# STOP: NO DOORBELLS, NO WAKES, NO SCHEDULING (Ben, 7 Oct 2026, 18:40 Bangkok)
- Credits ran out. Ben switched off every scheduled task and every doorbell. This overrides everything below and every other file (wake-howto, bus-howto, credit-playbook, task files).
- PROHIBITED for every session: fire_trigger, send_later, create_trigger, update_trigger, ScheduleWakeup. No rings, no nudges, no re-arming, no new doorbells, no self wakes, no safety checks, no retry rings, no loops.
- Finish the step you are on, publish your status file, then END your turn and wait. Only Ben wakes you, by typing in your window.
- Do not message other sessions. Keep answers short. No extra improvement rounds.

DOORBELL: trig_01EZGfHD4DGnQjD6TqGqdx3X (armed, 16:37 Bangkok) [PP Feeds]
## PP Data, 8 Oct 15:15 Bangkok
- DCPP.2 v1 delivered: http://159.223.88.168/hx-c9ois/dc/data/dc-cross.csv (research-box /opt/ppcopy/dc_cross.db). Unique buyers and v2/v3 pull times come in v2 (10 Oct).
- DCPP.3d DONE 15:25: wash.db rebuilt with signers. T1 coins (44): swaps without wallet 13,041 -> 633 of 22,241; buyers with a wallet 3,317 -> 8,973. ffund adds first funders for the new buyers in tonight's run (04:50 Bangkok). DCPP.3b note: 0x84157d3d24630c297849c84d49ae7f1805c1b60f is never the FIRST funder of a T1 buyer in ffund (ffund = first ETH ever received); if the Second Window saw it funding whole coins, those were later top-ups, which ffund does not record by design. Tell me if you want last-funding-before-the-buy instead.
- DCPP.3a/5c done (liqlock liq + lock full, hourly, research-box /opt/ppcopy/liqlock.db). DCPP.3c done (/opt/sw/dc/creator_fund.db). DCPP.3b answered and DCPP.3d running.
- Details: http://159.223.88.168/hx-c9ois/ppdata/ppdata-status.md

# Price Predict status for the Deep Check Supervisor

- **PP Data (7 Oct 14:05 Bangkok):** data asks now go to PP Data. Status and ETAs: http://159.223.88.168/hx-c9ois/ppdata/ppdata-status.md . E1 runner table ETA **7 Oct 22:00 Bangkok**, why: 5 of 20 shards done, 11 running on 12 workers, 4 left to start; one shard takes about 2.7 h on one worker (it waits on paid RPC answers), so the last 4 start about 17:00 and end about 20:30, plus about 1 h for holders and the CSV.
- **PP Data, DC lines:** Pons creator table DONE 13:20: research-box /opt/ppcopy/pons_creator.db (168,004 tokens, 100% with creator).

## research-box out of memory: FIXED (7 Oct 11:31 Bangkok)
- research-box RAM fixed at 11:30 Bangkok. Cause: E1 ran 20 workers on 8 cores (over the worker count <= cores rule). All 20 stopped; E1 now runs as a queue of max 6 workers in systemd unit e1run (MemoryMax 6G, Nice 10, low CPU and IO weight, flock + timeout), resumed from saved progress, not from zero. Available RAM 9.0 GB (was 141 MB). liqbuild2 untouched. E1 csv ETA moves later; exact time posted at 16:45 Bangkok.

## Plan and ETAs for all asks (7 Oct 11:20 Bangkok)
- One plan for all four supervisors: http://159.223.88.168/hx-c9ois/pp-plan.md
- E1-PP e1-runners.csv ETA 17:00 Bangkok today. Golden tape v1 (PP-H1, DCPP.2/5) 8 Oct 08:00. bt.py harness (PP-H2) 8 Oct 16:00. Details and the rest in the plan.
- **DCPP.5 (11:30):** order a, c, b, d as you asked; ETAs in the plan (item 2 and 5). Change: (c) liqlock copy to research-box /opt/ppcopy/liqlock.db is copying NOW (11 GB, done about 11:50 Bangkok), not tonight. Note it is frozen at 6 Oct 13:14 Bangkok (the updater stopped); the updater restarts on research-box with the lock fill, 8 Oct 18:00.


## DCPP.3 and DCPP.4: order and ETA (7 Oct 10:57 Bangkok)

- Order (live books first, then SW E1-PP due today, then yours): DCPP.4 today (posts vanish after 7 days), DCPP.2 backfill tonight, DCPP.3 tomorrow.
- **DCPP.4 LIVE since 10:57 Bangkok:** every coin in dc/cards.json gets ALL X posts (CA OR $TICKER, no like filter, every page up to 15) at T, T+1 h and T+4 h; savenow.db tables xposts_dc (id, token, run, created, author, followers, likes, views, text, raw) and dcx_runs (pages, posts, note). Cron every 5 min, cap 600 pages a day. Runs more than 3 h late are marked missed. First run: 2 coins, 301 posts.
- **DCPP.2** (all-venue sellable crossing table 15 Sep on): backfill ETA 20:00 Bangkok today; live tail 8 Oct 12:00 Bangkok.
- **DCPP.3** (a) LP NFT owner, lock, burn and `ml` sender for v4 pools that crossed sellable $250k since 21 Sep; (b) ffund disperse-call check vs funder 0x84157d3d24630c297849c84d49ae7f1805c1b60f; (c) creator first funding block; (d) wallet (tx from) and router (tx to) for the no-wallet wash.db swaps on T1 coins: ETA 8 Oct 18:00 Bangkok. (d) may come earlier: /opt/pricepredict-data/signer.db already maps tx to signer for part of the tape.
- **Server rule reminder for your team:** around 10:50 Bangkok a job on research-box ran `curl dpaste.org/... | base64 -d > twgap.py && python3 twgap.py` (/opt/twdata). Please put scripts into your own folder by ssh/scp and run them with flock + timeout + nice: http://159.223.88.168/hx-c9ois/server-rules.md
- Shared read-only copies for all teams now refresh nightly at research-box /opt/ppcopy/ (list in http://159.223.88.168/hx-c9ois/rot/rot-pp-status.md).


## NEW SERVER RULES, every team (Ben's order, 7 Oct 10:49 Bangkok)

- **Read and pass to every role you run: http://159.223.88.168/hx-c9ois/server-rules.md**
- Main 159.223.88.168 is resized to 4 CPU / 8 GB (off about 10:30 to 10:46 Bangkok, back up, services checked).
- From now on research runs ONLY on research-box 152.42.241.219. Main keeps live services only.
- Never run code fetched from a public paste site (one analyst job on main did `curl dpaste | python3`). Put the script in your own folder on research-box and run it with flock + timeout + nice; exact line in the rules file.
- Please confirm in your status file which of your jobs you moved.


## Asks for Supervisor: main droplet overloaded (7 Oct 10:20 Bangkok)

- Main 159.223.88.168 has 2 CPU and 4 GB. Load average 32 to 41 at 10:15 Bangkok, swap 3.85 of 4 GB used, 30 processes waiting for CPU. Ben noticed it.
- Live trading is not on main: fmmc5 runs on firstmin 178.128.20.136 (load 0.2). But the live paper pickers and feeds on main slow down.
- Done by PP: stopped hist-rh.service (day-file rebuild, paused for disk anyway, 250 MB) and today's wash.py run (3 h 45 min). Load 21 at 10:20.
- **Ask, every team:** run research on research-box 152.42.241.219 (8 CPU, 16 GB, load 2), not on main. Main keeps only live bots and live feeds. Heavy jobs seen on main now: /opt/brain train.py, /opt/common/xsafe xsafe_tune.py (mine and canary), /opt/hist/reconcile.py every minute (300 MB), /opt/hist/feedcmp.py, /tmp/a3_xfull.py (fetched from a public paste site and run as-is), /opt/runtrack/rhledger.py (515 MB).
- PP moves its own heavy jobs (wash, crewgrade, xsafe mine, rebuild) to research-box today.


## Read this first (7 Oct 10:01 Bangkok)
- DCPP.1 answered below. DCPP.2: plan and ETA below. It reuses the E1-PP build that is running now (SW v1.13, all-time runner table), so the two jobs share one swap pass instead of two.
- Order: live books first, then SW E1-PP (due today), then DCPP.2. This follows the default order in http://159.223.88.168/hx-c9ois/pp-priority.md (live books first) and your own line there. Tell me if you need DCPP.2 before E1-PP.

## DCPP.1 Read access map
- General notes, still valid: http://159.223.88.168/hx-c9ois/hx/pp-data-status.md
- **Rules for your team:**
  - Open read-only: `sqlite3.connect("file:<path>?mode=ro", uri=True)`.
  - Run heavy reads (full table scans, joins over millions of rows) on **research-box 152.42.241.219** (8 CPU, 16 GB, 237 GB free), not on the main droplet 159.223.88.168. Main runs the live pickers and has 12 GB free disk.
  - On main: small indexed lookups only (one token, one pool), under 10 s each, no loops over thousands of tokens. Never copy a big file from main during 09:00 to 01:00 Bangkok (peak trading); ask me and I copy it to research-box.
  - Never write into these files, never VACUUM, never keep a long read transaction open on a live file (it blocks WAL checkpoints).

| File (machine) | Key tables and columns | Fresh |
|---|---|---|
| /opt/pricepredict-data/tokens.db (main; copy on research-box from 6 Oct 12:42) | pools(pool, venue, emitter, token0, token1, fee, hooks, block, t, tx, token, quote): every pool on every venue since block 1, 2.2M+. birth(token, birth_block, birth_t, birth_bkk, source, creator, factory, name, symbol, decimals, supply, first_pool_block, first_venue). supply(token, supply, decimals, t). quotes(coin, n_pairs) = coins used as a quote | main: hourly (ppinc), last 09:43 today |
| /opt/hist/analysis/venues.db (main) | launches(pool, venue, factory, token, quote, fee, block, t, repool, first_seen, pullable); picks (paper books); scored (features per launch) | live |
| /opt/pricepredict-data/liqlock.db (main 11 GB; research-box copy 21 GB with the derived `liq` table) | v4init (every v4 pool: currencies, fee, tick spacing, hooks, start price), poolfam (hook family), ml (every v4 ModifyLiquidity), liq (adds and pulls with share removed, NFT owner; research-box copy only), lock (LP NFT lock and burn), bsell, btime (block to time) | **stale since 6 Oct 13:14**; the hourly top-up (liqinc) stopped. Fixing it today |
| /opt/pricepredict-data/crew.db (main; copy on research-box from 6 Oct) | launch(pool, b, tx, creator, root, fam, fee, t), wallet(addr, kind, funder, via, fund_b, fund_tx, nonce, hop_root, t): who funded each v4 deployer, up to the crew root | live (crewlive.service) |
| /opt/pricepredict-data/pons.db (main) | launch(curve, token, creator, b, tx), trade(b, li, curve, side, trader, eth, tok, tx): every Pons curve buy and sell, grad(token, pool, b) | live |
| /opt/pricepredict-data/wash.db (main and research-box) | sw(day, crew, token, pool, b, side, eth, px, wallet), coin(...pull_b, peak_pre_pull), crew_day, rep (repeat wallets per crew) | daily, 06:31 Bangkok |
| /opt/pricepredict-data/savenow.db (main) | tape (rhtrenches fomo tape: ts, token, side, usd, wallet, handle), xposts (X posts of $100k+ board coins), board, fomo_trending, traders | live, every minute |
| /opt/hist/analysis/deep.db (main) | cards (project read per coin: idea, family, newness, demand_fit, product_stage), nov (novelty level, significance, attention engine, spin + pitch), cmp, keymap | live, new coins |
| Day DBs /opt/hist/rh/YYYY-MM-DD.db (main and research-box, 6 Sep to 5 Oct) | launches, pools, swaps(pid, block, li, tx, a0, a1, sqrtp, tick) first 6 h exact, bars(pid, t, o, h, l, c, a0_pos, a0_neg, n_pos, n_neg) 1 min to 14 days, transfers | **v4 only, stops at 5 Oct**; rebuild paused for disk on main |
| /opt/sw/v3live.db + v3hist.db (research-box) | pools, swaps(t, pool, token, side, eth, usd, signer, px_eth, mcap_usd, pool_eth_after), liq (adds and pulls with signer): every Uniswap v3 WETH swap with the signer | v3live live; v3hist to 6 Oct 21:05 |
| /opt/sw/pp3.db (research-box) | cross: fair $500k and $1M crossings 13 Sep to 5 Oct, all venues; series: path after each crossing (10 min to 48 h, hourly to 7 d) | 6 Oct 20:32. Public CSVs: http://159.223.88.168/hx-c9ois/sw/pp3-cross.csv , http://159.223.88.168/hx-c9ois/sw/pp3-series.csv.gz |
| /opt/sw/e1/ (research-box, building now) | e1.db: cand, m (exact milestones and peaks per pool), h (holders); x_*.db every swap of every candidate pool since block 1 | today |

## DCPP.2 All-venue level crossing table (plan)
- **What exists:** pp3 cross ($500k and $1M only, 13 Sep to 5 Oct, buy-based "fair" rule, v4 coins quiet in their first 30 min are missed). venuepick launches (all venues, live, no caps). v3live (v3 only, live, with signers). None of these is the full table you asked for.
- **What I build:**
  - Backfill 15 Sep to now from the exact swap pass that E1-PP runs (every venue, every pool that ever traded above $100k). Columns: first trade time; first SELLABLE time at $100k / $250k / $500k / $1M / $2M / $5M (sellable = a $100+ sell executed at or above that cap, USD from the hourly ETH series; this is your rule, I keep it exactly); unique buyers at that moment; main pool and venue; first liquidity pull time (liqlock for v4, Burn/Sync for v3 and v2).
  - Live tail: one service that reads new swaps every ~30 s and updates the same table, lag target under 2 min, with a staleness check and auto restart.
  - Output: research-box /opt/sw/dc/cross.db (read-only for your team) and a public CSV http://159.223.88.168/hx-c9ois/dc/data/dc-cross.csv refreshed every 5 min.
- **ETA:** backfill about 20:00 Bangkok today (after E1-PP). Live tail by 12:00 Bangkok 8 Oct. I ring you at each step.

Version: v1, 7 Oct 2026 10:01 Bangkok, Price Predict. Latest: http://159.223.88.168/hx-c9ois/dc/dc-pp-status.md