Crewai is not running properly

CrewAI Community Support
1 
.venv) paarttipaa@Paarttipaabhalajis-MacBook-Pro trail % crewai run
Running the Crew
warning: `VIRTUAL_ENV=/Users/paarttipaa/ProjectTask/GithubProj/agent-kit/tutorials/crewai/.venv` does not match the project environment path `.venv` and will be ignored; use `--active` to target the active environment instead
Using CPython 3.12.6 interpreter at: /Library/Frameworks/Python.framework/Versions/3.12/bin/python3
Creating virtual environment at: .venv
Built trail @ file:///Users/paarttipaa/ProjectTask/GithubProj/agent-kit/tutorials/crewai/trail
Installed 200 packages in 484ms
╭────────────────────────────────────────────────────────────────────────── Crew Execution Started ───────────────────────────────────────────────────────────────────────────╮
│ │
│ Crew Execution Started │
│ Name: crew │
│ ID: f783d384-4d96-490c-9cdc-1d496f8803a9 │
│ │
│ │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯

# Agent: AI LLMs Senior Data Researcher
## Task: Conduct a thorough research about AI LLMs Make sure you find any interesting and relevant information given the current year is 2025.

:rocket: Crew: crew
└── :clipboard: Task: a0508086-2bac-43f7-a62c-c420f39dd881
Status: Executing Task…
└── :brain: Thinking…

still is keep on thinking. crewai is not properly executing. in ipynb its runing fine. but in cli its not running properly.

Steps to follow:

  1. install the crewai and create a crew using the command.
  2. run the crew using crewai run command.
2

waiting for around half an hour . it not running.

3

Had the same problem with crewai version 0.121, rolled back to 0.120 and it works again.

4

You guys are absolutely right. There’s a bug that’s just freezing up the console. Right when you see that 🧠 Thinking... message, what’s actually happening is an OpenAIError exception popping up because the OPENAI_API_KEY is missing. The problem is, this exception isn’t being shown since the console is frozen, and the process can’t move forward either.

I’m digging into it now to gather more details for a report.

@tonykipkemboi, this is a critical bug I’ve been noticing in the new version. It now seems like whenever litellm.completion raises an error, the console fails to update and the code just hangs indefinitely.

1 Like
5

Recently, I have to fight against the following, when using o4-mini:

OpenAIException - Invalid prompt: your prompt was flagged as potentially violating our usage policy. Please try again with a different prompt.

Could that also cause 0.121 to freeze? Because, I am very sure, that my OPENAI_API_KEY has been correctly set.

6

While the console formatting bug is getting fixed, you should just roll back to the previous version:

cd /path/to/your/project
uv add "crewai<0.121.0"
crewai run
7

thanks for flagging. will look into this.

8

could you share your code so we can replicate? it seems it’s happening inconsistently at the moment.

9

Looking into it!
Thanks for reporting, sorry about the furstation

1 Like
10

Update: Fixed in commit 4e0ce9a.

1 Like
11

great catch, fix is in main already, new version will be out tomorrow

2 Likes
12

This topic was automatically closed 24 hours after the last reply. New replies are no longer allowed.