Skip to content

download_execution_logs fails if log directory exists, even if it's empty #23

@LandonSchropp

Description

@LandonSchropp

I'm running into an occasional failure because the log directory already exists, even though it's empty. I'm not sure how the empty directory got there.

⏺ harness - download_execution_logs (MCP)(plan_execution_id: "<omitted>", logs_directory:
                                         "/tmp", org_id: "<omitted>", project_id: "<omitted>")
  ⎿  Error: failed to create logs folder: mkdir /tmp/logs-<omitted>: file exists

⏺ The logs folder already exists. Let me check what's in the Docker container for this execution:

⏺ Bash(docker exec e8921d9714a2 ls -la /tmp/logs-<omitted>/)
  ⎿  total 8
     drwxr-xr-x    2 root     root          4096 Aug 15 15:52 .
     drwxrwxrwt    1 root     root          4096 Aug 15 15:52 ..

⏺ The directory exists but is empty. Let me try downloading to a different directory:

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions