segmentation fault (core dumped)

[NesterukGabriel]

Hello, after a while when running codebase-memory-mcp cli index_repository i got following log:

[1] 10790 segmentation fault (core dumped) codebase-memory-mcp cli index_repository

codebase-memory-mcp --version                                                              
codebase-memory-mcp 0.6.0

[DeusData]

Hey @NesterukGabriel, thanks for flagging this. Can you tell me on which OS you are running? Obviously this should not happen, sorry for this experience. Currently I am not available but will fix it after the 15th. But some more details would be good, maybe your local agent can try to generate a report

[NesterukGabriel]

I’m using WSL2 with Ubuntu 24.04 on Windows 11. I think the issue might be related to the files in the repository. I’m using LFS, and when the indexing process reaches a folder containing those large files, it crashes with a segmentation fault. It always happens at the same point in the process

[MaximGalkin-TomTom]

I think I encountered the same problem.

codebase-memory-mcp --version
codebase-memory-mcp 0.6.0

Running on Ubuntu 24.04. Error happening both with CLI and when run inside Claude Code. Crashing with:

Job 1, 'codebase-memory-mcp cli index_r…' terminated by signal SIGSEGV (Address boundary error)

CLI started with:

codebase-memory-mcp cli index_repository '{"repo_path": "private-repo-path-here"}'
level=info msg=mem.init budget_mb=32013 total_ram_mb=64026
level=info msg=pipeline.discover files=2413 elapsed_ms=42
level=info msg=pipeline.route path=full
level=info msg=pass.start pass=structure files=2413
level=info msg=pass.done pass=structure nodes=2676 edges=2499
level=info msg=pass.timing pass=structure elapsed_ms=5
level=info msg=pipeline.mode mode=parallel workers=16 files=2413
level=info msg=parallel.extract.start files=2413 workers=16
level=info msg=parallel.mem.budget total_mb=32013 per_worker_mb=2000

The repository is quite big (5-6M lines of code).