Re: djbdns/dnscache poisoning weakness
Jeff King <[email protected]>
| Newsgroups | gmane.network.djbdns |
|---|---|
| Message-ID | <[email protected]> |
On Wed, Feb 11, 2009 at 12:45:50PM -0600, David Nicol wrote: > On Tue, Feb 10, 2009 at 3:47 PM, Jeff King <[email protected]> wrote: > > I wanted to > > make clear why the patch was written the way it was. > > Treating dnscache as a black box, the incoming/outgoing distinction > evaporates. For some definition of black box. If it is: incoming outgoing queries queries clients -------> dnscache -------> authoritative servers black box then no, it's not the same. Each incoming query may generate an arbitrary number of outgoing queries with arbitrary contents. So limiting the incoming queries does _not_ limit the outgoing queries. But if you mean "dnscache thinks of queries from clients and queries it wants to make as a result of the resolution algorithm both as incoming", then yes, those are the same as the outgoing queries. -Peff