Repository navigation
T13 — Mint every box's id in the VM host and take it only from there #2397
Description
Task
| slice | size | difficulty | model | who | estimated lines |
|---|---|---|---|---|---|
| Names follow the id | M | unknown ground | strongest | an agent | 560 (320 code, 240 tests) |
A box in the VM carries the VM host's minted id.
Covers:
- BOX-002
Layers:
- vm-host:
- crates/minvmd/src/control.rs: passes the minted id
- crates/minvmd/tests/minimald_session_integration.rs: tests
cargo nextest run -p minvmd vm_host_mints_box_id_before_in_vm_create(deferred: vm)
- guest:
- crates/minimald/src/box_manager.rs: id only from VM host
- crates/minimald/src/box_manager/tests.rs: tests
- crates/minimald/src/env.rs: in-box id from record
- crates/minimald/src/net/listeners.rs: carry the minted id over the guest↔host report channel
- crates/minimald-rpc/src/lib.rs: the message or reply field that carries the id
- crates/minimald/src/box_actor/tests.rs: fixtures (
own_ip_handed_session_req) hand the id over the VM host's channel, notconfig.box_id cargo nextest run -p minimald guest_create_ignores_a_client_supplied_box_id
Widened by the maintainer (2026-10-30): the first run found no channel for the in-VM daemon to receive the id the VM host mints. The guest's report channel (net/listeners.rs) and the wire types (minimald-rpc) are now in the guest layer, so the id can travel host→guest at create without the client relaying it.
Split by the maintainer (2026-10-30): the second run found the client relay in create.rs and dash/rpc.rs, and the whole fix needed 11 layer files, above the 10-file cap. The client layer moved to its own task, which waits on this one; the box_actor test fixtures joined the guest layer. Until it lands the client still sends an id, which the guest now ignores.
Waits on:
Plan block
### T13 — Mint every box's id in the VM host and take it only from there
slice: S3
kind: task
covers: BOX-002
size: M
est: 320 code, 240 tests
executor: agent
difficulty: unknown ground
model: strongest
diagnostics: guest's daemon log names passed id
observability: guest's create span carries minted id
layers:
- vm-host:
- crates/minvmd/src/control.rs: passes the minted id
- crates/minvmd/tests/minimald_session_integration.rs: tests
test (deferred: vm): cargo nextest run -p minvmd vm_host_mints_box_id_before_in_vm_create
- guest:
- crates/minimald/src/box_manager.rs: id only from VM host
- crates/minimald/src/box_manager/tests.rs: tests
- crates/minimald/src/env.rs: in-box id from record
- crates/minimald/src/net/listeners.rs: carry the minted id over the guest<->host report channel
- crates/minimald-rpc/src/lib.rs: the message or reply field that carries the id
- crates/minimald/src/box_actor/tests.rs: fixtures hand the id over the VM host's channel
test: cargo nextest run -p minimald guest_create_ignores_a_client_supplied_box_id
produces: none
consumes: none
proof: A box in the VM carries the VM host's minted id.
depends on:
- blocked by #2394
Reacted by minimal-foundry
Activity
fix(minvmd): spell the ask audit box id as 32 hex
mitodrummer commentedon Oct 21, 2026
T45 (#2485) makes
BoxIdthe one type for both ids, but on a VM guest the in-VM store still mintsRecord::idinside the VM. Its docs name this issue as the owner of that interim. The interim ends whenDiskLoader::createkeys a record by the handedconfig.box_idinstead of minting one. Two follow-ups from the arch review of #2485 belong here too: give thebox_actortest fixtures distinct ids in place of the shared[0x42; 16], and give the duplicate-id refusal its own message rather than "A session with that name already exists".refactor(boxes): make the box id the one key for every record (gomini…
minimal-foundry commentedon Oct 30, 2026
Done. #2759 merged.
branch
agent/2397-mint-every-box-s-id-in-the-vm-host-and-take-it-o· epic #2236 · PR #2759f340a1f0827a0596d3123203905280e8Updated 2026-10-30T20:47:37.908Z.
mitodrummer commentedon Oct 30, 2026
/dispatch