Contact person: Fix AWS SSE Stream Stalling

Phone:Show

Email:Show

Location: Sangrur, India

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"My Server-Sent Events endpoints run flawlessly on my laptop, yet once the same code is behind Nginx → Gunicorn on an Ubuntu 20.04 EC2 instance they freeze after the initial connection and nothing ever reaches the browser. Everything else on the site is healthy, so the fault is confined to SSE.

Here’s the landscape you’ll step into: a Python web app, Gunicorn (systemd-managed) and Nginx acting as reverse proxy. I have already experimented with every obvious knob in the Nginx conf, tweaked worker classes and timeouts in Gunicorn, and checked AWS load-balancer and security-group settings, but the stream still stalls. Logs show no smoking gun.

What I need from you is a thorough trace that shows exactly where the data stops, a surgical fix in configuration or code, and proof that clients receive an uninterrupted event-stream while we hammer the endpoint with realistic load. You will have SSH access to the instance along with the existing Nginx site files and Gunicorn unit file, so feel free to bring curl, htop, strace, tcpdump, Siege, or whatever tooling you trust.

Deliverables
• Root-cause explanation with the line, setting, or network hop responsible
• Updated config or code (diff or replacement files) applied on the server
• Live demonstration—e.g. watch-curl or browser console video—showing a continuous stream sustaining load without stalls

Acceptance criteria: a minimum ten-minute uninterrupted event-stream to multiple concurrent clients, verified from an external network, with no 499/502 errors in Nginx and steady memory/CPU on Gunicorn.

If you know the quirks of proxy_buffering, Chunked-Encoding, EC2 idle timeouts, and long-polling pitfalls inside Gunicorn, this should be a quick win. Let’s get the stream flowing." (client-provided description)