BUG: moved-holder oracle regression fixture drops delegation indexes #641
Labels
No labels
bug
cli
core
docs
event
experiment
figure
invariant
metrics
oracle
phase-0
phase-1
phase-2
phase-3
phase-4
phase-5
phase-6
provenance
revocation
tests
workload
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
erikinkinen/AES#641
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Problem
The independent moved-state fixture added for #638 reconstructs EdgeStore through TestAccess::add_edge. That helper rebuilds holder/object adjacency, but does not rebuild delegation indexes. Moving a root with a surviving child therefore silently changes the topology consulted by eager resolvers. The production move primitive preserves those indexes; this is a regression-fixture defect, not an observed production movement failure.
At main
6f395749f9dae392173d04153f43c71a824b90ce, adding before/after delegation-index equality assertions makes aes_moved_holder_oracle_tests fail withmovement fixture erased delegation indexes(CTest exit 8). Logs and JUnit are preserved under/tmp/aes-phase2-execution/moved-oracle-fixture-indexes.Correction
Now that P2-08-A is merged, use the actual production movement primitive in this fixture. Assert preservation of incoming/delegation indexes and validity markers. Add hand-derived expected residue for all seven withdrawal mechanisms so agreement between the oracle and actual UseCap alone cannot conceal a corrupted resolver topology. Retain binding failures, birth/current-role separation, empty destinations and frozen-cohort checks.
Acceptance
Discovered during #98 snapshot-restoration review. Phase 2, milestone 3. No scientific data has been collected; prior #638 production-oracle correction remains applicable.