Repository navigation
fix(minimal): T42 — Rename on a native host prints no VM-host warning - #3333
mitodrummer merged 1 commit into
Conversation
move_held_box_name withdrew the old host row even with no control socket, so min box rename on a native host logged a warning about the VM host daemon. With no control socket it now asks for nothing. A VM host still withdraws the old row and moves the name hold.
|
Important Review skippedBot user detected. To trigger a single review, invoke the
|
|
Reviewed: 1 finding, none blocking at Address the comments on the diff. Clobber: none seen. |
There was a problem hiding this comment.
arch-review: PASS · the full ruling
The change conforms to the design. architecture.md line 487 says that a rename changes only the alias, and the rename path is unchanged. The early return in move_held_box_name removes a spurious warning on a native host. A native host has no VM host daemon, and the sibling helpers already skip silently without a control socket. Note: vm_host_control_sock also returns None on a VM host when the socket does not resolve, so the new doc comment is narrower than the code. The design states no rule for that log, so it is not a condition.
Ruled at 34d6d62 against gominimal/arch@15b603e6afb8 by claude-opus-5-5.
|
Round 1 of 10 on the runtime's review findings: the coder changed nothing. Revise round 1: no code change. The one review thread is invalid, and the reason is in its entry. files_changed lists the files from the earlier commit on this branch. Review threads the coder judged:
|
T42 — Rename on a native host prints no VM-host warning
slice: S7
kind: task
covers: BCLI-058
size: S
est: 15 code, 45 tests
executor: agent
difficulty: routine
model: standard
diagnostics: none new; the spurious warning goes away
observability: none new
layers:
- client:
- crates/minimal/src/cmd/boxes/vm_host.rs:
move_held_box_namewithdraws the old host row only where a VM host daemon holds one: with no control socket (a native host) it skips the withdrawal and prints nothing- crates/minimal/src/tests.rs: the in-process minimald test harness gains this layer's tests
test: cargo nextest run -p minimal box_rename_native_host_prints_no_vm_host_warning
produces: none
consumes: none
proof:
min box rename <box> <new>on a native host renames the box and prints no warning; on a VM host it still withdraws the old host row.Proof
min box rename <box> <new>on a native host renames the box and prints no warning about the VM host daemon. On a VM host, the rename still withdraws the old host row by the box id and moves the name hold.Notes from the agent: The warning came from minimal_client::box_registration::withdraw_box_row. move_held_box_name called it with no control socket and with the box id. Now move_held_box_name returns early when there is no control socket. The hold and release calls after it were already silent without a socket. The new test failed before the fix, with that exact warning in the log. The existing VM-host test a_rename_retires_nothing_and_a_stop_retires_as_stopped still passes, and it shows the withdraw, release and hold requests. cargo clippy -p minimal --all-targets and cargo fmt --check reported nothing. I did not run just ci.
Tests
cargo nextest run -p minimal box_rename_native_host_prints_no_vm_host_warningVerify
min task run agent-verifyexited 100, so this PR is a draft.Closes #3319