ingress-gce fails to create backend services when across zones with different balancing modes

[nimeshksingh]

We have a gke cluster with two zones, one of which was added significantly later than the other. The ingress controller added backend services with a balancing mode of UTILIZATION for the first zone, and when we added the second zone it set the balancing mode in the new zone to RATE. Now when we try to add new backend services, it goes into a failure loop where it tries the RATE balancing mode, which fails in one zone, then tries the UTILIZATION balancing mode, which fails in the second zone. This continues forever and we're blocked from adding new backends.

Image showing the initial state before trying to add a second backend service:

GKE Master version: 1.10.7-gke.1

[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.

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

[bowei]

/remove-lifecycle stale

[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.

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

[rramkumar1]

@nimeshksingh Is this still occurring? My understanding is that this should already be handled semi-gracefully but maybe you have hit an edge case.

[swetharepakula]

Closing as there has been no update since 2019.