GLBC: Each ingress sync updates resources for all ingresses

[bowei]

From @nicksardo on July 11, 2017 23:58

For every ingress, the controller runs resource synchronization for all ingresses, thus resulting in unnecessary work.

Furthermore, if a GCP error occurs when handling one of the ingress resources, an event will be raised for every ingress resource.

Copied from original issue: kubernetes/ingress-nginx#950

[bowei]

From @thockin on August 3, 2017 16:46

What event?

[bowei]

From @nicksardo on August 3, 2017 17:10

A K8s Event record noting an error/warning message.

Take a look at this code: https://github.com/kubernetes/ingress/blob/7b38e5a36ed7e097e1cd98aca0f37c504059ab4c/controllers/gce/controller/controller.go#L311-L320
Checkpoint (https://github.com/kubernetes/ingress/blob/7b38e5a36ed7e097e1cd98aca0f37c504059ab4c/controllers/gce/controller/cluster_manager.go#L141-L149) creates/manages the resources for all ingresses. If there's an error, it's returned and an event is recorded for the ingress which caused the sync, but not necessarily the ingress which caused the error.

[bowei]

From @thockin on August 3, 2017 17:12

blech!

On Thu, Aug 3, 2017 at 10:10 AM, Nick Sardo notifications@github.com
wrote:

A K8s Event record noting an error/warning message.

Take a look at this code: https://github.com/kubernetes/ingress/blob/
7b38e5a/controllers/gce/
controller/controller.go#L311-L320
Checkpoint (https://github.com/kubernetes/ingress/blob/
7b38e5a/controllers/gce/
controller/cluster_manager.go#L141-L149) creates/manages the resources
for all ingresses. If there's an error, it's returned and an event is
recorded for the ingress which caused the sync, but not necessarily the
ingress which caused the error.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
kubernetes/ingress-nginx#950 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AFVgVMFWnIrMtJ0fE_9BSeJJIEw2IWOTks5sUf74gaJpZM4OU9aH
.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

Prevent issues from auto-closing with an /lifecycle frozen comment.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or @fejta.
/lifecycle stale

[mooperd]

+1 Anyone looking at this? Our new ingresses can take upto 30 mins to be ready which is quite annoying in a dynamic environment.

[bowei]

We are actively working this issue. Expect an update in a few weeks.

[bowei]

/lifecycle frozen

[mooperd]

@nicksardo Hey, is this issue fixed? When is it being rolled out?

[nicksardo]

The logic was changed to single the relevant resources. Assuming testing goes well, it'll make K8s 1.10. Currently unknown which 1.9 or 1.8 patch releases will get this new version.