Very slow POD to POD throughput via vxlan when the Hypervisor has bnx2x_en NIC(s)

Solution In Progress - Updated -

Issue

  • On our OCP cluster, communication between pods running on different worker nodes is extremely slow. The cluster is installed on 25 GB Ethernet, yet inter pod communication has throughput only in 8-9 Mbit range.

Environment

  • Red Hat Enterprise Virtualization
  • Openshift
  • bnx2x_en NIC
    HPE Eth 10/25Gb 2p 631FLR-SFP28 Adptr
    Ethernet controller: Broadcom Inc. and subsidiaries BCM57414 NetXtreme-E 10Gb/25Gb RDMA Ethernet Controller (rev 01)

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In