Cases5 cases

1 to 5 of 5

1. IL-0204 An etcd memory spike overloaded the control plane of a Render Frankfurt cluster, and DNS then failed because CoreDNS ran on the rebooted control plane nodes and could not start without the API server

   Render2024

   Symptoms

   In January 2022 an etcd memory spike overloaded the Kubernetes control plane of one of Render's clusters in Frankfurt. At first only operations that need the control plane failed: new deploys and cron job runs.

   Control plane unresponsiveDeploys failingDNS resolution failuresHigh error rate

   Trigger

   A memory spike in etcd overloaded the cluster's control plane.

   Root cause

   Capacity exhaustion

   Component

   etcdCoreDNSKubernetes API server

   Contributing factor

   Unreviewed defaultsHarmful manual action

   Action items

   • Run CoreDNS on data plane nodes instead of control plane nodes
   • Run etcd on dedicated nodes

   Lessons

   CoreDNS can keep answering from its in-memory cache while the API server is down, but it cannot initialise without it. With CoreDNS on the control plane nodes, a reboot or restart during a control plane outage turned a failure of deploys and cron jobs into a failure of requests to services that were already running. The core problem was that the data plane was not isolated enough from the control plane: even without the reboot, the etcd memory spike alone could have forced CoreDNS to restart.

   View case
2. IL-0458 Toolforge Kubernetes API server failed to start after a restart, logging "etcd cluster is unavailable or misconfigured"

   Wikimedia Foundation2019

   Symptoms

   On 10 September 2019 the Kubernetes API server of Wikimedia's Toolforge cluster did not come back after the maintain-kubeusers process restarted it at 18:30. To monitoring and to users the whole cluster looked down: Toolforge users could not launch tools with the webservice command on the Kubernetes backend or use kubectl, and an Icinga check against the "All k8s nodes healthy" endpoint started alerting at 18:45.

   Control plane unresponsiveService outageTLS certificate errors

   Trigger

   At 18:30 the maintain-kubeusers process restarted the Kubernetes API server, which forced it to reconnect to etcd with a CA certificate that Puppet changes the night before had left mismatched with the etcd servers.

   Root cause

   Configuration change

   Component

   Kubernetes API serveretcdConfiguration management tool

   Contributing factor

   Change with a delayed effectHarmful manual actionNo runbook

   Detection

   Automated monitoring: an Icinga alert on the toolschecker check http://checker.tools.wmflabs.org/k8s/nodes/ready, about "All k8s nodes healthy", fired at 18:45.

   Remediation

   • Rebooted the Kubernetes master and the etcd cluster, which did not help
   • Took a backup of etcd and reduced the etcd cluster to one node to rule out raft election flaps
   • Brought etcd back to full capacity through a full disaster recovery from the backup
   • Swapped the API server's CA certificate for the one from the etcd servers, which restored services until Puppet reverted it
   • Placed the etcd Puppet CA certificate in /var/lib/puppet/client/ssl/certs/ca.pem so Puppet kept the right CA

   Action items

   • Document the etcd cluster processes
   • Upgrade Toolforge Kubernetes to improve etcd and Kubernetes error reporting and stop needing an API server restart to add a tool
   • Make kube2proxy more resilient to API server failures
   • Audit the Puppet CA certificates so the tools project has no more than one
   • Make the Kubernetes control plane highly available

   Lessons

   Components that often misbehave, such as etcd, drew the responders to the wrong fix, and working on etcd instead of the CA certificates made the outage worse. Puppet-managed certificates are risky, and very old Kubernetes versions log errors that are hard to read. The failure was hard to find because it surfaced at an API server restart rather than at the Puppet change, because the etcd cluster had long-standing undocumented quirks, and because no disaster recovery or node replacement procedure was written down for etcd.

   View case
3. IL-0076 Kubernetes masters went down after Dex filled etcd with AuthRequest objects and deleting its namespace started an endless list-and-timeout loop

   Fullstaq2019

   Symptoms

   Dex, the OIDC proxy that handled Kubernetes authentication, stored its data in the Kubernetes API (storage type kubernetes, inCluster: true) and had built up a huge number of AuthRequest objects. Listing all AuthRequests timed out, and without the list they could not be deleted.

   Control plane unresponsiveMonitoring outage

   Trigger

   The dex namespace was deleted so that Kubernetes would clean up the accumulated AuthRequest objects itself.

   Root cause

   Capacity exhaustion

   Component

   Kubernetes API serveretcdCluster authentication provider

   Contributing factor

   Unreviewed defaultsHarmful manual actionRetry stormLimited debug accessIgnored early signal

   Detection

   The speaker says he had got hints of the problem earlier. While the masters were down, plenty of alerts arrived in Slack, but monitoring could not be reached because authentication and port-forwarding no longer worked.

   Remediation

   • removed the kubernetes finalizer from the dex namespace by calling the API directly (kill-kube-ns)
   • deployed Dex to a new namespace with sane defaults

   Lessons

   Dex with Kubernetes as its storage keeps every AuthRequest as an API object, and with a long TTL they pile up until the API cannot list them. Deleting a namespace that holds that many objects makes Kubernetes list them in a loop and can overload the masters. Monitoring that is reached through cluster authentication is lost when authentication is what broke.

   View case
4. IL-0059 Single-master Kubernetes control plane kept dying because the kubelet restarted a healthy etcd on a failing kubeadm liveness probe

   NRE Labs2018

   Symptoms

   The daily NRE Labs statistics showed lesson activity collapsing, and kubectl could not reach the cluster: `The connection to the server 1.2.3.4:6443 was refused - did you specify the right host or port?`

   Control plane unresponsiveProbe failuresService outage

   Trigger

   Unknown. The liveness probe that kubeadm writes into the etcd static pod manifest (an `etcdctl get foo` against 127.0.0.1:2379, failureThreshold 8, timeoutSeconds 15) started failing. The author first suspected GCP maintenance the night before, but could not reproduce the probe failure by hand and does not know why it began.

   Root cause

   unknown

   Component

   etcdkubeletKubernetes API server

   Contributing factor

   Single point of failureUnreviewed defaultsHarmful manual actionIgnored early signalMonitoring blind spot

   Detection

   The author saw lesson activity drop in the daily NRE Labs statistics during his morning routine.

   Remediation

   • removed the liveness probe section from the kubeadm-generated etcd static pod manifest and applied it between etcd restarts

   Action items

   • Add monitoring, especially for the single-node master
   • Run a basic "down for maintenance" page in Google Cloud Functions that points to Twitter for updates, to redirect to during outages

   Lessons

   Because all of the infrastructure is managed as code, the cluster could be rebuilt quickly, but destroying it before finding the cause also destroyed the evidence. Monitoring matters, especially with a single-node master. Read the restart counts in kubectl output, not just the Running status. Remember the fundamentals instead of expecting the issue to be more complicated than it is, and do not let the urgency of an outage push you into hasty decisions. Put up a maintenance page instead of leaving the site failing halfway.

   View case
5. IL-0003 linkerd lost Kubernetes service discovery after an etcd reconfiguration, and restarting every linkerd instance turned failed payments into a full platform outage

   Monzo2017

   Symptoms

   At 14:10 an engineer deployed a change to a service that Monzo's current account needs to process payments, and every request to that service started to fail. Current account customers saw payment failures; the prepaid card was not affected at first because it does not use that service. The change was rolled back at 14:12, but the errors continued.

   High error rateStale endpoint routingService outage

   Trigger

   At 14:10 an engineer deployed a change to a service needed to process current account payments. Two weeks earlier the etcd cluster had been upgraded and grown from three to nine nodes, and one day earlier another team had scaled a new service down to zero replicas while leaving its Kubernetes service in place.

   Root cause

   Software defect

   Component

   etcdKubernetes API serverLinkerdkubelet

   Contributing factor

   Ineffective rollbackHarmful manual actionOverlapping failuresMonitoring blind spotSingle point of failure

   Detection

   Payment failures began right after the 14:10 deploy. The team rolled back at 14:12 and declared an outage internally at 14:16. At 15:13, when no service could serve requests, practically all automated alerts fired.

   Remediation

   • Rolled back the payments service change (the errors continued)
   • Activated internal backup processes so most customers could still use their card
   • Restarted all linkerd instances (this made the outage worse)
   • Restarted the three Kubernetes API server processes so replacement linkerd instances could start
   • Began deploying a newer linkerd version with the fix for empty services
   • Deleted the Kubernetes service that had no endpoints, which let linkerd load service discovery again

   Action items

   • Fix the bug in Kubernetes that can trigger timeouts after a cluster reconfiguration
   • Roll out a new version of linkerd that fixes the parsing error
   • Create better health checks, dashboards and alerts for the affected components so they give clearer signals about what is wrong and prevent human error
   • Improve procedures to communicate outages internally and externally as clearly and quickly as possible

   Lessons

   A large failure in a distributed system can be very hard to understand, and well-meant human action can compound it, as restarting every linkerd instance did here by exposing a second incompatibility. Because Monzo deploys often and its network changes constantly, the deployment and request-routing systems have to be trustworthy. Monzo committed to publishing a similar post-mortem for any incident of this severity.

   View case

By root cause4