. I just tried adding an IPv6 server which webwiz can see and also hit the error message re: checking NTP status.
This isn’t entirely surprising. Ask mentioned in the thread about completing the move of the central pool infrastructure off Equinix Metal that he had not tested NTP server addition.
Hopefully he’ll have time to take a look before long, but his schedule for pool work is mostly weekends.
In the wake of a recently completed infrastructure move, there currently still are some kinks that need ironing out, such as in the flow to add new servers.
Thanks everyone. This was pretty puzzling. I don’t think even it’s a new bug. There was a race condition in the code where the sub-system talking to the MQTT clients (the ntppool-agents) for ad-hoc checks could lock-up.
The relevant code hasn’t changed for 3 years, so I’m not entirely sure if something in the new cluster triggered it or if it’s just a co-incidence.
Thanks! Indeed working at least for me now. 
But there still is some occasional Could not decode NTP response from trace server. Will monitor…
Oh – the trace server is an old fallback that I didn’t setup in the new cluster. When you get that response, if you capture the “traceid” from the HTML (a hidden tag close to the top of the page) then that’ll help track it down.
@ask, here’s a “trace-id” for a “Could not decode NTP response from trace server” error:
b0a37e3dbfe66d5cec0603f5a207454e
Here some more:
43dcd543a183da10682472846260776d
8e502cf6d335c77726e05e23b12c9cc3
d1dc880fd21b6929636b08d57e4c366f
thanks, that’s helpful. I added some more diagnostics as the logs weren’t clear on what happened (the monitor-api returned NTP data): NP::NTP: log details and mark span error on JSON decode failure · abh/ntppool@b43862d · GitHub
Hi all,
Thanks for your work. It is my first intervention on the forum.
I have the same problem with my server, ntp.liant.cloud. It has a good record on NTP Server Online Diagnostics Tool (IPv4 & IPv6) yet, it is rejected on pool.ntp.org: the internet cluster of ntp servers (both IPv4 and IPv6). Both with the error described by @mehedims :
Results
- Could not check NTP status
All the best,
It’s working very well now; I can add it and all.
I’ll do it when it is protected with a secondary electrical power.