forked from phoenix/litellm-mirror
refactor: refactor key tests
This commit is contained in:
Krrish Dholakia
parent
bb04a340a5
commit
162f6f1ed3
2 changed files with 248 additions and 229 deletions
|
|
@ -1,63 +1,63 @@
|
|||
import sys, os, time
|
||||
import traceback
|
||||
from dotenv import load_dotenv
|
||||
# import sys, os, time
|
||||
# import traceback
|
||||
# from dotenv import load_dotenv
|
||||
|
||||
load_dotenv()
|
||||
import os, io
|
||||
# load_dotenv()
|
||||
# import os, io
|
||||
|
||||
# this file is to test litellm/proxy
|
||||
# # this file is to test litellm/proxy
|
||||
|
||||
sys.path.insert(
|
||||
0, os.path.abspath("../..")
|
||||
) # Adds the parent directory to the system path
|
||||
import pytest, logging, requests
|
||||
import litellm
|
||||
from litellm import embedding, completion, completion_cost, Timeout
|
||||
from litellm import RateLimitError
|
||||
# sys.path.insert(
|
||||
# 0, os.path.abspath("../..")
|
||||
# ) # Adds the parent directory to the system path
|
||||
# import pytest, logging, requests
|
||||
# import litellm
|
||||
# from litellm import embedding, completion, completion_cost, Timeout
|
||||
# from litellm import RateLimitError
|
||||
|
||||
|
||||
def test_add_new_key():
|
||||
max_retries = 3
|
||||
retry_delay = 1 # seconds
|
||||
# def test_add_new_key():
|
||||
# max_retries = 3
|
||||
# retry_delay = 1 # seconds
|
||||
|
||||
for retry in range(max_retries + 1):
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {
|
||||
"models": ["gpt-3.5-turbo", "gpt-4", "claude-2", "azure-model"],
|
||||
"aliases": {"mistral-7b": "gpt-3.5-turbo"},
|
||||
"duration": "20m",
|
||||
}
|
||||
print("testing proxy server")
|
||||
# for retry in range(max_retries + 1):
|
||||
# try:
|
||||
# # Your test data
|
||||
# test_data = {
|
||||
# "models": ["gpt-3.5-turbo", "gpt-4", "claude-2", "azure-model"],
|
||||
# "aliases": {"mistral-7b": "gpt-3.5-turbo"},
|
||||
# "duration": "20m",
|
||||
# }
|
||||
# print("testing proxy server")
|
||||
|
||||
# Your bearer token
|
||||
token = os.getenv("PROXY_MASTER_KEY")
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
# # Your bearer token
|
||||
# token = os.getenv("PROXY_MASTER_KEY")
|
||||
# headers = {"Authorization": f"Bearer {token}"}
|
||||
|
||||
staging_endpoint = "https://litellm-litellm-pr-1376.up.railway.app"
|
||||
main_endpoint = "https://litellm-staging.up.railway.app"
|
||||
# staging_endpoint = "https://litellm-litellm-pr-1376.up.railway.app"
|
||||
# main_endpoint = "https://litellm-staging.up.railway.app"
|
||||
|
||||
# Make a request to the staging endpoint
|
||||
response = requests.post(
|
||||
main_endpoint + "/key/generate", json=test_data, headers=headers
|
||||
)
|
||||
# # Make a request to the staging endpoint
|
||||
# response = requests.post(
|
||||
# main_endpoint + "/key/generate", json=test_data, headers=headers
|
||||
# )
|
||||
|
||||
print(f"response: {response.text}")
|
||||
# print(f"response: {response.text}")
|
||||
|
||||
if response.status_code == 200:
|
||||
result = response.json()
|
||||
break # Successful response, exit the loop
|
||||
elif response.status_code == 503 and retry < max_retries:
|
||||
print(
|
||||
f"Retrying in {retry_delay} seconds... (Retry {retry + 1}/{max_retries})"
|
||||
)
|
||||
time.sleep(retry_delay)
|
||||
else:
|
||||
assert False, f"Unexpected response status code: {response.status_code}"
|
||||
# if response.status_code == 200:
|
||||
# result = response.json()
|
||||
# break # Successful response, exit the loop
|
||||
# elif response.status_code == 503 and retry < max_retries:
|
||||
# print(
|
||||
# f"Retrying in {retry_delay} seconds... (Retry {retry + 1}/{max_retries})"
|
||||
# )
|
||||
# time.sleep(retry_delay)
|
||||
# else:
|
||||
# assert False, f"Unexpected response status code: {response.status_code}"
|
||||
|
||||
except Exception as e:
|
||||
print(traceback.format_exc())
|
||||
pytest.fail(f"An error occurred {e}")
|
||||
# except Exception as e:
|
||||
# print(traceback.format_exc())
|
||||
# pytest.fail(f"An error occurred {e}")
|
||||
|
||||
|
||||
test_add_new_key()
|
||||
# test_add_new_key()
|
||||
|
|
|
|||
|
|
@ -15,69 +15,66 @@ import litellm
|
|||
from litellm import embedding, completion, completion_cost, Timeout
|
||||
from litellm import RateLimitError
|
||||
|
||||
# Configure logging
|
||||
logging.basicConfig(
|
||||
level=logging.DEBUG, # Set the desired logging level
|
||||
format="%(asctime)s - %(levelname)s - %(message)s",
|
||||
)
|
||||
|
||||
import sys, os, time
|
||||
import traceback
|
||||
from dotenv import load_dotenv
|
||||
|
||||
load_dotenv()
|
||||
import os, io
|
||||
|
||||
# this file is to test litellm/proxy
|
||||
from concurrent.futures import ThreadPoolExecutor
|
||||
|
||||
# test /chat/completion request to the proxy
|
||||
from fastapi.testclient import TestClient
|
||||
from fastapi import FastAPI
|
||||
from litellm.proxy.proxy_server import (
|
||||
router,
|
||||
save_worker_config,
|
||||
startup_event,
|
||||
shutdown_event,
|
||||
) # Replace with the actual module where your FastAPI router is defined
|
||||
sys.path.insert(
|
||||
0, os.path.abspath("../..")
|
||||
) # Adds the parent directory to the system path
|
||||
|
||||
filepath = os.path.dirname(os.path.abspath(__file__))
|
||||
config_fp = f"{filepath}/test_configs/test_config.yaml"
|
||||
save_worker_config(
|
||||
config=config_fp,
|
||||
model=None,
|
||||
alias=None,
|
||||
api_base=None,
|
||||
api_version=None,
|
||||
debug=True,
|
||||
temperature=None,
|
||||
max_tokens=None,
|
||||
request_timeout=600,
|
||||
max_budget=None,
|
||||
telemetry=False,
|
||||
drop_params=True,
|
||||
add_function_to_prompt=False,
|
||||
headers=None,
|
||||
save=False,
|
||||
use_queue=False,
|
||||
)
|
||||
app = FastAPI()
|
||||
app.include_router(router) # Include your router in the test app
|
||||
import pytest, logging, requests
|
||||
import litellm
|
||||
from litellm import embedding, completion, completion_cost, Timeout
|
||||
from litellm import RateLimitError
|
||||
from github import Github
|
||||
import subprocess
|
||||
|
||||
|
||||
@app.on_event("startup")
|
||||
async def wrapper_startup_event():
|
||||
await startup_event()
|
||||
# Function to execute a command and return the output
|
||||
def run_command(command):
|
||||
process = subprocess.Popen(command, stdout=subprocess.PIPE, shell=True)
|
||||
output, _ = process.communicate()
|
||||
return output.decode().strip()
|
||||
|
||||
|
||||
@app.on_event("shutdown")
|
||||
async def wrapper_shutdown_event():
|
||||
await shutdown_event()
|
||||
# Retrieve the current branch name
|
||||
branch_name = run_command("git rev-parse --abbrev-ref HEAD")
|
||||
|
||||
# GitHub personal access token (with repo scope) or use username and password
|
||||
access_token = os.getenv("GITHUB_ACCESS_TOKEN")
|
||||
# Instantiate the PyGithub library's Github object
|
||||
g = Github(access_token)
|
||||
|
||||
# Provide the owner and name of the repository where the pull request is located
|
||||
repository_owner = "BerriAI"
|
||||
repository_name = "litellm"
|
||||
|
||||
# Get the repository object
|
||||
repo = g.get_repo(f"{repository_owner}/{repository_name}")
|
||||
|
||||
# Iterate through the pull requests to find the one related to your branch
|
||||
for pr in repo.get_pulls():
|
||||
print(f"in here! {pr.head.ref}")
|
||||
if pr.head.ref == branch_name:
|
||||
pr_number = pr.number
|
||||
break
|
||||
|
||||
print(f"The pull request number for branch {branch_name} is: {pr_number}")
|
||||
|
||||
|
||||
# Here you create a fixture that will be used by your tests
|
||||
# Make sure the fixture returns TestClient(app)
|
||||
@pytest.fixture(autouse=True)
|
||||
def client():
|
||||
from litellm.proxy.proxy_server import cleanup_router_config_variables
|
||||
def test_add_new_key():
|
||||
max_retries = 3
|
||||
retry_delay = 1 # seconds
|
||||
|
||||
cleanup_router_config_variables()
|
||||
with TestClient(app) as client:
|
||||
yield client
|
||||
|
||||
|
||||
def test_add_new_key(client):
|
||||
for retry in range(max_retries + 1):
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {
|
||||
|
|
@ -86,40 +83,37 @@ def test_add_new_key(client):
|
|||
"duration": "20m",
|
||||
}
|
||||
print("testing proxy server")
|
||||
|
||||
# Your bearer token
|
||||
token = os.getenv("PROXY_MASTER_KEY")
|
||||
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
response = client.post("/key/generate", json=test_data, headers=headers)
|
||||
|
||||
endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app"
|
||||
|
||||
# Make a request to the staging endpoint
|
||||
response = requests.post(
|
||||
endpoint + "/key/generate", json=test_data, headers=headers
|
||||
)
|
||||
|
||||
print(f"response: {response.text}")
|
||||
assert response.status_code == 200
|
||||
|
||||
if response.status_code == 200:
|
||||
result = response.json()
|
||||
assert result["key"].startswith("sk-")
|
||||
|
||||
def _post_data():
|
||||
json_data = {
|
||||
"model": "azure-model",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": f"this is a test request, write a short poem {time.time()}",
|
||||
}
|
||||
],
|
||||
}
|
||||
response = client.post(
|
||||
"/chat/completions",
|
||||
json=json_data,
|
||||
headers={"Authorization": f"Bearer {result['key']}"},
|
||||
break # Successful response, exit the loop
|
||||
elif response.status_code == 503 and retry < max_retries:
|
||||
print(
|
||||
f"Retrying in {retry_delay} seconds... (Retry {retry + 1}/{max_retries})"
|
||||
)
|
||||
return response
|
||||
time.sleep(retry_delay)
|
||||
else:
|
||||
assert False, f"Unexpected response status code: {response.status_code}"
|
||||
|
||||
_post_data()
|
||||
print(f"Received response: {result}")
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
print(traceback.format_exc())
|
||||
pytest.fail(f"An error occurred {e}")
|
||||
|
||||
|
||||
def test_update_new_key(client):
|
||||
def test_update_new_key():
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {
|
||||
|
|
@ -130,17 +124,23 @@ def test_update_new_key(client):
|
|||
print("testing proxy server")
|
||||
# Your bearer token
|
||||
token = os.getenv("PROXY_MASTER_KEY")
|
||||
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
response = client.post("/key/generate", json=test_data, headers=headers)
|
||||
print(f"response: {response.text}")
|
||||
|
||||
endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app"
|
||||
|
||||
# Make a request to the staging endpoint
|
||||
response = requests.post(
|
||||
endpoint + "/key/generate", json=test_data, headers=headers
|
||||
)
|
||||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
assert result["key"].startswith("sk-")
|
||||
|
||||
def _post_data():
|
||||
json_data = {"models": ["bedrock-models"], "key": result["key"]}
|
||||
response = client.post("/key/update", json=json_data, headers=headers)
|
||||
response = requests.post(
|
||||
endpoint + "/key/generate", json=json_data, headers=headers
|
||||
)
|
||||
print(f"response text: {response.text}")
|
||||
assert response.status_code == 200
|
||||
return response
|
||||
|
|
@ -151,101 +151,120 @@ def test_update_new_key(client):
|
|||
pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
|
||||
|
||||
# # Run the test - only runs via pytest
|
||||
# def test_add_new_key_max_parallel_limit():
|
||||
# try:
|
||||
# # Your test data
|
||||
# test_data = {"duration": "20m", "max_parallel_requests": 1}
|
||||
# # Your bearer token
|
||||
# token = os.getenv("PROXY_MASTER_KEY")
|
||||
# headers = {"Authorization": f"Bearer {token}"}
|
||||
|
||||
# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app"
|
||||
# print(f"endpoint: {endpoint}")
|
||||
# # Make a request to the staging endpoint
|
||||
# response = requests.post(
|
||||
# endpoint + "/key/generate", json=test_data, headers=headers
|
||||
# )
|
||||
# assert response.status_code == 200
|
||||
# result = response.json()
|
||||
|
||||
# # load endpoint with model
|
||||
# model_data = {
|
||||
# "model_name": "azure-model",
|
||||
# "litellm_params": {
|
||||
# "model": "azure/chatgpt-v-2",
|
||||
# "api_key": os.getenv("AZURE_API_KEY"),
|
||||
# "api_base": os.getenv("AZURE_API_BASE"),
|
||||
# "api_version": os.getenv("AZURE_API_VERSION")
|
||||
# }
|
||||
# }
|
||||
# response = requests.post(endpoint + "/model/new", json=model_data, headers=headers)
|
||||
# assert response.status_code == 200
|
||||
# print(f"response text: {response.text}")
|
||||
|
||||
|
||||
def test_add_new_key_max_parallel_limit(client):
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {"duration": "20m", "max_parallel_requests": 1}
|
||||
# Your bearer token
|
||||
token = os.getenv("PROXY_MASTER_KEY")
|
||||
# def _post_data():
|
||||
# json_data = {
|
||||
# "model": "azure-model",
|
||||
# "messages": [
|
||||
# {
|
||||
# "role": "user",
|
||||
# "content": f"this is a test request, write a short poem {time.time()}",
|
||||
# }
|
||||
# ],
|
||||
# }
|
||||
# # Your bearer token
|
||||
# response = requests.post(
|
||||
# endpoint + "/chat/completions", json=json_data, headers={"Authorization": f"Bearer {result['key']}"}
|
||||
# )
|
||||
# return response
|
||||
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
response = client.post("/key/generate", json=test_data, headers=headers)
|
||||
print(f"response: {response.text}")
|
||||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
# def _run_in_parallel():
|
||||
# with ThreadPoolExecutor(max_workers=2) as executor:
|
||||
# future1 = executor.submit(_post_data)
|
||||
# future2 = executor.submit(_post_data)
|
||||
|
||||
def _post_data():
|
||||
json_data = {
|
||||
"model": "azure-model",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": f"this is a test request, write a short poem {time.time()}",
|
||||
}
|
||||
],
|
||||
}
|
||||
response = client.post(
|
||||
"/chat/completions",
|
||||
json=json_data,
|
||||
headers={"Authorization": f"Bearer {result['key']}"},
|
||||
)
|
||||
return response
|
||||
# # Obtain the results from the futures
|
||||
# response1 = future1.result()
|
||||
# print(f"response1 text: {response1.text}")
|
||||
# response2 = future2.result()
|
||||
# print(f"response2 text: {response2.text}")
|
||||
# if response1.status_code == 429 or response2.status_code == 429:
|
||||
# pass
|
||||
# else:
|
||||
# raise Exception()
|
||||
|
||||
def _run_in_parallel():
|
||||
with ThreadPoolExecutor(max_workers=2) as executor:
|
||||
future1 = executor.submit(_post_data)
|
||||
future2 = executor.submit(_post_data)
|
||||
# _run_in_parallel()
|
||||
# except Exception as e:
|
||||
# pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
|
||||
# Obtain the results from the futures
|
||||
response1 = future1.result()
|
||||
response2 = future2.result()
|
||||
if response1.status_code == 429 or response2.status_code == 429:
|
||||
pass
|
||||
else:
|
||||
raise Exception()
|
||||
# def test_add_new_key_max_parallel_limit_streaming():
|
||||
# try:
|
||||
# # Your test data
|
||||
# test_data = {"duration": "20m", "max_parallel_requests": 1}
|
||||
# # Your bearer token
|
||||
# token = os.getenv("PROXY_MASTER_KEY")
|
||||
# headers = {"Authorization": f"Bearer {token}"}
|
||||
|
||||
_run_in_parallel()
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
# endpoint = f"https://litellm-litellm-pr-{pr_number}.up.railway.app"
|
||||
|
||||
# # Make a request to the staging endpoint
|
||||
# response = requests.post(
|
||||
# endpoint + "/key/generate", json=test_data, headers=headers
|
||||
# )
|
||||
# print(f"response: {response.text}")
|
||||
# assert response.status_code == 200
|
||||
# result = response.json()
|
||||
|
||||
def test_add_new_key_max_parallel_limit_streaming(client):
|
||||
try:
|
||||
# Your test data
|
||||
test_data = {"duration": "20m", "max_parallel_requests": 1}
|
||||
# Your bearer token
|
||||
token = os.getenv("PROXY_MASTER_KEY")
|
||||
# def _post_data():
|
||||
# json_data = {
|
||||
# "model": "azure-model",
|
||||
# "messages": [
|
||||
# {
|
||||
# "role": "user",
|
||||
# "content": f"this is a test request, write a short poem {time.time()}",
|
||||
# }
|
||||
# ],
|
||||
# "stream": True,
|
||||
# }
|
||||
# response = requests.post(
|
||||
# endpoint + "/chat/completions", json=json_data, headers={"Authorization": f"Bearer {result['key']}"}
|
||||
# )
|
||||
# return response
|
||||
|
||||
headers = {"Authorization": f"Bearer {token}"}
|
||||
response = client.post("/key/generate", json=test_data, headers=headers)
|
||||
print(f"response: {response.text}")
|
||||
assert response.status_code == 200
|
||||
result = response.json()
|
||||
# def _run_in_parallel():
|
||||
# with ThreadPoolExecutor(max_workers=2) as executor:
|
||||
# future1 = executor.submit(_post_data)
|
||||
# future2 = executor.submit(_post_data)
|
||||
|
||||
def _post_data():
|
||||
json_data = {
|
||||
"model": "azure-model",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": f"this is a test request, write a short poem {time.time()}",
|
||||
}
|
||||
],
|
||||
"stream": True,
|
||||
}
|
||||
response = client.post(
|
||||
"/chat/completions",
|
||||
json=json_data,
|
||||
headers={"Authorization": f"Bearer {result['key']}"},
|
||||
)
|
||||
return response
|
||||
# # Obtain the results from the futures
|
||||
# response1 = future1.result()
|
||||
# response2 = future2.result()
|
||||
# if response1.status_code == 429 or response2.status_code == 429:
|
||||
# pass
|
||||
# else:
|
||||
# raise Exception()
|
||||
|
||||
def _run_in_parallel():
|
||||
with ThreadPoolExecutor(max_workers=2) as executor:
|
||||
future1 = executor.submit(_post_data)
|
||||
future2 = executor.submit(_post_data)
|
||||
|
||||
# Obtain the results from the futures
|
||||
response1 = future1.result()
|
||||
response2 = future2.result()
|
||||
if response1.status_code == 429 or response2.status_code == 429:
|
||||
pass
|
||||
else:
|
||||
raise Exception()
|
||||
|
||||
_run_in_parallel()
|
||||
except Exception as e:
|
||||
pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
# _run_in_parallel()
|
||||
# except Exception as e:
|
||||
# pytest.fail(f"LiteLLM Proxy test failed. Exception: {str(e)}")
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue