IL-0485 Dropbox · 2014-01-10database
Dropbox went down when an OS upgrade script reinstalled active MySQL master-replica pairs and backup restores ran slow
Dropbox runs thousands of databases, each with one master and two replica machines, and keeps full and incremental backups in a separate environment. During a planned OS upgrade on Friday at 5:30 PM PT, the upgrade script was supposed to verify remotely that each machine had no active data before installing the new OS. A subtle bug in the script made it reinstall a small number of machines that were serving production traffic. Some of them were master-replica pairs, and the site went down.
To bring the service back as fast as possible, the team recovered the databases from backups. Most functionality was back within about three hours, but the standard tool for recovering MySQL data from backups was slow on the largest databases, and core service was fully restored only at 4:40 PM PT on Sunday. Afterwards Dropbox added a local check on each machine before it runs incoming commands, and built a tool that replays binary logs in parallel to speed up restores of large MySQL backups.
- Severity
- –
- Scope
- Dropbox service down for about three hours after scheduled maintenance; core service not fully restored until 4:40 PM PT on Sunday; user files were not at risk
- Users affected
- –
- Region
- –
- Time to detect
- –
- Time to resolve
- –
Symptoms
On Friday evening, during scheduled maintenance that started at 5:30 PM PT, the Dropbox service went down. Most functionality came back about three hours later, but core service was not fully restored until 4:40 PM PT on Sunday.
The affected databases did not hold file data, so user files were not at risk. They back features such as photo album sharing, camera uploads and some API features.
Symptom Service outage
Root cause
Root cause Software defect
- Trigger
- A planned maintenance on Friday at 5:30 PM PT upgraded the operating system on some machines with a script that was meant to check that a machine held no active data before reinstalling it.
Detection
–
Remediation
- Restored the affected databases from backups
Action items
- Add a check that makes each machine verify its own state locally before running incoming commands, so machines running critical processes refuse destructive operations
- Use a new tool that replays binary logs in parallel to recover large MySQL backups faster
- Open source the parallel binary log replay tool
Lessons
Scripts that verify a machine's production state remotely before upgrading or repurposing it can be wrong, so machines need to check their own state before accepting destructive commands. Replicas give redundancy, but when every replica is lost the restore from backup decides the length of the outage, and the standard MySQL restore tool is slow on large data sets.
Sources
Related cases
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 backups and snapshots, Database
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 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-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-0385GitHub MySQL read replicas hit a semaphore deadlock during a schema migration's table rename and kept crash-recoveringGitHub · 2021same 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-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-0485
- The cases most like it: https://incidentlib-api.benchlet.com/cases/IL-0485/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.