max.grobecker at ml.grobecker.info
Sat Jul 25 20:12:44 UTC 2015
Am 21.07.2015 um 23:54 schrieb Arnold Schekkerman:
> They have more control over NTP traffic (and possible abuse or other issues) if
> they support the pool with dedicated resources. Another reason might be that NTP
> traffic to an IP can stay for a long time after you stop being a customer. The next
> customer obtaining your former IP may not like the residue NTP traffic. In this,
> NTP is different from other traffic.
I used to provide a Stratum 1 time server (DCF77) to the pool, configured with 1M bandwidth.
Since the hardware is not working anymore, I removed this server from the pool. In September. Last year.
Today I can see frequent requests coming from about 15-20 different source IPs...
Even the host name does not resolve anymore, so I assume I'm seeing traffic from very long running systems (or those
with static configured IPs instead of host names).
That's not a real problem and it does not consume any data volume. But what if I configured the server to have 100M bandwidth?
My pool servers are virtualized with KVM on my very own KVM hosts. Monitoring says, the server is pretty accurate and provides stable time
for the last 2-3 years. But I wouldn't setup a pool server on a leased virtual server from another provider. I see, that some hosters are doing
weird things on limiting/blocking UDP traffic directed or originating from virtual servers...
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 819 bytes
Desc: OpenPGP digital signature
More information about the pool