moby: Containers name resolution timed out with embedded DNS

I have two docker containers started and linked via docker-compose.yml file:

version: '2'
services:
  srv1:
    image: test-net
    expose:
      - 8000
    command: python -m SimpleHTTPServer
    working_dir: /tmp

  srv2:
    image: test-net
    links:
      - srv1
    command: tail -f /var/log/dpkg.log

networks:
  default:
    driver: bridge
    driver_opts:
      'com.docker.network.enable_ipv6': 'true'

test-net is a simple Ubuntu image with some additional stuff preinstalled.

After I run docker-compose up, I can attach to srv2 container (testdockercompose_srv2_1 in my example) and it looks like everything is set properly:

# cat /etc/resolv.conf

search haze.yandex.net yandex.net yandex.ru
nameserver 127.0.0.11
nameserver <my:host:system:ipv6:dns::1>
options ndots:0

# ping -w 5 127.0.0.11

PING 127.0.0.11 (127.0.0.11) 56(84) bytes of data.
64 bytes from 127.0.0.11: icmp_seq=1 ttl=64 time=0.090 ms
64 bytes from 127.0.0.11: icmp_seq=2 ttl=64 time=0.086 ms

--- 127.0.0.11 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 3999ms
rtt min/avg/max/mdev = 0.065/0.081/0.090/0.011 ms

# ping -w 2 srv1

PING srv1 (172.19.0.2) 56(84) bytes of data.
64 bytes from testdockercompose_srv1_1.testdockercompose_default (172.19.0.2): icmp_seq=1 ttl=64 time=0.132 ms
64 bytes from testdockercompose_srv1_1.testdockercompose_default (172.19.0.2): icmp_seq=2 ttl=64 time=0.114 ms

--- srv1 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1000ms
rtt min/avg/max/mdev = 0.114/0.123/0.132/0.009 ms

But if I try to call name resolution for srv1 directly via host of nslookup I get “connection timed out”:

# host srv1

;; connection timed out; no servers could be reached

# nslookup srv1

;; Got recursion not available from 127.0.0.11, trying next server
;; connection timed out; no servers could be reached

At the same time name resolution works fine if I specify DNS server explicitly:

# nslookup srv1 127.0.0.11

Server:     127.0.0.11
Address:    127.0.0.11#53

Non-authoritative answer:
Name:   srv1
Address: 172.19.0.2

This leads to HTTP timeouts during calls to linked containers. A simple curl call to http://srv1:8000 takes 15 secs, while call to http://172.19.0.2:8000 is instant.

My host system is IPv6-only Linux with docker 1.10.1 installed.

› docker version

Client:
 Version:      1.10.1
 API version:  1.22
 Go version:   go1.5.3
 Git commit:   9e83765
 Built:        Thu Feb 11 19:27:08 2016
 OS/Arch:      linux/amd64

Server:
 Version:      1.10.1
 API version:  1.22
 Go version:   go1.5.3
 Git commit:   9e83765
 Built:        Thu Feb 11 19:27:08 2016
 OS/Arch:      linux/amd64

› sudo docker info

Containers: 9
 Running: 2
 Paused: 0
 Stopped: 7
Images: 88
Server Version: 1.10.1
Storage Driver: aufs
 Root Dir: /var/lib/docker/aufs
 Backing Filesystem: extfs
 Dirs: 93
 Dirperm1 Supported: true
Execution Driver: native-0.2
Logging Driver: json-file
Plugins:
 Volume: local
 Network: host bridge null
Kernel Version: 3.19.0-42-lowlatency
Operating System: Ubuntu 14.04.1 LTS
OSType: linux
Architecture: x86_64
CPUs: 1
Total Memory: 1.954 GiB
Name: var01f.vsfront.yandex.net
ID: YYOZ:TMTI:UE6N:6DUW:S2GP:KGNN:TN3G:CEQO:F2CA:ODQE:FRQH:Q356
Debug mode (server): true
 File Descriptors: 33
 Goroutines: 54
 System Time: 2016-02-15T23:48:08.593476359+03:00
 EventsListeners: 0
 Init SHA1: e1042dbb0bcf49bb9da188176d9a5063cdb92a01
 Init Path: /usr/lib/docker/dockerinit
 Docker Root Dir: /var/lib/docker
WARNING: No swap limit support
Labels:
 provider=generic