Cases96 cases

79 to 83 of 96

1. IL-0014 A GKE upgrade to a withdrawn version stalled on slow drains and a stuck node pool, then broke an ingress and confused the HPA

   loveholidays2019

   Symptoms

   On the evening of a long-overdue upgrade of a loveholidays production GKE cluster from 1.10.X to 1.12.5-gke.10, planned to take 2 to 2.5 hours, the master upgrade and the first two 7-node NodePools went smoothly. After that the NodePool upgrades slowed to a crawl. Each node is cordoned, drained and upgraded in turn. Even a 14-node cluster at about 5 minutes per node took over an hour, and with about 100 more nodes to go, some of them taking over 15 minutes to drain, the sequential drains were too slow to finish at a reasonable hour.

   Upgrade stallPods PendingScale-up failureService outageAutoscaler misbehaviour

   Trigger

   A manual upgrade of the GKE masters and NodePools from 1.10.X to 1.12.5-gke.10, two versions at once, on 19 March 2019; GKE's release notes of 14 March had already said that version was no longer available for new clusters or master upgrades because masters showed elevated error rates, but it could still be selected.

   Root cause

   Cloud provider outage

   Component

   GKE control planeNode poolCluster autoscalerIngress controllerHorizontal pod autoscaler

   Contributing factor

   Unrepresentative stagingOverlapping failuresIgnored early signal

   Detection

   The slow drains, zombie nodes and stuck NodePool were seen by the engineer running the upgrade; the broken Ingress was noticed after the upgrade and explained by Google support; the HPA problem was noticed the next day.

   Remediation

   • drained the remaining NodePool nodes ahead of the upgrade with kubectl drain --ignore-daemonsets --delete-local-data
   • deleted the zombie instances from the managed instance group with gcloud compute instance-groups managed delete-instances
   • cancelled the stuck UPGRADE_NODES operation with gcloud beta container operations cancel
   • used kubectl port-forward in place of the broken internal Ingresses
   • scaled down the older of two active ReplicaSets, or triggered a dummy deployment, to bring the HPA back

   Action items

   • Alert when a HorizontalPodAutoscaler is at 70% and at 100% of its capacity
   • Alert when a deployment has had two active ReplicaSets for 30 minutes and for 60 minutes
   • Re-evaluate termination grace periods that were set to arbitrary large values
   • Read the GKE release notes before every cluster upgrade
   • Use auto-upgrade in non-production environments, and manual cluster upgrades with automatic node upgrades in production
   • Keep environments as identical as possible, including resource names

   Lessons

   Read the release notes before upgrading: the target version had been withdrawn five days earlier. Drain time depends on pod configuration, so emptyDir volumes, long grace periods and missing disruption budgets turn a NodePool upgrade into hours. Pre-production only catches problems like the ingress bug if it matches production down to resource names. Auto-upgrades in non-production surface problems early, and a mix of manual cluster upgrades and automatic node upgrades in production keeps clusters current without much effort.

   View case
2. IL-0469 The managed Azure Kubernetes API server for the data100 hub stopped answering, so the stopped hub pod was not recreated

   UC Berkeley2019

   Symptoms

   On 25 February 2019 the Kubernetes API server of the managed Azure cluster running the data100 hub became unreachable. kubectl failed for every operation and requests to create new resources failed, so when the hub pod stopped, no new one was created and users saw a proxy error message.

   Control plane unresponsiveService outage

   Root cause

   Cloud provider outage

   Component

   AKS control planeKubernetes API serverkubelet

   Contributing factor

   Limited debug access

   Detection

   Infrastructure staff were notified in Slack at 11:34 and found that kubectl failed for all operations while the hub proxy was up.

   Remediation

   • Created a new cluster and requested a new NFS server in its resource group
   • Detached the data volumes from the old NFS server and moved them to the new cluster's resource group
   • Imported the ZFS pool into the new NFS server
   • Created the staging hub with helm and updated DNS
   • Created the prod hub with helm and updated DNS

   Action items

   • Monitor the API server endpoint remotely and alert when it is down

   View case
3. IL-0054 All ingress traffic lost while draining a GKE node pool because the load balancer only targeted new nodes without ingress-nginx pods

   DevOps Hof2019

   Symptoms

   While the old nodes of a GKE cluster were being drained to move workloads to a newly created node pool, none of the services in the cluster could be reached through ingress-nginx any more.

   Service outageProbe failures

   Trigger

   Cordoning and draining the old node pool removed those nodes from the GCP load balancer backend while the ingress-nginx pods were still running on them.

   Root cause

   Operator error

   Component

   Ingress controllerkube-proxyNode pool

   Contributing factor

   Unreviewed defaultsUnverified assumption

   Detection

   The team noticed during the drain that no service in the cluster was reachable through ingress-nginx.

   Remediation

   • deleted the nginx-ingress-controller pods still running on the old nodes so they were rescheduled on the new, ready nodes

   Lessons

   With externalTrafficPolicy: Local, the load balancer can only reach nodes that run an ingress pod, and cordoned nodes leave the load balancer backend immediately; if the ingress pods stay behind on cordoned nodes, all ingress traffic stops. Before any change that might touch traffic, look twice into the documentation.

   View case
4. IL-0064 CPU-heavy pods packed onto one Google Cloud node drove it to kernel panics, and the crash moved from node to node until the site was down

   Moonlight2019

   Symptoms

   On Friday, January 18, Moonlight's API began returning occasional errors that looked like connection problems with its Redis database. The API checks Redis on every authenticated request to validate the session. At the same time the Kubernetes monitoring reported some nodes and pods as unresponsive.

   High error rateIntermittent connection failuresNode NotReadyService outageSustained high node CPU

   Trigger

   The Kubernetes scheduler put several CPU-heavy pods on the same node, including all three replicas of the website and the server-side rendering pods.

   Root cause

   Capacity exhaustion

   Component

   Pod scheduling and placementLinux kernelDatabase

   Contributing factor

   Single point of failureUnverified assumptionAutomation amplification

   Detection

   API errors that looked like Redis connectivity problems, and Kubernetes monitoring that showed nodes and pods as unresponsive. Google Cloud Support later found the link between sustained 100% CPU and kernel panics.

   Remediation

   • added anti-affinity rules to all major deployments so their pods spread across nodes

   Action items

   • Keep working with Google Cloud support to find and fix the cause of the node kernel panics

   Lessons

   Three replicas give no fault tolerance if the scheduler places them all on one node; anti-affinity rules make the spread explicit. A node failure that comes from the workload itself moves with the workload when the scheduler reschedules it.

   View case
5. IL-0472 Node-to-pod traffic and DNS failed on Tinder's 605-node Flannel cluster when every node's ARP cache passed gc_thresh3

   Tinder2019

   Symptoms

   In the early morning of 8 January 2019 Tinder's platform had a persistent outage. Communication from nodes to pods failed, and so did DNS lookups, because DNS ran inside the Kubernetes cluster.

   Service outageNetwork unreachableDNS resolution failures

   Trigger

   Earlier that morning, pod and node counts on the cluster were scaled up in response to an unrelated increase in platform latency.

   Root cause

   Capacity exhaustion

   Component

   Linux kernelCNI plugin

   Contributing factor

   Unreviewed defaults

   Remediation

   • Raised gc_thresh1, gc_thresh2 and gc_thresh3 on the nodes
   • Restarted Flannel so it re-registered the missing networks

   Lessons

   Tinder lists the outage among its learnings about network fabric limits: with Flannel over VXLAN, every node, and every pair of nodes that exchange pod traffic, adds entries to each node's ARP table, so at 605 nodes the default gc_thresh3 hard cap was exceeded on all nodes.

   View case

By root cause11