notebooks/components/tensorboard-controller
Hao Xin fcc4786a49 Fix(manifests): Upgrade rbac.authorization.k8s.io from v1beta1 to v1 (kubeflow/kubeflow#6261) 2022-02-03 16:18:16 +00:00
..
api/v1alpha1 Tesorboard-Controller: use updateStatus instead of update (kubeflow/kubeflow#5644) 2021-03-10 07:46:24 -08:00
config Fix(manifests): Upgrade rbac.authorization.k8s.io from v1beta1 to v1 (kubeflow/kubeflow#6261) 2022-02-03 16:18:16 +00:00
controllers tensorboard-controller: fix binding issue (kubeflow/kubeflow#5925) 2021-05-25 07:30:09 -07:00
hack initial checkin of tensorboard controller (kubeflow/kubeflow#4312) 2019-10-29 09:12:44 -07:00
.gitignore Manifests for Tensorboard controller (kubeflow/kubeflow#5730) 2021-03-21 14:28:17 -07:00
Dockerfile Bump Golang version in PodDefaults, TensorBoard Controller and KFAM to 1.17 (kubeflow/kubeflow#6180) 2021-10-08 06:02:23 -07:00
Makefile Add roles for Tensorboard controller pod (kubeflow/kubeflow#5262) 2020-08-30 06:56:20 -07:00
OWNERS Add OWNERs file to tensorboard controller (kubeflow/kubeflow#5088) 2020-08-07 06:32:19 -07:00
PROJECT initial checkin of tensorboard controller (kubeflow/kubeflow#4312) 2019-10-29 09:12:44 -07:00
README.md Add RWO_PVC_SCHEDULING env var to the Tensorboard Controller deployment (kubeflow/kubeflow#5266) 2020-08-31 08:12:21 -07:00
go.mod Bump Golang version in PodDefaults, TensorBoard Controller and KFAM to 1.17 (kubeflow/kubeflow#6180) 2021-10-08 06:02:23 -07:00
go.sum Bump Golang version in PodDefaults, TensorBoard Controller and KFAM to 1.17 (kubeflow/kubeflow#6180) 2021-10-08 06:02:23 -07:00
main.go Tensorboard controller: Add scheduling functionality for Tensorboard servers that use RWO PVCs as log storages (kubeflow/kubeflow#5218) 2020-08-26 02:58:03 -07:00
skaffold.yaml Fix docker builds of notebook and tensorboard controller (kubeflow/kubeflow#4664) 2020-01-21 17:54:34 -08:00

README.md

GSoC 2020 - TENSORBOARD CONTROLLER

Prequisites to build and run the controller:

  1. GO

  2. Docker

  3. kustomize

  4. kubectl

RUN TENSORBOARD CONTROLLER LOCALLY

Steps:

  1. Clone the repository

  2. Change directories to components/tensorboard-controller

  3. Generate and install manifests and build the controller: make install

  4. Run the controller locally: make run

If you want to enable the scheduling functionality for Tensorboard servers that use ReadWriteOnce PVCs as log storages, then set the RWO_PVC_SCHEDULING to true and run: RWO_PVC_SCHEDULING="true" make run

BUILD TENSORBOARD CONTROLLER IMAGE AND DEPLOY TO CLUSTER

  1. Clone the repository

  2. Change directories to components/tensorboard-controller

  3. Generate and install manifests and build the controller: make manifests

  4. Build and push the docker image: make docker-build docker-push IMG=YOUR_IMAGE_NAME

  5. Deploy the Tensorboard controller: make deploy IMG=YOUR_IMAGE_NAME

If you want to enable the scheduling functionality for Tensorboard servers that use ReadWriteOnce PVCs as log storages, then:

  1. Change directories to components/tensorboard-controller/config/manager

  2. Modify the manager.yaml file by navigating to the deployment.spec.template.spec field and manually setting the value of the RWO_PVC_SCHEDULING env var to "true" in the manager container.

  3. Run: make deploy IMG=YOUR_IMAGE_NAME