With the memory manager static policy:
- start multiple guaranteed pods and verify that pods succeeded to start
- start workload pod on each NUMA node to load the memory and start the
pod that requested more memory than each NUMA node have, the pod should fail
to start with the admission error, because no single NUMA node has enough
memory to start the pod and also each NUMA node already used for single
NUMA node allocation
The test requires at least two NUMA nodes
Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
Provides basic tests e2e to verify that pod succeeds
to start with MemoryManager enabled.
Verifies both MemoryManager policies and when the node has
multiple NUMA nodes it will verify the memory pinning.
Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
Due to the dockerhub rate limiting, we had to find an alternative solution. We've mirrored the dockerhub
images into our own.
Additionally, our own busybox, httpd, and nginx images also have Windows support.
Current gcRegistry configuration is responsible for both etcd and pause images. We should
use the upstream pause image in testing.
The etcd image doesn't have Windows support yet, so we need to have a separate configuration
for it.
We've added Windows support to the resource-consumer image and 1.8 tag is already promoted.
We need to bump the VERSION, so we can promote the new image.
The PHP release page tends to get updated and changed every time there's a
new release, removing the old ones. Because of this, the PHP link in the
httpd and httpd-new images may become invalid.
Updating the links to the archives solves this issue.
These are the latest stable releases. We should test with those.
The newer external-provisioner no longer needs (and doesn't support)
the --provisioner parameter.
This adds a call to createBalancedPods during the ubernetes_lite scheduling e2es,
which are prone to improper score balancing due to unbalanced utilization.
Instead of allowing the cloud provider to guess at the zones that
should be applied for a cluster under test, allow the explicit list
of zones to consider to be passed as a new test context flag -gce-zones.
Only the GCE test cloud provider recognizes this value because only
the GCE test cloud provider makes assumptions about zones for verifying
values, and the default assumptions for GKE do not always match non-GKE
providers.
A number of e2e tests are useful to run after the system has been
disrupted or is in the progress of being disrupted, but the current
suite and test logic blocks progress waiting for all nodes to be
healthy.
By passing -1 to --minStartupPods or --allowed-not-ready-nodes flags
the caller can bypass wait logic before and after test suites that
would prevent running e2e during disruption. This allows use of parts
of the e2e suite during cluster duress to verify that controllers or
components still function.
Adds the httpd, nginx, images that are used in tests.
Two different versions of nginx have to build, and thus, the have
different folders. An ALIAS file was added to nginx-new in order to
keep the same image name.
