IL-0385 GitHub · 2021-11-27database
GitHub MySQL read replicas hit a semaphore deadlock during a schema migration's table rename and kept crash-recovering
GitHub's MySQL clusters have a primary node for writes, several read replicas for production traffic, and several replicas that serve internal reads for backups and analytics. During the final rename step of a schema migration on a large table, a significant portion of the read replicas entered a semaphore deadlock, a failure mode GitHub had not seen before. Those replicas went into crash recovery, the remaining replicas took their load, and the cascade left too few active read replicas for production requests, which degraded core GitHub services for 2 hours and 50 minutes.
To add capacity, the team moved every healthy internal replica into the production path, but that was not enough. Replicas serving production traffic kept recovering and crashing again under load, so the team chose data integrity over availability and took production traffic off the broken replicas until each one had processed the table rename. Once the replicas recovered they were put back into production and capacity returned to normal. Writes stayed healthy and GitHub verified that no data was corrupted. GitHub paused schema migrations while it investigated the failure.
- Severity
- –
- Scope
- Core GitHub services were degraded: GitHub Actions, API Requests, Codespaces, Git Operations, Issues, GitHub Packages, GitHub Pages, Pull Requests and Webhooks; write operations stayed healthy
- Users affected
- –
- Region
- –
- Time to detect
- –
- Time to resolve
- 2h 50m
Symptoms
On 27 November 2021 at 20:40 UTC, a significant part of GitHub's MySQL read replicas entered a semaphore deadlock while a schema migration on a large MySQL table ran its final step, the rename that moves the updated table into place. Core GitHub services, including GitHub Actions, API Requests, Codespaces, Git Operations, Issues, GitHub Packages, GitHub Pages, Pull Requests and Webhooks, became degraded.
The replicas that hit the deadlock went into crash recovery, which put more load on the healthy read replicas. Too few read replicas were left active to handle production requests. Replicas serving production traffic recovered from crash recovery for a while and then crashed again under the load. Write operations stayed healthy throughout.
Root cause
Root cause Software defect
- Trigger
- The final rename step of a schema migration on a large MySQL table. Schema migrations are a common task at GitHub and often take weeks to complete.
Detection
–
Remediation
- Promoted every healthy internal replica into the production path to add read capacity
- Removed production traffic from the broken replicas until each had processed the table rename
- Moved the recovered replicas back into production to restore capacity
Action items
- Continue functional partitioning so migrations can run in canary mode on a single shard
- Update internal procedures to increase how much each cluster is over-provisioned
- Pause schema migrations until it is known how to safeguard against this failure
- Keep investigating the specific failure scenario
- Test the migration tooling and classify ways to improve it for this kind of scenario
Lessons
–
Sources
Related cases
IL-0451Setting indexes INVISIBLE on Aurora MySQL 3.10 spiked CPU and ReadIOPS on all three readers but not on the writerDaangn · 2026same root cause · shared Schema migration, Database
IL-0313Decagon voice requests stalled for exactly 300 seconds behind self-managed PgBouncer and exhausted the SQLAlchemy poolDecagon · 2026same root cause · shared Database
IL-0281PostHog's Logs ClickHouse cluster silently deleted almost all its S3 data, losing US logs older than 3 daysPostHog · 2026same root cause · shared Database
IL-0310An Aurora PostgreSQL failover to an upsized reader reversed twice and backend writes failed in read-only transactionsHightouch · 2025same root cause · shared Database
IL-0150A latent race condition in DynamoDB's DNS automation left an empty DNS record for dynamodb.us-east-1.amazonaws.com, and the DynamoDB and EC2 launch failures that followed caused EKS, ECS and Fargate container launch failures and cluster scaling delays in us-east-1Amazon Web Services · 2025same root cause · shared Database
IL-0436A supremum lock left by ROLLBACK TO SAVEPOINT in a long transaction blocked inserts and filled Aurora MySQL connectionsfreee · 2025same root cause · shared Database
IL-0527OpenStreetMap.org made no replication diffs for 16 hours as PostgreSQL logical decoding failed on a memory alloc errorOpenStreetMap Foundation · 2025same root cause · shared Database
IL-0329A routine migration left PGAudit stuck holding Postgres locks, causing intermittent timeouts and a 2-minute outageincident.io · 2025same root cause · shared Database
IL-0347About 25% of Scaleway FR-PAR-1 hypervisors lost VPC connectivity when both VPC BGP route reflectors got stuckScaleway · 2024same root cause · shared Database
IL-0373Heroku Postgres Essential-tier databases in the EU region ran out of memory and refused connections, twice in four daysHeroku · 2024same root cause · shared Database
IL-0024Expensive cache-backed API calls and uncancelled database requests overloaded MongoDB and took the Hugging Face Hub down three timesHugging Face · 2024same root cause · shared Database
IL-0217Honeycomb's main MySQL database locked up under cache-miss reads and ran out of connections, a 68-minute total outageHoneycomb · 2023same root cause · shared Database
IL-0330Requests timed out for two weeks waiting on Go's database/sql connection pool, one for almost 20 secondsincident.io · 2023same root cause · shared Database
IL-0462Some MySQL pods on Shopify's GKE clusters took up to 30 minutes to start, slowing instance replacementShopify · 2021same root cause · shared Database
IL-0196Adding a second Redis replica and lowering memory limits in one RedisFailover change restarted the only Redis pod that held data, and Sentinel promoted the empty new replica, erasing the dataPalark · 2020same root cause · shared Database
IL-0486Figma's PostgreSQL 9 database on RDS slowed under high write IOPS and lock contention until the site was unavailableFigma · 2020same root cause · shared Database
IL-0241A database shard could not elect a new primary, and the Stripe API failed most requests in two periods on one dayStripe · 2019same root cause · shared Database
IL-0439Database access slowed after each block device and filesystem expansion because the Linux kernel dropped the page cacheCybozu · 2018same root cause · shared Database
IL-0388A Google Cloud live migration took a Discord Redis primary offline, and the failover broke API instances and then guildsDiscord · 2017same root cause · shared Database
IL-0454Baemin billing servers could not open new MariaDB connections at the start of a card-discount eventWoowa Brothers · 2016same root cause · shared Database
IL-0536Amazon SimpleDB in US East returned HTTP 500 to nearly all API calls for two hours after its nodes left the clusterAmazon Web Services · 2014same root cause · shared Database
IL-0485Dropbox went down when an OS upgrade script reinstalled active MySQL master-replica pairs and backup restores ran slowDropbox · 2014same root cause · shared Database
IL-0589Every Kickstarter MySQL replica on Amazon RDS stopped replicating on a duplicate-key error in the backings tableKickstarter · 2013same root cause · shared Database
IL-0303GitHub's MySQL master failed over to a node with a cold buffer pool during a schema migration and GitHub.com went downGitHub · 2012same root cause · shared Database
Check your cluster for this
For AI tools and scripts
- This case as JSON: https://incidentlib-api.benchlet.com/cases/IL-0385
- The cases most like it: https://incidentlib-api.benchlet.com/cases/IL-0385/similar
- MCP server (Streamable HTTP, no key):
https://incidentlib-mcp.benchlet.com/mcp
The API needs no key. The MCP server offers search_incidents, get_incident, list_incidents, get_topic and vocab; both are explained on API and MCP.