Is there an existing issue for this?

• I have searched the existing issues

What happened?

Each time a node needs more IP, the operator will log a warning indicating that the pod store is not initialized.
As a result the surge allocation is skipped and the operator will not allocate more than the pre-allocate value for the node (defaults to 8).

Cilium Version

1.15.0

Kernel Version

Linux xxx 5.10.215-203.850.amzn2.x86_64 #1 SMP Tue Apr 23 20:32:19 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux

Kubernetes Version

v1.28.8-eks-adc7111

Regression

1a9a1e7#diff-feb41af31e6ee7149118809ccc6fe42b0603dbb5e8f9c16a72bc2fe492c0b921L56-L61

Sysdump

No response

Relevant log output

level=warning msg="Unable to compute pending pods, will not surge-allocate" error="pod store uninitialized" instanceID=i-xxx name=xxx subsys=ipam
level=info msg="Resolving IP deficit of node" available=16 availableForAllocation=208 emptyInterfaceSlots=3 instanceID=i-xxx maxIPsToAllocate=1 name=xxx neededIPs=1 remainingInterfaces=4 selectedInterface=eni-xxx selectedPoolID=subnet-xxx subsys=ipam used=9

Anything else?

The pod store in operator/watchers/pod.go is created by the function PodsInit which is never called anywhere in the code.

Cilium Users Document

• Are you a user of Cilium? Please add yourself to the Users doc

Code of Conduct

• I agree to follow this project's Code of Conduct