Cases3 cases

1 to 3 of 3

1. IL-0481 RoCEv2 GIDs changed after RNIC port flaps, collapsing the communication collective and crashing LLM training jobs

   Infrawaves2026

   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.

   Link flappingTraining job failure

   Trigger

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

   Root cause

   Software defect

   Component

   NIC, cable and switch port linkInfiniBand fabricNCCLHost routes and routing rules

   Contributing factor

   Unreviewed defaults

   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.

   View case
2. IL-0344 Linode's US-West data centers in Fremont and Santa Clara lost connectivity as the fibre circuits between them failed

   Linode2023

   Symptoms

   On 12 February 2023, customers in Linode's US-West region could not reach their Linodes. Santa Clara customers lost Internet connectivity after a change made at 8:45 UTC until it was reverted at 10:13 UTC, and later in the day the connection between the Fremont and Santa Clara data centers was cut completely, leaving both with degraded service or a full outage until 16:25 UTC; customers could also have trouble provisioning, powering down or rebooting services.

   Network unreachable

   Trigger

   A single link of the A-side fibre circuit between Fremont and Santa Clara went down on 11 February, the whole B-side circuit followed on 12 February, and during troubleshooting with the hardware vendor a non-intrusive command triggered a rare bug that took down all A-side links.

   Root cause

   Software defect

   Component

   Backbone network (WAN)NIC, cable and switch port link

   Contributing factor

   Overlapping failuresHarmful manual action

   Detection

   The response team began investigating when the A-side link went down on 11 February. The status page opened at 10:22 UTC on 12 February to investigate reports of connectivity issues in the US-West region.

   Remediation

   • Reverted the change that removed internet traffic from the Santa Clara border routers
   • Performed an emergency code upgrade of both circuits, restoring connectivity between the sites
   • Restored the B-side path with the hardware vendor and fiber provider, regaining full redundancy

   View case
3. IL-0263 GitHub was slow and erroring all day and down for 18 minutes as links between access and aggregation switches saturated

   GitHub2012

   Symptoms

   On Friday, 30 November 2012, GitHub had sporadic bursts of slow responses and intermittent errors for the whole day and was completely unavailable for 18 minutes. One fileserver pair was offline for a large part of the day, which affected a small percentage of repositories; this was a side effect of the network problems on the high-availability clustering between the fileserver nodes.

   High latencyHigh error rateService outage

   Trigger

   On the morning of 30 November GitHub moved the A switches in its server cabinets onto the new aggregation switches, the second half of a migration whose B-switch half had been done the day before.

   Root cause

   Software defect

   Component

   Network switchNIC, cable and switch port link

   Contributing factor

   Overlapping failuresUnverified assumptionHarmful manual action

   Detection

   Engineers saw the problems during the maintenance itself: connectivity to the B switch was lost as the last A switch was connected, and afterwards performance stayed low and the links between access and aggregation switches showed as completely saturated.

   Remediation

   • Fixed the bridge-loop problem and returned the affected B switch to service
   • Removed the partial link failure detection setting from the access switches
   • Escalated to the network vendor, which raised the case to its highest severity level
   • Restarted core processes on the aggregation switches in a Saturday 0600 Pacific maintenance window so they learned MAC addresses again

   Action items

   • Deploy the vendor's software update that fixes the MAC learning issue to the switches
   • Closely monitor aggregation-to-access capacity and keep a workaround process ready if the problem returns
   • Build a duplicate of the network stack, from the routers through the access switches, as a staging environment for testing such changes
   • Add automated network monitoring that alerts sooner on similar issues
   • Run scheduled incident response exercises with scenarios that counter tunnel vision
   • Contact the network vendor's support at the first sign of trouble in the network

   Lessons

   GitHub had designed the maintenance to have no impact on customers and clearly failed, so it plans to invest in a staging copy of its network to test such changes and catch bugs like this one. During the response the team fixated for a long time on a bridge loop, which kept it from considering other causes. GitHub had a very positive experience with the vendor's support staff, so it will contact that support team at the first sign of network trouble.

   View case

By root cause1