Error running remote compact task: timeout waiting for child process to exit

What version of Codex CLI is running?

0.120.0

What subscription do you have?

Pro

Which model were you using?

gpt-5.4

What platform is your computer?

Linux 6.17.0-20-generic x86_64 x86_64

What terminal emulator and version are you using (if applicable)?

Ghostty & t3code

What issue are you seeing?

Once thread is about to compact for its first time it gets errored. Can't run /compact, /feedback. Sometimes it goes through tho, but not very often.

<img width="1124" height="236" alt="Image" src="https://github.com/user-attachments/assets/8c95a8f8-e8f8-4b40-91cf-ac20d546eaab" />

What steps can reproduce the bug?

Fill the context to the top.

What is the expected behavior?

Compact.

Additional information

Codex said:
The failing turn shows a large Neon MCP doc fetch, not a ctx7 CLI/plugin call:

• The last big payload before compaction was server":"Neon","tool":"get_doc_resource" for docs/auth/production-checklist.md at /home/kox/.t3/userdata/logs/provider/aa59955c-ee49-4b6d-8a39-07ae0234c689.log:3006.
• Compaction started immediately after at /home/kox/.t3/userdata/logs/provider/aa59955c-ee49-4b6d-8a39-07ae0234c689.log:3007.
• Then Codex failed with compact_error=timeout waiting for child process to exit at /home/kox/.t3/userdata/logs/provider/aa59955c-

ee49-4b6d-8a39-07ae0234c689.log:3008.

So the mechanism is:

• Large tool output gets added to the thread’s model-visible history.
• Codex notices the thread is getting too large.
• Codex runs internal contextCompaction.
• That internal compaction worker times out.