Alok Singh

@benoitc Is it safe to assume that thundering herd problem wont occur in gunicorn, at least with gevent worker type, if using linux 4.5+ ?

this is what the profiler shows when the number of requests are low

@tilgovi Here is a snapshot of a gunicorn process from a small test: Infra: 2 aws ecs containers: (with 2vCPU and 6GB RAM each) gunicorn workers: 4(each container) worker class:...

gevent hub is sleeping, that could be because it is idle a lot. If that's true, why CPU is going beyond 100% i'm not understanding. By low throughput I mean,...

Plus, in my datadog profiler, I'm not seeing any cpu intensive functions(from my own code or django project) that's taking too much time apart from gevent sleep(). So not sure...

This is my gunicorn command that I'm running in each container: gunicorn -n "testapp" --access-logfile gunicorn.access.log --error-logfile gunicorn.error.log --log-level 'debug' -w 4 --worker-connections=200 --reuse-port -k gevent -t 60 -b 0.0.0.0:8000...

For clarity, adding screenshots

@benoitc @tilgovi can you please provide some suggestion.

this is what the profiler shows when the number of requests are low 

Here is a snapshot of a gunicorn process from a small test: Infra: 2 aws ecs containers: (with 2vCPU and 6GB RAM each) gunicorn workers: 4 worker class: gevent My...