launch-node: ignore failure to restart eth0

eth0 might not exist, such as on Xenial hosts with interfaced-based
names.  Since this is a bit of platform/provider specific hack, just
ignore failures.

Change-Id: Ie18b7f49ea2f1b72b496c61ac2576ae53f5ad3eb
This commit is contained in:
Ian Wienand 2017-03-28 13:50:26 +11:00
parent fa76dff67d
commit 8791e9fafa

View File

@ -121,7 +121,7 @@ def bootstrap_server(server, key, name, volume_device, keep,
# Something up with RAX images that they have the ipv6 interface in # Something up with RAX images that they have the ipv6 interface in
# /etc/network/interfaces but eth0 hasn't noticed yet; reload it # /etc/network/interfaces but eth0 hasn't noticed yet; reload it
ssh_client.ssh('ifdown eth0 && ifup eth0') ssh_client.ssh('(ifdown eth0 && ifup eth0) || true')
if server.public_v6: if server.public_v6:
ssh_client.ssh('ping6 -c5 -Q 0x10 review.openstack.org ' ssh_client.ssh('ping6 -c5 -Q 0x10 review.openstack.org '