[rdo-list] Overcloud Deployment failsm not br-int

[Marius Cornea]

On Sun, Jul 31, 2016 at 11:10 AM, Gabriel Peral <gbrlperal at gmail.com> wrote:
> Hi everyone!
>
> I've been trying to deploy OpenStack using TripleO for 2 weeks and I am get
> stucked in the same step.
>
> But let's see first my environment, 3 baremetal nodes, one will be a compute
> node, the other two will be compute nodes. In each pyhiscal node I have two
> pyhiscal NICs, one for the PXE+DHCP provisioning, and the other pyhsical NIC
> connected to a pyhiscal switch which has the rest of the vlans for the
> deployment (api, storage, tenant, exeternal, etc). I am deploying OpeStack
> Kilo version.
> The first step of the installation was good, the undercloud and the
> introspection was done in the right way, but the problem cames when I try to
> deploy the overcloud, I set up the network-environment.yaml template in
> order to set the vlans to the networks, and the compute.yaml and
> controller.yaml templates to set the right NICs to the right networks. And
> then I launch the deploy command like this (vlan mode) :
>
> $ openstack overcloud deploy --templates ~/templates/my-overcloud/ -e
> ~/templates/network-environment.yaml -e ~/templates/storage-environment.yaml
> -e ~/templates/my-overcloud/environments/network-isolation.yaml
> --control-scale 1 --compute-scale 2 --control-flavor control
> --compute-flavor compute --ntp-server 10.20.10.3 --neutron-network-type vlan
> --neutron-disable-tunneling --neutron-bridge-mappings datacentre:br-ex
> --neutron-network-vlan-ranges datacentre:30:100 --timeout 60 --verbose
> --log-file overcloud_vlan.log
>
> And all seems go well untill it fails due to timeout, the
>
> heat resource-list --nested-depth 5 overcloud | grep FAILED
>
> Show that something is wrong in the NetworkDeploy, and when I log into the
> controller node, if I type sudo ovs-vsctl list-br it only shows me the br-ex
> bridge, so it wasn't able to create the br-int, and I think that's the
> reason because the deployment has been failed, when I've tryied to deploy
> more i log into the controller node, and check the sudo tail -f
> /var/log/messages and this is the only message that appears:
>
>
> Jul 31 04:59:53 overcloud-controller-0 os-collect-config: 2016-07-31
> 04:59:53.337 4606 WARNING os-collect-config [-] Source [ec2] Unavailable.
> Jul 31 04:59:53 overcloud-controller-0 os-collect-config: 2016-07-31
> 04:59:53.624 4606 WARNING os-collect-config [-] Source [request]
> Unavailable.
> Jul 31 04:59:53 overcloud-controller-0 os-collect-config: 2016-07-31
> 04:59:53.625 4606 WARNING os_collect_config.local [-]
> /var/lib/os-collect-config/local-data not found. Skipping
> Jul 31 04:59:53 overcloud-controller-0 os-collect-config: 2016-07-31
> 04:59:53.625 4606 WARNING os_collect_config.local [-] No local metadata
> found (['/var/lib/os-collect-config/local-data'])
> Jul 31 05:00:26 overcloud-controller-0 os-collect-config: 2016-07-31
> 05:00:26.615 4606 WARNING os_collect_config.ec2 [-] ('Connection aborted.',
> error(113, 'No route to host'))

This generally indicates that it couldn't reach the metadata server (
route to 169.254.169.254). This gets set in the nic templates that
you've used.
Next debugging steps would be to run ip r get 169.254.169.254, see
what's the next hop for it - it should be the ctlplane address of the
undercloud. Then do some connectivity checks for it such as ping or
curl http://169.254.169.254

> Jul 31 05:00:26 overcloud-controller-0 os-collect-config: 2016-07-31
> 05:00:26.615 4606 WARNING os-collect-config [-] Source [ec2] Unavailable.
> Jul 31 05:00:26 overcloud-controller-0 os-collect-config: 2016-07-31
> 05:00:26.833 4606 WARNING os-collect-config [-] Source [request]
> Unavailable.
> Jul 31 05:00:26 overcloud-controller-0 os-collect-config: 2016-07-31
> 05:00:26.833 4606 WARNING os_collect_config.local [-]
> /var/lib/os-collect-config/local-data not found. Skipping
> Jul 31 05:00:26 overcloud-controller-0 os-collect-config: 2016-07-31
> 05:00:26.834 4606 WARNING os_collect_config.local [-] No local metadata
> found (['/var/lib/os-collect-config/local-data'])
>
> So something is wrong with the network, but I don't know what because I can
> ping the external network and gateway from the controller node.
>
> If someone knows what could be I would be so glad, because I've been trying
> to deploy and all the times it fails
>
> Kind regards,
>
> Gabriel
>
> _______________________________________________
> rdo-list mailing list
> rdo-list at redhat.com
> https://www.redhat.com/mailman/listinfo/rdo-list
>
> To unsubscribe: rdo-list-unsubscribe at redhat.com