Trying to create a pod using a resourcequota whose status has not been
populated will result in api admission rejecting it with a 403
Forbidden. The pod resize tests do not wait/check to make sure the
resourcequota status is populated before trying to use the quota when
creating a pod, so in scenarious where either the test runs too fast, or
the resourcequota controller is too slow, the tests will fail.
This just makes the test wait up to a minute for the quota status to
populate. The status generation timing/logic gets tested separately in
test/e2e/apimachinery/resource_quota.go, and it waits a minute, so we
wait a minute, but in practice it takes at worst seconds.
The controller is available only with SELinuxChangePolicy FG enabled.
Kubernetes e2e jobs do the right --focus & --skip explicitly in the SELinux
jobs, this just helps other test runners to figure out what FGs are needed
to run the tests.
When at it, I noticed SELinuxMountReadWriteOncePod tag is always required, so
add it on the common level and not in each test.
In a downstream test job, I've seen an e2e test failing to create a
directory for Pod logs when running an e2e test. Shorten them to 255
characters.
For example, consider this test: "[sig-storage] CSI Mock selinux on mount metrics and SELinuxWarningController SELinuxMount metrics [LinuxOnly] [Feature:SELinux] [Serial] error is not bumped on two Pods with a different policy RWX volume (MountOption + MountOption) [FeatureGate:SELinuxMountReadWriteOncePod] [Beta] [FeatureGate:SELinuxChangePolicy] [Beta] [FeatureGate:SELinuxMount] [Beta] [Feature:OffByDefault] [sig-storage, Feature:SELinux, Serial, FeatureGate:SELinuxMountReadWriteOncePod, Beta, FeatureGate:SELinuxChangePolicy, FeatureGate:SELinuxMount, Feature:OffByDefault, BetaOffByDefault]"
During execution, the test will create a directory
`error_is_not_bumped_on_two_Pods_with_Recursive_policy_and_a_different_context_on_RWX_volume_FeatureGate_SELinuxMountReadWriteOncePod_Beta_FeatureGate_SELinuxChangePolicy_Beta_FeatureGate_SELinuxMount_Beta_Feature_OffByDefault_`,
which has 226 characters and it's close to the limit (256).
The existing TrafficDistribution test didn't really distinguish "same
zone" from "same node". Add another test that makes sure there are at
least 2 nodes in each zone so it can do that.
(Keep the original test as well to avoid losing coverage in CI systems
with single-schedulable-node-per-zone clusters.)
