Incident Library
Applied filters ComponentHost routes and routing rulesFactorUnreviewed defaultsClear allEvery case under the filters, with no sentence; newest first.
Suggested filters Search, and the values found in your words appear here.

Graph

  • Root cause
  • Component
  • Symptom
  • Contributing factor
  • Case

Cases1 case

1 to 1 of 1

  1. Symptoms

    In Infrawaves' RoCEv2 GPU training clusters, LLM training jobs crashed because their communication collective collapsed after an RNIC port flapped. The Global Identifier (GID) of the RDMA network card's port, which in RoCEv2 is expected to stay the same after initialisation, had changed: with the default OS network manager, NetworkManager, the port briefly lost its IP address while it was down, and the GID number was incremented even though the original IP address came back. The paper describes the GID shift as non-deterministic.

    Trigger

    Ports of the RDMA network cards (RNICs) on the GPU servers flapped and were briefly down.

    Root cause
    Component
    Contributing factor
    Remediation
    • Configured the IP addresses with the Ubuntu Networking service instead of NetworkManager, so a port down no longer removed the address or changed the GID
    Lessons

    Host networking configuration is critical to the reliability of the collective communication library. In RoCEv2 a GID is expected to stay fixed after initialisation, but an IP configuration method that briefly drops the address when an RNIC port goes down can change it and bring down the training job. Changing the IP configuration method avoided the GID shift without the compatibility risks that driver-level patches carry in virtualised networks.

Check your cluster for this

This case involves Host routes and routing rules and InfiniBand fabric. kubegap audits a Kubernetes cluster for gaps like these, read-only, from Claude Code or Codex; its checks cite public reports like the ones in this library.

By root cause1