Cases7 cases

1 to 5 of 7

1. IL-0471 GitHub Actions runs failed or sat queued for 9 hours after a routine deployment's pod replacement saturated capacity

   GitHub2026

   Symptoms

   From 15:05 UTC on 6 August 2026 GitHub Actions workflow runs failed to start, failed partway through or stayed queued for a long time, on both GitHub-hosted and self-hosted runners. Some requests to the Actions REST API returned errors, and some customers saw unexpected rate limiting in their workflows.

   High error rateHigh latency

   Trigger

   A routine deployment to an internal Actions service that processes events and generates Actions jobs replaced pods, and the remaining capacity became saturated.

   Root cause

   Capacity exhaustion

   Component

   CI system

   Contributing factor

   No headroomExcessive blast radiusHarmful manual action

   Remediation

   • Expanded capacity of the affected services
   • Throttled incoming webhook-triggered work
   • Increased processing capacity for the backlog of affected events
   • Deployed changes that stopped runners from repeatedly trying to acquire invalid jobs
   • Deployed a fix for self-hosted runners that were not picking up jobs and enabled it incrementally
   • Paused migrations with GitHub Enterprise Importer
   • Restored webhook-triggered Actions workflows to full throughput
   • Rolled back the mitigation that left some ARC runners offline, and recovered those runners by hand
   • Shared CLI and UI steps for customers to clear jobs that could not be retried or canceled

   Action items

   • Improve deployment and capacity safeguards for the affected services
   • Strengthen monitoring for the conditions that preceded the incident
   • Improve the resiliency and recovery of queued work and runner assignment
   • Add automatic recovery for self-hosted runners in upcoming Runner and ARC releases
   • Reduce the risk of cascading failures and speed up recovery during large Actions disruptions

   View case
2. IL-0409 AssemblyAI's async and LLM Gateway APIs returned 5xx after its auth service could not get ENIs and scaled to zero

   AssemblyAI2026

   Symptoms

   On 29 January 2026 AssemblyAI's Async Transcription API and LLM Gateway on the NA API endpoints went down for 25 minutes, returning 500 errors, and then ran degraded for two hours. The status page first reported the investigation at 17:54 UTC.

   Service outageHigh error rateHigh latencyCloud API rate limit exhausted

   Trigger

   An influx of batch transcription jobs higher than usual made autoscaling add workers until spot instances ran out, on-demand instances were used, and the AWS Elastic Network Interface limit was reached.

   Root cause

   Capacity exhaustion

   Component

   Amazon ECS

   Contributing factor

   Excessive blast radiusMonitoring blind spotAutomation amplificationIgnored early signal

   Remediation

   • Worked through the backlog of requests until turnaround times were back to normal

   Action items

   • Add alerts on ENI utilisation
   • Run the authentication service on dedicated capacity, separate from batch workers
   • Fix the Lambda function that rate-limited the AWS API during recovery
   • Run everything in multiple regions

   Lessons

   AssemblyAI calls the outage preventable: ENI utilisation was not monitored, and because the authentication service shared resources with batch workers, a traffic spike in one starved the other. The warning signs were there but were not connected, and a tool built for recovery ended up blocking recovery.

   View case
3. IL-0341 Traffic between Cloudflare and AWS us-east-1 had high latency, packet loss and origin failures for almost four hours

   Cloudflare2025

   Symptoms

   From 16:27 UTC on 21 August 2025, users connecting to Cloudflare from servers in AWS us-east-1, or receiving connections from Cloudflare there, saw high latency, packet loss and failures to reach origins. Cloudflare's network team was alerted to internal congestion in Ashburn (IAD) at 16:44 UTC.

   Packet lossHigh latencyIntermittent timeouts

   Trigger

   At about 16:27 UTC a single customer began sending many requests from AWS us-east-1 to Cloudflare for objects in Cloudflare's cache, doubling the total traffic from Cloudflare to AWS.

   Root cause

   Capacity exhaustion

   Component

   Transit gateway or network interconnectBGP routingCDN and edge network

   Contributing factor

   No headroomHarmful manual actionExcessive blast radius

   Detection

   Cloudflare's network team was alerted to internal congestion in Ashburn (IAD) at 16:44 UTC, and an incident was raised for customer impact at 17:45 UTC.

   Remediation

   • Rate limited the single customer whose traffic caused the surge
   • Took further traffic engineering actions that fully resolved the congestion
   • Asked AWS to revert its BGP prefix withdrawals, which AWS finished at 20:07 UTC

   Action items

   • Build a mechanism to deprioritise a customer's traffic when it congests the network enough to affect others
   • Speed up the Data Center Interconnect upgrades to add network capacity
   • Work with AWS so the two companies' BGP traffic engineering actions do not conflict
   • Build a traffic management system that gives each customer a network resource budget and automates the manual actions used here

   Lessons

   Cloudflare concludes that it must isolate customers better so that no single customer can take over shared network resources and degrade the platform for others, and that its network needs enough capacity to meet demand. It describes the outage as the result of insufficient network congestion management during an unusual high-traffic event.

   View case
4. IL-0312 Gadget's main Postgres did not start again after a restart to raise max_connections, and the platform went down

   Gadget2024

   Symptoms

   At 11:25 AM EST on 13 December 2024, Gadget restarted its main PostgreSQL server in place to raise max_connections. Postgres shut down but did not start again, and every request to Gadget applications failed, including frontend requests, GraphQL API requests, webhook processing and syncs.

   Service outageHigh latency

   Trigger

   An emergency in-place restart of the main Postgres server at 11:25 AM EST, decided at 11:00 AM, to raise max_connections, which Postgres cannot change without a full restart.

   Root cause

   Capacity exhaustion

   Component

   DatabaseLinux kernelDatabase proxy and connection pooler

   Contributing factor

   Unverified assumptionUnrepresentative stagingAutomation amplificationExcessive blast radius

   Detection

   The team carrying out the planned restart saw at 11:26 AM EST that Postgres had shut down and did not start again.

   Remediation

   • Restarted the database server with a hardware reset, which cleared the fragmented memory so Postgres could start
   • Provisioned almost 4x the compute after the autoscalers had scaled it down during the outage

   Action items

   • Use only the replica-failover approach for disruptive database operations such as restarts
   • Run a more aggressive capacity planning process before Black Friday / Cyber Monday, preparing for a 6x increase in load and proving it with load tests well in advance
   • Announce planned downtime ahead of time on the status page and community Discord, and make the operating procedures tell planned maintenance apart from service disruptions
   • Finish the database sharding project so apps can use different databases and configuration changes can go through canary and blue/green setups
   • Finish building a zero-downtime database failover system
   • Set much higher scaling floors on the compute autoscalers

   Lessons

   Gadget names three deeper factors. Choosing the in-place restart put the database into a state the team had not seen before and was not confident in. Capacity planning for Black Friday / Cyber Monday assumed a doubling of traffic instead of the almost 5x that came, so settings such as max_connections stopped fitting and had to be changed in an emergency instead of in a maintenance window practised in staging. And there was no way to test database configuration changes in a production-like environment: integration tests run against a synthetic Postgres database that never goes through the months of production load behind the memory fragmentation or the memory usage issues.

   View case
5. IL-0456 Inflearn went down six times during a January 2022 free-course event as slow queries piled up on Aurora PostgreSQL

   Inflab2022

   Symptoms

   During a 100% discount event that made 125 courses free, Inflearn received 5 to 20 times its normal traffic, and on four days, 3, 4, 7 and 17 January 2022, the service became unusable around 9 a.m. and 10 p.m. There were six outages in all.

   Service outageDatabase overloadHigh latency

   Trigger

   A 100% discount event on 125 courses, from 3 to 17 January and 18 to 24 January 2022, went viral and brought 5 to 20 times the usual traffic.

   Root cause

   Capacity exhaustion

   Component

   DatabasePostgreSQL vacuum

   Contributing factor

   Inefficient queryMonitoring blind spotExcessive blast radius

   Detection

   The team followed the outages through hourly transaction counts, RDS metrics, monitoring and slow query alerts.

   Remediation

   • Added an index to the interested-companies query on the course page
   • Changed how often the cache data stored in the database was reset and regenerated
   • Moved the read queries that loaded the writer DB to the reader DBs
   • Scaled the database to 8 r5.8xlarge instances, which bought only a few tens of minutes
   • Rewrote slow course queries or added indexes so they use indexes, and switched off features whose queries could not be fixed
   • Removed the large course description column from the select * course queries
   • Split the IDs of the IN clause into batches of 20 run with Promise.all
   • Upgraded PostgreSQL from 10.16 to 11.13
   • Set a 5-second query timeout
   • Raised max connections
   • Ran VACUUM by hand on the course table
   • Moved the large course description column into a separate table

   Action items

   • Change the structure in which one table's failure spreads to the whole service
   • Improve the table and cache structures that are hard to tune
   • Make the SQL code easier to follow
   • Improve code that gives no confidence to change, such as code without types or test code
   • Build a stable batch environment
   • Keep a monolith of reasonable size and separate the domains through which failures spread most

   Lessons

   Because every database bottleneck appeared as slow queries on the reader DBs, watching only the readers hid the root cause on the writer. In PostgreSQL, frequent updates to a table with large columns produce large dead tuples in bulk and slow every query on that table. Before the second event the team could not run a meaningful performance test because the first week showed that almost any page could fail under load. The event exposed high sensitivity to failure in every area, one table's failure spreading to the whole service, hard-to-tune table and cache structures, unintuitive SQL code, code without types or tests, no stable batch environment, and many hidden failure points not yet found.

   View case

By root cause1