David J Taylor wrote:
Thanks for your reply. So I presume it's just a matter of time before
it is automatically sorted out?
Yes, it should be.
As caching is almost mandatory for DNS servers, I hope the cause of
the glitch is soon fixed.
Well, I think caching of negative replies (for more than a few seconds)
is a definate no-no, but it is of course a matter of opinion.
About the glitches: I think they occur because of the way the pool DNS
is implemented.
They just use a standard DNS server and dynamically build a zone file
using a script and load it into the DNS.
During the load, it may reply with nonexisting domain for a very short
interval. When that reply is cached somewhere, you are SOL.
The DNS should be replaced by something that is able to generate dynamic
replies query-by-query.
So, for every query from a user (possibly via a caching server) the DNS
should generate a reply from all information it has available at that time.
This includes not only the database of pool members, but also the
uptodate reachability information, the recent history of replies sent to
users, the source network of the query, etc.
That would make it possible to distribute the load more evenly and to
give out server addresses that are reasonably close to the requester
without requiring all those different names to be figured out by the
clients.
I am not sure how much coding would be required to do this, as I have
not investigated what is already available. Ideally one would pick an
existing "dynamic DNS server" into which the application-specific
address resolver would be plugged using Perl or Python.
Rob
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
