Ask Your Question
0

networking problems in running instances

asked 2011-05-09 18:32:11 -0500

diego-lalo gravatar image

Hi everyone, it's me again :) This time I have issues with networking of the instances. I'm running openstack in a two node configuration, there are three instances running in a node and one in the other node, here is the output of euca-describe-instances:

euca-describe-instances -a 3b96cc4b-6cba-45b0-bded-2a0356726a5e:proj1 -s b5dd676e-dacc-4cec-bb16-1ebf2d15018b --url http://192.168.1.208:8773/services/Cloud RESERVATION r-qgwxc38z proj1 default INSTANCE i-00000002 ami-00000003 192.168.0.2 192.168.0.2 running None (proj1, openstack-in-a-box-1rev85) 0 m1.tiny 2011-05-09T17:57:04Z nova RESERVATION r-0w1f16dn proj1 default INSTANCE i-00000004 ami-00000003 192.168.0.4 192.168.0.4 running None (proj1, eucalyptus) 0 m1.tiny 2011-05-09T17:57:28Z nova RESERVATION r-20qqjtse proj1 default INSTANCE i-00000003 ami-00000003 192.168.0.3 192.168.0.3 running None (proj1, openstack-in-a-box-1rev85) 0 m1.tiny 2011-05-09T17:57:18Z nova

here is the nova.conf of the cloud controller:

--dhcpbridge_flagfile=/etc/nova/nova.conf --dhcpbridge=/usr/bin/nova-dhcpbridge --logdir=/var/log/nova --state_path=/var/lib/nova --lock_path=/var/lock/nova --verbose --s3_host=192.168.1.208 --rabbit_host=192.168.1.208 --cc_host=192.168.1.208 --ec2_url=http://192.168.1.208:8773/services/Cloud --fixed_range=10.0.0.0/12 --network_size=88 --network_manager=nova.network.manager.FlatManager --routing_source_ip=192.168.1.208 --FAKE_subdomain=ec2 --image_service=nova.image.glance.GlanceImageService --glance_host=192.168.1.208 --glance_port=9292 --sql_connection=mysql://root:nova@192.168.1.208/nova

here is the nova.conf of the compute node:

--dhcpbridge_flagfile=/etc/nova/nova.conf --dhcpbridge=/usr/bin/nova-dhcpbridge --logdir=/var/log/nova --state_path=/var/lib/nova --lock_path=/var/lock/nova --verbose --s3_host=192.168.1.208 --rabbit_host=192.168.1.208 --cc_host=192.168.1.208 --ec2_url=http://192.168.1.208:8773/services/Cloud --fixed_range=10.0.0.0/12 --network_size=18 --network_manager=nova.network.manager.FlatManager --image_service=nova.image.glance.GlanceImageService --glance_host=192.168.1.208 --glance_port=9292 --sql_connection=mysql://root:nova@192.168.1.208/nova

here is the ifconfig output of the cloud controller:

br100 Link encap:Ethernet HWaddr bc:ae:c5:c1:d1:c3
inet addr:192.168.1.208 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::beae:c5ff:fec1:d1c3/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1304169 errors:0 dropped:0 overruns:0 frame:0 TX packets:1881333 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:426295659 (426.2 MB) TX bytes:2294268253 (2.2 GB)

eth0 Link encap:Ethernet HWaddr bc:ae:c5:c1:d1:c3
inet6 addr: fe80::beae:c5ff:fec1:d1c3/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1313062 errors:0 dropped:0 overruns:0 frame:0 TX packets:1885918 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:446334788 (446.3 MB) TX bytes:2294658231 (2.2 GB) Interrupt:42 Base address:0xc000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255 ...
(more)

edit retag flag offensive close merge delete

20 answers

Sort by » oldest newest most voted
0

answered 2011-05-12 20:51:12 -0500

vishvananda gravatar image

\o/ On May 12, 2011, at 1:45 PM, Diego Lalo wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Solved

Diego Lalo confirmed that the question is solved: I reconfigured the cloud and the problem has gone, ty for your help Vish :)


You received this question notification because you are a member of Nova Core, which is an answer contact for OpenStack Compute (nova).

edit flag offensive delete link more
0

answered 2011-05-12 20:41:29 -0500

vishvananda gravatar image

The promisc on is supposed to fix the public internet problem. you might want to make sure you set --my_ip=192.168.1.100 # (or whatever the ip is for each server) sometimes it can't discover the key properly. Afterwards you may have to flush all of the iptables rules on both hosts and restart the services and rerun instances.

Vish

On May 12, 2011, at 12:16 PM, Diego Lalo wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Open

Diego Lalo is still having a problem: The console output is fine, you are right I can't access the instance from within the compute node but I can access it through the cloud controller :) The only problem that remains is that the instance doesn't have internet, I tried to do a yum update from within the instance and nothing happened, I also tried to do a ping www.free.fr and didn't work:

-bash-3.2# ping www.free.fr PING www.free.fr (212.27.48.10) 56(84) bytes of data.

this is the only line that appears in the instance console

Thanks a lot for the help you are giving to me Vish!! Diego


You received this question notification because you are a member of Nova Core, which is an answer contact for OpenStack Compute (nova).

edit flag offensive delete link more
0

answered 2011-05-12 19:15:51 -0500

diego-lalo gravatar image

The console output is fine, you are right I can't access the instance from within the compute node but I can access it through the cloud controller :) The only problem that remains is that the instance doesn't have internet, I tried to do a yum update from within the instance and nothing happened, I also tried to do a ping http://www.free.fr and didn't work:

-bash-3.2# ping http://www.free.fr PING http://www.free.fr (212.27.48.10) 56(84) bytes of data.

this is the only line that appears in the instance console

Thanks a lot for the help you are giving to me Vish!! Diego

edit flag offensive delete link more
0

answered 2011-05-12 19:04:34 -0500

vishvananda gravatar image

Are you pinging from the network host? You won't be able to get to it from the compute host. If you get console output on the instance are there any errors? On May 12, 2011 2:01 PM, "Diego Lalo" question156783@answers.launchpad.net wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Open

Diego Lalo is still having a problem: Thanks again for your answer Vish, I restarted both machines and I set promisc only in the host that is running nova-network, now in the compute node the instance gets the correct IP but it can't be pinged or accessed via SSH, I executed ifconfig in the compute node after launching the instance and this is the output:

br100 Link encap:Ethernet HWaddr 00:24:8c:88:a2:9d inet addr:192.168.1.50 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::c448:7aff:fef8:fa05/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:18055 errors:0 dropped:0 overruns:0 frame:0 TX packets:34659 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:1542409 (1.5 MB) TX bytes:11416258 (11.4 MB)

eth0 Link encap:Ethernet HWaddr 00:24:8c:88:a2:9d inet6 addr: fe80::224:8cff:fe88:a29d/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:26926 errors:0 dropped:0 overruns:0 frame:0 TX packets:50887 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:2928673 (2.9 MB) TX bytes:13152301 (13.1 MB) Interrupt:42 Base address:0x8000

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:935 errors:0 dropped:0 overruns:0 frame:0 TX packets:935 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:187503 (187.5 KB) TX bytes:187503 (187.5 KB)

virbr0 Link encap:Ethernet HWaddr 42:9c:11:d8:13:d0 inet addr:192.168.122.1 Bcast:192.168.122.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

vnet0 Link encap:Ethernet HWaddr fe:16:3e:2f:d6:3b inet6 addr: fe80::fc16:3eff:fe2f:d63b/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:33 errors:0 dropped:0 overruns:0 frame:0 TX packets:248 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:500 RX bytes:4835 (4.8 KB) TX bytes:18541 (18.5 KB)

Cheers Diego


You received this question notification because you are ...

(more)
edit flag offensive delete link more
0

answered 2011-05-12 18:59:23 -0500

diego-lalo gravatar image

Thanks again for your answer Vish, I restarted both machines and I set promisc only in the host that is running nova-network, now in the compute node the instance gets the correct IP but it can't be pinged or accessed via SSH, I executed ifconfig in the compute node after launching the instance and this is the output:

br100 Link encap:Ethernet HWaddr 00:24:8c:88:a2:9d
inet addr:192.168.1.50 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::c448:7aff:fef8:fa05/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:18055 errors:0 dropped:0 overruns:0 frame:0 TX packets:34659 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:1542409 (1.5 MB) TX bytes:11416258 (11.4 MB)

eth0 Link encap:Ethernet HWaddr 00:24:8c:88:a2:9d
inet6 addr: fe80::224:8cff:fe88:a29d/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:26926 errors:0 dropped:0 overruns:0 frame:0 TX packets:50887 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:2928673 (2.9 MB) TX bytes:13152301 (13.1 MB) Interrupt:42 Base address:0x8000

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:935 errors:0 dropped:0 overruns:0 frame:0 TX packets:935 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:187503 (187.5 KB) TX bytes:187503 (187.5 KB)

virbr0 Link encap:Ethernet HWaddr 42:9c:11:d8:13:d0
inet addr:192.168.122.1 Bcast:192.168.122.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

vnet0 Link encap:Ethernet HWaddr fe:16:3e:2f:d6:3b
inet6 addr: fe80::fc16:3eff:fe2f:d63b/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:33 errors:0 dropped:0 overruns:0 frame:0 TX packets:248 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:500 RX bytes:4835 (4.8 KB) TX bytes:18541 (18.5 KB)

Cheers Diego

edit flag offensive delete link more
0

answered 2011-05-12 18:18:11 -0500

vishvananda gravatar image

Promisc only needs to be set on the network host. You will have to restart all of the workers after setting flat interface. On May 12, 2011 12:25 PM, "Diego Lalo" question156783@answers.launchpad.net wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Open

Diego Lalo is still having a problem: I tried to do what you said: ip link set promisc on dev br100 on the compute node but br100 only appears when some instance is launched on that node but it doesn't exist right now. I also set the flag --flat_interface on both nodes.

Cheers Diego


You received this question notification because you are a member of Nova Core, which is an answer contact for OpenStack Compute (nova).

edit flag offensive delete link more
0

answered 2011-05-12 17:09:07 -0500

vishvananda gravatar image

If you want to use flatdchp with multiple hosts you will need to use --flat_interface=eth0 # or whatever interface they share

If you only have one interface on all of your hosts, you also may need to do: ip link set promisc on dev br100

Vish

On May 12, 2011, at 9:41 AM, Diego Lalo wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Open

Diego Lalo is still having a problem: Ty for your answer Vish, but how I configure the extra compute node to make it work? I mean when I launch an instance that will run in the new compute node it doesn't get a valid IP because the bridge has no IP on that node, how I should create the network for the project and how should the flags be configured for each node (the Cloud Controller and the Compute Node).

Cheers Diego


You received this question notification because you are a member of Nova Core, which is an answer contact for OpenStack Compute (nova).

edit flag offensive delete link more
0

answered 2011-05-12 16:37:49 -0500

diego-lalo gravatar image
   Ty for your answer Vish, but how I configure the extra compute node to make it work? I mean when I launch an instance that will run in the new compute node it doesn't get a valid IP because the bridge has no IP on that node, how I should create the network for the project and how should the flags be configured for each node (the Cloud Controller and the Compute Node).

Cheers Diego

edit flag offensive delete link more
0

answered 2011-05-12 16:22:02 -0500

vishvananda gravatar image

You have a couple options here.

  1. set up a route on the external node specifying the host as the gateway. route add -net 10.0.0.0/24 gw 192.168.1.100

  2. use floating ips and euca-allocate-address and euca-associate-address nova-manage floating create <hostname> 192.168.1.128/25 # hostname is the hostname of the host running nova-network euca-allocate-address euca-associate-address -i i-1 192.168.1.128 # for example

either way you will need to use euca-authorize to open up specific ports to be accessible: euca-authorize -P tcp -p 22 default

Vish

On May 12, 2011, at 7:26 AM, Diego Lalo wrote:

Question #156783 on OpenStack Compute (nova) changed: https://answers.launchpad.net/nova/+q...

Status: Answered => Solved

Diego Lalo confirmed that the question is solved: Is working nooooowwww!!!!! Thanks for your help Vish, the instance gets the right IP and can be accessed via SSH from within the physical machine that is running, for the moment I can't access the instance from other machine inside the local network, the instance IP is 10.0.0.2 and the local machines of our network are in the net 192.168.1.0 how can I do to acces the instance from my machine for example? My IP is 192.168.1.100

Cheers Diego


You received this question notification because you are a member of Nova Core, which is an answer contact for OpenStack Compute (nova).

edit flag offensive delete link more
0

answered 2011-05-12 16:07:07 -0500

diego-lalo gravatar image

I have another question about multinode configuration, I have two compute nodes, the network manager is FlatDHCPManager and the network is 10.0.0.0/12, I configured both nodes with the same flags:

--dhcpbridge_flagfile=/etc/nova/nova.conf --dhcpbridge=/usr/bin/nova-dhcpbridge --logdir=/var/log/nova --state_path=/var/lib/nova --lock_path=/var/lock/nova --verbose --s3_host=192.168.1.234 --rabbit_host=192.168.1.234 --cc_host=192.168.1.234 --ec2_url=http://192.168.1.234:8773/services/Cloud --fixed_range=10.0.0.0/12 --network_size=2048 --network_manager=nova.network.manager.FlatDHCPManager --routing_source_ip=192.168.1.234 --FAKE_subdomain=ec2 --image_service=nova.image.glance.GlanceImageService --glance_host=192.168.1.234 --glance_port=9292 --sql_connection=mysql://root:nova@192.168.1.234/nova

The problem is that one of the nodes has no IP in it's bridge, this is the output of ifconfig in the CloudController:

br100 Link encap:Ethernet HWaddr fe:16:3e:18:b4:41
inet addr:10.0.0.1 Bcast:10.0.3.255 Mask:255.255.252.0 inet6 addr: fe80::5c06:fdff:fef5:bef0/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1046 errors:0 dropped:0 overruns:0 frame:0 TX packets:1101 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:175557 (175.5 KB) TX bytes:193852 (193.8 KB)

eth0 Link encap:Ethernet HWaddr 00:24:8c:2e:50:88
inet addr:192.168.1.234 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::224:8cff:fe2e:5088/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1272863 errors:0 dropped:0 overruns:0 frame:0 TX packets:1852857 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:433340720 (433.3 MB) TX bytes:2296477103 (2.2 GB) Interrupt:42 Base address:0x6000

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:3224420 errors:0 dropped:0 overruns:0 frame:0 TX packets:3224420 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:4916528700 (4.9 GB) TX bytes:4916528700 (4.9 GB)

virbr0 Link encap:Ethernet HWaddr 82:af:25:15:73:00
inet addr:192.168.122.1 Bcast:192.168.122.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

vnet0 Link encap:Ethernet HWaddr fe:16:3e:18:b4:41
inet6 addr: fe80::fc16:3eff:fe18:b441/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:431 errors:0 dropped:0 overruns:0 frame:0 TX packets:630 errors ...
(more)

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

1 follower

Stats

Asked: 2011-05-09 18:32:11 -0500

Seen: 238 times

Last updated: May 12 '11