[BUG] Out of range - command line dies
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
I am in the process of conducting a test review; it is a larger codebase, and I have over 600 tests, in various test stages, that have been or need to be tested to pass the CI/CD. When I ask Claude to help monitor the test results and generate a statistic, it stops working after a while with the following error:
RangeError: Out of memory
at join (unknown)
at hPG (/$bunfs/root/claude:1657:6270)
at O (/$bunfs/root/claude:1657:4516)
at Z (/$bunfs/root/claude:1657:4771)
at Ai5 (/$bunfs/root/claude:1667:1544)
at <anonymous> (/$bunfs/root/claude:1668:781)
at useMemo (/$bunfs/root/claude:229:25924)
at Qz (/$bunfs/root/claude:1667:324)
at AD (/$bunfs/root/claude:229:19415)
at Ag (/$bunfs/root/claude:231:42805)
ERROR Out of memory
- at join (unknown)
- hPG (/$bunfs/root/claude:1657:6270)
- O (/$bunfs/root/claude:1657:4516)
- Z (/$bunfs/root/claude:1657:4771)
- Ai5 (/$bunfs/root/claude:1667:1544)
- <anonymous> (/$bunfs/root/claude:1668:781)
- useMemo (/$bunfs/root/claude:229:25924)
- Qz (/$bunfs/root/claude:1667:324)
- AD (/$bunfs/root/claude:229:19415)
- Ag (/$bunfs/root/claude:231:42805)
I work on a Mac and install Claude Code with "brew". It is a Django code base, all with the newest version. I neither use Redis, PG, nor other services during the tests. If I run them alone, all pass without issue. If I use a Python script and the API, with a step-by-step approach (one test at a time), it might work, but I haven't tested it yet. It is, first of all, a statistics page I generate, and I don't want to write a script for that reason (my initial thought).
When a task is completed successfully or a test passes, the memory/context window is not released as it should be. Is it possible to instruct the command-line version of Claude accordingly?
What else would help as information?
What Should Happen?
It should not stop working.
Error Messages/Logs
RangeError: Out of memory
at join (unknown)
at hPG (/$bunfs/root/claude:1657:6270)
at O (/$bunfs/root/claude:1657:4516)
at Z (/$bunfs/root/claude:1657:4771)
at Ai5 (/$bunfs/root/claude:1667:1544)
at <anonymous> (/$bunfs/root/claude:1668:781)
at useMemo (/$bunfs/root/claude:229:25924)
at Qz (/$bunfs/root/claude:1667:324)
at AD (/$bunfs/root/claude:229:19415)
at Ag (/$bunfs/root/claude:231:42805)
ERROR Out of memory
- at join (unknown)
- hPG (/$bunfs/root/claude:1657:6270)
- O (/$bunfs/root/claude:1657:4516)
- Z (/$bunfs/root/claude:1657:4771)
- Ai5 (/$bunfs/root/claude:1667:1544)
- <anonymous> (/$bunfs/root/claude:1668:781)
- useMemo (/$bunfs/root/claude:229:25924)
- Qz (/$bunfs/root/claude:1667:324)
- AD (/$bunfs/root/claude:229:19415)
- Ag (/$bunfs/root/claude:231:42805)
Steps to Reproduce
You would need my code base, and then run pytest with Claude monitoring 600+ tests. Sorry that I can not help more for that reason. I updated from version 1.100 to 1.0.108 and will monitor if the behavior disappears; if it does, I will close the issue before Monday morning.
Claude Model
Sonnet (default)
Is this a regression?
I don't know
Last Working Version
1.0.100
Claude Code Version
1.0.100 (Claude Code)
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗