Node not downloading image from Glance on Cloud Controller

Hello,
I have 1 cloud controller where is installed all services (nova+glance) and 3 nodes with just the compute service. See below :
================================================================
> euca-describe-availability-zones verbose

AVAILABILITYZONE nova available
AVAILABILITYZONE |- krypton.inrialpes.fr
AVAILABILITYZONE | |- nova-compute enabled :-) 2011-11-15 15:55:11
AVAILABILITYZONE | |- nova-network enabled :-) 2011-11-15 15:55:15
AVAILABILITYZONE | |- nova-scheduler enabled :-) 2011-11-15 15:55:16
AVAILABILITYZONE | |- nova-vncproxy enabled :-) 2011-11-15 15:55:17
AVAILABILITYZONE |- cadmium.inrialpes.fr
AVAILABILITYZONE | |- nova-compute enabled :-) 2011-11-15 15:55:18
AVAILABILITYZONE |- fermium.inrialpes.fr
AVAILABILITYZONE | |- nova-compute enabled :-) 2011-11-15 15:55:14
AVAILABILITYZONE |- molecule.inrialpes.fr
AVAILABILITYZONE | |- nova-compute enabled :-) 2011-11-15 15:55:15
================================================================
(krypton is 194.199.25.36)

Each install are Diablo 2011.3 on CentOS 6
My nova.conf on each machine is almost unmodified from default (only ip and verbose modified) :
================================================================
--verbose=true
--ec2_url=http://194.199.25.36:8773/services/Cloud
--s3_host=194.199.25.36
--cc_host=194.199.25.36
--rabbit_host=194.199.25.36
--sql_connection=mysql://nova:nova@194.199.25.36/nova
--glance_api_servers=194.199.25.36:9292
--use_s3=true
--libvirt_type=kvm
--use_syslog=false
--node_availability_zone=nova
--logdir=/var/log/nova
--logging_context_format_string=%(asctime)s %(name)s: %(levelname)s [%(request_id)s %(user)s %(project)s] %(message)s
--logging_default_format_string=%(asctime)s %(name)s: %(message)s
--logging_debug_format_suffix=
--use_cow_images=true
--auth_driver=nova.auth.dbdriver.DbDriver
--network_manager=nova.network.manager.VlanManager
--scheduler_driver=nova.scheduler.zone.ZoneScheduler
--image_service=nova.image.glance.GlanceImageService
--use_ipv6=false
--ca_path=/var/lib/nova/CA
--keys_path=/var/lib/nova/keys
--images_path=/var/lib/nova/images
--buckets_path=/var/lib/nova/buckets
--instances_path=/var/lib/nova/instances
--networks_path=/var/lib/nova/networks
--dhcpbridge_flagfile=/etc/nova/nova.conf
--dhcpbridge=/usr/bin/nova-dhcpbridge
--injected_network_template_dir=/usr/share/nova/interfaces/
--libvirt_xml_template=/usr/share/nova/libvirt.xml.template
--vpn_client_template=/usr/share/nova/client.ovpn.template
--credentials_template=/usr/share/nova/novarc.template
--state_path=/var/lib/nova
--lock_path=/var/lib/nova/tmp
--vnc_enabled=true
--vncproxy_url=http://194.199.25.36:6080
--vncserver_host=0.0.0.0
--vnc_token_ttl=300
--vlan_interface=eth0
================================================================

When I try to launch instances (images stored on Glance) :
- if the compute service is the one on the controller, instance is created.
- if instances are to be launch on nodes, VM stay on "pending" state.
Note : Ip is allocated in both cases.

Log on node are as follow :
================================================================
2011-11-15 16:39:40,782 nova.compute.manager: Updating host status
2011-11-15 16:41:10,943 nova.rpc: received {u'_context_roles': [], u'_context_request_id': u'13da7fb6-40ff-4c85-91ec-0fefb574ce20', u'_context_read_deleted': False, u'args': {u'request_spec': {u'num_instances': 1, u'image': {u'status': u'active', u'name': u'MachineTest', u'deleted': False, u'container_format': u'ami', u'created_at': u'2011-11-15 14:09:12.017433', u'disk_format': u'ami', u'updated_at': u'2011-11-15 14:09:12.368728', u'properties': {u'kernel_id': u'1', u'min_ram': u'0', u'ramdisk_id': u'2', u'min_disk': u'0', u'owner': None, u'type': u'machine'}, u'location': u'file:///var/lib/glance/images/3', u'checksum': u'10047a119149e08fb206eea89832eee0', u'is_public': True, u'deleted_at': None, u'id': 3, u'size': 25165824}, u'filter': None, u'instance_type': {u'rxtx_quota': 0, u'flavorid': 1, u'name': u'm1.tiny', u'deleted': False, u'created_at': None, u'updated_at': None, u'memory_mb': 512, u'vcpus': 1, u'rxtx_cap': 0, u'extra_specs': {}, u'swap': 0, u'deleted_at': None, u'id': 2, u'local_gb': 0}, u'blob': None, u'instance_properties': {u'vm_state': u'building', u'availability_zone': None, u'ramdisk_id': u'2', u'instance_type_id': 2, u'user_data': u'', u'vm_mode': None, u'reservation_id': u'r-y1rt8m7r', u'user_id': u'bertrand', u'display_description': None, u'key_data': u'ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAIEAlfPhqBFvvtSpcJ6sbVjajuROOorAAwiszpNi8RyIqEyOMzbk+TlY2aHCAWkh0ajx9ayz97LqVzJfwvzylTo5oXKs4xieHYXXgyozpm2R9zutAuRswwPMQrpf60kw05Cz9JxDKtn/6WK64/NqROfikMhWpG78Id6h7k92rqu2/f8= <email address hidden>\n', u'power_state': 0, u'project_id': u'projettest', u'metadata': {}, u'access_ip_v6': None, u'access_ip_v4': None, u'kernel_id': u'1', u'key_name': u'CleTest', u'display_name': None, u'config_drive_id': u'', u'local_gb': 0, u'architecture': None, u'locked': False, u'launch_time': u'2011-11-15T15:41:10Z', u'memory_mb': 512, u'vcpus': 1, u'image_ref': 3, u'root_device_name': None, u'os_type': None, u'config_drive': u''}}, u'requested_networks': None, u'availability_zone': None, u'instance_id': 15, u'admin_password': None, u'injected_files': None}, u'_context_auth_token': None, u'_context_strategy': u'noauth', u'_context_is_admin': True, u'_context_project_id': u'projettest', u'_context_timestamp': u'2011-11-15T15:41:10.580369', u'_context_user_id': u'bertrand', u'method': u'run_instance', u'_context_remote_address': u'127.0.0.1'}
2011-11-15 16:41:10,943 nova.rpc: unpacked context: {'user_id': u'bertrand', 'roles': [], 'timestamp': u'2011-11-15T15:41:10.580369', 'auth_token': None, 'msg_id': None, 'remote_address': u'127.0.0.1', 'strategy': u'noauth', 'is_admin': True, 'request_id': u'13da7fb6-40ff-4c85-91ec-0fefb574ce20', 'project_id': u'projettest', 'read_deleted': False}
2011-11-15 16:41:11,218 nova.rpc: Making asynchronous call on network ...
2011-11-15 16:41:11,219 nova.rpc: MSG_ID is d4d7cde953dd4f609f89a0eedd8125ca
2011-11-15 16:41:12,504 nova.compute.manager: instance network_info: |[[{u'bridge': u'br100', u'multi_host': False, u'bridge_interface': u'eth0', u'vlan': 100, u'id': 1, u'injected': False, u'cidr': u'10.0.0.0/24', u'cidr_v6': None}, {u'should_create_bridge': True, u'dns': [], u'vif_uuid': u'479cda54-9b53-4203-9cbd-7a0489996220', u'label': u'private', u'broadcast': u'10.0.0.255', u'ips': [{u'ip': u'10.0.0.3', u'netmask': u'255.255.255.0', u'enabled': u'1'}], u'mac': u'02:16:3e:7c:80:f4', u'rxtx_cap': 0, u'should_create_vlan': True, u'dhcp_server': u'10.0.0.1', u'gateway': u'10.0.0.1'}]]|
2011-11-15 16:41:12,728 nova.virt.libvirt_conn: instance instance-0000000f: starting toXML method
2011-11-15 16:41:12,729 nova.virt.libvirt.vif: Ensuring vlan 100 and bridge br100
2011-11-15 16:41:12,729 nova.utils: Attempting to grab semaphore "ensure_vlan" for method "ensure_vlan"...
2011-11-15 16:41:12,729 nova.utils: Attempting to grab file lock "ensure_vlan" for method "ensure_vlan"...
2011-11-15 16:41:12,730 nova.utils: Running cmd (subprocess): ip link show dev vlan100
2011-11-15 16:41:12,743 nova.utils: Attempting to grab semaphore "ensure_bridge" for method "ensure_bridge"...
2011-11-15 16:41:12,744 nova.utils: Attempting to grab file lock "ensure_bridge" for method "ensure_bridge"...
2011-11-15 16:41:12,744 nova.utils: Running cmd (subprocess): ip link show dev br100
2011-11-15 16:41:12,759 nova.utils: Running cmd (subprocess): sudo brctl addif br100 vlan100
2011-11-15 16:41:12,781 nova.utils: Result was 1
2011-11-15 16:41:12,782 nova.utils: Running cmd (subprocess): sudo route -n
2011-11-15 16:41:12,804 nova.utils: Running cmd (subprocess): sudo ip addr show dev vlan100 scope global
2011-11-15 16:41:12,840 nova.virt.libvirt_conn: block_device_list []
2011-11-15 16:41:12,840 nova.virt.libvirt_conn: block_device_list []
2011-11-15 16:41:13,017 nova.virt.libvirt_conn: instance instance-0000000f: finished toXML method
2011-11-15 16:41:13,017 nova: called setup_basic_filtering in nwfilter
2011-11-15 16:41:13,018 nova: ensuring static filters
2011-11-15 16:41:13,061 nova.virt.libvirt.firewall: iptables firewall: Setup Basic Filtering
2011-11-15 16:41:13,061 nova.utils: Attempting to grab semaphore "iptables" for method "_do_refresh_provider_fw_rules"...
2011-11-15 16:41:40,945 nova.compute.manager: Updating host status
2011-11-15 16:41:41,031 nova.compute.manager: Found 1 in the database and 0 on the hypervisor.
2011-11-15 16:42:41,123 nova.compute.manager: Found 1 in the database and 0 on the hypervisor.
2011-11-15 16:43:41,124 nova.compute.manager: Updating host status
================================================================

Seems that network stuff is done.

There is however a weird thing in the line :
=====
2011-11-15 16:41:10,943 nova.rpc: unpacked context: {'user_id': u'bertrand', 'roles': [], 'timestamp': u'2011-11-15T15:41:10.580369', 'auth_token': None, 'msg_id': None, 'remote_address': u'127.0.0.1', 'strategy': u'noauth', 'is_admin': True, 'request_id': u'13da7fb6-40ff-4c85-91ec-0fefb574ce20', 'project_id': u'projettest', 'read_deleted': False}
=====
This "127.0.0.1"

Moreover, looking in /var/lib/nova/instances/_base/ :
- On the controller node, I can see 3 files, with size different from 0
- On nodes, nothing.

The weird thing is that I can do the following and see the file displaying :
curl http://194.199.25.36:9292/v1/images/1

Has anyone ever had this kind of error?