Locust RPS stuck at 0 / all requests failing in CI

When Locust shows 0 RPS or a 100% failure column, no request is completing successfully. In CI this almost always means the --host is wrong or unreachable, the target is not up, or requests are rejected for missing auth.

What this error means

The Locust stats table shows all requests under the "# fails" column, RPS near 0, and a failures list dominated by ConnectionError or a repeated 401/403/404.

Locust

Type     Name    # reqs  # fails  |  Avg  ...  # RPS
GET      /       0       412(100%) |  0    ...  0.00
ConnectionError(... Connection refused ...)  412

Common causes

Wrong or unreachable host

A --host that points at the wrong address, or a target not yet listening, makes every request raise a connection error.

Missing auth or wrong path

If the endpoint needs a token or the path is wrong, every request returns 401/403/404 and counts as a failure.

How to fix it

Confirm host reachability before the run

Curl the target from the same job to prove it is up and correct.
Pass the confirmed URL as --host.
Wait for readiness so the run does not start against a down target.

Terminal

curl -sf http://localhost:8080/health
locust --headless -u 100 -r 10 --run-time 1m --host http://localhost:8080

Send required auth in the task

Attach the token or session the endpoint requires so requests succeed instead of returning 401/403.

locustfile.py

from locust import HttpUser, task
class ApiUser(HttpUser):
    @task
    def get(self):
        self.client.get("/", headers={"Authorization": "Bearer " + TOKEN})

How to prevent it

Verify the target is up and the host is correct before load.
Provide auth headers or a login flow in the locustfile.
Use correct request paths so responses are not all failures.