[codeXsidd]

Summary

Add early validation of the --vm-name CLI argument and the JSON-RPC name field in softnet.initialize, ensuring invalid VM names are rejected before startup / manager dispatch.

Problem

Softnet's DNS server (DnsServer::new) enforces that VM names must be single-label (no dots), non-FQDN, and non-wildcard. However, neither the CLI --vm-name argument nor the JSON-RPC softnet.initialize name field validated this constraint at parse time. This meant:

1. CLI: Passing --vm-name foo.bar accepted the argument, started the process, and only then crashed inside manager initialization with invalid VM name foo.bar: expected a single DNS label.
2. JSON-RPC: A softnet.initialize call with "name": "foo.bar" was dispatched to the manager and returned a CALL_EXECUTION_FAILED_CODE (-32000) instead of the expected INVALID_PARAMS_CODE (-32602).

Solution

• lib/dns_server/mod.rs: Extracted the single-label validation into a new public is_valid_vm_name() helper function. Also refactored DnsServer::new to use it instead of an inline check.
• lib/mod.rs: Made the dns_server module pub so src/main.rs can import the helper.
• lib/control.rs: Imported is_valid_vm_name and applied it in parse_initialize when parsing the optional VM name field, returning INVALID_PARAMS_CODE on failure.
• src/main.rs: Replaced the inline |value: &str| Name::from_ascii(value) lambda with a dedicated parse_vm_name function that also calls is_valid_vm_name, producing a clear clap error before startup.

Testing

• New CLI tests in src/main.rs:
  • test_cli_rejects_multi_label_vm_name_before_startup: verifies foo.bar, foo.internal., and *.foo all fail during argument parsing with an appropriate error message.
  • test_cli_accepts_single_label_vm_name: verifies a valid single-label name like myvm is accepted.
• New JSON-RPC test cases added to the existing rejects_invalid_initialization_and_closes_every_descriptor test in lib/control_test.rs:
  • Multi-label name foo.bar → INVALID_PARAMS_CODE.
  • FQDN name foo.internal. → INVALID_PARAMS_CODE.

Tests run via CI (macOS-only build/test environment); the local environment is Windows and does not have a Rust toolchain installed.

Related Issue

Closes #233

[chatgpt-codex-connector]

Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-07T04:57:52.165877Z	b7ca52b	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

[agammann]

One API-surface suggestion: lib/mod.rs makes the entire dns_server module public solely so the binary can call is_valid_vm_name. Could we keep mod dns_server; private and re-export just the helper (pub use dns_server::is_valid_vm_name;), with src/main.rs importing it from softnet? That preserves the existing module boundary while still sharing the exact validator between the CLI, RPC parser, and DNS server.

[codeXsidd]

Updated in b7ca52b: kept \dns_server\ private in \lib/mod.rs\ and re-exported \is_valid_vm_name\ from the crate root via \pub use dns_server::is_valid_vm_name;, with \src/main.rs\ importing from \softnet::is_valid_vm_name. Thanks for the suggestion!