← Back to team overview

openstack team mailing list archive

fixed_range='' results in critical error when starting nova-network


Hello all,

I am trying to resolve a critical error when fixed_range='' is specified in

The error is resolved if I set fixed_range manually, i.e. fixed_range=

VM launching via Horizon and guest networking is fully operational with the
above fix.

Similar error but different cause,

I am using FlatDHCP networking between two Openstack nodes running the
Folsom release (from EPEL) on a base installation of Scientific Linux 6.4.
I followed the guide from Red Hat Openstack Getting Started guide to
prepare Keystone, Glance and Horizon and the Compute manual for setting up
the networking.

Controller node (Keystone, Nova, Glance, Horizon) "sl1":
Compute only node "compute1":
Iptables disabled (service iptables stop on both nodes).

Every service except nova-network seems to be operational, judging from
openstack-status and service --status-all | grep openstack.

nova.conf snippet on the controller node:

network_manager = nova.network.manager.FlatDHCPManager
flat_interface = eth2
public_interface = eth1
my_ip =
flat_network_bridge = br100
flat_injected = False

nova.conf snippet on the compute only node:

network_manager = nova.network.manager.FlatDHCPManager
flat_interface = eth1
public_interface = eth0
my_ip =
#fixed_range =
#fixed_range = ''
flat_network_bridge = br100
flat_injected = False

s3_host =
ec2_host =
qpid_hostname =
metadata_host =
ec2_dmz_host =

nova-network.log (identical on both nodes):

2013-05-22 16:21:17 18884 AUDIT nova.service [-] Starting network node
(version 2012.2.3-1.el6)
2013-05-22 16:21:22 18884 CRITICAL nova [-] Unexpected error while running
Command: sudo nova-rootwrap /etc/nova/rootwrap.conf iptables-restore -c
Exit code: 2
Stdout: ''
Stderr: "Bad argument `SNAT'\nError occurred at line: 27\nTry
`iptables-restore -h' or 'iptables-restore --help' for more information.\n"

Any suggestions on how to resolve this issue would be much appreciated!