[email protected] said:
> Multiple monitoring should reject more servers. 

That depends.  If you use a logical OR rather than AND, you would keep more 
servers.

I think it depends on what you want to do.  If the problem is somewhere like 
a trans-Atlantic link, you probably want to keep the server.  That's the idea 
that started this discussion.  People who are using nearby servers won't use 
that link so they will benefit by having more servers in the part of the pool 
they do use.


> A different problem is a good way to investigate these kind of funnies and
> try to work out what is going on and what the real problem is.

That would work better if you had servers located closer to the ends of the 
links you want to monitor.  If you setup both servers to use the other as a 
server with noselect, ntpd will exchange packets and log data to rawstats and 
show it in ntpq -p but not use it for timekeeping.  If you graph the data, 
you can see jumps in round trip time.  If you assume the clocks on both 
servers are good, you can measure the transit time in each direction.




-- 
These are my opinions.  I hate spam.



_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to