On 4/30/21 10:07 AM, Bjoern Franke wrote:
I would like to test if the issue persists with
disabled qname
minimization
I forgot to mention that the approach probably won't help. We still
have a generic fallback that tries non-minimized QNAME instead of
finishing with negative answer, and there are some additional cases when
minimization gets turned off. I think it covers the typical issues,
and
it's also triggered in this case, as shown in subset of my verbose log:
[21597.06][iter] '_25._tcp.smtp-relay-in-s1.neusta.de.' type 'A' new uid
was assigned .10, parent uid .00
[21597.10][resl] => id: '42106' querying:
'nis-dns-01.neusta-is.de.'(a)'185.48.220.2#00053' zone cut:
'neusta.de.' qname: 'SMTP-reLay-in-s1.NeusTA.De.' qtype: 'NS'
proto: 'udp'
[21597.10][iter] <= rcode: NOERROR
[21597.10][iter] <= retrying with non-minimized name
[21597.11][resl] => id: '44657' querying:
'nis-dns-01.neusta-is.de.'(a)'185.48.220.2#00053' zone cut:
'neusta.de.' qname: '_25._tcP.SMtP-reLay-iN-S1.NeUstA.dE.' qtype:
'A' proto: 'udp'
[21597.11][iter] <= truncated response, failover to TCP
[21597.12][resl] => id: '46774' querying:
'nis-dns-01.neusta-is.de.'(a)'185.48.220.2#00053' zone cut:
'neusta.de.' qname: '_25._tcP.SmTp-rElaY-IN-S1.nEusta.dE.' qtype:
'A' proto: 'tcp'
[21597.12][iter] <= rcode: NXDOMAIN