Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

set termination policy to worker 0 by default when no master exists #444

Merged
merged 1 commit into from
Mar 19, 2018
Merged

Conversation

raddaoui
Copy link
Contributor

@raddaoui raddaoui commented Mar 17, 2018

most tensorflow training examples use worker 0 as the chief worker instead of master.
using ksonnet tf-job prototype to create training jobs dosen't allow that
and will make the tfjob fails if we set num_masters to 0.


This change is Reviewable

@jlewi
Copy link
Contributor

jlewi commented Mar 17, 2018

Thanks!
/lgtm
/approve

@k8s-ci-robot
Copy link
Contributor

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: jlewi

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@jlewi
Copy link
Contributor

jlewi commented Mar 19, 2018

/ok-to-test

@k8s-ci-robot k8s-ci-robot merged commit 5dca627 into kubeflow:master Mar 19, 2018
@jlewi jlewi mentioned this pull request Mar 20, 2018
jlewi added a commit to jlewi/kubeflow that referenced this pull request Mar 20, 2018
* kubeflow#444 add an argument for the termination policy but the TFCnn prototype
  isn't setting it.

* The TFCnn prototype overrides the termination policy so we can just pass null
  for the TP.
k8s-ci-robot pushed a commit that referenced this pull request Mar 21, 2018
* #444 add an argument for the termination policy but the TFCnn prototype
  isn't setting it.

* The TFCnn prototype overrides the termination policy so we can just pass null
  for the TP.
yanniszark added a commit to arrikto/kubeflow that referenced this pull request Nov 1, 2019
* tests: add script to add images to kustomizations

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* remove tabs from yaml files

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* e2e: check that kustomization is filled correctly

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* format all kustomizations the same way

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* kustomizations: add missing images

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* regenerate tests

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>

* fix run-tests script

Signed-off-by: Yannis Zarkadas <yanniszark@arrikto.com>
yanniszark pushed a commit to arrikto/kubeflow that referenced this pull request Nov 1, 2019
* Update the notebook-controller image.

* It looks like the code may not have actually changed since we last
  built the image. Nonethless, lets update the tag to use the sha
  as opposed to a tag like v20190911-e8193317-dirty-cd2831

* Update the tests.

* Disable the new test for making sure the images are set because it doesn't
handle digests.
  * Related to kubeflow#444
yanniszark pushed a commit to arrikto/kubeflow that referenced this pull request Feb 15, 2021
elenzio9 pushed a commit to arrikto/kubeflow that referenced this pull request Oct 31, 2022
This pull request was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants