Ask Your Question
0

Can't access VM after successfully created, IP is not assigned

asked 2013-10-06 21:49:42 -0500

Chien Q. Vu gravatar image

Hi all, I followed the guide posted here https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide/blob/OVS_MultiNode/OpenStack_Grizzly_Install_Guide.rst (OVS_MutliNode) for 3 machines (1 controller, 1 network and 1 compute), and successfully created my instance from the Horizon dashboard.

However, I can't access the vm (ping/ssh) from the network node directly

vuquangchien@network:~$ ping 50.50.1.1
connect: Network is unreachable
vuquangchien@network:~$ ping 50.50.1.2
connect: Network is unreachable

I also try to follow the pinging method from the http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_operate.html (official OpenStack documentation - Basic Installation Guide for Ubuntu 12.04 (LTS) and Debian Wheezy): Section 3. Create Your First VM; Procedure 3.4. To SSH into the launched instance. This time I was able to ping the interface, but not the VM itself

vuquangchien@network:~$ ip netns
qrouter-26f355ef-cad2-4ccd-a69e-233aabe66393
qdhcp-404bcb04-842b-499e-ba7a-72780e45edec

vuquangchien@network:~$ sudo ip netns exec qdhcp-404bcb04-842b-499e-ba7a-72780e45edec ping 50.50.1.1
PING 50.50.1.1 (50.50.1.1) 56(84) bytes of data.
64 bytes from 50.50.1.1: icmp_req=1 ttl=64 time=0.099 ms
64 bytes from 50.50.1.1: icmp_req=2 ttl=64 time=0.078 ms

vuquangchien@network:~$ sudo ip netns exec qdhcp-404bcb04-842b-499e-ba7a-72780e45edec ping 50.50.1.2
PING 50.50.1.2 (50.50.1.2) 56(84) bytes of data.
From 50.50.1.3 icmp_seq=1 Destination Host Unreachable
From 50.50.1.3 icmp_seq=2 Destination Host Unreachable
From 50.50.1.3 icmp_seq=3 Destination Host Unreachable

So my thought the VM doesn't get a valid IP. I also follow this thread https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide/issues/127 (Can't ping virtual machine. Virtual Machine doesn't get IP). My symtons are exactly the same as described in the thread:

  • At first the openvswitch version was 1.4.0: I updated it to 1.4.2
  • I also had the permission error, and then I granted the quantum user with sudo access. It seems to go away

However, after that point, I still couldn't ping my vm from the network node. While the questioner(llastra) got the issue fixed and move on to the next part.

Any help will be appreciated as I've been try for weeks to connect to my vm. I include some information as below for diagnostics. Many thanks, Chien

1) ifconfig from compute node

eth0      Link encap:Ethernet  HWaddr 6c:3b:e5:1f:c0:3c
          inet addr:10.20.20.11  Bcast:10.20.20.255  Mask:255.255.255.0
          inet6 addr: fe80::6e3b:e5ff:fe1f:c03c/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:468 (468.0 B)

eth1      Link encap:Ethernet  HWaddr 00:e0:4c:3d:3b:e9
          inet addr:10.10.10.11  Bcast ...
(more)
edit retag flag offensive close merge delete

Comments

Enable debugging on both nova and quantum and see if you have any error messages there. If nova list doesn't show an IP, then there should be an error message in the log file of the compute node.

fetahi gravatar imagefetahi ( 2013-10-07 01:38:46 -0500 )edit

Thanks for your help but how can I switch to the debugging mode of nova and quantum?

Chien Q. Vu gravatar imageChien Q. Vu ( 2013-10-07 03:52:38 -0500 )edit

I have the same question In Control Node /var/log/quantum/server.log: 2013-10-08 17:28:05 WARNING [quantum.db.agentschedulers_db] Fail scheduling network {'status': u'ACTIVE', 'subnets': [u'44ab5a2f-3f49-44be-bae5-b9d0e20dfe44'], 'name': u'net_proj_one', 'provider:physical_network': None, 'admin_state_up': True, 'tenant_id': u'3f7da9c28b6b4ca1bad72511d61461f6', 'provider:network_type': u'gre', 'router:external': False, 'shared': False, 'id': u'e2c71b11-6e3d-4b16-9dc7-5d308237e461', 'provider:segmentation_id': 1L} Do you have the same status?

Albert gravatar imageAlbert ( 2013-10-08 05:34:27 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2013-10-07 05:40:03 -0500

sr0056850 gravatar image

Please answer below questions..

1.) Are you able view virtual machine's console from horizon ? If yes then what is the output of ifconfig command in virtual machine. can you able to ping 50.50.1.1 from VM ?

2.) 50.50.1.0 network is VM's private network, so only a virtual router (qrouter) running in network namespace can directly ping that network . you did ip netns exec command on qdhcp, instead of qdhcp can run the same command on router as shown below and show the output here.

sudo ip netns exec qrouter-26f355ef-cad2-4ccd-a69e-233aabe66393 ping 50.50.1.1

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

2 followers

Stats

Asked: 2013-10-06 21:49:42 -0500

Seen: 2,289 times

Last updated: Oct 07 '13