Fails to record results for remote endpoints

[s1na]

Version
v0.3.1

Platform
Darwin Mac-mini 22.6.0

Description

I have two remote RPC endpoints. I have installed flood and vegeta both locally and on those machines. When I try the following command:

flood eth_getBlockByNumber geth=ubuntu@IP1:localhost:8545 reth=ubuntu@IP2:localhost:8545

it proceeds to execute the tests, however at the end when it wants to record the results I get this exception:

Traceback (most recent call last):
  File "/usr/local/Cellar/[email protected]/3.11.6/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/process.py", line 314, in _bootstrap
    self.run()
  File "/usr/local/Cellar/[email protected]/3.11.6/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/Users/sina/Library/Python/3.11/lib/python/site-packages/flood/tests/load_tests/load_test_runs.py", line 179, in run_load_test
    result = _run_load_test_remotely(
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sina/Library/Python/3.11/lib/python/site-packages/flood/tests/load_tests/load_test_runs.py", line 380, in _run_load_test_remotely
    subprocess.check_output(cmd.split(' '), stderr=subprocess.DEVNULL)
  File "/usr/local/Cellar/[email protected]/3.11.6/Frameworks/Python.framework/Versions/3.11/lib/python3.11/subprocess.py", line 466, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/Cellar/[email protected]/3.11.6/Frameworks/Python.framework/Versions/3.11/lib/python3.11/subprocess.py", line 571, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ssh', 'ubuntu@IP1', 'bash', '-c', "'source", '~/.profile;', 'python3', '-m', 'flood', '/tmp/flood__09f2f3ce-07d0-4b3d-b8af-597cf415626a', 'reth=http://localhost:8545', '--output', '/tmp/flood__09f2f3ce-07d0-4b3d-b8af-597cf415626a', '--no-figures', "'"]' returned non-zero exit status 1.

(I get one such exception from each endpoint)

[kamuik16]

Happens on Ubuntu 24.04 as well.

[sslivkoff]

are you using the same exact version of flood on each machine? try installing latest flood commit on each

[s1na]

As per @mattsse suggestion I tried running flood locally on those machines and it was failing with this error:

lxml.html.clean module is now a separate project lxml_html_clean.
Install lxml[html_clean] or lxml_html_clean directly.

This was fixed by pip install lxml_html_clean. However still the crash persists when I try to do the comparison benchmark.

For reference they are both running flood 0.3.1.

[s1na]

To shine more light on this. What happens is the remote server command sort-of fails (it produces generates results) but for some reason I'm not aware prints get_border_chars() got an unexpected keyword argument 'indent' and then exits with status 1. Which makes the ssh command fail and panic. So the results are not retrieved to be processed locally.

I worked around it for now by silencing the panic:

diff --git a/flood/tests/load_tests/load_test_runs.py b/flood/tests/load_tests/load_test_runs.py
index 0bd1d09..1e497f4 100644
--- a/flood/tests/load_tests/load_test_runs.py
+++ b/flood/tests/load_tests/load_test_runs.py
@@ -377,7 +377,12 @@ def _run_load_test_remotely(
         extra_kwargs=extra_kwargs.lstrip(),
     )
     cmd = cmd.strip()
-    subprocess.check_output(cmd.split(' '), stderr=subprocess.DEVNULL)
+    try:
+        subprocess.check_output(cmd.split(' '), stderr=subprocess.DEVNULL)
+    except subprocess.CalledProcessError as cpe_error:
+        print(f"CalledProcessError: {cpe_error}")
+        decoded_output = cpe_error.output.decode('utf-8', errors='replace')
+        print(f"Output (stdout + stderr): {decoded_output}")

     # retrieve benchmark results
     if verbose: