Stack Overflow
Stack Overflow

Questions tagged [karpenter]

20 questions
2
votes
1 answer

How to update EKS nodes version created by Karpenter?

Initially, I was planning to use AWS Managed Node Groups with Cluster Autoscaler, but then I started exploring Karpenter. It sounds promising, but I also care about the simplicity of overall architecture. I know if I use AWS Managed Node Groups the…
jumpy
  • 317
  • 1
  • 12
1
vote
1 answer

Monitoring Kubernetes POD scheduling (movement)

We have recently introduced Karpenter autoscaler in our Kubernetes infrastructure. One of its strength is the deprovisioning mechanism. However given that we have a lot of stateful workload, this can cause some serious issue at time. We are taking…
MaatDeamon
  • 9,532
  • 9
  • 60
  • 127
1
vote
1 answer

ERROR controller.provisioning Could not schedule pod, incompatible with provisioner "default", incompatible requirements, key karpenter.sh/provisioner

I read through the karpenter document at https://karpenter.sh/v0.16.1/getting-started/getting-started-with-terraform/#install-karpenter-helm-chart. I followed instructions step by step. I got errors at the end. kubectl logs -f -n karpenter -l…
Gary
  • 71
  • 9
1
vote
1 answer

AWS EKS - use spot nodes with failover to on-demand

Can you please kindly suggest. I run GPU-based workload, and my instances are g4dn.large. I would like to use spot instances, since on-demand cost a lot :-) But it is very usual that spot GPU instances are not available for long periods. I have…
dmitrii
  • 31
  • 1
0
votes
0 answers

Using Karpenter to autoscale spark on Kubernetes cluster

We are currently using EMR to run our Spark service and we want to migrate to Kubernetes. One of the things we thought of doing is using karpenter as the autoscaler of the Spark Kubernetes cluster we want to create. I want to know, is there a way to…
idan ahal
  • 707
  • 8
  • 21
0
votes
0 answers

Karpenter node/machine creation metrics not present on karpenter.karpenter.svc.cluster.local:8000/metrics on scaling up

I am using karpenter to handle autoscaling in my EKS cluster. As per the karpenter metrics doc karpenter exposes its metrics on karpenter.karpenter.svc.cluster.local:8000/metrics . I could see the metrics on this port. Expected behavior is that when…
Abhinay Chauhan
  • 11
  • 3
0
votes
1 answer

Karpenter consolidation without any downtime for pods

I have karpenter deployed in eks and for karpenter provisioner I have consolidation enabled true consolidation: enabled: true which means for any underutilized node, the workload (pods) will be deployed in the existing nodes. but during this…
saireddy bonthu
  • 1
0
votes
1 answer

How to enable hugepages on eks cluster?

How to deploy hugepages in EKS aws cluster? Ideally, managed nodes using Karpenter. Can't find any docs anywhere.
Pav K.
  • 2,548
  • 2
  • 19
  • 29
0
votes
0 answers

Is it the intended behaviour of Karpenter to record huge amount of EC2 fleet resources in AWS Config?

**Info:**When using Karpenter as an autoscaler in my EKS cluster, tens of thousands of entries is recorded in AWS Config with the resource type "resourceType":"AWS::EC2::EC2Fleet". Goal: I would like it to not do that, because the cost of AWS Config…
uggl
  • 31
  • 3
0
votes
1 answer

Karpenter autoscaling eks

How to ensure Karpenter spin up instances equally in all availability zones? Used the below code, but it will not spin up pods in all availability zone's at the same time.Rather it picks random availability zone for provisioning new nodes. - key:…
Bharath Nadigoti
  • 1
  • 1
0
votes
0 answers

Karpenter - half of nodes state are not ready after simulate spot disruption with aws fis

I have an EKS cluster running with Karpenter provisioning. Everything worked as expected, but when I used AWS FIS to simulate spot instances interruption, I faced a weird behavior - new nodes provisioned, but half of the new nodes were stuck in not…
TanIkemen
  • 53
  • 1
  • 7
0
votes
1 answer

How to troubleshoot Karpenter using k9s?

I have a kubernetes cluster with Karpenter installed in it. For some reasons, Karpenter doesn't launch the right nodes for my scheduled pods, and I want to troubleshoot it. How to debug karpenter log and get karpenter configuration in my cluster…
Brainless
  • 1,522
  • 1
  • 16
  • 30
0
votes
1 answer

Can I configure HPA and/or Karpenter to remove pods located on Karpenter-provisioned nodes with higher priority?

I am using HPA(Horizontal Pod Autoscaler) and Karpenter in my AWS EKS cluster to increase the number of pods and provision new nodes when my application encounters high traffic, respectively. My application is a simple API serving pod, that receives…
SHM
  • 61
  • 1
  • 8
0
votes
0 answers

Connectivity issues to EKS Fargate pod

I'm running an EKS cluster with several regular EC2 nodes, and a single pod running in fargate (karpenter). My problem is that I can't seem to connect from any of the EC2 nodes into the fargate pod. Here's what I've tried: Started ubuntu pod in one…
kenske
  • 2,235
  • 2
  • 20
  • 27
0
votes
1 answer

What's the difference between creating one fargate profile for each availability zone, and creating a single profile that spans multiple AZs?

I'm in the process of creating fargate profiles for my AWS EKS cluster using terraform. In this example for the Terraform Karpenter module, they have a loop that creates one profile for each of the 3 availability zones used in the example: …
kenske
  • 2,235
  • 2
  • 20
  • 27
1
2