Run kubeadm job as a standalone job #2568
Conversation
k8s-ci-robot
added
the
cncf-cla: yes
|
kubernetes-security :P |
|
I should really learn to run these tests before creating the PR |
|
/lgtm It's tough to verify this sort of change, as @pipejakob knows, but this looks right. |
k8s-ci-robot
added
the
lgtm
|
this seems murdered the job |
|
nvm, the periodic version was not running normally before anyway |
|
Can you explain the motivation behind this move? |
|
We don't want this job to run on every PR. And we don't want the delays in this job to slow down batches merges. |
|
It doesn't make sense to have a job that runs_after_success of a presubmit but has always_run set to false. |
|
This probably isn't the best place to discuss what to do with this job, so I'll follow up elsewhere, but if I understand the change, this will effectively permanently break this pull job, since it was designed to only run after the bazel job (which builds/uploads CI .debs that are used by this kubeadm job). I've had a separate PR open to fix this job for a while now (#2509), so maybe that's the best place to discuss how to move forward. |
|
always_run set to false means its manually triggered. Just trigger it after the bazel job completes. |
|
Sure. Once the job is fully working and we actually want to trigger it more generally (either on every PR, or on PRs that change |
|
Yes, then it makes sense to do so. |
|
Got it, thanks! |
This had been previously disabled by kubernetes#2568. Adding the job back to the bazel pipeline and reenabling.
This had been previously disabled by kubernetes#2568. Adding the job back to the bazel pipeline and reenabling. Also, remove sporadic trailing whitespace.
This had been previously disabled by kubernetes#2568. Adding the job back to the bazel pipeline and reenabling. Also, remove sporadic trailing whitespace.
/assign @spxtr @pipejakob