| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Kubernetes no longer uses 'kubecfg', but rather 'kubectl'.
Therefore, the way that pods and services are being created
needs to be changed.
In order to create a pod, the parameter 'kind' needs to be set
to 'pod' in the yaml file and the api version specified.
The stop script now reads from k8s/{pod,service,replication}/ in order
to stop all the started pods, services, and replication controllers.
Change-Id: Ibd39fa402e9df883df83272c3aefbb69009dfbd2
|
|
Neutron is not yet ready, so use flat nova networking instead.
Change-Id: I4905f2a54af2c0c06cc8cf95d21f5831b02fbe5e
|
|
Kubernetes does not have a horizontal scaling feature that works per
node. A workaround is to use hostPort to cause the scheduler not to schedule
several replicationController replicas to each node. By doing so, only one
nova-compute is launched per node up to a total of 6 nodes which is the most
we can generate in testing.
Change-Id: I1621f15cf1ec502418b454300645b46bc796dc8b
|
|
The nova-compute POD now starts and produces nearly correct output.
It may be that the libvirt container needs additional attention as it
is spitting out errors related to the firewall toolchain on the container.
Change-Id: I5bcd21c59f341494206a09e882631185ce66f336
|
|
Yaml is the future - json is the past
\o/
Change-Id: I36a9cbf8db034d1e8deba040a22e556d791c65ed
|
|
We want the nova-compute controller to run on each minion
Change-Id: If0355034927e195d1dd5482dab1026418185e9c6
|
