Skip to content

Multiple ztunnel getting stuck in crashloopbackoff with error wrokloadmanager failed to start proxy: can't assign requested address #1707

@karan56625

Description

@karan56625

Hi folks, we are running istio 1.28 on very high load AKS cluster. Recently we on-boarded Istio on the cluster and put logging components under service mesh, we saw 4-5 ztunnel pods getting crashedloopbaxkoff with the following error:
"
2026-01-07T17:51:33.579503Z error inpod::workloadmanager failed to start proxy: Bind([::1]:15053, Os { code: 99, kind: AddrNotAvailable, message: "Cannot assign requested address" }) uid=2a6140db-b27a-4f11-a7ff-58643b6ce31f
"
Due to this, workload inside the servicemesh as well got crashedloopbackoff.
Even restarting the ztunnel didn't help there. Once we put the workload outside the mesh, both ztunnels and workload pods started running fine.

Istio; 1.28
Platform: AKS 1.32 with Azure linux 3.0

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions