Cases14 cases

6 to 10 of 14

1. IL-0114 An unpinned AWS VPC CNI plugin was updated automatically during a capacity expansion, leaving new pods Pending without IP addresses and stopping CoW Protocol settlements for about 3.5 hours

   CoW Protocol2024

   Symptoms

   At 16:40 UTC on 19 April 2024 CoW Protocol's mainnet production environment stopped producing settlements; the outage officially began then.

   Pods PendingIP address exhaustionPod evictionService outage

   Trigger

   During a cluster capacity expansion, the Infrastructure as Code automation updated the AWS VPC CNI plugin, because the CNI version was not hardcoded in the IaC and the IaC library version had recently changed.

   Root cause

   Configuration change

   Component

   AWS VPC CNINode pool

   Contributing factor

   Unpinned versionOverlapping failures

   Detection

   The CNI version mismatch was identified as the cause at 17:03 UTC, 23 minutes after settlements stopped.

   Remediation

   • upgraded the Kubernetes cluster
   • set up new node groups with a new version of the networking plugin

   Lessons

   Versions in IaC scripts should be specified explicitly so an automation run cannot update a component such as the CNI plugin unintentionally. Changes in production need change management that is easy to revert.

   View case
2. IL-0135 A power source switch during UPS maintenance cut power to a row of GPU servers and shut down Crusoe Cloud VMs in us-northcentral1

   Crusoe Cloud2023

   Symptoms

   At 20:31 UTC on 15 November 2023, some GPU virtual machines in us-northcentral1-a became unavailable. One row of servers in a Crusoe Cloud data center in the us-northcentral1 region had lost power, and every server in the row powered off together with the VMs running on it.

   GPU unavailableService outage

   Trigger

   While the row's UPS was in bypass mode for routine maintenance, power switched from the primary source to a backup source for reasons unrelated to the maintenance.

   Root cause

   Dependency failure

   Component

   Data center power and cooling

   Contributing factor

   Single point of failureOverlapping failures

   Detection

   The power failure happened at 20:31 UTC. The first status post, at 21:07 UTC, said certain VMs in us-northcentral1-a were unavailable, and at 21:31 UTC the status page said the affected VMs had been identified.

   Remediation

   • identified the VMs on the row that lost power
   • restarted the affected VMs after power was restored

   Action items

   • Review the standard operating procedures for UPS maintenance to minimise customer impact during future maintenance
   • Build future data centers with redundant UPSs
   • Improve recovery procedures to reduce mean time to resolution

   Lessons

   When the only UPS for a row is in bypass for maintenance, the row is exposed to any power event during that window, even one unrelated to the work. Crusoe Cloud's response is redundant UPSs in future data centers, a review of maintenance procedures, and faster recovery of VMs after power returns.

   View case
3. IL-0112 The daily node rollover put DFDS's EKS nodes on an AMI later recalled for a kernel issue, causing "PLEG is not healthy" NodeNotReady flaps and pod network timeouts

   DFDS2022

   Symptoms

   On 2022-10-31 a planned upgrade of the vpc-cni EKS add-on, started from the deployment pipeline at 13:45 UTC, timed out at 14:10. Only about 10 of the 45 aws-node pods had upgraded, 5 were stuck in Terminating and the other 30 had not been attempted.

   Node NotReadyUpgrade stallIntermittent timeoutsIntermittent connection failures

   Trigger

   At 07:30 UTC on 2022-10-31 the daily node rollover replaced the existing nodes with nodes built from AMI v20221027, the latest AWS EKS node AMI, released on 2022-10-29 and later recalled because of a kernel issue.

   Root cause

   Software defect

   Component

   Node operating system imageLinux kernelkubeletAWS VPC CNINode launch templateTerraform

   Contributing factor

   No staged rolloutAutomation amplificationOverlapping failuresMonitoring blind spotNo health probes

   Detection

   The team noticed nodes flapping between NodeNotReady and NodeReady ("KubeletNotReady PLEG is not healthy") at 14:30 UTC while watching the stalled vpc-cni upgrade, about seven hours after the rollover. Time to detection was 7 hours. Alerting did not catch the NodeNotReady events or the socket growth.

   Remediation

   • cordoned, drained and deleted 15 stuck nodes one at a time to complete the vpc-cni upgrade
   • changed the Terraform infrastructure-modules so a specific AMI could be set
   • rolled the EC2 Auto Scaling group launch configuration back to the previous known good AMI through the pipeline

   Action items

   • Reach out to teams with suggestions about readiness and liveness probes
   • Create a dashboard for readiness and liveness probe coverage
   • Write a wiki article on readiness and liveness probes and their usefulness
   • Modify infrastructure-modules to allow specifying a particular AMI
   • Schedule a discussion on whether to take the latest AMI immediately and how to test new AMIs before they are deployed
   • Implement alerting on abnormal NodeNotReady events and on increased socket use
   • Create a learning path for Kubernetes within the team
   • Collect logs from nodes centrally
   • Create a Prometheus exporter that retrieves the latest available AMI, with a dashboard and alerts

   Lessons

   When a new node AMI has just been rolled out, upgrading other cluster components on the same day makes it hard to tell which change is causing trouble; here the vpc-cni upgrade on the same day confused matters. Taking the latest AMI automatically with no way to pin one meant a rollback first needed a Terraform change. The team also concluded that it needs better communication paths with application teams during incidents.

   View case
4. IL-0115 One of OpenAI's three production AKS clusters went offline through a human error, and rebuilding it took 10½ hours because an AKS Kubernetes upgrade broke the ingress and AKS cannot be downgraded

   OpenAI2022

   Symptoms

   At 18:56 PDT on 23 June 2022 (01:56 UTC on 24 June) several models went offline at once: some fine-tuned curie and davinci models, the Codex beta models code-davinci-001 and code-cushman-001, the legacy GPT models curie, babbage and ada, and all Embeddings models.

   Service outageHigh error rateIntermittent timeouts

   Trigger

   A destructive operation on one of the three production clusters, caused by human error that the existing safeguards did not catch, took the whole cluster offline.

   Root cause

   Operator error

   Component

   Ingress controllerModel hub

   Contributing factor

   No safeguard on destructive changeNo headroomDeferred upgradesOverlapping failures

   Remediation

   • re-allocated the models from the lost cluster to the remaining two clusters
   • rebuilt the VPN link with Azure support after losing the connection to AKS during cluster setup
   • built workarounds with Azure support after losing the connection to Azure Private Link
   • tore down the cluster and rebuilt it from scratch after the upgraded Kubernetes version broke the ingress
   • moved traffic back into the rebuilt cluster between 05:30 and 06:30 PDT

   Action items

   • Add checks and locks so that this and similar production-critical infrastructure cannot be modified by accident
   • Add better visibility into the operations being performed and make parts of the system more lexicographically distinct
   • Keep working with Azure on the AKS network connectivity issues so they do not disrupt new cluster operations
   • Update playbooks with the new cluster failure modes and change the setup steps for new clusters and infrastructure
   • Upgrade Kubernetes systematically and set aside time so the clusters do not fall behind again
   • Shard fine-tuned models across multiple clusters and work towards more automatic cross-cluster replication
   • Distribute and replicate Azure Blob Storage so fine-tuned models load faster and more reliably
   • Finish rolling out node-to-node caching to relieve Azure Blob Storage bottlenecks
   • Change how model weights are stored so they are more efficient and faster to load

   Lessons

   Losing one of three clusters exhausted the GPU headroom, and GPUs could not be added on short notice, so full recovery depended on bringing the same cluster back. Rebuilding it exposed a deferred Kubernetes upgrade: the newer version broke the ingress, and because AKS cannot downgrade there was no way back except a full rebuild. Recovering many large models at once is limited by the bandwidth of the storage they load from, and the caches that normally hide this do not help when everything moves to a different cluster.

   View case
5. IL-0205 The GKE cluster autoscaler could not add nodes to GitLab.com's production cluster because Google Cloud ran out of SSD persistent disks in us-east1

   GitLab2021

   Symptoms

   At 08:05 UTC on 2021-06-21 the API service's Apdex (the share of requests that finish within a target time) and Rails request queuing began trending the wrong way. At 09:01 the team found the API deployment unhealthy: it needed more pods for the day's growing traffic, but no new nodes could be added to give it pod capacity. Users saw slow responses; the review found no sign that error rates rose.

   Scale-up failureHigh latencyPods PendingDeploys failing

   Trigger

   Google Cloud ran out of SSD persistent disk stock in us-east1, the region GitLab.com runs in, while the API service needed more nodes for rising user traffic.

   Root cause

   Cloud provider outage

   Component

   Cluster autoscalerNode pool

   Contributing factor

   Monitoring blind spotLimited debug accessOverlapping failures

   Detection

   There was no alert on resource availability. Apdex and Rails queuing began trending the wrong way at 08:05 UTC and an engineer declared the incident in Slack at 08:43; finding the cause took a lot of log digging: cluster autoscaler errors about node pools in backoff, then ZONE_RESOURCE_POOL_EXHAUSTED at the instance-group level, and finally Google Cloud support naming SSD as the exhausted resource.

   Remediation

   • added a node pool that uses pd-standard disks
   • scaled the new node pool by hand across all three zones to relieve the API
   • brought more pd-standard node pools online for the other blocked workloads
   • stopped production deployments and opened a Google Cloud support case about the cluster autoscaler not working with the new node pools
   • changed node pools and nodeSelectors so the remaining workloads ran on pd-standard nodes
   • removed the SSD-backed node pools after a day without problems

   Action items

   • Move Kubernetes services to pd-standard disks (delivery#1837)
   • Capture resource exhaustion errors as a metric

   Lessons

   A cloud region can run out of one specific resource, here SSD persistent disks, and the only error may be a generic ZONE_RESOURCE_POOL_EXHAUSTED that does not name it. GitLab could neither monitor nor reserve resource availability, so detection took 22 hours and the cause came from Google Cloud support. The workloads ran fine on pd-standard disks. A second problem found on the way, a cluster autoscaler that does not fall back to another matching node pool when one cannot scale, slowed the move to the new node pools.

   View case

By root cause7