RDMA _ RoCE Connectivity Issue Between Two GPU Servers (Ping & ib_write_bw Failing).pdf (121.3 KB)
Hi,
I need help on connecting two DGX spark GB10 server.
Two GPU servers connected via NVIDIA/Mellanox 200 Gbps NICs show link UP and RDMA devices ACTIVE, but no IP connectivity (ping) and RDMA tests (ib_write_bw) fail between the hosts on the RoCE interfaces. Management network works normally.
🔹 Issues / Problem Statement
- Physical link is UP at 200 Gbps on both servers
- rdma link shows ACTIVE / LINK_UP for all RoCE devices
- Ping fails between servers on RDMA IP subnets
- ib_write_bw fails with socket connection error
- Correct IP addressing and routing present
- MTU tested with 1500 (no jumbo frames)
- Issue persists across all RDMA interfaces
🔹 Impact
- Unable to validate or use RDMA / RoCE between GPU servers
- Blocking GPU workload performance testing and deployment
🔹 Environment (Short)
- OS: Ubuntu Linux
- NIC: NVIDIA/Mellanox 200 Gbps (DAC, back-to-back)
- RDMA Type: RoCE
- Management network: 10.3.3.0/24 (working)
- RDMA networks: 192.168.100.0/24 – 192.168.103.0/24 (not working)
