As an experiment I removed a couple of my servers from the pool. And as expected; even after a couple of weeks, quite a few queries are still coming in:
That is why I wonder if it wouldn’t be better to recommend the pool directive first in pool.ntp.org: How do I setup NTP to use the pool? and only describe the server directive as an alternative method.
What do you think?
Hi Marco,
fork https://github.com/abh/ntppool
sed -i 's/server/pool/g' ntppool/blob/main/docs/ntppool/use/sample-config.html
create an issue and PR
Yea, i think it would be better to update the docs to use the pool directive. Maybe need to check if my github account still exists…
This is very normal, and even if most Linux systems used the pool directive (most new installs do this by default; I can attest to Ubuntu, Debian, & CentOS doing it), there are lots of embedded devices out there which are rarely rebooted and won’t pick up a new IP from the pool. Updating the docs would be a great step, but I don’t think it will stop traffic fully.
Operators of Linux mirrors will be used to this sort of behaviour: there are disturbing numbers of systems out there still requesting updates for distro versions from 2008.
I agree. In the spirit of what various Linux distributions currently do, the instructions should probably be changed into:
pool 2.ntp.pool.org
It wouldn’t hurt to mention that the instructions also apply to chronyd, the default NTP provider on Fedora et al.
Patches welcome; I agree the pool configuration is better (and hopefully nobody is touching configuration anymore on systems so old they don’t support that).
Well, there has been no activity on the PR for over a week now. All suggested changes have been integrated. Do I need to request a final review from anybody?