Skip to content

Missing IPC_LOCK capability to server, causes major page faults #1138

Description

@Stogas

Describe the bug
Helm chart does not add IPC_LOCK capability to server container, as documented in https://developer.hashicorp.com/vault/tutorials/kubernetes/kubernetes-security-concerns#ensure-mlock-is-enabled

To Reproduce
Install chart default server.statefulSet.securityContext.

For the Raft leader server, observe non-zero major page faults (I monitor from the kubernetes node with command pidstat -r -C vault 5).

In my case, the raft leader server created on average at least ~50 major page faults per second.

Expected behavior
IPC_LOCK is added as a capability as per the documentation, and Vault pods do not generate unecessary page faults

Environment

  • Kubernetes version: v1.32.3
    • Distribution or cloud vendor (OpenShift, EKS, GKE, AKS, etc.): kubeadm, self-managed
    • Other configuration options or runtime services (istio, etc.): N/A
  • vault-helm version: 0.30.0

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions