Closed acr92 closed 7 years ago
This looks like an internal GCP error having nothing to do with the router other than a coincidence that the IP it's unable to reserve happens to be for the router.
I "Fixed" it by reconfiguring deis-router as type: ClusterIP
, and then as type: LoadBalancer
(Which created a new load balancer).
Could be related to my experimentation. I'll close this issue, and hope that the issue doesn't surface again.
Hey, I'm running on Google Container Engine. Initially on 1.5.6, but then I upgraded to 1.6.1 (which I afterwards learned that it would break all
git push
, but...gave me an excuse to setup docker build for all images, so something positive came out of it).Latest version of Workflow.
First issue I saw was that suddenly some of my requests would hang forever. I manually added a healthcheck to the load balancer, and that worked around the issue:
Then I decided now to check the cloud console, and I saw a lot of errors related to Deis Router:
Any suggestions on how to fix this would be appreciated. I'm currently migrating everything from Deis 1.13 on AWS to Deis Workflow on GKE.