DHCP check fails after env deployment

[Anastasia Palkina]

"build_id": "2014-08-21_02-01-17",
"ostf_sha": "c6ecd0137b5d7c1576fa65baef0fc70f9a150daa",
"build_number": "464",
"auth_required": true,
"api": "1.0",
"nailgun_sha": "25eba6fbb2047f26d9da4d27ffdb742c9c27832a",
"production": "docker",
"fuelmain_sha": "25a0c228d998707f90e90877559f17817a749d2f",
"astute_sha": "efe3cb3668b9079e68fb1534fd4649ac45a344e1",
"feature_groups": ["mirantis"],
"release": "5.1",
"fuellib_sha": "52f3ebfa968f0338e0584edf47cff10911109de5"

1. Create new environment (Ubuntu, simple mode)
2. Choose neutron, vlan
3. Choose Ceilometer
4. Add 1 controller+mongo, 3 compute+cinder
5. Network verification was successul
6. Start deployment. It was successful
7. Start network verification

Verification failed.
Node Untitled (1f:07) discovered DHCP server via eth0 with following parameters: IP: 192.168.111.3, MAC: fa:16:3e:a7:69:b1. This server will conflict with the installation.
Node Untitled (f6:7e) discovered DHCP server via eth0 with following parameters: IP: 192.168.111.3, MAC: fa:16:3e:a7:69:b1. This server will conflict with the installation.
Node Untitled (b0:0d) discovered DHCP server via eth0 with following parameters: IP: 192.168.111.3, MAC: fa:16:3e:a7:69:b1. This server will conflict with the installation.

I have no node in 2 clusters with MAC 69:b1

Node Untitled (1f:07), Untitled (f6:7e), Untitled (b0:0d) are compute+cinder nodes (node-2,3,4)

root@node-2:~# ip n
10.20.0.2 dev br-fw-admin lladdr 08:00:27:17:6d:0d REACHABLE
172.16.0.1 dev br-ex lladdr 0a:00:27:00:00:01 STALE
192.168.1.4 dev br-storage lladdr aa:5e:51:e5:0c:43 REACHABLE
192.168.0.1 dev br-mgmt lladdr d6:f0:f6:d1:2b:45 REACHABLE
172.16.0.2 dev br-ex lladdr 5a:16:db:b2:94:4f STALE

root@node-3:~# ip n
192.168.0.1 dev br-mgmt lladdr d6:f0:f6:d1:2b:45 REACHABLE
172.16.0.2 dev br-ex lladdr 5a:16:db:b2:94:4f STALE
10.20.0.2 dev br-fw-admin lladdr 08:00:27:17:6d:0d REACHABLE
172.16.0.1 dev br-ex lladdr 0a:00:27:00:00:01 STALE

root@node-4:~# ip n
192.168.1.2 dev br-storage lladdr 3e:1a:17:6c:1f:4b REACHABLE
172.16.0.1 dev br-ex lladdr 0a:00:27:00:00:01 STALE
172.16.0.2 dev br-ex lladdr 5a:16:db:b2:94:4f STALE
192.168.0.1 dev br-mgmt lladdr d6:f0:f6:d1:2b:45 REACHABLE
10.20.0.2 dev br-fw-admin lladdr 08:00:27:17:6d:0d REACHABLE

[root@fuel ~]# ip l | grep 3e:a7:69:b1
[root@fuel ~]#

Logs are here: https://drive.google.com/a/mirantis.com/file/d/0B6SjzarTGFxaWWk1dDJ6ODNPU0E/edit?usp=sharing

[Sergey Kolekonov (skolekonov)]

This problem seems to be connected with your physical network configuration, I haven't reproduced the described issue on my environment.

[Dmitry Mescheryakov (dmitrymex)]

The issue is not related to MOS, moving to Fuel project

[Fabrizio Soppelsa (fsoppelsa)]

I am facing the same problem.

5.0.1
CentOS
HA
Neutron VLAN
Ceph
1 controller node
3 compute nodes
3 storage nodes

Verification failed.
Node Untitled (8d:2b) discovered DHCP server via eth0 with following parameters: IP: 172.17.0.3, MAC: 52:54:00:2c:53:4e. This server will conflict with the installation.
Node Untitled (04:64) discovered DHCP server via eth0 with following parameters: IP: 172.17.0.3, MAC: 52:54:00:2c:53:4e. This server will conflict with the installation.
Node Untitled (bb:ee) discovered DHCP server via eth0 with following parameters: IP: 172.17.0.3, MAC: 52:54:00:2c:53:4e. This server will conflict with the installation.
Node Untitled (12:f0) discovered DHCP server via eth0 with following parameters: IP: 172.17.0.3, MAC: 52:54:00:2c:53:4e. This server will conflict with the installation.

[root@fuel ~]# ip a | grep -B 1 52:54:00:2c:53:4e
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 52:54:00:2c:53:4e brd ff:ff:ff:ff:ff:ff

[Dima Shulyak (dshulyak)]

Anastasia,
192.168.111.0/24 is private tenant network, and 192.168.111.3 is dnsmasq running in namespace on controller.
The only way we can receive answer from this dhcp server, is to generate dhcpdiscover with vlan tag from neutron range, e.g (1000-1030) - and we are not doing this.

I've noticed this happens rarely on virtualbox, maybe it is related somehow to virtualbox network setup.. i am not sure, though

Fabrizio, can you provide more details about your setup?

172.17.0.3 - i guess this is cobbler container ip, but for some reason verification erred.
Please do next thing, and post your results, please
[root@nailgun ~]# cat /etc/fuel/astute.yaml
ADMIN_NETWORK:
  dhcp_pool_end: 10.108.15.254
  dhcp_pool_start: 10.108.15.128
  interface: eth0
  ipaddress: 10.108.15.2
  mac: 64:e2:24:28:32:03
  netmask: 255.255.255.0
  static_pool_end: 10.108.15.127
  static_pool_start: 10.108.15.3

[Fabrizio Soppelsa (fsoppelsa)]

Dima, here there are some additional info:

[root@fuel ~]# cat /etc/fuel/astute.yaml
ADMIN_NETWORK:
  dhcp_pool_start: 10.20.0.128
  netmask: 255.255.255.0
  static_pool_start: 10.20.0.3
  mac: 52:54:00:2c:53:4e
  dhcp_pool_end: 10.20.0.254
  interface: eth0
  static_pool_end: 10.20.0.127
  ipaddress: 10.20.0.2

172.17.0.3 was cobbler

Reproduced on libvirtd (libvirt) 1.1.3.5 starting from a customer's snapshot. Since we later have discovered he made wrong network configurations and I replied them verbatim, I think it's just due to that.