Conflict between "DNS zone name" in Cluster controller and "Default domain for new nodes" in settings
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
MAAS |
Fix Released
|
High
|
Raphaël Badin |
Bug Description
Running 1.2+bzr1360+
If MAAS manages DHCP and DNS, and under Cluster controller we leave the default DNS zone name to "master" and then in the settings panel, down under Network configuration we leave the default "local" for "Default domain for new nodes" we cannot use the constraint "maas-name=
juju deploy --repository=. --config=
And then in /var/log/maas:
ERROR 2013-03-15 12:30:06,214 maas.maasserver #######
ERROR 2013-03-15 12:30:06,215 maas.maasserver Traceback (most recent call last):
File "/usr/lib/
response = callback(request, *callback_args, **callback_kwargs)
File "/usr/lib/
response = func(*args, **kwargs)
File "/usr/lib/
result = self.error_
File "/usr/lib/
result = meth(request, *args, **kwargs)
File "/usr/lib/
return function(self, request, *args, **kwargs)
File "/usr/lib/
raise NodesNotAvailab
NodesNotAvailable: No matching node is available.
Even though controller.master CNAME is added to the maas zone file in bind.
I checked in the django models and the reason is that we use the "local" and not the "master" domain:
$ sudo maas shell
Python 2.7.3 (default, Aug 1 2012, 05:14:39)
[GCC 4.6.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
(InteractiveCon
>>> from maasserver.models import Node
>>> Node.objects.
u'controller.local'
>>>
If we specify the constraint maas-name with the "local" domain then it works.
I'm not sure if this is expected or if it would be better to give preference to the DNS name set by the maas-dns config with bind.
Thanks.
Related branches
- Jeroen T. Vermeulen (community): Approve
-
Diff: 63 lines (+34/-1)2 files modifiedsrc/maasserver/node_constraint_filter_forms.py (+14/-1)
src/maasserver/tests/test_node_constraint_filter_forms.py (+20/-0)
summary: |
- "DNS zone name" in Cluster controller and "Default domain for new - nodes" in settings should be the same + Conflict between "DNS zone name" in Cluster controller and "Default + domain for new nodes" in settings |
Changed in maas: | |
status: | New → Triaged |
importance: | Undecided → High |
tags: | added: trivial ui |
Changed in maas: | |
milestone: | none → 13.10 |
Changed in maas: | |
assignee: | nobody → Raphaël Badin (rvb) |
status: | Triaged → In Progress |
tags: | removed: trivial ui |
Changed in maas: | |
status: | In Progress → Fix Committed |
Changed in maas: | |
status: | Fix Committed → Fix Released |
Actually, I don't think this is a simple ui problem. This is a side-effect of the mix-up between a node's hostname and a node's fqdn.
We could change what "maas-name" refers to (from node.hostname to node.fqdn) but that could break existing scripts and thus I would be a bit reluctant to do that. One backward-compatible solution would be to add a new tag named something like "maas-fqdn" and make that tag point to node.fqdn.