Cases3 cases

1 to 3 of 3

1. IL-0237 A Kubernetes patch upgrade evicted every Ray Serve head and worker pod at once, and Lucidworks hosted LLMs went down

   Lucidworks2026sev1

   Symptoms

   From 16:18 UTC on 28 May 2026 the Lucidworks AI hosted LLMs in the us-southcarolina region were unavailable: queries to llama-3-8b-instruct, llama-3v2-3b-instruct and phi-4-multimodal-instruct returned 500 or 429 errors. Search, embedding models, the Lucidworks Platform UI and passthrough LLM calls kept working.

   Service outagePod evictionImage pull failures

   Trigger

   A routine Kubernetes patch upgrade on a cluster in us-southcarolina cycled its node pools one after another in quick succession.

   Root cause

   Faulty deployment

   Component

   Node poolModel inference server

   Contributing factor

   No staged rolloutOverlapping failures

   Remediation

   • Deleted the degraded replacement head node that could not pull container images
   • Waited for the LLM image pulls to finish on the replacement nodes
   • Launched new nodes with the required LLM images and redirected Lucidworks AI traffic to them

   Action items

   • Upgrade clusters that host LLM workloads in sequence, with each node pool fully healthy before the next one is upgraded
   • Investigate image pre-loading, such as pre-baked disks or image streaming, for large ML model images
   • Open a support ticket with the cloud provider about the abnormal Docker pull times
   • Coordinate Kubernetes maintenance windows with LLM service owners so they do not clash with ongoing deployments
   • Build tooling to force a Ray cluster failover to the blue or green state instead of waiting for Ray to settle it

   View case
2. IL-0236 After an Istio 1.15 to 1.24 upgrade, half of Wikimedia's ML model servers returned 5XX for about a day

   Wikimedia Foundation2026

   Symptoms

   From 17:30 UTC on 23 February 2026, right after the ml-serve-codfw Kubernetes cluster was upgraded, 5XX responses appeared in the metrics of a subset of Wikimedia's machine learning model servers and alerts started to fire. About half of the running model servers were affected and stayed unreachable for roughly 24 hours.

   High error rate

   Trigger

   The two production Machine Learning Kubernetes clusters were upgraded to version 1.31, which included upgrading Istio from 1.15 to 1.24; ml-serve-codfw was done at 17:21 UTC on 23 February and ml-serve-eqiad between 08:00 and 11:00 UTC on 24 February.

   Root cause

   Breaking change in an upgrade

   Component

   IstioModel inference server

   Contributing factor

   Ignored early signal

   Detection

   Users of the affected model servers first reported the issue in the talk-to-machine-learning Slack channel. Alerts for HTTP 5XX had fired in the Machine Learning IRC channel right after the codfw upgrade, but nobody on the team noticed them.

   Remediation

   • Deployed an Istio hot patch that restored the broken functionality and fixed most of the HTTP 500 errors
   • Deployed the permanent fix to all affected clusters after verifying it on ml-staging-codfw

   Action items

   • Deliver alerts to the relevant Slack channels and watch IRC or alerts.wikimedia.org
   • Add more SLOs and fix the existing ones, whose Istio metrics changed

   Lessons

   The problem could have been caught on the staging cluster first, or by ad-hoc testing within the team after each cluster upgrade, which had been requested when the upgrades were announced. Not everyone involved knew that httpbb tests existed, and nobody on the Machine Learning team saw the alerts that fired right after the codfw upgrade. Recovery was fairly quick once users reported the problem, helped by having the right person online to find the Istio hot patch.

   View case
3. IL-0413 LLM inference engines timed out or ran out of resources under load in about 60% of 156 severe incidents at Microsoft

   Microsoft2025

   Symptoms

   In Microsoft's cloud LLM inference service, the inference engines themselves failed in 94 of the 156 high-severity incidents from April to June 2025 (≈60%). Requests timed out or ran into resource exhaustion under load, and callers got HTTP 408, 429 and 5xx errors.

   Intermittent timeoutsHigh error rateHigh latency

   Trigger

   Bursty load, traffic with unusually high sampling parameters and long-running requests pushed the inference engines into resource exhaustion and timeouts.

   Root cause

   Capacity exhaustion

   Component

   Model inference server

   Contributing factor

   Monitoring blind spot

   Detection

   Across the 156 incidents, 115 (≈74%) were auto-detected, meaning health probes or DRI pipelines alerted before customers reported the problem; the others were not.

   Remediation

   • Enabled dynamic pushback at the load balancer with caps on pending tokens and requests
   • Tuned request batching and GPU/CPU replica counts
   • Improved gRPC backoff and error mapping
   • Limited sampling parameters and throttled per client
   • Enabled streaming, tuned request batching and capped pending tokens against engine connection timeouts on long-running requests
   • Increased capacity, routed traffic to healthy endpoints and restarted nodes, the actions applied mostly to inference engine failures

   Action items

   • Optimise serving for long-running requests and high sampling regimes, with connection liveness and GPU capacity-aware routing as the main levers
   • Expand health-probe-driven failover and GPU capacity-aware routing so traffic moves off degraded endpoints on its own
   • Forecast token and throughput growth and automate capacity increase decisions
   • Deploy adaptive alerting and expand engine-level telemetry to close detection gaps

   Lessons

   Hotfixes restore correctness but lengthen time to mitigate, while traffic routing and node rebalancing mitigate faster under overload and timeout patterns. Even with AIOps tooling a meaningful share of incidents still needs manual detection or judgement, so health-probe-driven traffic routing, restarts and capacity increases are the decisions to automate first.

   View case

By root cause3