Roger Lynn kirjoitti:
Hi,
My pool server[0] usually agrees with the monitoring server to within
+-2 ms. But at midday GMT on Tuesday there is a sudden 3 ms step in my
monitoring graph, and also on those for several other pool servers that
I've looked at. My server is happy with only small offsets from its
upstreams, but since then there has consistently been a systematic error
of about 2.5 ms on all the graphs I have looked at. A copy of my graph
from about 2130GMT on Wednesday, which clearly shows this, is at
http://rilynn.me.uk/offset.png
I realise this is a very small and insignificant error, well within the
tolerance of the pool system, but it does look a bit odd. Has something
happened to the clock on the monitoring system?
Hi, I don't see such a step in the graphs of my servers (several in fi,
fr, tr). The most probable reason for the step is that routing between
you and the monitoring server has changed, ie. the packets are routed
through some other ISP somewhere along the path. My best guess is that
the monitoring server's clock is OK, and there's very little you can do
about it, unless you happen to work at some of the ISPs along the path.
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool
