Cases4 cases

1 to 4 of 4

1. IL-0543 Twingate API traffic piled onto one backend behind the global load balancer, failing core services for 29 minutes

   Twingate2025

   Symptoms

   At 20:45 UTC on 28 May 2025, most of Twingate's API traffic started going to a single backend service, which was overwhelmed and returned elevated error rates, so API calls failed widely across core services.

   Misrouted requestsLoad imbalanceHigh error rate

   Trigger

   A rollout that updated the routing rules of Twingate's global load balancer.

   Root cause

   Configuration change

   Component

   Cloud load balancer

   Contributing factor

   Unrepresentative stagingCircular dependency

   Detection

   The incident was detected at 20:48 UTC, three minutes after API traffic was first misrouted, and triage began.

   Remediation

   • Started a rollback of the routing change at 20:52 UTC; the automated rollback stalled on infrastructure dependencies
   • Applied a manual fix with local tooling at 21:09 UTC, which restored service by 21:14 UTC

   Action items

   • Correct the routing rule
   • Add alerts that detect backend traffic anomalies in lower environments
   • Improve resource scaling thresholds
   • Decouple infrastructure tooling from service dependencies so recovery has faster options
   • Introduce backend sharding to isolate customer workloads and reduce the impact of incidents

   View case
2. IL-0285 Clever Cloud's Pulsar brokers stalled in a minor upgrade, then hypervisors overloaded and crashed with kernel panics

   Clever Cloud2024

   Symptoms

   On 2 August 2024, two minutes after Clever Cloud's data team started a minor rolling upgrade of its Apache Pulsar messaging cluster to release 3.3.1, monitoring showed connection problems with the messaging platform. Producing data into the event bus failed, the backlog grew while no new producers or consumers succeeded, and agents on hypervisors, log forwarders on load balancers and deployments were affected. At 10:20 CEST the Pulsar cluster was declared defective.

   Deploys failingSustained high node CPUMemory pressureNode kernel panicOOMKilledAutoscaler misbehaviourMonitoring outage

   Trigger

   At 10:06 CEST the data team began a minor rolling upgrade of Apache Pulsar to release 3.3.1 that also switched on a new balancer for bundle placement: the extensible load manager (loadManagerClassName=org.apache.pulsar.broker.loadbalance.extensions.ExtensibleLoadManagerImpl) with the TransferShedder load shedding strategy instead of the native OverloadShedder.

   Root cause

   Configuration change

   Component

   Message brokerTelemetry agentLinux kernelAutomated capacity manager

   Contributing factor

   No configuration testsMonitoring blind spotIneffective rollbackUnreviewed defaultsAutomation amplificationRetry stormCircular dependency

   Detection

   Monitoring showed connection issues to the messaging platform at 10:08 CEST, two minutes after the upgrade began, and engineers confirmed at 10:09 that producing data into the event bus failed. Monitoring did not catch the first broker's failure to create the system topic, because that broker stayed up.

   Remediation

   • Rolled back the new unload strategy while keeping Pulsar 3.3.1
   • Shut down the client proxies to release the brokers
   • Switched deployment notification events to RabbitMQ as a workaround
   • Stopped all brokers and started only one, which loaded the conflicting system topic so it could be deleted, then started all brokers
   • Restarted the messaging proxies, after which messages flowed again
   • Stopped the watchdog so it would not reload every failed component, and started components one by one
   • Disabled Metrics API ingestion to protect its storage cluster
   • Stopped the long-term storage for observability metrics (the Warp10/FoundationDB cluster) to free resources
   • Put all monitoring instances in all regions into maintenance so they stopped sending events to the orchestrator
   • Killed the overloaded small VMs one by one
   • Flagged 12 hypervisors to drain workload and shut down some load balancers to free CPU and memory
   • Cleaned zombie VMs to speed up recovery
   • Cleared the monitoring alert queue so queued events would not flood the orchestrator
   • Added parked hypervisors to EU-FR-1 to relieve the overloaded ones
   • Removed an overloaded RabbitMQ node from the cluster used by the orchestrator
   • Rolling-restarted every regional L7 load balancer and every L4 load balancer
   • Restarted all HV agents and the add-on provider APIs
   • Destroyed all dangling VMs stuck in the stopping state
   • Re-enabled Metrics API ingestion with a ramp-up QoS policy

   Action items

   • Track KPIs for Pulsar broker and cluster health, hypervisor load, I/O and kernel panic reboots, zombie VMs, RabbitMQ broker and cluster health and deployment queue consumption, each by time of event, detection, decision and resolution
   • Deal with too many monitoring alerts
   • Look at a backup bastion, because a regional bastion is vulnerable
   • Rework the control plane recovery process
   • Rework the correlation between monitoring and orchestration
   • Revisit fault domain separation against multi-tenancy
   • Rethink runtime images and the role of the hypervisor
   • Isolate waved release upgrades of Pulsar
   • Redesign runtime images with better hypervisor integration
   • Regionalise observability for redundancy
   • Set Vector's buffer.max_events and buffer.max_size for small VMs
   • Switch Vector's buffer.type from memory, the default, to disk
   • Change Vector's buffer.when_full from block, the default, to drop_newest
   • Tune journald verbosity and SystemMaxUse on small VMs
   • Add cgroups restrictions for better process control
   • Add I/O restrictions for the VM manager
   • Detect zombie VMs better

   Lessons

   The simulation process that validates non-minor releases was not applied to minor upgrades or to small configuration changes, and this one went straight to production. Running Clever Cloud on Clever Cloud turned into a chicken-and-egg situation: recovery needed the scheduler to carry out large actions while the control plane was itself slowed and the infrastructure had no free resources. Measuring each key indicator by time of event, detection, decision and resolution showed where the team had lacked visibility or reaction and challenged established rules.

   View case
3. IL-0254 Square's internal DNS servers failed and Square products, services and payments were unavailable for about 15 hours

   Square2023

   Symptoms

   From 1:54 PM ET on 7 September 2023, Square products and services were unavailable, and at 2:02 PM ET engineers were notified of infrastructure issues and began investigating DNS. Square Online websites stayed up, but their customers could not make payments.

   DNS resolution failuresService outage

   Trigger

   At 11:04 AM ET a host-based firewall rule change was deployed to enable region communication, which increased the size of the firewall rules on nodes. Combined with a DNS service upgrade, it put unexpected load on the internal DNS servers. The timeline also lists a DNS zone change at 1:56 PM ET.

   Root cause

   Configuration change

   Component

   Authoritative DNS servers and recordsNode iptables rules

   Contributing factor

   No staged rolloutCache-masked failureCircular dependency

   Detection

   Engineers were notified of infrastructure issues at 2:02 PM ET, eight minutes after services became unavailable.

   Remediation

   • Shed network traffic to the DNS servers and started bringing up new DNS servers by hand
   • Deployed networking changes by hand to re-enable the authorization and access services
   • Deployed the networking changes by hand to all data centers
   • Rebuilt the DNS servers
   • Built a new configuration from the reverted firewall ruleset and pushed it to the DNS hosts

   Action items

   • Move the DNS infrastructure onto isolated infrastructure
   • Add monitoring and optimizations for critical networking infrastructure
   • Optimize dependencies between the deploy and platform infrastructure where feasible
   • Expand support for Offline Mode and improve communication about when it is available

   View case
4. IL-0272 New Cassandra nodes served reads for data they did not hold, so Monzo records looked missing and payments failed

   Monzo2019

   Symptoms

   On 29 July 2019 at 13:14 BST, four minutes after six new servers were added to Monzo's Cassandra cluster at 13:10, automated alerts showed a small percentage of Mastercard card transactions failing, and the tool Customer Operations uses to talk to customers through in-app chat stopped working. Customers could not log in, send or receive payments or withdraw cash, and saw inaccurate balances and transactions.

   Inconsistent reads from a data storeHigh error rate

   Trigger

   At 13:10 BST Monzo added six new servers to its 21-server production Cassandra cluster to add capacity, with auto_bootstrap set to false.

   Root cause

   Configuration change

   Component

   Database

   Contributing factor

   Unverified assumptionUnrepresentative stagingMonitoring blind spotExcessive blast radiusCircular dependency

   Detection

   At 13:14 automated alerts reported a small percentage of Mastercard card transactions failing, and Customer Operations reported that its customer communication tool was not working; an incident was declared at 13:15.

   Remediation

   • Deployed a fix for an error-handling bug in a Mastercard service through a planned fallback mechanism, because the deployment tooling was failing
   • Decommissioned the six new Cassandra servers one by one so data ownership returned to the original servers
   • Reconciled the data written while the new servers were serving reads and writes, by replaying externally stored events and running consistency checks

   Action items

   • Fix the use of auto_bootstrap
   • Review and document the decisions behind the other Cassandra settings and write more extensive runbooks for operations such as scale-ups and restarts
   • Expose more Cassandra metrics and consider alerting on strong shifts in metrics such as 'row not found'
   • Split the single Cassandra cluster into several smaller clusters

   Lessons

   Monzo had gaps in its knowledge of rarer Cassandra operations such as adding servers. The scale-up test with one server could not show the problem, because quorum still came from the original replicas, so it did not match adding six servers in production. The lack of metrics pointing at Cassandra delayed the response, and a single shared cluster both let one change have a far-reaching impact and made the failing part harder to find.

   View case

By root cause1