| 答复: 答复: [chrony-users] about nts |
[ Thread Index |
Date Index
| More chrony.tuxfamily.org/chrony-users Archives
]
Yes, this test requires a description document. In addition, during the test, I find that if the interval of the ntsrefresh parameter is set to a small value, the NTS function of the client will be abnormal. Therefore, I suggest that the minimum value be limited in the man manual.
Another problem is that after ntsdumpdir is configured on the client, cookie files are not generated sometimes. Why?
-----邮件原件-----
发件人: chrony-users@xxxxxxxxx [mailto:chrony-users@xxxxxxxxx]
发送时间: 2022年11月7日 17:27
收件人: chrony-users@xxxxxxxxxxxxxxxxxxxx
主题: Re: 答复: [chrony-users] about nts
Invariably there are people who have to setup an NTP time synchronization system for their company, and they have to make a test plan. Usually that test plan includes configuring everything correctly and watching it sync, then yank the time reference by a big offset and expecting the clients to follow the leader.
I have seen the same discussion on mailinglists or newsgroups about ntpd.
It is always the same thing: rather than realizing that their test plan was wrong (which is probably difficult because it was approved before), they continue tweaking the system to make it perform in the way that was written in the test plan.
Which is then inferior to how it would have performed with the default settings.
It is sad, but what can be done about it? Maybe a "standard test plan" should be written?
Rob
On 11/7/22 09:00, Miroslav Lichvar wrote:
> If you intentionally set your clock to wrong time to see if chronyd
> can bring it back, you need to allow the step to happen.
>
--
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "help" in the subject.
Trouble? Email listmaster@xxxxxxxxxxxxxxxxxxxx.