Every pre-existing runtime surface is per-process. gitea_get_runtime_context
and gitea_assess_master_parity describe only the server answering the call,
and gitea_assess_mcp_namespace_health accepts process, probe_result and
registered_tools from the caller, so it cannot constrain the caller. The
control-plane sessions table records allocator task sessions, not server
processes. Five self-reports of one revision therefore never proved that five
processes exist, that no sixth exists, or that all five share one client
cohort.
Add gitea_assess_fleet_inventory, a strictly read-only capability that takes
no evidence parameters. It combines two sources that must agree:
- a control-plane runtime registry row each server writes about itself,
from the official entrypoint, immediately after the native transport bind;
- a process observation the answering server performs, never the caller.
A member is running only when both agree and the observed process predates
its registration, so configuration alone never counts and a recycled PID
cannot impersonate an exited server. Classification is a pure function of the
snapshot, so gitea-controller and gitea-reconciler return the same verdict.
Missing, duplicate, unexpected, stale and unregistered members are reported
in separate fields rather than collapsed into one error, because each needs a
different operator action. single_cohort is derived only from recorded cohort
identity and stays null when unknown, so matching revisions never establish a
cohort. Any unreadable registry, unavailable listing, unregistered process,
unknown cohort or unknown revision sets inventory_complete and
mutation_gate_satisfied false with a specific blocked_reason.
The capability performs no restart, reconnect, drain, lease mutation, issue
mutation or process termination, never terminates a duplicate, and never
manufactures restart evidence. The only signal sent is signal 0.
Also resolves the fleet namespace from the expected roster:
role_namespace_gate.infer_mcp_namespace recognises only author and reviewer
and echoes the profile name otherwise, which would have mislabelled the
controller, merger and reconciler members. Widening that shared helper is
controller role-metadata work owned by #950 and is deliberately not done here.
Schema v5 to v6 is additive and idempotent: one new table, no existing table,
tool signature or result field changed. Two control-plane tests that pinned
the schema version to the literal 5 now assert against SCHEMA_VERSION.
#950 (controller role metadata), #951 (restart receipts) and #952 (stale-lease
consistency) remain separate and untouched.
Tests: 118 new across tests/test_mcp_fleet_inventory.py,
tests/test_control_plane_db_server_runtimes.py and
tests/test_fleet_inventory_tool.py, covering every acceptance criterion
including the multi-LLM duplicate-server regression that motivated the issue.
Closes#949
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01NBaQKcRRKeKbZuhk72MhEf
Add a versioned, redacted, reconcile-on-boot session checkpoint store to
control_plane_db so a restart can recover session identity, stage, lease
ownership, and next valid action instead of forcing human reconstruction
(umbrella #655; #628 autonomous-handoff goal).
- Bump SCHEMA_VERSION 4->5. New `session_checkpoints` table + two indexes,
added via `CREATE TABLE IF NOT EXISTS` so table creation is itself the
additive, idempotent v4->v5 migration (dependency_edges precedent).
- Writer `write_session_checkpoint` upserts the current recoverable state
per (remote, org, repo, session_id, work_kind, work_number); stage
transitions audit to `events`. Readers `get_session_checkpoint` /
`list_session_checkpoints`.
- Every free-text and JSON field is passed through `gitea_audit.redact`
before storage — no tokens/credential URLs can land in a checkpoint (AC4).
- `reconcile_session_checkpoint` is pure and never restores: it diagnoses a
stored checkpoint against live head/lease state and flags staleness (AC3).
- Drain gate: `require_complete=True` fails closed (writes nothing) when a
checkpoint is missing a drain-required field, so drain cannot complete on
an unrecoverable record.
- `lease_id`/`assignment_id` are soft references (no enforced FK) so a
checkpoint survives deletion of the lease it names; `work_number=0` is the
NULL-safe session-level sentinel.
Tests: 13 new cases (schema/version, multi-role fixtures, upsert+audit,
JSON round-trip, secret redaction, reconcile stale-head/dead-lease/
reassigned-lease/clean/unknown, drain fail-closed, sentinel key). Existing
schema_version assertion updated 4->5. Full tests/test_control_plane_db.py
suite: 33/33 pass.
Links #652#653#655.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Umbrella #628 scope item 6 requires dependencies to be durable structured
state carrying source, target, type, blocking condition, completion
condition, current state, and evidence. Nothing stored any of that.
Dependency knowledge existed only as a per-run computation:
allocator_dependencies re-parsed the Depends: declaration out of every
issue body on every allocation, _allocator_candidates_from_gitea resolved
each reference against live issue state, and the result collapsed into two
in-memory WorkCandidate fields that classify_skip consumed and discarded.
Three consequences followed. Nothing could answer "what is waiting on #N"
without re-listing every open issue and re-parsing every body, so the
reverse edge automatic resumption needs did not exist in any form. Only
issue-blocked-by-issue was expressible, leaving the other six #628
relationships with nowhere to live. And no observation was recorded, so a
transient lookup failure and a real block were indistinguishable after the
fact.
Add the store:
- dependency_edges table under schema v4. Creating the table is itself the
v3 to v4 migration: additive, idempotent, and it never touches the
existing tables. Uniqueness is (scope, source, target, edge_type), so
re-observation updates one row rather than appending duplicates.
- dependency_graph.py owns the vocabulary: the seven #628 relationship
types, the three states, and fail-closed normalization for both plus
endpoint kinds. An unrecognized value writes nothing rather than landing
as unqueryable free text. Evidence is sanitized before storage, so no
credential or endpoint URL can be persisted or read back.
- upsert_dependency_edge, list_dependency_edges, and
record_dependency_edge_observation on ControlPlaneDB. Filtering by target
makes reverse lookup a single query. State transitions append to the
existing events table rather than a parallel audit table.
- The allocator persists what it already resolved. States map one-to-one
from the resolver's met/unmet/unavailable partitions, so nothing is
re-classified and unavailable evidence is never recorded as met.
- gitea_list_dependency_edges exposes stored edges read-only, gated on
gitea.read, and is added to the documented inventory the #781 drift guard
checks.
Selection is deliberately untouched: classify_skip still consumes the
in-memory dependency_unmet field. The write is best-effort and reports
failures through reasons, so a broken or absent store leaves allocation
behaving exactly as it did before — proven by allocating the same candidate
set through a store whose writes all raise and comparing the selection,
skip set, and candidate count.
Automatic blocking and resumption (#628 item 7), non-issue edge creation,
and defect auto-linking are later slices; this one only makes the graph
durable and queryable.
Tests: 31 new cases covering fresh-schema creation, a real v3-to-v4
migration with row retention, idempotent re-migration, enum rejection,
upsert idempotence, forward and reverse lookup, scope isolation, transition
events, redaction at rest, live allocation-run ingestion, write-failure
tolerance, and the tool's permission gate.
Verification: 4126 passed in the branch worktree. The 11 failures in
test_commit_payloads, test_issue_702_review_findings_f1_f6, test_mcp_server,
test_post_merge_moot_lease, and test_reconciler_supersession_close reproduce
identically on a clean detached checkout of master at 300e8acd, so they are
pre-existing and proven by baseline run, not introduced here.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Legacy NULL-scope dedupe only compared Gitea targets, so duplicate rows
with the same provider key and issue but different fingerprint/status/
event_count/etc. collapsed to the lowest link_id and silently dropped
observation data. Migration now compares all meaningful observation
fields and refuses to discard conflicts (#619 RC3 / #613).
Address remaining PR #619 blockers on head 783ea88:
- Deduplicate legacy NULL-scope incident_links before normalizing to ''
- Fail closed when duplicate rows disagree on Gitea target
- Require non-empty expected_head_sha for PR assign and mutation
Address REQUEST_CHANGES on PR #619:
- require_valid_assignment rejects terminal work states and head drift
- incident_links scope keys normalize NULL/blank to '' for UNIQUE
- remove trailing whitespace in control-plane-db-substrate.md
Add SQLite single-writer MVP control plane per the allocator ADR:
sessions, work_items (issue/pr only), atomic assign+lease, mutation
gates, terminal-lock index, and provider-neutral incident_links.
DB coordinates concurrency; Gitea remains assignable work; raw
Sentry/GlitchTip incidents are never work items. #600 and #612 build on
this substrate.
sysadminandClaude Opus 5