Open Stack

Maybe worth posting on operators, but it looks like the scheduling of
the action fails, which let me think that nova is not running fine
somewhere.

Why is the restart in a random order? That can cause issues, and
that's the whole reason why we are orchestrating the deploys/upgrade
with ansible.
Also, why don't you follow our operations guide for recovering for a
failure? Is there something wrong there?

Regards,
JP

On 17 April 2018 at 14:07, Periyasamy Palanisamy
<periyasamy.palanisamy at ericsson.com> wrote:
> Hi,
>
>
>
> I’m trying to migrate controller and compute VMs installed with
> Openstack-Ansible across systems with following approach.
>
> This is mainly to minimize the deployment time in the Jenkins CI
> environment.
>
>
>
> Export steps:
>
> Power off the VMs gracefully.
> virsh dumpxml ${node} > $EXPORT_PATH/${node}.xml
> cp /var/lib/libvirt/images/${node}.qcow2 $EXPORT_PATH/$node.qcow2
> create a tar ball for the xml’s and qcow2 images.
>
>
>
> Import steps:
>
> cp ${node}.qcow2 /var/lib/libvirt/images/
> virsh define ${node}.xml
> virsh start ${node}
>
>
>
> After the import of the VMs, The openstack services (neutron-server, DHCP
> agent, Metering agent, Metadata agent, L3 agent, Open vSwitch agent,
> nova-conductor and nova-comute) are started in random order.
>
> This causes neutron and nova is not able to find DHCP agent and compute
> accordingly to bring up the tenant VM and throws the error [1].
>
>
>
> I have also tried to boot compute VM followed by controller VM. It also
> doesn’t help.
>
> Could you please let me know what is going wrong here ?
>
>
>
> [1] https://paste.ubuntu.com/p/YNg2NnjvpS/ (fault section)
>
>
>
> Thanks,
>
> Periyasamy
>
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>