Hi, thank you for this framework. I was wondering if you can update the version of locsut please? Thank you!

According to @vamega.

We probably need to update the code here.

Hi mates,

Following the readme tutorial to deploy on AWS lambda I got the following error when I'm trying to invoke

[ERROR] 2020-07-14T18:04:00.80Z bf187cde-4c12-46ed-b185-b0be7fe04c1f Locust exception TypeError("create_settings() got an unexpected keyword argument 'num_requests'")

Any idea?

Hi :)

I have a use case that I need to add environment variables with some settings to use on Locust.
I'm using at moment LambdaLoadTest method to run my lambda function, there is any way to do it?

Thanks 🙏

I'm trying to understand the example provided at the bottom of the Readme about running the load test in parallel.

Unfortunately I do not fully grasp how the called lambda_locust function should look like.

def handler(event=None):
 # create settings from payload 
 settings = invokust.create_settings(event )
 loadtest = invokust.LocustLoadTest(settings)
 loadtest.run()

Should it just accept the JSON payload and run a standard load test ? Or do I have to change the settings according to the number of threads passed to the LambdaLoadTest class?

# Default settings that are not to be changed
    settings.no_web = True
    settings.master = False
    settings.show_task_ratio_json = False
    settings.list_commands = False
    settings.loglevel = 'INFO'
    settings.slave = False
    settings.only_summary = True
    settings.logfile = None
    settings.show_task_ratio = False
    settings.print_stats = False

In particular it would be useful to be able to deploy AWS lambda slaves all communicating with a master host.

Would I be able to run this on OpenFaaS' Python runtime?

https://github.com/openfaas-incubator/python3-debian

I can do a build following the instructions but it seems to reach out to a Lambda/AWS region.

Hi,

I was running my invokust with the below settings,

settings = invokust.create_settings(
    classes=[WebsiteUser],
    host='https://google.com',
    num_clients=1,
    hatch_rate=1,
    run_time='10s'
)

WIth the above settings i should see 10 requests being made, as the hatch_rate is 1 and time limit is 10s.

But i see 51 requests being made in this 10s duration.

{
  "requests": {
    "GET_/": {
      "request_type": "GET",
      "num_requests": 51,
      "min_response_time": 139.55307006835938,
      "median_response_time": 180,

Hi,

I'm using the following code to trigger the load test:

import logging
import os
import json
from invokust.aws_lambda import LambdaLoadTest, results_aggregator

if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    run_time = os.environ.get("LOCUST_RUN_TIME")
    threads = max(1, int(os.environ.get("LOCUST_NUM_USERS")) / 10)
    load_test = LambdaLoadTest(
        lambda_function_name="load-test",
        threads=threads,
        ramp_time=0,
        time_limit=30,
        lambda_payload={"run_time": run_time}
    )
    load_test.run()
    logging.info(load_test.get_summary_stats())
    logging.info(json.dumps(results_aggregator(load_test.get_locust_results()), indent=4))

load-test lambda is very simple (128MB memory and timeout of 15mins), hitting one endpoint every 5s.

For 10m and 5 users and I get:

[2021-10-06 13:21:43,035] cedb38dd3bda/INFO/root: Waiting for all Lambdas to return. This may take up to 10m.
[2021-10-06 13:37:23,942] cedb38dd3bda/CRITICAL/root: Lambda invocation failed: ConnectionClosedError('Connection was closed before we received a valid response from endpoint URL: "https://lambda.eu-central-1.amazonaws.com/2015-03-31/functions/load-test/invocations".')
[2021-10-06 13:37:25,950] cedb38dd3bda/INFO/root: thread finished
[2021-10-06 13:37:26,037] cedb38dd3bda/INFO/root: {'lambda_invocation_count': 0, 'total_lambda_execution_time': 0, 'requests_total': 0, 'request_fail_ratio': 0, 'invocation_error_ratio': 0}
[2021-10-06 13:37:26,041] cedb38dd3bda/INFO/root: {
    "requests": {},
    "failures": {},
    "num_requests": 0,
    "num_requests_fail": 0,
    "total_lambda_execution_time": 0,
    "lambda_invocations": 0,
    "approximate_cost": 0.0
}

Would you consider adopting semantic versioning?

I have seen your the release version was bumped up from 0.61 to 0.7 yesterday.

Chronologically 0.7 is newer than 0.61, but mathematically isn't. While I understand the logic behind it, it will cause problems with dependency management.

I'd like to specify the requirements as invokust>=0.7, this would not change as 0.61 is larger as pip interprets it.

HI, are you planning to publish 0.75 to Pypi?
Currently, it seems that only the 0.73 is published, the 0.73 is linked to an old version of locust.
Thank you

Running into an error AttributeError: 'StatsError' object has no attribute 'to_dict' when there are failures present in the load test. Coming from locust_stats = loadtest.stats() Any ideas? Thanks!

def lambda_handler(event, context=None):
    try:
        if event:
            settings = create_settings(**event)
        else:
            settings = create_settings(from_environment=True)

        loadtest = LocustLoadTest(settings)
        loadtest.run()
    except Exception as e:
        logging.error("Locust exception {0}".format(repr(e)))

    else:
        locust_stats = loadtest.stats()
        lambda_runtime_info = get_lambda_runtime_info(context)
        loadtest_results = locust_stats.copy()
        loadtest_results.update(lambda_runtime_info)
        json_results = json.dumps(loadtest_results)

        logging.info(json_results)
        return json_results

is there any option to save the generated report in s3 or somewhere else?

Is there a way to configure what is logged/stdout?

It now runs with locust 0.9, but locust 0.11 is already out and it does not work with it.
Is it possible to bring it to support locust 0.11?

Can you run a LoadTestShape class in lambda? Specifically a case for ramping up and ramping down users at different stages?

Locust exception ImportError("cannot import name 'User'",)
END RequestId: 0fec6b92-8c39-42a7-8c14-16629a464c67

NOTE: when i used the import with HttpUser, it gives the same error.
I'm trying to use the invokust with AWS Lambda, but no luck so far, below are the steps that i followed,

• installing invokust in to the python target folder (sudo pip3 install invokust --target=python-packages)
• Created the locustfile and lambda_locust(handler).
• Created a zip with all 3 files (zip -q -r lambda_locust.zip lambda_locust.py princess_locustfile.py python-packages )
• Created the lambda
• and invoked from the console.

Hi, I followed the instructions under Running Locust on AWS Lambda but all I get is this error when I run invokr.py:

[2022-01-05 07:33:57,245] LM-C02C775WMD6R/ERROR/invokust.aws_lambda.lambda_load_test: error Unhandled: b'{"errorMessage":"RequestId: bb804730-df31-47d0-bc5f-ac734d0aa70e Error: Runtime exited with error: signal: segmentation fault (core dumped)","errorType":"Runtime.ExitError"}'

Any pointers what I might be doing wrong? I'm building the Lambda package on MacOS Big Sur using these commands:

docker run -it --volume=$PWD:/temp python:3.6 bash -c "pip install /temp --target=/temp/python-packages"
zip -q -r lambda_locust.zip lambda_locust.py locustfile_example.py python-packages
aws lambda create-function --function-name lambda_locust --timeout 300 --runtime python3.6 --role arn:aws:iam::XXXXXX --handler lambda_locust.handler --zip-file fileb://lambda_locust.zip

We can add in a feature which will help developers to manage invokust lambda through aws-cdk toolkit instead of relying upon local lambda bundle script

@events.init.add_listener
def on_locust_init(environment, **_kwargs):

invokust is a breakthrough.
But, I couldn't get add_listner to work.

Hi,

I'm testing invokust locally with localstack, and haven't deployed to AWS yet.
Everything works well, but I'm getting a negative execution time.

This is a test result; the key total_lambda_execution_time is assigned to this negative value: -582398. It should probably be 15.6032865047 (1663692383.5552292 - 1663692367.9519427), which is consistent with the duration I set for this test (15s).

{
  "summary_stats":{
    "lambda_invocation_count":1,
    "total_lambda_execution_time":-582398,
    "requests_total":1392,
    "request_fail_ratio":0.0,
    "invocation_error_ratio":0.0
  },
  "results":[
    {
      "requests":{
        "GET_/":{
          "request_type":"GET",
          "num_requests":1392,
          "min_response_time":716.8549480848014,
          "median_response_time":2300.0,
          "avg_response_time":3316.3983788552537,
          "max_response_time":14563.939598971047,
          "response_times":{...},
          "response_time_percentiles":{...},
          "total_rps":93.70657030509588,
          "total_rpm":5622.3942183057525
        }
      },
      "failures":{},
      "num_requests":1392,
      "num_requests_fail":0,
      "start_time":1663692367.9519427,
      "end_time":1663692383.5552292,
      "remaining_time":882398,
      "function_name":"locust-function",
      "function_version":"$LATEST",
      "invoked_function_arn":"arn:aws:lambda:us-east-2:000000000000:function:locust-function",
      "memory_limit":"1536",
      "aws_request_id":"2483a0c5-3c47-150a-6b27-c9535e8eeacb",
      "log_group_name":"/aws/lambda/locust-function",
      "log_stream_name":"2022/09/20/[$LATEST]4aff2d740c665b04f425e1c3239ea985"
    }
  ]
}

This is the code I'm using to launch the test:

        load_test: LambdaLoadTest = LambdaLoadTest(
            lambda_function_name=locust_lambda_name,
            threads=threads,
            ramp_time=0,
            time_limit=duration,
            lambda_payload={
                "locust_modules": modules,
                "host": host,
                "num_users": users,
                "spawn_rate": spawn_rate,
                "run_time": f"{duration}s",
            },
        )
        load_test.run()
        result: TestResult = TestResult(
            summary_stats=load_test.get_summary_stats(),
            results=load_test.get_locust_results(),
        )

This is latest release, the pypi 0.44 is pinned to locustio 0.8a2?

Running py3.8, I was able to create a lambda not not invoke it

Getting usage errrors locally

`
aws lambda invoke --function-name lambda_locust --invocation-type RequestResponse --payload '{"locustfile": "my_locustfile.py", "host":"https://miamioh-gradify.com", "num_requests":"20", "num_clients": "1", "hatch_rate": "1", "run_time":"3m"}' output.txt

Unknown options: num_requests:20,, num_clients:, 1,, hatch_rate:, 1,, run_time:3m}', output.txt, host:https://miamioh-gradify.com,`

Same via aws-shell

Tried creating test in Lambda Console.

{
"locustfile": "my_locustfile.py",
"host": "https://miamioh-gradify.com",
"num_requests": "20",
"num_clients": "1",
"hatch_rate": "1",
"run_time": "3m"
}

but got
{
"errorMessage": "Unable to import module 'lambda_locust': No module named 'gevent.__hub_local'",
"errorType": "Runtime.ImportModuleError"
}

Hi,

I created a lambda python function with a lambda runtime environment set to python 3.6. I made a python 3.6 layer where I installed invokust and added it to the lambda function.

The code in my lambda function is:

import json
from invokust.aws_lambda import get_lambda_runtime_info

def lambda_handler(event, context):
# TODO implement
return {
'statusCode': 200,
'body': json.dumps('Hello from Lambda!')
}

When I execute it, I get the following error:

Response
{
"errorMessage": "module initialization error"
}

If I remove the invokust import, everything works fine. Does anybody have an idea how to fix this?

The command used to create the zip package for the lambda layer is: pip3 install --platform manylinux2014_x86_64 --target=python --implementation cp --python 3.6 --only-binary=:all: --upgrade invokust

After that, I zipped the python folder and uploaded it as a lambda layer.

Hi,

I'm working my way through the instructions in the README for running the sample code on AWS Lambda. I've worked my way through a variety of errors, but have bumped up on one that has me stymied. I'm hoping you can help.

I'm working on a MacBook Pro, running Catalina (10.15.7), so I'm using docker to compile the python package, as described in the README:

docker run -it --volume=$PWD:/temp python:3.6 bash -c "pip install /temp --target=/temp/python-packages"

Then I'm zipping and creating the lambda, no problem.

zip -q -r lambda_locust.zip lambda_locust.py locustfile_example.py python-packages

aws --profile sandbox lambda create-function --function-name lambda_locust --timeout 300 --memory-size 256 --runtime python3.6 --role arn:aws:iam::039069484348:role/service-role/sdg-test-function-role-54kv5be3 --handler lambda_locust.handler --zip-file fileb://lambda_locust.zip

As an aside, you have a note in your TODO.md about Process exited before completing request errors. I was getting that too, but I fixed it by doubling the memory size. Also, that role in my create-function command is a basic lambda execution role.

Finally, invoking the lambda gets me this:

(invokust) SGALLIVER-MBP:invokust sgalliver$ aws --profile sandbox lambda invoke --function-name lambda_locust --invocation-type RequestResponse --payload '{"locustfile": "locustfile_example.py", "host":"http://www.google.com", "num_users": "1", "spawn_rate": "1", "run_time":"1m"}' output.txt --log-type Tail --query 'LogResult' --output text | base64 -d
START RequestId: c8d8bbbf-4e5c-4f20-bd56-f00c3bf000eb Version: $LATEST
[2021-01-13 17:47:35,008] 169.254.244.125/ERROR/root: Locust exception ImportError("/lib64/libc.so.6: version `GLIBC_2.28' not found (required by python-packages/gevent/libuv/_corecffi.abi3.so)",)
END RequestId: c8d8bbbf-4e5c-4f20-bd56-f00c3bf000eb
REPORT RequestId: c8d8bbbf-4e5c-4f20-bd56-f00c3bf000eb	Duration: 3.73 ms	Billed Duration: 4 ms	Memory Size: 256 MB	Max Memory Used: 121 MB	Init Duration: 1272.77 ms

Any clues about that import error? I did some poking around and it appears that gevent needs to be compiled on an Amazon Linux instance. I suppose I could go spin up an EC2 instance, but I figured the README would have mentioned it if that were necessary. So I thought I'd ask if you have hit this error and how you solved it, before I go running down the EC2 rabbithole.

Many thanks!

Hey Guys,
I have a question regarding your Locust-wrapper. Is it possible to load test Odoo applications over jsonrpc ?
I currently have the problem that my Locust tests are running but not returning any test results on my CLI. If someone knows something about this just hit me up!

Hi, I am trying to run the invokust file locally, but I encounter the module 'invokust' has no attribute 'create_settings' error on the settings = invokust.create_settings( line.
I am using the second example in the readme against my own endpoints. Am I not able to run this locally on a windows machine?

import invokust

from locust import HttpLocust, TaskSet, task

class Task(TaskSet):
    @task()
    def get_home_page(self):
        '''
        Gets /
        '''
        self.client.get("/")

class WebsiteUser(HttpLocust):
    task_set = Task

settings = invokust.create_settings(
    classes=[WebsiteUser],
    host='http://example.com',
    num_requests=10,
    num_clients=1,
    hatch_rate=1
    )

loadtest = invokust.LocustLoadTest(settings)
loadtest.run()
loadtest.stats()

I have a following directory structure.

myapp.py is a simple flask application running locally
locustfile.py is my locust file
run_locust.py is where i have put the invokust code.

All 3 files attached.
Files.zip

When i run run_locust.py like this:

python run_locust.py, I keep getting this error

/Users/kapilmathur/virtual_envs/audit_trail_auto/bin/python3.8 /Users/kapilmathur/repos/Load-tools/Load-test/run_locust.py
Traceback (most recent call last):
  File "/Users/kapilmathur/repos/Load-tools/Load-test/run_locust.py", line 8, in <module>
    settings = create_settings(
  File "/Users/kapilmathur/virtual_envs/audit_trail_auto/lib/python3.8/site-packages/invokust/settings.py", line 68, in create_settings
    docstring, classes = load_locustfile(settings.locustfile)
ValueError: too many values to unpack (expected 2)

Process finished with exit code 1

Can you plz help with the issue and tell me if i am using it correctly. is there some version dependency/conflict I am hitting ?

I get this error believe whenever I try to run on zipped dependencies on AWS.

{
"errorMessage": "Unable to import module 'lambda_locust': No module named 'gevent.__hub_local'",
"errorType": "Runtime.ImportModuleError"
}

Packaging code with with in Makefile:
PYTHON_VERSION=3.7
ZIP_PATH=$(abspath ./build.zip)

zip:
$(eval VENV=$(shell pipenv --venv))
cd ${VENV}/lib/python${PYTHON_VERSION}/site-packages/ && zip -r ${ZIP_FILE} .
zip -r ${ZIP_PATH} . -x .git

I've gotten my handler paired down to only printing the event, but it fails at importing dependencies:

import sys
import os
sys.path.insert(0, "python-packages")
import logging
import json

from invokust.aws_lambda import get_lambda_runtime_info
from invokust import LocustLoadTest, create_settings

logging.basicConfig(level=logging.INFO)

def handler(event=None, context=None):
print(str(event))
# try:
# if event:
# settings = create_settings(**event)
# else:
# settings = create_settings(from_environment=True)
#
# loadtest = LocustLoadTest(settings)
# loadtest.run()
#
# except Exception as e:
# logging.error("Locust exception {0}".format(repr(e)))
#
# else:
# locust_stats = loadtest.stats()
# lambda_runtime_info = get_lambda_runtime_info(context)
# loadtest_results = locust_stats.copy()
# loadtest_results.update(lambda_runtime_info)
# json_results = json.dumps(loadtest_results)
#
# logging.info(json_results)
# return json_results

We've just successfully run our first load test with invokust.
Just wondering if there is a way in the config to output CSV or locust HTML reports?

Hi.

I really like the approach you've taken here.
Would you be willing to change the license to a license that is compatible with locustio?

As it stands it's not possible to use this code as the basis of something that I can submit as a PR to locust with.

Thanks

Running the default example results in
Unable to import module 'lambda_locust': No module named 'greenlet'

I have the standard example files (host changed) and have ran
pip3 install invokust --target=python-packages
and zipped them all together

I've included a tree of the files it downloads which end up in the zip.

file.txt

Any idea on how to solve this?