-
Notifications
You must be signed in to change notification settings - Fork 675
Timeout error when performing step to install flux #63
Comments
@ZillaG When helm install In your EKS console, the screenshot that you posted is consistent with that. Can you probe any further to see what has caused them to fail becoming ready? Some candidate reasons are: the deployment was not able to pull images from outside of a private network In |
I see the following.
Note I created a Fargate cluster with the following Fargate profiles
|
I created the following Fargate profile to see if it helps
|
Sorry, the deployment describe doesn't provide much useful information. Could you see if there are pods in any state with |
Here you go. Thanks!
|
The important info is there at the bottom, failed scheduling: Warning FailedScheduling 3m22s (x137 over 3h24m) default-scheduler 0/2 nodes are available: 2 Too many pods. I don't know how fargate clusters work unfortunately so not sure how I can help troubleshoot this further. I think you are on the right track, this appears to be a fargate concern of some kind, somehow you are getting a too many pods error on a fargate cluster? Check if your fargate profile is scheduling t2.micro can be increased to a larger profile, googling of this error leads me to believe the network ENIs have been exhausted and this problem will go away if a larger instance profile is used. https://stackoverflow.com/questions/64965832/aws-eks-only-2-pod-can-be-launched-too-many-pods-error |
I doubt that the stackoverflow link is similar to my issue. With Fargate, I shouldn't have to care the node type, which is the whole purpose of Fargate. |
Changing the fargate profile to this made the difference.
|
I'm blindly following the walkthrough, and when I get to this step I get a timeout error. I did replace the repo URL to my forked repo, and am using my company's AWS EKS cluster that I set up, using Kubernetes v1.18.
What do I need to do to proceed?
The text was updated successfully, but these errors were encountered: