Dave, See my comments below.
David L. Mills wrote: > Henk, > > Mitigation code is already in the daemon, but it is really useful only > for the manycast scheme. There is even a new pool command, but I haven't > checked it out. > > There is a mysterious server option you might not know about called > preempt. Very crudely, at least now, configure say ten pool members and > specify this option on the server command lines. What should happen is > that all ten will be mobilized and the usual pruning process will select > several of the best. A system of timers is used to manage the > population at a selected level. To do that requires that we keep more than one IP address returned by the DNS. That doesn't happen today. I'm just about to start work on fixing that and I will need your advice later on when we get to the point that we have them in a data structure. > Don't take this seriously just yet. While it's running now with > manycast, and relatively short timers, to make it work properly with the > pool, it needs an asynchronous resolver or at least a way to call the > resolver during regulat operation. We will be able to do this but the asynch resolver is still an issue that we have not resolved. > Sachim was working on these issues and was to make a candidate list from > the all the poolsters in the DNS reply, not just the first one. The > paint is still wet. There are more refined ways to skin the cat, but > having said that, the cat perched next to me is giving a very strange look. > Yes, but this is not simple and I don't think Sachim got this started. Danny > Dave _______________________________________________ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
