Follow and record chain reorganisations, pausing on deep ones (B-85, ELC-09, ELC-22, ELC-23, ELC-24, AUD-38, DAT-69)

A reorganisation was neither recorded nor acted upon, and a competing header at the tip's
own height was ignored for as long as the tip stood, leaving the listener on an orphan.

- A header at the height we hold with a different hash is followed through the same
  fork-point search as any other reorg (ELC-09) instead of being ignored; it still has to
  bring valid work and a connected chain.
- ElectrumListener._record_reorg writes `chain_reorg_detected` (depth, fork height, old and
  new tip height and hash) before the window moves, so a failed write ends the session and
  the next server's announcement of the same fork is recorded then. Depth 1 is a
  transaction of its own; depth >= 2 also pauses the lottery with paused_reason 'reorg'
  and writes `lottery_paused` in the same transaction (begin_immediate first).
- Docs: ELC-09/22/23/24, AUD-03/11/38, DAT-30/69, OPS-30, RND-04, ADR-26/52 lose their
  B-85 "da implementare" lines.

Tests changed because the spec changed: test_apply_header_ignores_a_competing_header_at_
the_current_tip_height encoded the B-64 behaviour that ELC-09 now reverses; it is replaced
by test_a_competing_header_at_the_tip_height_is_a_reorg_to_follow and the depth tests.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

BUGS.md

@@ -166,14 +166,6 @@ winner's own bet is gone the payout is held (`payout_blocked`) and the lottery p
 
 ## High
 
-### B-85 — Reorgs are neither followed nor surfaced
-
-Nothing detects a reorg; a 1-block reorg is treated as a hostile server (B-74).
-
-Fix: the listener keeps the last 10 headers; a fork is followed to its fork point, a
-`chain_reorg_detected` audit row records depth and heights, and a depth ≥ 2 pauses
-the lottery (`paused_reason = 'reorg'`) until the operator resumes it.
-
 ### B-86 — A connected but stalled server is never replaced
 
 `ping` keeps succeeding while the server stops following the chain; nothing compares

CLAUDE.md

@@ -8,7 +8,7 @@ The user communicates in Italian in chat — reply to them in Italian. Everythin
 
 ## Project status
 
-All 10 stages of the original build order are code-complete and unit-tested — 570 tests, all under `tests/unit/` (`tests/integration/` is an empty package). Beyond them: Docker + Caddy deployment, admin dashboard (`/admin`), static test UI (`/`), pending-inclusive balance display, an SSE push channel layered over the original polling, self-service password change + admin password reset, and the reconciliation/corroboration machinery below.
+All 10 stages of the original build order are code-complete and unit-tested — 576 tests, all under `tests/unit/` (`tests/integration/` is an empty package). Beyond them: Docker + Caddy deployment, admin dashboard (`/admin`), static test UI (`/`), pending-inclusive balance display, an SSE push channel layered over the original polling, self-service password change + admin password reset, and the reconciliation/corroboration machinery below.
 
 Verified on mainnet with real money: registration + address derivation, deposit crediting (1-conf), a real 10 PLM bet (broadcast → confirmed → change credited back), and one full round cycle (close → draw on a real block hash → 70/30 payout with sat math checked against the broadcast tx → confirmation → close → next round auto-opened). **Withdrawal and the RBF bump path have never been exercised against a live broadcast** — unit-tested only.
 
@@ -49,7 +49,7 @@ PYTHONPATH=. python scripts/decrypt_master_key.py     # ops recovery: decrypt+pr
 PYTHONPATH=. python scripts/encrypt_master_key.py     # ops bootstrap: import an externally-generated xprv (--overwrite to replace)
 PYTHONPATH=. python scripts/electrum_smoke_test.py    # manual check: connect, handshake, subscribe to headers, print the tip
 
-python -m pytest                                      # all 570 tests
+python -m pytest                                      # all 576 tests
 python -m pytest tests/unit/test_hd.py                # one file
 python -m pytest tests/unit/test_hd.py::test_derivation_is_deterministic   # one test
 ```
@@ -157,7 +157,7 @@ One connection serves everything — deposit credits, broadcasts, confirmations,
 - **Rotation.** `ELECTRUM_HOST`/`PORT` is primary, `ELECTRUM_FALLBACK_SERVERS` a comma-separated `host:port[:notls][:pin=<sha256>]` list (`client.py:parse_endpoints` rejects malformed entries at startup, not during the outage when the fallback is needed; `pin=` fixes the server's DER certificate hash, `ELECTRUM_CERT_PIN` does the same for the primary, B-84). The server **refuses to start with fewer than 2 fallbacks** (`config.validate_quorum_servers`, B-84; `ALLOW_NO_QUORUM=true` overrides it, development only). After any failed or dropped session the next server is tried immediately; the backoff (1s doubling to 30s) only kicks in once every server has had a turn.
 - **Bounded requests** (`_REQUEST_TIMEOUT_SECONDS` = 15s); a timeout tears the connection down. Unbounded waits used to hang `POST /bets` *while holding the per-user lock*, and could stall the confirmation poller permanently.
 - **The drop is observable**: `client.wait_closed()` resolves when the read loop dies, and `_run_once` races it against the notification consumers and a 60s `server.ping`. Without it the listener sat on queues nobody would ever fill while `listener.client` still looked alive.
-- **Headers are validated, not trusted** (`_apply_header`, async since B-74): the tip never regresses, a header must meet the difficulty target it claims *and* that target must be no easier than `draw.POW_LIMIT` (PLM's powLimit; `tests/unit/conftest.py` swaps in a regtest limit so tests can mine headers, and the tests of the limit itself put `MAINNET_POW_LIMIT` back), and every advance must be linked to the tip: a single block by its `prev_hash`, a jump by downloading the headers in between (`blockchain.block.headers`, ≤2016 per request) from the active server and checking each. Failure raises `HeaderValidationError`, which ends the session like a dropped connection and rotates away — that header is the draw's only entropy, so a fabricated one picks the winner. The listener keeps the last 10 headers (`_window`, built from the first header plus its 9 predecessors); a header that does not join the tip is a reorg, walked back to its fork point inside the window and followed (`_follow_reorg`), a fork older than the window ends the session. Two more headers are refused *without* ending the session, since neither implies a hostile server (B-64): one at the tip's own height with a different hash (until B-85 follows it; `corroborate_header` is what catches us holding an orphan), and one carrying no `hex` at all (nothing to validate or draw from, and applying the height alone would break the `tip_height`/`tip_header_hex` pairing). `_run_once` separately refuses to publish the client while *no* tip is known, so the ignore-don't-kill choice can't reopen B-63.
+- **Headers are validated, not trusted** (`_apply_header`, async since B-74): the tip never regresses, a header must meet the difficulty target it claims *and* that target must be no easier than `draw.POW_LIMIT` (PLM's powLimit; `tests/unit/conftest.py` swaps in a regtest limit so tests can mine headers, and the tests of the limit itself put `MAINNET_POW_LIMIT` back), and every advance must be linked to the tip: a single block by its `prev_hash`, a jump by downloading the headers in between (`blockchain.block.headers`, ≤2016 per request) from the active server and checking each. Failure raises `HeaderValidationError`, which ends the session like a dropped connection and rotates away — that header is the draw's only entropy, so a fabricated one picks the winner. The listener keeps the last 10 headers (`_window`, built from the first header plus its 9 predecessors); a header that does not join the tip is a reorg, walked back to its fork point inside the window and followed (`_follow_reorg`), a fork older than the window ends the session. A followed reorg writes `chain_reorg_detected` *before* the window moves (`_record_reorg`, B-85), and a depth of 2 or more pauses the lottery with `paused_reason = 'reorg'` in the same transaction; a header at the tip's own height with a different hash is a reorg too, not ignored. One more header is refused *without* ending the session (B-64): one carrying no `hex` at all (nothing to validate or draw from, and applying the height alone would break the `tip_height`/`tip_header_hex` pairing). `_run_once` separately refuses to publish the client while *no* tip is known, so the ignore-don't-kill choice can't reopen B-63.
 - **A quorum corroborates every money-moving decision** (`_corroborate_majority`, 10s per server, asking only the *other* endpoints — never the active one, which is what a MITM controls): `corroborate_header` before a block seeds the draw (B-28), `corroborate_utxo_spent` before a UTXO missing from one `listunspent` is written off as externally spent (B-29), and `corroborate_utxo_credit` before a new outpoint credits a balance — same outpoint, same amount, confirmed (B-59). Balances move in both directions, so both directions need the same quorum — and since B-83 that includes `corroborate_utxo_reinstatement` before a UTXO written off as externally spent is put back (same outpoint, amount, confirmed; `deposits/service.py:find_reinstate_candidates` only reads, `reinstate_reappeared_utxos` writes what the quorum agreed on), and `confirmation_proven` before a tx is marked confirmed: `tx/confirmation.py:poll_once` takes the height the active server reports only as a *claim*, and the row confirms when a strict majority of at least 2 of the *other* servers return a `get_merkle` proof that recomputes to the merkle root of the block's header (`electrum/merkle.py`; the header is the tip we hold, or for an older block the one the quorum agrees on and that meets its own target). With no `network` nothing confirms. The one decision still on the active server's word is the pool-coin listing for a payout (B-82). It needs **at least 2 answers** and a strict majority of those that answered (B-84): fewer → returns **False**, since one voice decides nothing independently and an unreachable network proves nothing. Only `ALLOW_NO_QUORUM=true` keeps the old reduced rule (no other server → True, otherwise a majority of whoever answers). A failed credit corroboration only *delays*: `find_new_credit_candidates` re-offers the outpoint on the next refresh or `DepositReconciler` sweep.
 
 On reconnect `_subscribe_all_users` runs as its own task with bounded concurrency (`_RESUBSCRIBE_CONCURRENCY` = 20) rather than inline and serially — otherwise a large user base froze `tip_height`, and with it an in-flight draw, for the whole sweep (B-31); one user's failure is logged and skipped. `address_for_new_user` (called right after registration) is best-effort by design: on failure that address stays unsubscribed until the next reconnect or `DepositReconciler` sweep.

README.md

@@ -62,7 +62,7 @@ python -m pytest                    # all tests
 python -m pytest tests/unit/test_hd.py   # one file
 ```
 
-570 unit tests cover HD derivation, PSBT building, the Electrum client, bets,
+576 unit tests cover HD derivation, PSBT building, the Electrum client, bets,
 deposits, withdrawals, the round/draw engine, RBF fee-bumping, admin config,
 the pending-inclusive balance calculation, and the SSE push channel. No
 automated integration tests against a live Electrum connection — mainnet

app/electrum/listener.py

@@ -7,6 +7,8 @@ from embit.transaction import Transaction
 from sqlalchemy import select
 from sqlalchemy.ext.asyncio import async_sessionmaker
 
+from app.audit.log import write_audit_log
+from app.db.base import begin_immediate
 from app.db.models import User
 from app.deposits.service import (
     credit_confirmed_utxos,
@@ -28,6 +30,7 @@ from app.rounds.draw import (
     header_prev_hash,
 )
 from app.rounds.events import broadcaster
+from app.rounds.pause import pause_automatically
 
 logger = logging.getLogger(__name__)
 
@@ -52,6 +55,10 @@ _RESUBSCRIBE_CONCURRENCY = 20
 # a hostile server. Also the deepest fork it will follow.
 _HEADER_WINDOW = 10
 
+# A reorganisation at least this deep pauses the lottery (B-85, ELC-24): one block is
+# ordinary on any chain, two or more is worth a look before a new round opens.
+_REORG_PAUSE_DEPTH = 2
+
 # `blockchain.block.headers` serves at most this many headers per request (ELC-12).
 _MAX_HEADERS_PER_REQUEST = 2016
 
@@ -339,14 +346,11 @@ class ElectrumListener:
         if self.tip_header_hex is not None and height == self.tip_height:
             if header_hex == self.tip_header_hex:
                 return  # the same block announced again
-            # B-64: a second header for the height we hold one for — a reorg at the tip, or a
-            # server disagreeing with the rest. Not replaced under us (B-85 follows it,
-            # ELC-09); corroborate_header refuses to seed a draw from an orphan anyway.
-            logger.warning(
-                "ignoring a competing header at the current tip height %s "
-                "(reorg at the tip, or a server disagreeing with the rest)",
-                height,
-            )
+            # B-85 (ELC-09): another block at the height we hold — a reorg at the tip. It is
+            # followed like any other fork (it must bring the work and the chain to show for
+            # it), not ignored: ignoring it left us on an orphan for as long as the tip
+            # stood, and a draw seeded from it would have been refused by the quorum.
+            await self._follow_reorg(height, header_hex, client)
             return
 
         if not header_meets_its_own_target(header_hex):
@@ -460,17 +464,37 @@ class ElectrumListener:
             raise HeaderValidationError(f"the server's branch does not end at the header it announced at height {height}")
         self._check_chained(branch, header_hex_to_block_hash(self._window[fork_height]))
         old_height, old_hex = self.tip_height, self.tip_header_hex
-        self._extend_window([(fork_height + 1 + i, h) for i, h in enumerate(branch)])
-        await self._reorg_followed(old_height, old_hex, fork_height)
-
-    async def _reorg_followed(self, old_height: int, old_hex: str, fork_height: int) -> None:
-        logger.warning(
-            "followed a chain reorganisation: tip %s -> %s, fork at height %s (depth %s)",
-            old_height,
-            self.tip_height,
-            fork_height,
-            old_height - fork_height,
+        # Recorded *before* the window moves: if the write fails the session ends, the next
+        # server announces the same fork against an unchanged window, and the event is
+        # written then — a reorg is never followed without a trace.
+        await self._record_reorg(
+            depth=old_height - fork_height,
+            fork_height=fork_height,
+            old_tip_height=old_height,
+            old_tip_hash=header_hex_to_block_hash(old_hex),
+            new_tip_height=height,
+            new_tip_hash=header_hex_to_block_hash(header_hex),
         )
+        self._extend_window([(fork_height + 1 + i, h) for i, h in enumerate(branch)])
+
+    async def _record_reorg(self, **payload) -> None:
+        """B-85 (ELC-24, AUD-38): the audit row of a followed reorganisation; a depth of 2
+        or more also pauses the lottery (`paused_reason = 'reorg'`), in the same transaction
+        as the row and as the `lottery_paused` one. Depth 1 is only observed: a one-block
+        reorganisation is ordinary on any chain and touches no deposit."""
+        logger.warning(
+            "followed a chain reorganisation of depth %s: fork at height %s, tip %s -> %s",
+            payload["depth"],
+            payload["fork_height"],
+            payload["old_tip_height"],
+            payload["new_tip_height"],
+        )
+        async with self._session_factory() as session:
+            await begin_immediate(session)
+            await write_audit_log(session, "chain_reorg_detected", payload)
+            if payload["depth"] >= _REORG_PAUSE_DEPTH:
+                await pause_automatically(session, "reorg")
+            await session.commit()
 
     async def _corroborate_majority(
         self,

docs/adr/ADR-26-validazione-degli-header-concatenata-senza-ricalcolo.md

@@ -1,6 +1,6 @@
 # ADR-26 — Validazione degli header concatenata, senza ricalcolo della difficoltà
 
-- **Stato:** Accettata — implementazione da completare (B-85)
+- **Stato:** Accettata
 - **Area:** Round ed estrazione
 - **Fonte di verità per:** la scelta di progetto descritta qui sotto; i requisiti (`PREFISSO-NN`) che la attuano stanno nei documenti di dominio, indicati nelle *Conseguenze*.
 
@@ -32,7 +32,4 @@ in [connessione-electrum.md](../connessione-electrum.md#4-header-validati-non-cr
   quorum ([ADR-07](ADR-07-ogni-decisione-sposta-denaro-base-risposta.md)) e il fatto che il seme dipende anche da `server_seed`.
 
 *Codice:* `app/electrum/listener.py:ElectrumListener._apply_header` · `app/rounds/draw.py:header_meets_its_own_target` ·
-*Test:* `tests/unit/test_electrum_listener.py::test_a_multi_block_jump_is_validated_by_fetching_the_headers_in_between`, `tests/unit/test_electrum_listener.py::test_a_fork_is_walked_back_and_followed_instead_of_ending_the_session`, `tests/unit/test_draw.py::test_a_header_below_the_pow_limit_is_rejected_even_if_it_meets_its_own_target`
-*Stato:* da implementare (B-85), per l'header concorrente alla stessa
-altezza · *Oggi:* un header concorrente alla stessa altezza della punta è
-ignorato invece di essere seguito.
+*Test:* `tests/unit/test_electrum_listener.py::test_a_multi_block_jump_is_validated_by_fetching_the_headers_in_between`, `tests/unit/test_electrum_listener.py::test_a_fork_is_walked_back_and_followed_instead_of_ending_the_session`, `tests/unit/test_electrum_listener.py::test_a_competing_header_at_the_tip_height_is_a_reorg_to_follow`, `tests/unit/test_draw.py::test_a_header_below_the_pow_limit_is_rejected_even_if_it_meets_its_own_target`

docs/adr/ADR-52-riorganizzazioni-catena-contengono-invece-ignorarle.md

@@ -1,6 +1,6 @@
 # ADR-52 — Le riorganizzazioni della catena si contengono invece di ignorarle
 
-- **Stato:** Accettata — implementazione da completare (B-82, B-85)
+- **Stato:** Accettata — implementazione da completare (B-82)
 - **Area:** Conferme e fiducia nella rete
 - **Fonte di verità per:** la scelta di progetto descritta qui sotto; i requisiti (`PREFISSO-NN`) che la attuano stanno nei documenti di dominio, indicati nelle *Conseguenze*.
 
@@ -42,9 +42,7 @@ e [ciclo-di-vita-del-round.md](../ciclo-di-vita-del-round.md#4--pagamento)
   irrecuperabile e limitano il danno a ciò che l'attaccante ha rimosso
   ([LIM-24]).
 
-*Codice:* `app/electrum/listener.py:ElectrumListener._apply_header` · `app/rounds/scheduler.py:RoundScheduler._trigger_payout` · `app/rounds/service.py:open_new_round_if_needed` ·
-*Test:* —
-*Stato:* da implementare (B-82, B-85) · *Oggi:* una riorganizzazione è
-seguita ma non registrata e non mette in pausa la lotteria; un header
-concorrente alla stessa altezza è ignorato; il pagamento non verifica le
-monete del round.
+*Codice:* `app/electrum/listener.py:ElectrumListener._record_reorg` · `app/rounds/scheduler.py:RoundScheduler._trigger_payout` · `app/rounds/service.py:open_new_round_if_needed` ·
+*Test:* `tests/unit/test_electrum_listener.py`
+*Stato:* da implementare (B-82) · *Oggi:* il pagamento non verifica le monete
+del round: le prime due misure (rilevare, fermare) sono attive.

docs/architettura.md

@@ -528,7 +528,7 @@ implementare» non esistono ancora nel codice (vedi [ARC-60] e [ARC-22]).
   - il listener conserva gli ultimi 10 header ricevuti; un salto di più
     blocchi scarica e valida gli header intermedi, e un header che si
     concatena a uno di quelli conservati ma non alla punta è una
-    riorganizzazione da seguire, non la fine della sessione (B-74, B-85);
+    riorganizzazione da seguire, non la fine della sessione;
   - una sessione connessa la cui punta non avanza da 360s, mentre un altro
     server ne riporta una più alta, viene chiusa e si passa al server
     successivo (B-86);
@@ -539,9 +539,7 @@ implementare» non esistono ancora nel codice (vedi [ARC-60] e [ARC-22]).
   `app/electrum/listener.py:ElectrumListener._apply_header` ·
   `app/electrum/client.py:ElectrumClient.connect` ·
   *Test:* `tests/unit/test_electrum_listener.py`, `tests/unit/test_electrum_client.py`
-  *Stato:* da implementare (B-85, B-86) · *Oggi:* una riorganizzazione
-  è seguita ma non registrata né segnalata con una pausa, un header
-  concorrente alla stessa altezza è ignorato, e il listener non confronta la
+  *Stato:* da implementare (B-86) · *Oggi:* il listener non confronta la
   propria punta con gli altri server.
 - **[ARC-27]** **`RoundScheduler`.** A ogni giro (5s) chiama
   `open_new_round_if_needed` e poi avanza il round attivo, in qualunque stato

docs/audit-log.md

@@ -43,9 +43,10 @@ ogni tipo di evento. Le colonne della tabella `audit_log` sono descritte in
   `app/rounds/scheduler.py:RoundScheduler._log_payout_failure`,
   `app/tx/broadcast.py:RbfBumper._tick`, `app/main.py:lifespan`,
   `app/rounds/scheduler.py:RoundScheduler._tick`,
-  `app/electrum/listener.py:ElectrumListener._apply_header`
-  *Stato:* da implementare (B-85, B-87, B-90, B-96), per i quattro eventi
-  nuovi · *Oggi:* gli eventi osservativi sono solo i primi quattro.
+  `app/electrum/listener.py:ElectrumListener._record_reorg`
+  *Stato:* da implementare (B-87, B-90, B-96), per tre degli eventi
+  nuovi · *Oggi:* gli eventi osservativi sono solo i primi quattro e
+  `chain_reorg_detected`.
 
 - **[AUD-04]** Il `payload` è un dizionario serializzato in JSON nella colonna
   `payload_json`. `created_at` è impostato dall'applicazione, non dal
@@ -210,14 +211,14 @@ flowchart LR
   ogni pausa, anche se la lotteria era già in pausa. Stessa transazione di
   `paused = true`.
   *Codice:* `app/api/routes/admin.py:pause_lottery`, `app/main.py:lifespan`,
-  `app/electrum/listener.py:ElectrumListener._apply_header`,
+  `app/electrum/listener.py:ElectrumListener._record_reorg`,
   `app/rounds/scheduler.py:RoundScheduler._trigger_payout`,
   `app/tx/reconcile.py:reconcile_once` ·
-  *Test:* `tests/unit/test_pause_reason.py`
-  *Stato:* da implementare (B-82, B-85, B-90) · *Oggi:* scritto da
-  `POST /admin/pause` (`operator`) e dal riconciliatore (`resurrected_tx`);
-  `reorg`, `payout_blocked` e l'avvio con `START_PAUSED` non lo scrivono
-  ancora.
+  *Test:* `tests/unit/test_pause_reason.py`, `tests/unit/test_electrum_listener.py`
+  *Stato:* da implementare (B-82, B-90) · *Oggi:* scritto da
+  `POST /admin/pause` (`operator`), dal riconciliatore (`resurrected_tx`) e dal
+  listener (`reorg`); `payout_blocked` e l'avvio con `START_PAUSED` non lo
+  scrivono ancora.
 
 - **[AUD-12]** `lottery_resumed` — `POST /admin/resume`. Payload
   `{"reason"}`: il `paused_reason` che la ripresa ha tolto (`null` se la
@@ -632,10 +633,8 @@ flowchart LR
   `lottery_paused` con `reason = "reorg"` ([AUD-11]). Una biforcazione più
   profonda di 10 header non viene seguita e non scrive questo evento (la
   sessione si chiude). Nessuna logica rilegge la riga.
-  *Codice:* `app/electrum/listener.py:ElectrumListener._apply_header` ·
-  *Test:* —
-  *Stato:* da implementare (B-85) · *Oggi:* nessuna riorganizzazione è
-  rilevata; l'evento non esiste.
+  *Codice:* `app/electrum/listener.py:ElectrumListener._record_reorg` ·
+  *Test:* `tests/unit/test_electrum_listener.py`
 
 ## Cosa non viene registrato
 

docs/ciclo-di-vita-del-round.md

@@ -97,9 +97,8 @@ richiesta di scommessa (vedi
 
   *Codice:* `app/rounds/service.py:open_new_round_if_needed`, `app/rounds/service.py:rounds_can_open` ·
   *Test:* `tests/unit/test_rounds_service.py`
-  *Stato:* da implementare (B-85, B-79) · *Oggi:* l'unica pausa è quella
-  manuale (`paused`) e `rounds_can_open` controlla solo che `fee_address` non
-  sia vuoto.
+  *Stato:* da implementare (B-79) · *Oggi:* `rounds_can_open` controlla solo
+  che `fee_address` non sia vuoto.
 - **[RND-05]** All'apertura il round **fotografa** i parametri correnti che lo
   governano: durata (`round_duration_seconds`, 600s predefiniti), pausa
   successiva (`round_cooldown_seconds`, 30s predefiniti), costo della

docs/connessione-electrum.md

@@ -169,9 +169,6 @@ chiudere la sessione, perché non indicano un server ostile:
   (lo stesso blocco riannunciato). Un header per l'altezza già nota con un
   hash **diverso** non è ignorato: è una riorganizzazione in punta e si
   segue secondo [ELC-23].
-  *Stato:* da implementare (B-85), per il caso con hash diverso · *Oggi:*
-  anche un hash diverso è ignorato e quello già adottato non viene mai
-  sostituito.
 - **[ELC-10]** un header privo del contenuto binario: non c'è nulla da
   validare né da usare per l'estrazione.
 
@@ -236,9 +233,6 @@ passare a un altro server.
   un'altra catena), la sessione si chiude come per [ELC-11].
   *Codice:* `app/electrum/listener.py:ElectrumListener._follow_reorg` ·
   *Test:* `tests/unit/test_electrum_listener.py`
-  *Stato:* da implementare (B-85), per l'header con hash diverso alla stessa
-  altezza della punta · *Oggi:* un header concorrente alla stessa altezza è
-  ignorato ([ELC-09]); gli altri casi sono seguiti.
 - **[ELC-24]** Ogni riorganizzazione seguita scrive l'evento
   `chain_reorg_detected` con la profondità (numero dei nostri blocchi
   sostituiti, cioè vecchia altezza della punta meno altezza del punto di
@@ -248,9 +242,12 @@ passare a un altro server.
   [modello-dati.md](modello-dati.md) B-85) finché l'operatore non la
   riprende: come ogni pausa, impedisce solo l'apertura del round successivo.
   Con profondità 1 si scrive solo l'evento.
-  *Codice:* `app/electrum/listener.py:ElectrumListener._apply_header`, `app/db/models.py:RoundConfig` · *Test:* —
-  *Stato:* da implementare (B-85) · *Oggi:* l'evento non esiste e nulla mette
-  in pausa la lotteria da solo.
+  L'evento è scritto **prima** che la finestra si sposti: se la scrittura
+  fallisce la sessione termina e la stessa biforcazione, annunciata dal server
+  successivo, viene registrata allora (nessuna riorganizzazione è seguita
+  senza traccia).
+  *Codice:* `app/electrum/listener.py:ElectrumListener._record_reorg`, `app/electrum/listener.py:ElectrumListener._follow_reorg`, `app/db/models.py:RoundConfig` ·
+  *Test:* `tests/unit/test_electrum_listener.py`
 
 ## 5. Il quorum
 

docs/limiti-noti.md

@@ -36,7 +36,7 @@ implementato il [piano delle correzioni](piano-correzioni.md).
 | [LIM-02] RBF di una sola forma (e pagamento senza resto) | **Sì, in caso di congestione della rete** | Chiuso da B-79 e B-87 |
 | [LIM-03] Attesa in `closing` senza limite di tempo | **Sì, in caso di congestione** | Resta senza limite di tempo per scelta ([ADR-51](adr/ADR-51-dubbio-liberano-mai-fondi.md)); con B-78 e B-87 finisce quando la rete decide |
 | [LIM-04] Prelievo e RBF mai provati in rete reale | **Sì, come verifica** | Da chiudere con la prova su rete reale ([operativita.md](operativita.md#prova-su-rete-reale)) prima di aprire al pubblico |
-| [LIM-10] Riorganizzazioni non gestite | **Sì** — una scommessa sparita rende il round impagabile | Chiuso da B-82 e B-85; resta [LIM-24] |
+| [LIM-10] Riorganizzazioni non gestite | **Sì** — una scommessa sparita rende il round impagabile | Chiuso da B-82 e B-85 (la pausa e il registro sono attivi; resta il pagamento, B-82); resta [LIM-24] |
 | [LIM-11], [LIM-12] Quorum facoltativo e parziale | **Sì** — un solo server decide conferme e abbandoni | Chiusi da B-83 e B-84 (resta la lettura delle monete del montepremi, B-82) |
 | [LIM-13] Estrazione manipolabile | **Sì** — per un giocatore che mina | Chiuso da B-73; resta [LIM-25] |
 | [LIM-14] Header senza difficoltà minima | **Sì** — un header falso congela ogni estrazione | Chiuso da B-74 per il blocco; resta il costo basso di un header a difficoltà minima |
@@ -250,7 +250,7 @@ implementato il [piano delle correzioni](piano-correzioni.md).
     su una moneta scomparsa è rifiutata dalla rete e annullata; per una
     scommessa già confermata non esiste alcuna mitigazione (la versione
     precedente di questa voce diceva il contrario).
-  - *Stato:* da chiudere con B-82 e B-85.
+  - *Stato:* da chiudere con B-82; le riorganizzazioni sono già rilevate, registrate e (se profonde) in pausa (B-85).
   - *Per chiuderlo:* depositi a `deposit_confirmations` conferme e coinbase a
     120 ([ADR-06](adr/ADR-06-depositi-richiedono-deposit-confirmations-conferme-6.md)); riorganizzazioni rilevate, registrate e, se profonde,
     con pausa; monete del round verificate prima del pagamento, con i

docs/modello-dati.md

@@ -826,8 +826,8 @@ senza riavviare (vedi [limiti-noti.md](limiti-noti.md)).
   [DAT-69], che scrivono la propria voce di audit.
   *Codice:* `app/api/routes/admin.py:pause_lottery` ·
   `app/api/routes/admin.py:resume_lottery` · *Test:* `tests/unit/test_admin.py`
-  *Stato:* da implementare (B-82, B-85) · *Oggi:* l'unica pausa automatica è
-  `resurrected_tx` (B-78); quelle di B-82 e B-85 non esistono ancora.
+  *Stato:* da implementare (B-82) · *Oggi:* le pause automatiche sono
+  `resurrected_tx` (B-78) e `reorg` (B-85); quella di B-82 non esiste ancora.
 
 - **[DAT-62]** `deposit_confirmations` (predefinito 6) è il numero di
   conferme dopo cui un output che paga un utente si accredita ([DAT-10]);
@@ -864,12 +864,12 @@ senza riavviare (vedi [limiti-noti.md](limiti-noti.md)).
   successivo.
   *Codice:* `app/api/routes/admin.py:pause_lottery` ·
   `app/api/routes/admin.py:resume_lottery` ·
-  `app/electrum/listener.py:ElectrumListener._apply_header` ·
+  `app/electrum/listener.py:ElectrumListener._record_reorg` ·
   `app/rounds/scheduler.py:RoundScheduler._trigger_payout` ·
-  `app/tx/reconcile.py:reconcile_once` · *Test:* `tests/unit/test_pause_reason.py`
-  *Stato:* da implementare (B-82, B-85, B-90) · *Oggi:* si scrivono solo
-  `operator` (`POST /admin/pause`) e `resurrected_tx` (il riconciliatore);
-  `reorg`, `payout_blocked` e `START_PAUSED` non hanno ancora chi li scrive.
+  `app/tx/reconcile.py:reconcile_once` · *Test:* `tests/unit/test_pause_reason.py`, `tests/unit/test_electrum_listener.py`
+  *Stato:* da implementare (B-82, B-90) · *Oggi:* si scrivono
+  `operator` (`POST /admin/pause`), `resurrected_tx` (il riconciliatore) e
+  `reorg` (il listener); `payout_blocked` e `START_PAUSED` non hanno ancora chi li scrive.
 
 - **[DAT-31]** Con `fee_address` vuoto (o di soli spazi) non si apre nessun round
   (B-66). Una volta impostato, l'API non permette di svuotarlo di nuovo: la

docs/operativita.md

@@ -815,13 +815,13 @@ con `docker compose logs app` ([architettura.md](architettura.md) ARC-09).
   *Codice:* `app/api/routes/admin.py:pause_lottery`,
   `app/api/routes/admin.py:resume_lottery`,
   `app/rounds/service.py:open_new_round_if_needed`,
-  `app/electrum/listener.py:ElectrumListener._apply_header`,
+  `app/electrum/listener.py:ElectrumListener._record_reorg`,
   `app/tx/confirmation.py:poll_once`,
   `app/rounds/scheduler.py:RoundScheduler._trigger_payout` ·
-  *Test:* —
-  *Stato:* da implementare (B-85, B-82, B-90) · *Oggi:* il sistema mette in
-  pausa la lotteria solo per `resurrected_tx` (B-78); `reorg` e
-  `payout_blocked` non esistono ancora.
+  *Test:* `tests/unit/test_electrum_listener.py`
+  *Stato:* da implementare (B-82, B-90) · *Oggi:* il sistema mette in
+  pausa la lotteria per `resurrected_tx` (B-78) e `reorg` (B-85);
+  `payout_blocked` non esiste ancora.
 
 ### Intervento manuale sui fondi
 

docs/sicurezza.md

@@ -92,8 +92,8 @@ Restano **accettati**, e descritti in
   riorganizzazioni con pausa sopra la profondità 1 e verifica delle monete
   del round prima del pagamento ([ADR-52](adr/ADR-52-riorganizzazioni-catena-contengono-invece-ignorarle.md), B-82,
   B-85). Il denaro che un attaccante riesce a sottrarre è perso per la
-  piattaforma. *Oggi:* depositi a 1 conferma, nessuna riorganizzazione
-  rilevata, nessuna verifica prima del pagamento.
+  piattaforma. *Oggi:* depositi a 1 conferma e nessuna verifica prima
+  del pagamento (B-82); le riorganizzazioni sono rilevate e, da profondità 2, mettono in pausa.
 
 ## Inventario dei segreti
 

docs/storico-bug.md

@@ -233,7 +233,7 @@ cifrava di nuovo il file sbagliato (vedi [operativita.md](operativita.md)).
 | B-82 | Un deposito annullato da una riorganizzazione dopo la scommessa rende il round impagabile; coinbase accreditate a 1 conferma | Critica | **aperto** | — | DEP, RND |
 | B-83 | Conferme e reintegri si basano solo sul server attivo | Alta | *da registrare* | `test_confirmation_proof.py`, `test_confirmation.py` | ELC, TX, DEP |
 | B-84 | Il quorum può ridursi a un solo server o a nessuno senza avvisi; certificati TLS non verificati | Alta | *da registrare* | `test_electrum_client.py`, `test_electrum_listener.py` | ELC, SEC |
-| B-85 | Le riorganizzazioni della catena non vengono né seguite né segnalate | Alta | **aperto** | — | ELC, RND |
+| B-85 | Le riorganizzazioni della catena non vengono né seguite né segnalate | Alta | *da registrare* | `test_electrum_listener.py` | ELC, RND |
 | B-86 | Un server connesso ma fermo non viene mai sostituito | Alta | **aperto** | — | ELC, OPS |
 | B-87 | L'RBF può lasciare un resto sotto la soglia di dust | Alta | **aperto** | — | TX |
 | B-88 | Le transazioni SQLite non sono quelle che il codice presuppone; `synchronous=NORMAL` può perdere gli ultimi commit | Alta | *da registrare* | `test_sqlite_transactions.py`, `test_db_base.py` | DAT, ARC |
@@ -252,9 +252,9 @@ cifrava di nuovo il file sbagliato (vedi [operativita.md](operativita.md)).
 
 - **Corretti:** B-01 … B-70, tranne B-16, che è chiuso nella sostanza (il
   crash è eliminato e `/guida` serve un segnaposto) ma lascia la guida vera
-  come limite accettato; dalla revisione del 2026-09-28: B-88, B-75, B-97, B-80, B-84, B-76, B-77, B-78, B-83, B-74.
+  come limite accettato; dalla revisione del 2026-09-28: B-88, B-75, B-97, B-80, B-84, B-76, B-77, B-78, B-83, B-74, B-85.
 - **Aperti:** B-71, B-72 (gravità bassa, operativi o di documentazione) e
   B-73 … B-98 dalla revisione del 2026-09-28 tranne i corretti (4 critici,
-  6 alti, 6 medi), da implementare nell'ordine di
+  5 alti, 6 medi), da implementare nell'ordine di
   [piano-correzioni.md](piano-correzioni.md).
 - **Prossimo numero libero:** `B-99`.

tests/unit/test_electrum_listener.py

@@ -18,7 +18,7 @@ from sqlalchemy import select
 from sqlalchemy.ext.asyncio import async_sessionmaker, create_async_engine
 
 from app.db.base import Base
-from app.db.models import User, UtxoEvent
+from app.db.models import AuditLog, RoundConfig, User, UtxoEvent
 from app.electrum.client import ElectrumEndpoint
 from app.electrum.listener import ElectrumListener
 from app.rounds.draw import HeaderValidationError, header_hex_to_block_hash, header_meets_its_own_target
@@ -247,26 +247,6 @@ async def test_apply_header_rejects_one_that_does_not_chain_from_the_tip(session
     assert (listener.tip_height, listener.tip_header_hex) == (100, header_100)  # untouched
 
 
-async def test_apply_header_ignores_a_competing_header_at_the_current_tip_height(session_factory, caplog):  # B-64
-    """The linkage check only fires on a single-block advance, so a header at the
-    height we already hold one for used to be applied on nothing but its own
-    self-consistency — replacing the very hash a draw may be about to be seeded
-    with. Whichever it is (a reorg at the tip, or a server disagreeing with the
-    rest), the hash committed to for a height is not swapped under us; if ours is
-    the orphan, corroborate_header refuses to draw from it anyway."""
-    listener = ElectrumListener(lambda endpoint: None, session_factory, _ENDPOINTS)
-    header_100 = _mine_header("00" * 32)
-    await listener._apply_header({"height": 100, "hex": header_100})
-
-    competing_100 = _mine_header("11" * 32)  # same height, well-formed, different block
-    assert competing_100 != header_100
-    with caplog.at_level(logging.WARNING):
-        await listener._apply_header({"height": 100, "hex": competing_100})
-
-    assert (listener.tip_height, listener.tip_header_hex) == (100, header_100)  # untouched
-    assert "competing header" in caplog.text
-
-
 async def test_apply_header_treats_the_same_header_re_announced_as_a_no_op(session_factory):  # B-64
     listener = ElectrumListener(lambda endpoint: None, session_factory, _ENDPOINTS)
     header_100 = _mine_header("00" * 32)
@@ -1115,3 +1095,162 @@ async def test_a_header_easier_than_the_pow_limit_ends_the_session(session_facto
     with pytest.raises(HeaderValidationError):
         await fresh._apply_header({"height": 100, "hex": chain[100]}, client)
     assert (listener.tip_height, fresh.tip_height) == (100, 0)
+
+
+# --- B-85 (ELC-09, ELC-24, AUD-38): reorganisations are recorded, deep ones pause -------
+
+
+async def _reorg_events(session_factory) -> list[dict]:
+    import json
+
+    async with session_factory() as session:
+        rows = (await session.scalars(select(AuditLog).where(AuditLog.event_type == "chain_reorg_detected"))).all()
+        return [json.loads(row.payload_json) for row in rows]
+
+
+async def _config(session_factory) -> RoundConfig:
+    from app.rounds.config import get_round_config
+
+    async with session_factory() as session:
+        return await get_round_config(session)
+
+
+def _fork_at(chain_a: dict[int, str], fork_height: int, length: int, salt: int = 1) -> dict[int, str]:
+    """The server's chain: ours up to `fork_height`, then `length` blocks of its own."""
+    branch = _mine_chain(fork_height + 1, length, header_hex_to_block_hash(chain_a[fork_height]), salt=salt)
+    return {**{h: v for h, v in chain_a.items() if h <= fork_height}, **branch}
+
+
+async def test_a_one_block_reorg_is_followed_and_only_audited(session_factory):
+    """B-85, ELC-24: the tip 100 turns orphan and the server announces 101 on a branch from
+    99 — depth 1. The listener follows it and writes `chain_reorg_detected`; nothing is
+    paused, since a one-block reorganisation is ordinary."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    chain_b = _fork_at(chain_a, 99, 2)
+    client.chain = chain_b
+
+    await listener._apply_header({"height": 101, "hex": chain_b[101]}, client)
+
+    assert (listener.tip_height, listener.tip_header_hex) == (101, chain_b[101])
+    assert await _reorg_events(session_factory) == [
+        {
+            "depth": 1,
+            "fork_height": 99,
+            "old_tip_height": 100,
+            "old_tip_hash": header_hex_to_block_hash(chain_a[100]),
+            "new_tip_height": 101,
+            "new_tip_hash": header_hex_to_block_hash(chain_b[101]),
+        }
+    ]
+    config = await _config(session_factory)
+    assert (config.paused, config.paused_reason) == (False, None)
+
+
+async def test_a_competing_header_at_the_tip_height_is_a_reorg_to_follow(session_factory):
+    """B-85, ELC-09: a header for the height we hold with a *different* hash used to be
+    ignored (B-64), leaving the tip on an orphan. It is now followed like any fork."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    chain_b = _fork_at(chain_a, 99, 1)
+    assert chain_b[100] != chain_a[100]
+    client.chain = chain_b
+
+    await listener._apply_header({"height": 100, "hex": chain_b[100]}, client)
+
+    assert (listener.tip_height, listener.tip_header_hex) == (100, chain_b[100])
+    [event] = await _reorg_events(session_factory)
+    assert (event["depth"], event["fork_height"]) == (1, 99)
+
+
+async def test_a_competing_header_without_work_or_chain_is_not_followed(session_factory):
+    """ELC-09 / ELC-23: the competing header has to prove itself — a forged one that does
+    not join the server's own chain ends the session and changes nothing."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    stranger = _mine_chain(100, 1, "ee" * 32, salt=5)[100]
+    client.chain = {**chain_a, 100: stranger}
+
+    with pytest.raises(HeaderValidationError):
+        await listener._apply_header({"height": 100, "hex": stranger}, client)
+    assert listener.tip_header_hex == chain_a[100]
+    assert await _reorg_events(session_factory) == []
+
+
+async def test_a_reorg_of_depth_two_pauses_the_lottery_with_reason_reorg(session_factory):
+    """B-85, ELC-24: two of our blocks replaced (fork at 98, tip 100 -> 101) pauses the
+    lottery with `paused_reason = 'reorg'`, and the `lottery_paused` row is written with
+    the event. The round in progress is untouched; only the next one waits for the
+    operator."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    chain_b = _fork_at(chain_a, 98, 3)  # 99', 100', 101'
+    client.chain = chain_b
+
+    await listener._apply_header({"height": 101, "hex": chain_b[101]}, client)
+
+    assert listener.tip_height == 101
+    [event] = await _reorg_events(session_factory)
+    assert (event["depth"], event["fork_height"]) == (2, 98)
+    config = await _config(session_factory)
+    assert (config.paused, config.paused_reason) == (True, "reorg")
+    async with session_factory() as session:
+        reasons = [
+            row.payload_json
+            for row in (await session.scalars(select(AuditLog).where(AuditLog.event_type == "lottery_paused"))).all()
+        ]
+    assert reasons == ['{"reason": "reorg"}']
+
+
+async def test_a_deeper_reorg_overrides_an_earlier_operator_pause_reason(session_factory):
+    """DAT-69: an automatic reason is written even over a pause that is already active."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    async with session_factory() as session:
+        from app.rounds.config import get_round_config
+
+        config = await get_round_config(session)
+        config.paused, config.paused_reason = True, "operator"
+        await session.commit()
+    client.chain = _fork_at(chain_a, 96, 5)
+
+    await listener._apply_header({"height": 101, "hex": client.chain[101]}, client)
+
+    config = await _config(session_factory)
+    assert (config.paused, config.paused_reason) == (True, "reorg")
+    [event] = await _reorg_events(session_factory)
+    assert event["depth"] == 4
+
+
+async def test_a_fork_beyond_the_window_writes_no_event_and_pauses_nothing(session_factory):
+    """AUD-38: a fork deeper than 10 headers is not followed and writes no event — the
+    session ends instead."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    other = _mine_chain(91, 11, "dd" * 32, salt=2)
+    client.chain = other
+    with pytest.raises(HeaderValidationError):
+        await listener._apply_header({"height": 101, "hex": other[101]}, client)
+    assert await _reorg_events(session_factory) == []
+    assert (await _config(session_factory)).paused is False
+
+
+async def test_the_window_does_not_move_when_the_reorg_cannot_be_recorded(session_factory):
+    """A reorg is never followed without a trace: if the audit write fails the session
+    ends, and the next server announces the fork against an unchanged window."""
+    chain_a = _mine_chain(91, 10)
+    listener, client = await _listener_at(session_factory, chain_a, 100)
+    client.chain = _fork_at(chain_a, 98, 3)
+
+    def broken_session_factory():
+        raise RuntimeError("database is locked")
+
+    listener._session_factory = broken_session_factory
+    with pytest.raises(RuntimeError):
+        await listener._apply_header({"height": 101, "hex": client.chain[101]}, client)
+    assert (listener.tip_height, listener.tip_header_hex) == (100, chain_a[100])
+
+    listener._session_factory = session_factory  # the database is back: same fork, now recorded
+    await listener._apply_header({"height": 101, "hex": client.chain[101]}, client)
+    assert listener.tip_height == 101
+    assert len(await _reorg_events(session_factory)) == 1