[
https://issues.apache.org/jira/browse/SUBMARINE-347?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17017947#comment-17017947
]
Wanqiang Ji commented on SUBMARINE-347:
---------------------------------------
[~tangzhankun], thanks for interest to this.
> We should mention this in the document
We mentioned at
[https://github.com/apache/submarine/blob/master/docs/submarine-server/setup-kubernetes.md#setup-kubernetes]
and
[https://github.com/apache/submarine/blob/master/docs/submarine-server/ml-frameworks/tensorflow.md#deploy-tf-operator]
> The curl command is not correct. The line break "\" is not working. And the
>"`" character should be "'".
Good catch, *`* should be replaced with *'*
**> We should have a note for user to check the running job in document. Like
"kubectl get TFJob". etc.
Good idea, I will add it later.
> Refine the TF job on K8s document of submarine server
> -----------------------------------------------------
>
> Key: SUBMARINE-347
> URL: https://issues.apache.org/jira/browse/SUBMARINE-347
> Project: Apache Submarine
> Issue Type: Improvement
> Components: Doc
> Reporter: Zhankun Tang
> Assignee: Wanqiang Ji
> Priority: Major
>
> 1. When trying to deploy the TF-operator following the document:
> {code:java}
> $ kubectl kustomize ./dev-support/k8s/tfjob/operator | kubectl apply -f -
> clusterrole.rbac.authorization.k8s.io/kubeflow-tfjobs-admin created
> clusterrole.rbac.authorization.k8s.io/kubeflow-tfjobs-edit created
> clusterrole.rbac.authorization.k8s.io/kubeflow-tfjobs-view created
> clusterrole.rbac.authorization.k8s.io/tf-job-operator created
> clusterrolebinding.rbac.authorization.k8s.io/tf-job-operator created
> Error from server (NotFound): error when creating "STDIN": namespaces
> "submarine" not found
> Error from server (NotFound): error when creating "STDIN": namespaces
> "submarine" not found
> Error from server (NotFound): error when creating "STDIN": namespaces
> "submarine" not found
> Error from server (NotFound): error when creating "STDIN": namespaces
> "submarine" not found{code}
> We should mention this in the document:
> {code:java}
> kubectl create namespace submarine{code}
> 2. The curl command is not correct. The line break "\" is not working. And
> the "`" character should be "'".
> {code:java}
> curl -H "Content-Type: application/json" --request POST --data
> '{"name":"mnist","librarySpec":{"name":"TensorFlow","version":"2.1.0","image":"gcr.io/kubeflow-ci/tf-mnist-with-summaries:1.0","cmd":"python
> /var/tf_mnist/mnist_with_summaries.py --log_dir=/train/log
> --learning_rate=0.01
> --batch_size=150","envVars":{"ENV_1":"ENV1"}},"submitterSpec":{"type":"k8s","configPath":null,"namespace":"submarine","kind":"TFJob","apiVersion":"kubeflow.org/v1"},"taskSpecs":{"Ps":{"name":"tensorflow","replicas":2,"resources":"cpu=4,memory=2048M,nvidia.com/gpu=1"},"Worker":{"name":"tensorflow","replicas":2,"resources":"cpu=4,memory=2048M"}}}'
> http://127.0.0.1:8080/api/v1/jobs
> {code}
> 3. We should have a note for user to check the running job in document. Like
> "kubectl get TFJob". etc.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@submarine.apache.org
For additional commands, e-mail: dev-h...@submarine.apache.org
