nova-deployment-tool

network unreachable to the instance

Asked by Soramichi AKIYAMA

Hello.

I used deployment tool to install Nova into a cluster, which consists of five Ubuntu 10.10s.
When I call `euca-describe-instances' command right after the installation,
there seems to be one instance running.

nii@nova01:~$ euca-describe-instances
RESERVATION r-akj0oymc IRT default
INSTANCE i-00000001 ami-hgr5s3pm 10.0.0.3 10.0.0.3 running mykey (IRT, nova03) 0 m1.tiny 2011-02-01 19:25:52 nova

However, I cannot establish connection to the instance.

nii@nova01:~$ ping 10.0.0.3
PING 10.0.0.3 (10.0.0.3) 56(84) bytes of data.
From 10.0.0.1 icmp_seq=1 Destination Host Unreachable

nii@nova01:~$ ssh 10.0.0.3
ssh: connect to host 10.0.0.3 port 22: No route to host

The network circumstances and deploy.conf are as follows:
(It seems that nova01 and i-00000001 are directly connected by br100).

nii@nova01:~$ ifconfig
br100 Link encap:Ethernet HWaddr 00:21:85:ce:bd:11
          inet addr:10.0.0.1 Bcast:10.0.0.15 Mask:255.255.255.240
          inet6 addr: fe80::221:85ff:fece:bd11/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:60 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B) TX bytes:2736 (2.7 KB)

eth1 Link encap:Ethernet HWaddr 00:21:85:ce:bd:11
          inet addr:10.1.2.128 Bcast:10.1.2.255 Mask:255.255.255.0
          inet6 addr: fe80::221:85ff:fece:bd11/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:126931490 errors:0 dropped:0 overruns:0 frame:0
          TX packets:50129422 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:10079531230 (10.0 GB) TX bytes:4438950677 (4.4 GB)
          Interrupt:19 Memory:dfb80000-dfba0000

lo Link encap:Local Loopback
          inet addr:127.0.0.1 Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING MTU:16436 Metric:1
          RX packets:74774446 errors:0 dropped:0 overruns:0 frame:0
          TX packets:74774446 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:5303999549 (5.3 GB) TX bytes:5303999549 (5.3 GB)

vlan100 Link encap:Ethernet HWaddr 00:21:85:ce:bd:11
          inet6 addr: fe80::221:85ff:fece:bd11/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:9796 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B) TX bytes:645336 (645.3 KB)

nii@nova01:~$ sudo brctl show
bridge name bridge id STP enabled interfaces
br100 8000.002185cebd11 no vlan100

nii@nova01:~$ sudo route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
10.0.0.0 0.0.0.0 255.255.255.240 U 0 0 0 br100
10.1.2.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1
0.0.0.0 10.1.2.2 0.0.0.0 UG 100 0 0 eth1

nii@nova01:~$ less DeploymentTool/conf/deploy.conf
nova_api=nova01
nova_objectstore=nova01
nova_compute=nova02,nova03,nova04
nova_scheduler=nova01
nova_network=nova01
nova_volume=nova05
euca2ools=nova01

mysql=nova01
puppet_server=nova01

libvirt_type=kvm
network_manager=nova.network.manager.VlanManager

network_interface=eth1

ssh_user=nii

Question information

Language:
English Edit question
Status:
Answered
For:
nova-deployment-tool Edit question
Assignee:
Shigetoshi Yokoyama Edit question
Last query:
Last reply:
Revision history for this message
Shigetoshi Yokoyama (yoko) said : 		#1

Could you follow the following procedure to make the problem?

1. Direct cause
According to the following messages in console.txt, the IP address couldn't be gotten from DHCP server(dnsmasq on nova-network server).
console.txt
{{{
udhcpc (v1.17.2) started
Sending discover...
Sending discover...
Sending discover...
No lease, forking to background
}}}

2. Underlying cause
As to the reason that IP address couldn't be gotten from dnsmasq on nova-network server, the following ones may be possible.

2.1 Service dnsmasq wasn't started.
With command "sudo ps aux | grep dns", you can confirm the dnsmasq process. There should be two processes.

2.2 The setting of interfaces was wrong.
The value of network_interface in file "conf/deploy.conf" is eth1, but you should confirm nova-network and nova-compute servers are using eth1.

2.3 The setting of bridge or vlan was wrong.
Please confirm the following interfaces exist in nova-computer server(nova03).
br100
vlan100
vnet0

And please confirm the following bridge exists too.
bridge name................interfaces
br100 vlan100 vnet0

2.4 The switch couldn't handle packets with vlan tag.
Please do the following vlan test.

a. Execute the following command in nova-network server.
vconfig add eth1 10
ifconfig eth1.10 192.168.3.1 up

b. Execute the following command in nova-compute server(nova03).
vconfig add eth1 10
ifconfig eth1.10 192.168.3.2 up

c. Execute ping command in nova-network server.
ping 192.168.3.2

d. Execute ping command in nova-compute server(nova03).
ping 192.168.3.1

Can you help with this problem?

Provide an answer of your own, or ask Soramichi AKIYAMA for more information if necessary.

To post a message you must log in.