cleanPodPolicy Set to Running should clean Running pod

[xrmzju]

https://github.com/kubeflow/pytorch-operator/blob/047cf0f41e68e030158f532017a226c18827a660/pkg/controller.v1/pytorch/job.go#L160 we just ignore running policy for now

[issue-label-bot[bot]]

Issue-Label Bot is automatically applying the labels:

Label	Probability
bug	0.57

Please mark this comment with :thumbsup: or :thumbsdown: to give our bot feedback! Links: app homepage, dashboard and code for this bot.

[xrmzju]

@gaocegege

[gaocegege]

When the PyTorchJob is failed, all replicas should be failed. Then there is no difference between none and running. Then we ignore it. Do you have problem with it?

/cc @johnugeorge

[xrmzju]

When the PyTorchJob is failed, all replicas should be failed. Then there is no difference between none and running. Then we ignore it. Do you have problem with it?

/cc @johnugeorge

but in my condition it seems not like this...

pytorch-test-master-0                                               1/1     Running                0          4m15s
pytorch-test-worker-0                                               0/1     Error                  0          4m15s
pytorch-test-worker-1                                               0/1     Error                  0          4m15s

[gaocegege]

@johnugeorge WDYT