Codex CLI hangs indefinitely on prompt (no response) on Fedora with Node 22 – affects `codex` and `codex exec`
What version of Codex CLI is running?
codex-cli 0.112.0
What subscription do you have?
Plus
Which model were you using?
gpt-5.4
What platform is your computer?
Linux 6.18.5-200.fc43.x86_64 x86_64
What terminal emulator and version are you using (if applicable)?
Kitty
What issue are you seeing?
Codex CLI hangs indefinitely after accepting a prompt.
The prompt appears in the session output, but no response ever starts streaming and the CLI never finishes.
This happens both in:
interactive mode (codex)
non-interactive mode (codex exec)
Example command:
codex exec "hello"
Output:
OpenAI Codex v0.112.0 (research preview)
-------
workdir:
model: gpt-5.4
provider: openai
approval: never
sandbox: workspace-write [workdir, /tmp, $TMPDIR, *]
reasoning effort: medium
reasoning summaries: none
session id: 019cd2fb-b2a4-70a3-8390-22b8aba4ca37
-------
user
hello
After this point nothing else happens.
No response starts streaming and the CLI hangs indefinitely.
Additional observations:
Ctrl+C does not terminate the process.
I must kill the process or close the terminal.
The issue occurs even with a clean shell environment.
Example clean environment reproduction:
env -i HOME="$HOME" PATH="$PATH" TERM="$TERM" bash --noprofile --norc
codex exec "hello"
The behavior is the same.
What steps can reproduce the bug?
Install Codex CLI
npm install -g @openai/codex
Run a simple command:
codex exec "hello"
The CLI prints session metadata and the prompt:
user
hello
The process hangs indefinitely and no model response is produced.
This also occurs when running interactive mode:
codex
and submitting a prompt like:
hello
The prompt is displayed but no response is ever generated.
What is the expected behavior?
The CLI should send the request to the model and stream the response.
For example, running:
codex exec "hello"
should produce a response such as:
Hello! How can I help?
instead of hanging indefinitely.
Additional information
Environment:
OS: Fedora Linux
Terminal: kitty
Node: v22.20.0
npm: 10.9.3
Codex CLI version: 0.112.0
TERM: xterm-kitty
Notes:
The issue also occurred with Codex CLI v0.110.0.
7 Comments
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
I am currently seeing the same issue: https://github.com/openai/codex/issues/14084
and this shows others with the issue: https://www.reddit.com/r/codex/comments/1roxj50/codex_down_5min_wait_time_between_tool_calls/
+1 here, same behaviour reported both on my personal account and company enterprise account.
same for me
Same here... Rocky Linux 10.1... but After putting my system to sleep and waking it up again... causes all the text to appear at once
We're investigating the issue. I'm going to close this in favor of #14048. I'll post updates in that thread as they become available.