I few days ago i started having issues deploying my Openstack bundle. The issue seems to be focused around DNS.
Specifically my ha clustered nova-cloud-controller is unable to resolve my host “h14” by that name. Checking myself it does not work until i add the search domain in netplan or by using FQDN.
I realize that i may have caused this because i added a second authoritative domain in maas, after that i started having problems.
I was wondering if anyone knows where juju has it netplan templates so that i can modify it to include the searchdomain?
Or if there is another better way to add the search domain.
Maybe i have stumbled on a bug?
2019-02-06 18:23:54 DEBUG juju-log identity-service:29: 0 section(s) found
2019-02-06 18:23:54 INFO juju-log identity-service:29: HAProxy context is incomplete, this unit has no peers.
2019-02-06 18:23:55 INFO juju-log identity-service:29: HAProxy context is incomplete, this unit has no peers.
2019-02-06 18:23:55 INFO juju-log identity-service:29: Generating template context from neutron api relation
2019-02-06 18:23:55 INFO juju-log identity-service:29: identity-service relation incomplete. Peer not ready?
2019-02-06 18:23:55 INFO juju-log identity-service:29: Registered config file:
2019-02-06 18:23:55 DEBUG juju-log identity-service:29: 0 section(s) found
2019-02-06 18:23:56 DEBUG juju-log identity-service:29: Generating template context for identity-service
2019-02-06 18:23:56 INFO juju-log identity-service:29: Missing required data: service_port service_host auth_host auth_port admin_tenant_name admin_user admin_password
2019-02-06 18:23:56 INFO juju-log identity-service:29: identity relation’s interface, identity-service, is related awaiting the following data from the relationship: service_port, service_host, auth_host, auth_port, admin_tenant_name, admin_user, admin_password.
root@juju-692f12-0-lxd-7:/var/log/juju# tail unit-nova-cloud-controller-0.log
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed File “/var/lib/juju/agents/unit-nova-cloud-controller-0/charm/hooks/nova_cc_utils.py”, line 1040, in ssh_compute_add
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed if ch_ip.ns_query(short):
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed File “/var/lib/juju/agents/unit-nova-cloud-controller-0/charm/charmhelpers/contrib/network/ip.py”, line 478, in ns_query
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed answers = dns.resolver.query(address, rtype)
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed File “/usr/lib/python3/dist-packages/dns/resolver.py”, line 1132, in query
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed raise_on_no_answer, source_port)
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed File “/usr/lib/python3/dist-packages/dns/resolver.py”, line 947, in query
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed raise NoNameservers(request=request, errors=errors)
2019-02-06 18:47:28 DEBUG cloud-compute-relation-changed dns.resolver.NoNameservers: All nameservers failed to answer the query h14. IN A: Server 127.0.0.53 UDP port 53 answered SERVFAIL
2019-02-06 18:47:28 ERROR juju.worker.uniter.operation runhook.go:132 hook “cloud-compute-relation-changed” failed: exit status 1
I’m having the same issue despite adding maas to my nameservers, can I see your entire netplan yaml? I think it’s an issue of me not setting my maas region+rack controller to reference itself as a dns server but I’m a little stumped.
Except that you have specified the default gateway twice you config seems OK.
Here is mine for reference.
I have a separate network for PXE booting because i am bonding the interfaces on the compute nodes.
The br-private is used for all things Openstack in my case.