Catalogue
- Cards
- 165
- Records
- 486
- Workloads
- 19
- Corpus version
- 74847648e25a
175 of 486 records match; the first 20 are shown.
| Kind | Record | Organisation | Year | Setup |
|---|---|---|---|---|
| Failure | A dormant topology error and non-staged legacy rollout withdrew the BGP routes of Cloudflare's 1.1.1.1 Resolver globally, making it unavailable for 62 minutes | Cloudflare | 2025 | Multi-region |
| Failure | Automatic removal of IP bindings on 23 percent of edge servers delayed full recovery after BGP readvertisement | Cloudflare | 2025 | Multi-region |
| Decision | Replace hard-coded prefix-to-datacenter lists with service topologies that support staged rollouts | Cloudflare | 2025 | Multi-region |
| Failure | Search compaction invalidations triggered a WebSocket reconnect storm that was prolonged by a manual deploy onto default free-tier hardware | Convex | 2025 | |
| Failure | NOTIFY inside transactions serialized all commits globally on Postgres, causing three downtimes | Recall.ai | 2025 | |
| Decision | Replace Postgres LISTEN/NOTIFY with application-layer tracking for bot configuration updates | Recall.ai | 2025 | |
| Failure | Users who follow hundreds of thousands of accounts made hot shards that slowed the timelines sharing them | Bluesky | 2025 | Sharded |
| Decision | Drop timeline writes with a probability set by how many accounts a user follows, accepting imperfect timelines | Bluesky | 2025 | Sharded |
| Decision | Cache high-follow accounts in a Redis sorted set loaded into memory every 30 seconds instead of reading follow counts per write | Bluesky | 2025 | Sharded |
| Decision | Move the public sites from bare-metal data centers to Kubernetes in one GCP region, using Windows Kubernetes for legacy .NET apps | Stack Overflow | 2025 | One site |
| Decision | Run SQL Server on dedicated high-memory GCP VMs with half the CPU cores disabled instead of managed cloud SQL | Stack Overflow | 2025 | One site |
| Decision | Use GCP MemoryStore for Redis with key prefixes instead of self-hosting Redis on Kubernetes | Stack Overflow | 2025 | One site |
| Decision | Self-host Elasticsearch on Kubernetes with the ECK operator instead of using Elastic Cloud | Stack Overflow | 2025 | One site |
| Decision | Load-test the cloud stack in isolated Terraform-built environments matching production before migrating | Stack Overflow | 2025 | One site |
| Decision | Plan to put public assets behind Cloudflare CDN and process its logs in Snowflake, retiring HAProxy logs and TPS | Stack Overflow | 2025 | One site |
| Decision | Package cross-cloud deployments in CNAB containers deployed via Octopus Deploy and GitHub Actions | Stack Overflow | 2025 | One site |
| Decision | Shrink a 57 GB uncleaned staging table by cloning, copying recent rows, renaming and dropping the old table | Fireside.fm | 2024 | |
| Decision | Build a two-level global and per-location container scheduler on Workers, Durable Objects, and KV | Cloudflare | 2024 | Multi-region |
| Decision | Keep the container scheduler runtime-agnostic across gVisor, Firecracker microVMs, and QEMU | Cloudflare | 2024 | Multi-region |
| Decision | Serve large GPU images via Workers serverless-registry on R2 with a custom zstd and 500 MB chunking CLI | Cloudflare | 2024 | Multi-region |
155 more records. Sign in to see all 175.