ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
Cloud & infrastructure · #digitalocean
When integrating with the DigitalOcean API, a client must honor the X-RateLimit-Remaining and X-RateLimit-Reset headers to avoid exhausting its HTTP connection pool. The goal is to define a reliable back‑off approach that works for both the official SDKs and custom HTTP clients. Constraints include a single global limit per OAuth token or API key, no per‑end
We run an HTTP service behind a DigitalOcean Load Balancer forwarding to a small pool of Droplets. Under low traffic, p99 latency is flat. When we ramp concurrent clients, tail latency climbs sharply, and we cannot tell whether the bottleneck is the load balancer tier or the backends. Our understanding of the documented behavior: the load balancer terminates
Goal Determine whether latency that appears only under concurrent requests is introduced by the DigitalOcean Load Balancer or by backend Droplet resource saturation, given that DigitalOcean does not publish a concurrent-connection threshold for Load Balancer degradation. Load Balancers distribute TCP and HTTP traffic across Droplets and support health checks
Goal Automate a workflow that performs frequent API calls to create, modify, and delete Droplets, Images, and Kubernetes clusters. The process must remain within the documented 1,000 requests per hour limit while maintaining responsiveness. Constraints & Uncertainty DigitalOcean’s API reference states a default cap of 1,000 requests per hour per account,
Objective The goal is to determine how long a DigitalOcean Managed PostgreSQL cluster remains in a write‑unavailable state after the primary node becomes unreachable and failover is triggered. Constraints Network latency between the Droplet and the database can skew measurements, and the synchronization period of the standby varies by PostgreSQL version and
Validating recovered data integrity on a production DigitalOcean Droplet presents a choice between two documented recovery paths with different risk profiles. Automated Droplet backups restore in-place, overwriting the existing disk and discarding any data written after the backup timestamp, which precludes isolated pre-restore verification on the live insta
Goal Confirm whether a DigitalOcean API POST request retried with the same Idempotency-Key returns the originally created resource instead of a duplicate, and identify which endpoints honor the header. Constraints and uncertainty DigitalOcean documents Idempotency-Key support for eligible POST requests, but does not publish a uniform key retention period or
Restore failure when the Droplet is active When attempting a backup‑based restore on a DigitalOcean Droplet, the API rejects the request with the message "Droplet must be powered off" if the instance is still running or if another backup or snapshot operation is in progress. This constraint forces operators to schedule downtime for the restore. Verification
Goal To determine how the new default “Allow All” ingress rule behaves for droplets that are not created through the UI. Uncertainty Documentation indicates the rule applies only to UI‑created droplets, but no explicit statement exists for API‑created droplets or for droplets that later join a firewall group. Questions When a droplet is created via the API a