Tags: dreamryx/tf-operator
Tags
add ActiveDeadlineSeconds and BackoffLimit features (kubeflow#963) * add ActiveDeadlineSeconds and BackoffLimit features * fix goimport and unassign variable in test * fix test and delete the package added by dep ensure * fix goimports * fix ActiveDeadlineSeconds unit test * add logger for test * add logger for test * add logger for test * add BackoffForOnFailure test * fix test * fix test * fix unit test
Update tf-operator v1beta1 documentation and examples (kubeflow#870) * Update tf-operator v1beta1 documentation and examples * Fix typo
Fix postusbmit builds; registry is incorrect. (kubeflow#832) * Related to kubeflow/kubeflow#1599
*: Add cleanpod policy for v1alpha2 (kubeflow#691) * *: Support cleanpod policy Signed-off-by: Ce Gao <gaoce@caicloud.io> * Gopkg: Keep code-generator Signed-off-by: Ce Gao <gaoce@caicloud.io> * vendor: Update Signed-off-by: Ce Gao <gaoce@caicloud.io> * test: Fix test cases Signed-off-by: Ce Gao <gaoce@caicloud.io> * test: Fix Signed-off-by: Ce Gao <gaoce@caicloud.io> * test: Fix style Signed-off-by: Ce Gao <gaoce@caicloud.io> * defaults: Update Signed-off-by: Ce Gao <gaoce@caicloud.io>
Don't leave pods running when a job completes. (kubeflow#512) * Don't leave pods running when a job completes. * We originally did this to preserve the logs. * But this ends up leaving pods running consuming resources. * The fix is straightforward * Transition to the cleanup phase before transitioning to the done phase. Fix kubeflow#128 * Don't teardown the cluster. * Don't set phase to cleanup when job is running. * Should only call get status if we are in creating or running phase. * Update the E2E test * Check that pod/service event creations are recorded * Check that pods are deleted when job ends. * Fix lint.