aws-load-balancer-controller icon indicating copy to clipboard operation
aws-load-balancer-controller copied to clipboard

Published 20 hours ago • verified publisher icon kubernetes-sigs

Pod readiness gates not coming up for new pods

Open Somnath-Pandey-Vested opened this issue 1 year ago • 0 comments

Describe the bug A concise description of what the bug is. I have injected the readiness gate at the namespace level. When i try to deploy the pod using helm the readiness gate is coming as 1/1 but when i execute the below command `kubectl get po invest-test-66db578799-gcfg5 -n readiness -o yaml | grep -B7 'type: target-health' path: namespace status: conditions:

  • lastProbeTime: null lastTransitionTime: "2024-04-25T12:29:53Z" message: Target Group Binding is deleted status: "True" type: target-health.elbv2.k8s.aws/k8s-readines-investte-a2b6b3aaf3 when i do an upgrade of the same helm chart then the readiness gate is not coming upkubectl get po -o wide -n readiness NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES invest-test-66db578799-z44xk 1/1 Running 0 4m51s 10.3.18.153 ip-10-3-18-70.ap-south-1.compute.internal 0/1 invest-test-68f44889b4-ndx84 1/1 Running 0 9m 10.3.18.36 ip-10-3-18-70.ap-south-1.compute.internal ` Steps to reproduce

Expected outcome A concise description of what you expected to happen. When the readiness gate is installed at namespace level all the upcoming pods should get the readiness gate. Environment

  • AWS Load Balancer controller version - v2.4.3
  • Kubernetes version - 1.28
  • Using EKS (yes/no), if so version? yes v1.28

Additional Context:

Somnath-Pandey-Vested avatar Apr 25 '24 12:04 Somnath-Pandey-Vested