Result of the repro:
#2 ipv6 fails and gets into a loop dig @10.241.3.1 -t AAAA math.lab #2a - Good case (dnsmasq 2.85) root@j-dnsmasq-release:~# dig @10.241.3.1 -t AAAA math.lab dnsmasq: query[AAAA] math.lab from 10.241.3.1 dnsmasq: config math.lab is NODATA-IPv6 ; <<>> DiG 9.16.15-Ubuntu <<>> @10.241.3.1 -t AAAA math.lab ; (1 server found) ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 48650 ;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1 ;; OPT PSEUDOSECTION: ; EDNS: version: 0, flags:; udp: 4096 ;; QUESTION SECTION: ;math.lab. IN AAAA ;; Query time: 0 msec ;; SERVER: 10.241.3.1#53(10.241.3.1) ;; WHEN: Tue Jan 11 13:09:21 UTC 2022 ;; MSG SIZE rcvd: 37 #2b - Bad case (dnsmasq 2.86) root@j-dnsmasq-proposed:~# dig @10.241.3.1 -t AAAA math.lab dnsmasq: query[AAAA] math.lab from 10.241.3.1 dnsmasq: forwarded math.lab to 127.0.0.53 dnsmasq: query[AAAA] math.lab from 10.241.3.1 dnsmasq: forwarded math.lab to 127.0.0.53 dnsmasq: query[AAAA] math.lab from 10.241.3.1 ... dnsmasq: Maximum number of concurrent DNS queries reached (max: 150) dnsmasq: config error is REFUSED ... repeats infinitely There is no ipv6 answer configured, so it is ok to return an empty/bad answer. But dnsmasq should not fall into a forwarding loop right? Or does this test make a hard configuration mistake (always did) that now in 2.86 exposes an issue, but was always wrong? The original test command of `resolvectl query math.lab` gets into a "wait until timeout" with the new behavior. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1957086 Title: Autopkgtest systemd fail against dnsmasq 2.86 (22.04) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/dnsmasq/+bug/1957086/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs