| Age | Commit message (Collapse) | Author |
|---|
|
Automatic merge from submit-queue.
Fix OpenStack playbooks
This shuffles a few playbooks around to fix the ordering between various
facts and prerequisites.
|
|
Automatic merge from submit-queue.
[openstack] custom user commands for cloud-init
Allow to specify additional user commands executed on all Nova servers
provisioned via Heat.
An example use case is installing and starting os-collect-config agents
to put Nova servers under the configuration management driven via the
host openstack cloud Heat services. This allows to integrate with another
deployment tools like TripleO.
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
Automatic merge from submit-queue.
Removing prefix, replacing with cidr, pool_start and pool_end vars
The heat template was hardcoded with a /24 cidr and that limited customers to 251 ip addresses in the OpenStack subnet. This allows the user to configure the cidr and the allocation pool start and end.
Addresses issue #6829 that I created last week.
@tomassedovic please take a look
|
|
This shuffles a few playbooks around to fix the ordering between various
facts and prerequisites.
|
|
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
Document use cases for custom post-provision ansible hooks
vs cloud-init runcmd shell commands. Rename to
openshift_openstack_cloud_init_runcmd.
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
Allow to specify additional user commands executed on all Nova servers
provisioned via Heat.
An example use case is installing and starting os-collect-config agents
to put Nova servers under the configuration management driven via the
host openstack cloud Heat services. This allows to integrate with another
deployment tools like TripleO.
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
|
|
This commit will allow the user to login as root. By default, the
user is set to openshift and disable_root is set to true.
|
|
Automatic merge from submit-queue.
Fix: e2e tests failing due to :1936/metrics unaccessible.
This PR allows access to tcp/1936 ingress infra-secgrp policy, so that e2e test do not fail looking for router metrics (:1936/metrics).
|
|
|
|
|
|
|
|
This PR allows access to tcp/1936 ingress infra-secgrp policy,
so that e2e test do not fail looking for router metrics port
:1936/metrics
|
|
Fix provider network support at openstack playbook
|
|
When deploying on OpenStack with internal DNS configured, this will set
`openshift_hostname` to the Nova server name instead of its IP address.
Without those two matching, the OpenStack cloud provider configuration
will fail and the OpenShift nodes will not start.
|
|
It ensures no floating ips are attached if a provider
network is used
|
|
|
|
|
|
Support private/public hostnames suffixes for DNS records.
Real hostnames, Inventory variables, Nova servers and ansible
hostnames will ignore the custom suffixes. Those are only for
nsupdates sent to external DNS servers.
Related change: add openshift_openstack_public_dns_domain to
the role defaults to not rely on the group vars example only.
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
Co-authored-by: Tomas Sedovic <tsedovic@redhat.com>
|
|
|
|
* Do not manage hostnames for openstack cloud provider,
let cloud-init to do its job.
* Make python-dns / dnspython dependency check conditional.
* Drop not used dns node flavor and image.
* Do not manage dns nodes and sec groups in heat stacks.
* Keep supporting dynamic updates for private DNS records, yet only
limited to an external DNS managed elsewhere (not deployed by
the openshift_openstack provider). So users may still benefit from
this feature, sending nsupdates to private and public servers as
they want it.
* Fix openstack default for external nsupdate keys. It should be
undefined by default as the dns-populate logic is based on that.
* Fix dns records generation for openstack provider's populate-dns
* Update docs
|
|
Remove references for bastion, ssh UI tunnek and static inventory.
Update docs.
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
|
|
When a package install/update fails due to network blips or other spotty
availability, retry it. If the failure is a real failure (e.g. package
is really not there) it still fails after 3 tries (Ansible default).
|
|
|
|
|
|
This makes sure that all the variables used in the `openshift_openstack`
role are prefixed with `openshift_openstack_` as is the convention.
|
|
The `openstack_*_network_name` vars are strings, not booleans, so the
absense shouldn't really be marked by `False`.
|
|
Because the templates are present in a role, the `template` module is
able to look them up directly, without having to use `{{ role_path
}}/templates`.
|
|
|
|
|
|
Most of the vars in `roles/openshift_openstack/defaults/main.yml` are
now prefixed with `openstack_`.
|
|
It's no longer being used.
|
|
The contents of roles/openshift_openstack/vars/main.yml were moved to
the defaults/main.yml file instead.
There are now duplication warnings we need to address, but the
deployment does still work.
|
|
This will mostly not work but it's a starting point.
|
|
All the tasks that were previously in playbooks are now under
`roles/openshift_openstack`.
The `openshift-cluster` directory now only contains playbooks that
include tasks from that role. This makes the structure much closer to
that of the AWS provider.
|
