Cases9 cases

2 to 6 of 9

1. IL-0063 VPN outage after recreating a Redis StatefulSet restarted the whole Redis cluster without readiness checks

   AdGuard VPN2025

   Symptoms

   The day began with an alert that several instances of the Redis cluster were running out of disk space. That part was fixed quickly. The outage came an hour later: right after the Redis StatefulSet was deleted and recreated at 8:10 UTC, Kubernetes restarted every Redis pod within about five minutes, and problems started to appear.

   Service outageOOMKilledProbe failuresHigh latency

   Trigger

   To make Git match the volumes that had been resized by hand, the Redis StatefulSet was deleted with orphan (non-cascading) deletion and recreated with the new storage template; Kubernetes reconciled the new spec by restarting the entire Redis cluster.

   Root cause

   Configuration change

   Component

   Persistent volumes and stateful setsIn-memory cacheContainer memory limit

   Contributing factor

   No health probesUnverified assumptionSingle point of failureRetry stormNo headroom

   Detection

   An alert about Redis instances running out of disk space started the work; the outage itself showed up as backend errors about Redis connections right after the StatefulSet was recreated.

   Remediation

   • made the Redis masters forget obsolete node identities, reset the affected replicas and rejoined them, then waited for replication to catch up
   • identified the component that was actually running out of memory, scaled the backend service and raised memory limits for that component
   • temporarily relaxed probes so backend instances took traffic immediately instead of flapping between ready and not ready
   • reverted the temporary workarounds at 10:00 UTC

   Action items

   • Add startup probes so a Redis pod is not marked ready until its dataset is loaded
   • Build an own control plane for Redis that handles restarted pods stuck to an old master address
   • Reduce dependence on the single Redis cluster: move some important data to other stores, such as a separate Redis cluster per server, and use Kafka for some backend components

   Lessons

   The disk resize was not the problem; the cleanup afterwards was. An immutable field pushes operators toward deleting and recreating the controller, and that can set off a full reconcile. Startup and readiness probes are what slow a rolling restart down to a safe pace; without them a distributed store restarts in the most fragile way. Recovery plans must expect a cluster to heal slowly, and dependent services need to degrade gracefully while it does.

   View case
2. IL-0101 After Google forced a GKE control plane upgrade to 1.33.2 at a deprecation deadline, no new node could initialise, so autoscaling and spot node replacement stopped for 15 hours

   Wealthy2025sev1

   Symptoms

   Nodes that already existed kept running, but every new node, spot or standard, on either v1.32 or v1.33, failed to initialise and could not join the cluster. Spot nodes that were preempted and recreated never came back, so services degraded progressively: background workers on spot instances, reporting services and order allocation stopped completely, and the APIs on existing nodes kept answering but could not scale for load spikes.

   Scale-up failureContainer crash loopService outageDNS resolution failures

   Trigger

   Google upgraded the GKE control plane from v1.32.x to v1.33.2 at the version deprecation deadline, although auto-upgrade was disabled on the node pools.

   Root cause

   Breaking change in an upgrade

   Component

   GKE control planeNode poolClient-side DNS resolver and cachePersistent volumes and stateful setsConfigMaps and Secrets

   Contributing factor

   Deferred upgradesSingle point of failureIneffective rollbackNo headroom

   Remediation

   • Created a new GKE cluster at v1.33, the version of the forced upgrade
   • Upgraded Istio to 1.27.1 for Kubernetes 1.33 compatibility
   • Updated the cert-manager configuration for Kubernetes 1.33
   • Migrated all production workloads to the new cluster and verified pod scheduling, DNS resolution, PVC mounting and external integrations
   • Decommissioned the broken cluster

   Action items

   • Configure all clusters to use the GKE stable release channel
   • Implement a separate management cluster for Argo CD, monitoring and other control components (Q4 2025)
   • Review spot instance usage for critical workloads and add on-demand fallback pools
   • Document the SRE upgrade procedure: management cluster first, production after a monitoring period of several days
   • Document a Kubernetes, Istio and cert-manager compatibility matrix
   • Create an automated upgrade testing pipeline
   • Reserve static IPs for quick migrations
   • Pin versions of Istio, cert-manager, CSI and DNS addons
   • Stay current with stable releases and regularly check GKE alerts and announcements to avoid deprecation deadlines

   Lessons

   GKE upgrades the control plane at a version's deprecation deadline even when auto-upgrade is disabled, so staying inside the supported version window is how to avoid a forced upgrade. A control plane upgrade can leave existing nodes healthy while every new node fails, and spot nodes that cannot rejoin after preemption turn this into cascading failures. Relying heavily on spot instances for critical workers turned a node join failure into a progressive outage. Kubernetes upgrades have to be planned together with dependent components such as Istio and cert-manager, and without a standby cluster on the same provider there was no immediate failover; the only recovery left was building a new cluster.

   View case
3. IL-0020 During an EKS 1.30 upgrade the descheduler tried to reschedule every pod on the cordoned nodes at once and production workloads went down

   UK Ministry of Justice2025

   Symptoms

   At 09:36 on 18 March 2025, during the upgrade of the live cluster to EKS 1.30, users reported that their pods were being terminated and were not restarting automatically, in both production and non-production environments. Automated infrastructure alerts fired as well, and production services had outages.

   Pod evictionPods PendingService outageUpgrade stall

   Trigger

   Upgrading the live EKS cluster from 1.29 to 1.30, during which the descheduler attempted to reschedule all pods on the cordoned nodes simultaneously.

   Root cause

   Faulty deployment

   Component

   DeschedulerNode poolCluster autoscaler

   Contributing factor

   No headroomNo alert on limitUnreviewed defaults

   Detection

   Users reported at 09:36 that pods were being terminated and not restarting, and automated infrastructure alerts fired.

   Remediation

   • Removed the cordon annotation so pods could be rescheduled
   • Deleted stuck 1.30 nodes to force their recreation
   • Added a new node group of 20 nodes running 1.30 by hand
   • Cordoned and drained all old 1.29 nodes onto the new 1.30 nodes

   Action items

   • Raise an AWS Production System Down support case (Case ID: 174229643100496)
   • Get guidance from AWS support on querying control-plane logs to find the root cause of similar issues
   • Document a runbook for emergency scaling of a new node group in case of unexpected capacity shortages
   • Alert on an imbalance between the autoscaling group and the Kubernetes nodes
   • Review the descheduler configuration to understand why it throws pods off cordoned nodes, and fix it where possible
   • Alert on spikes in the pending pod count

   Lessons

   The write-up lists review actions rather than lessons. They show the team had no documented way to add an emergency node group, no alert on a mismatch between the autoscaling group and the nodes Kubernetes sees, no alert on a spike in pending pods, and no clear understanding of how the descheduler treats cordoned nodes.

   View case
4. IL-0084 Packet drops on a bare metal provider's vSwitch forced Namespace to rebuild its fra3 region and left global capacity short, and the next day slow NVMe writes on a control plane node drove the Kubernetes API server to very high CPU

   Namespace2024

   Symptoms

   From 17 October 2024, packets through the vSwitch that joined Namespace's bare metal machines at one provider in the fra3 region were dropped at random. Retries in TCP, QUIC and the Namespace control plane hid most of it, but the SLO error budget for latency targets on key interactions began to burn.

   Intermittent connection failuresHigh latencyControl plane lagService outage

   Trigger

   Random packet drops on the bare metal provider's vSwitch from 17 October 2024; for the second outage, a rate of Kubernetes changes that the slow NVMe on one control plane node could no longer keep up with.

   Root cause

   Cloud provider outage

   Component

   Kubernetes API server

   Contributing factor

   No capacity planningNo headroomOverlapping failures

   Detection

   SLO-based monitoring of latency targets showed the error budget burning while retries hid the individual packet drops.

   Remediation

   • diverted all customer traffic from fra3 to other regions
   • connected each provider machine to the zrh1 cluster over Wireguard instead of the vSwitch
   • repaved every fra3 machine with a new identity, certificates and encryption-at-rest keys, dropping regional caches to recover faster
   • cloned existing datasets in a more distributed way to speed up repaving the machines
   • added machines back to the new region in batches and returned the region to the compute pool
   • redirected workloads away from zrh1 during the second outage
   • cloned the failing control plane node's key state to a higher-capacity node that took over its role and IP address

   Action items

   • Credit all instances that ran during the outages back to customers
   • Build up backup regions and solutions so a similar event has options that do not affect customers

   Lessons

   Running hot with a delayed hardware delivery left no room to take a region out, so fixing a provider network fault turned into a global capacity shortage. After one outage the team was anchored on its cause and first looked for network problems in the second outage, which had a different cause: a control plane node whose disk could not keep up with the Kubernetes change rate.

   View case
5. IL-0024 Expensive cache-backed API calls and uncancelled database requests overloaded MongoDB and took the Hugging Face Hub down three times

   Hugging Face2024

   Symptoms

   At 08:45 CET on 22 April 2024 the Hugging Face Hub became unstable, and by 09:30 it was unresponsive. For most users the website was slow or could not be reached at all.

   Service outageHigh latencyDatabase overload

   Trigger

   Bursts of requests to revision endpoints that read heavily from the *_info_cache collections: a call to https://huggingface.co/api/datasets/lighteval/mmlu/revision/main rose sharply right before the first crash, and a call to https://huggingface.co/api/models/sentence-transformers/all-mpnet-base-v2/revision/main, a model with 300+ associated spaces, lined up with the crash at 21:30 on 23 April.

   Root cause

   Software defect

   Component

   Application databaseModel hubCluster autoscaler

   Contributing factor

   No headroomRetry stormOverlapping failures

   Detection

   Instability started at 08:45 on 22 April and the Hub was unresponsive by 09:30; the team followed MongoDB Atlas metrics and a MongoDB support case. The root cause was found days later from a new graph of requests per URL.

   Remediation

   • Added a firewall rule blocking all requests with a Java/* user agent
   • Disabled the quicksearch feature on the Hub
   • Scaled the MongoDB Atlas cluster by hand from tier 2 to tier 3
   • Switched the Hub to maintenance mode with a 503 page through AWS WAF and blocked incoming traffic
   • Scaled all Hub pods down to 0 and brought them back
   • Set an aggressive rate limit
   • Added timeouts to requests and reworked database queries that were slow or returned large bodies

   Action items

   • Add timeouts to requests based on their usual completion time
   • Optimise requests that took too long or returned a large body, and rework poorly designed database queries
   • Remove old unused indexes from the database
   • Build new dashboards and monitoring metrics
   • Move the *_info_cache collections to Redis
   • Make it possible to abort any database request when the client aborts theirs

   Lessons

   Finding the root cause took two to three days: the existing charts did not show it, and a per-URL request breakdown did. Not cancelling database work when a client goes away turns a strained database into a thundering herd. A close inspection of the database exposed architectural flaws and bottlenecks, such as a collection with large unused indexes, and taught the team which metrics need closer watching to anticipate crashes. Unrelated failures happening at the same time added to the stress and made restoring service harder.

   View case

By root cause7