diff --git a/03-DESIGN/01-to-be/47-milestones/G-22-lead-answers-2.md b/03-DESIGN/01-to-be/47-milestones/G-22-lead-answers-2.md index 084a4781..aef01a10 100644 --- a/03-DESIGN/01-to-be/47-milestones/G-22-lead-answers-2.md +++ b/03-DESIGN/01-to-be/47-milestones/G-22-lead-answers-2.md @@ -1845,3 +1845,9 @@ the principle that settles it, cited as "design 47 G A-n". `validate-tries` tries in a row ends could-not-validate with its words, so a request that keeps moving is loud. Could-not-validate keeps ยง713's other causes (the step could not run, the program exited other than 0, the answer not both cases), asked again by a later merge or `indexer_validate_again`. +- **A-461d** (A-461b; principles 7 and 2) A module placed on another node and not heard running there counts as one + whose running build has no entry: the composer lays it from the build its placement names, or, when the placement + names none, from its trunk build through the seat `mesh-catalog`'s verb `module`, exactly as A-461b says. Being heard + running is never a condition for laying another node's module; the assignment is the mesh's truth of what is placed + there. (From the develop of the store fetchers' fix: shanks heard no `running` of `distribution` on novox, so it laid + the store's holder name-only and found no store address.)