On 4-Jun-2009, at 09:10, W.C.A. Wijngaards wrote: > Attila Nagy wrote: >> Hello, >> I've made a capture and according to that, it was about 150 ms >> between >> unbound's query and the response. I think it's pretty normal for >> nameservers on the other side of the globe. > > Yes, however unbound keeps a rtt profile for each server. So if the > server has a very fast RTT then unbound will consider 150msec to be > very > late (for that server), and retries before that time is up. This suggests to me that for large, anycast-distributed servers there is an opportunity for unbound to give up to early whenever services are moved between far-distant anycast nodes. What is the actual algorithm? For example, if the RTT to an anycast server suddenly changes from reliably <20ms to something substantially higher, what is the threshold RTT at which we can expect unbound to stop being able to get answers? Joe