← Back to team overview

openstack team mailing list archive

[Grizzly] cannot attach volume to instance due to wrong iscsi target

 

Hi all,

I installed three nodes like this topology:
http://docs.openstack.org/trunk/openstack-network/admin/content/connectivity.html

Here are my subnets:
management: 192.168.203.X/24
data: 192.168.201.X/24
external & API: 192.168.202.X/24

I'm running ubuntu 12.04.

When I try to attach a volume to a VM, I get the following error in nova-compute.log: 2013-04-26 16:27:59.439 WARNING nova.virt.libvirt.volume [req-f5b4e121-a3ac-456d-b5cb-ba389c7fb409 6d72da42f39648c48d3dfb4cd190107d 93a48de7ef674f07a96e169383c34399] ISCSI volume not yet found at: vdr. Will rescan & retry. Try number: 0 2013-04-26 16:27:59.490 ERROR nova.compute.manager [req-f5b4e121-a3ac-456d-b5cb-ba389c7fb409 6d72da42f39648c48d3dfb4cd190107d 93a48de7ef674f07a96e169383c34399] [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Failed to attach volume d9424219-33f6-40c8-88d9-ecba4c8aa6be at /dev/vdr 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Traceback (most recent call last): 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2859, in _attach_volume 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] mountpoint) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line 957, in attach_volume 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] disk_info) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line 943, in volume_driver_method 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] return method(connection_info, *args, **kwargs) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/openstack/common/lockutils.py", line 242, in inner 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] retval = f(*args, **kwargs) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/volume.py", line 245, in connect_volume 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] self._run_iscsiadm(iscsi_properties, ("--rescan",)) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/volume.py", line 179, in _run_iscsiadm 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] check_exit_code=check_exit_code) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] File "/usr/lib/python2.7/dist-packages/nova/utils.py", line 239, in execute 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] cmd=' '.join(cmd)) 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] ProcessExecutionError: Unexpected error while running command. 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Command: sudo nova-rootwrap /etc/nova/rootwrap.conf iscsiadm -m node -T iqn.2010-10.org.openstack:volume-d9424219-33f6-40c8-88d9-ecba4c8aa6be -p _/*192.168.202.103*/_:3260 --rescan 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Exit code: 255 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Stdout: '' 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] Stderr: 'iscsiadm: No portal found.\n' 2013-04-26 16:27:59.490 30092 TRACE nova.compute.manager [instance: 05141f81-04cc-4493-86da-d2c05fd8a2f9] 2013-04-26 16:27:59.849 ERROR nova.openstack.common.rpc.amqp [req-f5b4e121-a3ac-456d-b5cb-ba389c7fb409 6d72da42f39648c48d3dfb4cd190107d 93a48de7ef674f07a96e169383c34399] Exception during message handling


As we can see, the compute node tries to reach the API interface (192.168.202.103) of the controller node. Of course, it cannot, since the compute node only knows the data and the management subnets.

Is this the default behaviour and I'm missing a parameter somewhere?
I set this key in nova.conf:
iscsi_ip_address=192.168.203.103

but still no luck

any clue?

yours,

michaël


--
Michaël Van de Borne
R&D Engineer, SOA team, CETIC
Phone: +32 (0)71 49 07 45 Mobile: +32 (0)472 69 57 16, Skype: mikemowgli
www.cetic.be, rue des Frères Wright, 29/3, B-6041 Charleroi