This is part of the steps to remove usage of six library, which is no
longer needed since python 2 support was removed.
Change-Id: I683a86eeab858c3f57c587e653ae53aff052fe9c
The mock third party library was needed for mock support in py2
runtimes. Since we now only support py36 and later, we can use the
standard lib unittest.mock module instead.
Change-Id: Ib3ca91e93db260e081c43d236d5d6fae5a2adfd7
Now that we no longer support py27, we can use the standard library
unittest.mock module instead of the third party mock lib.
Change-Id: I6245a207ec65df0ef11a798b4cffe3157abccb73
Signed-off-by: Sean McGinnis <sean.mcginnis@gmail.com>
Co-Authored-By: Spyros Trigazis <spyridon.trigazis@cern.ch>
The original design of k8s cluster health status is allowing
the health status being updated by Magnum control plane. However,
it doesn't work when the cluster is private. This patch supports
updating the k8s cluster health status via the Magnum cluster
update API by a 3rd party service so that a controller (e.g.
magnum-auto-healer) running inside the k8s cluster can call
the Magnum update API to update the cluster health status.
Task: 38583
Story: 2007242
Change-Id: Ie7189d328c4038403576b0324e7b0e8a9b305a5e
This adds the support for creating and deleting worker nodegroups
using different stack per nodegroup. In order to be backwards
compatible, default nodegroups will remain in one stack.
Having this in mind cluster status is now calculated aggregating the
statuses of the underlying stacks.
Change-Id: I97839ab8495ed5d860785dff1f6e3cc59b6a9ff7
The existing drivers are adapted to get node_count and master_count
information from the cluster's nodegroups. At the same time the
output mappings were updated to reflect the changes in the stack to
the nodegroups.
story: 2005266
Change-Id: I725413e77f5a7bdb48131e8a10e5dc884b5e066a
This changes the existing cluster APIs and the cluster conductor to
take into consideration nodegroups:
* create: now creates the default nodegroups for the cluster
* update: updates the default nodegroups of the cluster
* delete: deletes also the nodegroups that belong to the cluster
* cluster_resize: takes into account the nodegroup provided by the API
story: 2005266
Change-Id: I5478c83ca316f8f09625607d5ae9d9f3c02eb65a
Now an OpenStack driver for Kubernetes Cluster Autoscaler is being
proposed to support autoscaling when running k8s cluster on top of
OpenStack. However, currently there is no way in Magnum to let
the external consumer to control which node will be removed. The
alternative option is calling Heat API directly but obviously it
is not the best solution and it's confusing k8s community. So with
this patch, we're going to add a new API:
POST <ClusterID>/actions/resize
And the post body will be:
{
"node_count": 3,
"nodes_to_remove": ["dd9cc5ed-3a2b-11e9-9233-fa163e46bcc2"],
"nodegroup": "production_group"
}
The API will be working in a declarative way. For example, there
are 3 nodes in the cluser now, user can propose an API request
like above. Magnum will call Heat to remove the node
dd9cc5ed-3a2b-11e9-9233-fa163e46bcc2 firstly, then bring the node
count back to 3 again.
Task: 29563
Story: 2005052
Change-Id: I7e36ce82c3f442976cc498153950b19c56a1759f
For k8s cluster, the loadbalancers created for LoadBalancer type
services should be deleted before the cluster deletion.
Change-Id: I75f44187b7be7d0ffb6a8f195f755de4b1564335
Closes-Bug: #1712062
This commit addresses multiple potential vulnerabilities in
Magnum. It makes the following changes:
* Permissions for /etc/sysconfig/heat-params inside Magnum
created instances are tightened to 0600 (used to be 0755).
* Certificate retrieval is modified to work without the need
for a Keystone trust.
* The cluster's Keystone trust id is only passed into
instances for clusters where that is actually needed. This
prevents the trustee user from consuming the trust in cases
where it is not needed.
* The configuration setting trust/cluster_user_trust (False by
default) is introduced. It needs to be explicitely enabled
by the cloud operator to allow clusters that need the
trust_id to be passed into instances to work. Without this
setting, attempts to create such clusters will fail.
Please note, that none of these changes apply to existing
clusters. They will have to be deleted and rebuilt to benefit
from these changes.
Change-Id: I643d408cde0d6e30812cf6429fb7118184793400
This commit changes the incorrect behavior of cluster create workflow.
Now db record with status CREATE_IN_PROGRESS is created right after
related API request.
Change-Id: I11692c4126823d49672ba5172fa45774bf0ce544
Closes-bug: #1640729
This is an alternative implementation to:
https://review.openstack.org/#/c/397961
This version implements an earlier proposal from the
spec that adds a driver method for synchronizing
cluster state. This method is optional so that drivers
that do not wish to leverage the existing periodic
synchronization task can do so in whatever manner
they wish and Magnum will not force them to do anything
unnecessarily.
1. add an update_cluster_status method to the driver
interface
2. implment update_cluster_status for Heat drivers
using the existing tested logic
3. Remove cluster status updates from the cluster conductor
in favor of the periodic sync_cluster_status task - this
should avoid timeouts and race conditions possible in the
previous implementation
4. Update the periodic sync_cluster_status method to use
the driver to update cluster status rather than calling
Heat directly
Change-Id: Iae0ec7af2542343cc51e85f0efd21086d693e540
Partial-Blueprint: bp-driver-consolodation
Refactor driver interface to encapsulate the orchestration
strategy. This first patch only refactors the main driver
operations. A follow-on will handle the state synchronization
and removing the poller from the conductor.
1. Make driver interface abstract
2. Move external cluster operations into driver interface
3. Make Heat-based driver abstract and update based on
driver interface changes
4. Move Heat driver code into its own module
5. Update existing Heat drivers based on interface changes
Change-Id: Icfa72e27dc496862d950ac608885567c911f47f2
Partial-Blueprint: bp-driver-consolodation
We currently allow Magnum to scale down mesos cluster by removing nodes
from the cluster's ResourceGroup by updating the heat stack that created
the cluster. The problem with this approach is that Heat decides which
nodes to delete, and all containers on that node will also be deleted.
The smart cluster scale down feature has been implemented for k8s bays(
for k8s cluster, we'll ask Heat to delete nodes that have NO CONTAINERS
on them).
This patch proposes a similar implementation for a mesos cluster.
Change-Id: I00cda7f35c9db978bdc604cf86603ef58e339256
Implements: blueprint mesos-smart-bay-scale-down
After changing create to async operation, all the exceptions goes hidden
and no db entry is created to show the actual failure. One has to go through
the logs to find the actual error. This patch creates a db entry in cluster
table to show the actual error in column 'status_reason'.
Change-Id: Iad6e8bfce7326b34dea04914e4552f87d2796e86
Closes-bug: #1623387
This is patch 3 of 3 to change the internal usage of the terms
Bay and BayModel. This patch updates Bay to Cluster in DB and
Object as well as all the usages. No functionality should be
changed by this patch, just naming and db updates.
Change-Id: Ife04b0f944ded03ca932d70e09e6766d09cf5d9f
Implements: blueprint rename-bay-to-cluster
Michal Nasiadka