fix job active count; count terminating pod as failed (#214)

Signed-off-by: yowenter <wenter.wu@gmail.com>

Upgrade PriorityClass apiVersion to scheduling.k8s.io/v1 (#208)

Signed-off-by: Yuki Iwai <yuki.iwai.tz@gmail.com>

Signed-off-by: Yuki Iwai <yuki.iwai.tz@gmail.com>

Remove deprecated labels (#200)

update k8s dependencies to 1.25 (#198)

* update k8s dependencies to 1.25

* update k8s dependencies to 1.25

* update k8s dependencies to 1.25

* Fix new line

Fix exitCode check (#190)

The exitCode check works differently then described in the API: https://github.com/kubeflow/common/blob/master/pkg/apis/common/v1/types.go\#L163-L168

In addition the document exitCodes for Docker will not appear in a Kubernetes pod status (anymore?)

Update all k8s dependencies to 0.24.1 (#194)

* Update all k8s dependencies to 0.24.1

* Update go flags

* Update go flags

* Fix lint

* Fix mod

* Fix deps

feat: Add LabelSelector (#177)

* feat: Add LabelSelector

Signed-off-by: Ce Gao <ce.gao@outlook.com>

* fix: Update mod

Signed-off-by: Ce Gao <ce.gao@outlook.com>

* chore: Update go version

Signed-off-by: Ce Gao <ce.gao@outlook.com>

Revert #135: change rtype to commonv1.ReplicaType (#158)

#135 This change brings some extra side effects and make training operator dependency upgrade fail. Since this change is kind of refactor, we determine to revert it at this moment.

Fix ReplicaType key for filtering pods and services (#152)

fix: Add uppercase support for GenGeneralName (#146)

Signed-off-by: cegao <cegao@tencent.com>