Thus said Ken Hornstein on Thu, 26 Oct 2017 14:22:37 -0400: > There is some variance here. It seems that the DNS TTL for > 'cyrus.andrew.cmu.edu' is set to 1 (!) second, so if it has to do a > DNS query it takes noticably longer:
That's a bit extreme... I see that they have an MX record that is considerably longer: $ dnsqr mx cyrus.andrew.cmu.edu 15 cyrus.andrew.cmu.edu: 212 bytes, 1+6+0+0 records, response, noerror query: 15 cyrus.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-06.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-02.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-03.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-01.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-04.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 21519 MX 10 andrew-mx-05.andrew.cmu.edu And the A record for at least one of those hosts is similarly much longer: $ dnsqr a andrew-mx-06.andrew.cmu.edu 1 andrew-mx-06.andrew.cmu.edu: 61 bytes, 1+1+0+0 records, response, noerror query: 1 andrew-mx-06.andrew.cmu.edu answer: andrew-mx-06.andrew.cmu.edu 21455 A 128.2.157.74 But the A record for that name is only 1 second... strange: $ dnsqr a cyrus.andrew.cmu.edu 1 cyrus.andrew.cmu.edu: 86 bytes, 1+3+0+0 records, response, noerror query: 1 cyrus.andrew.cmu.edu answer: cyrus.andrew.cmu.edu 1 A 128.2.158.26 answer: cyrus.andrew.cmu.edu 1 A 128.2.105.45 answer: cyrus.andrew.cmu.edu 1 A 128.2.158.29 I wonder what the purpose of such a configuration might be. Clearly there's a difference between their MX for the domain and the A for the domain itself so one wonders what is on cyrus.andrew.cmu.edu that they don't want DNS clients to cache those records, but don't care if the MX records are cached much longer. Andy -- TAI64 timestamp: 4000000059f235a2 _______________________________________________ Nmh-workers mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/nmh-workers
