On Feb 11, 2006, at 7:29 PM, Jeffrey B.Ferland wrote:
Things I've noticed in random boredom:
* The global server count and the total pool server listings differ
by 30.....
About 30 servers aren't in the global pool (but only in the pool for
their respective country and usually continent).
I made this setting triggered by your "netspeed" about a week ago
(see the list archives) which is also when I made the "all servers"
graph to track the difference better.
* The number of registered servers took a very sharp downward spike
about a week ago...
I deleted the backlog of servers that had not been working for months.
If you got an alert about your server not working 62 or more days ago
and haven't fixed it in the meantime (and you got the last "server is
bad" warning more than 5 days ago) then the server is deleted from
the system.
- ask
(the SQL is this:
SELECT DISTINCT s.id, s.admin
FROM
servers s
JOIN scores sc ON(sc.server=s.id)
LEFT JOIN server_alerts sa ON
(sa.server=s.id)
WHERE
sc.score < 0
AND s.in_pool = 1
AND (sa.first_email_time < DATE_SUB(NOW
(), INTERVAL 62 DAY))
AND (sa.last_email_time < DATE_SUB(NOW
(), INTERVAL 5 DAY))
AND (sa.last_score+10) >= sc.score
from here https://svn.develooper.com/projects/ntppool/trunk/lib/
NTPPool/Server.pm
)
--
http://www.askbjoernhansen.com/
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
