Hour audit: 2026-09-23T15
back to the files for this hour · how the archive is audited
Every number on this page is a stored measurement read from this hour's own records: its
manifest.json, its witness
statistics sidecar, and the corpus attestation. Nothing here is computed from the served
bytes at the time you load the page, so no line below claims the bytes are correct. The
digest and the one command that checks them are printed instead.
No findings
Every check that could be evaluated for this hour passed. That is not the same as the hour being perfect: it means the records this archive keeps about it agree with each other and with the corpus ledger.
The report
POLYMARKET ORDERBOOK ARCHIVE — HOUR AUDIT
hour 2026-09-23T15 · corpus v3 · verdict AUDIT PASS
COVERAGE minutes present 60/60 · first event 2026-09-23T15:00:00.000004718+00:00 · last event 2026-09-23T15:59:59.999880807+00:00
VOLUME 123,391,739 rows · 1,372,424,658 bytes (1.28 GiB) · 7 products, all holding rows
WITNESSES cfghik — 6 machines contributed rows
captured h: 121,776,674 · i: 120,676,954 · g: 116,301,368 · f: 115,580,322 · c: 110,739,604 · k: 107,630,996 (each machine's own feed; these overlap and do not sum to the file)
supplied h: 5,398,750 (4.38%) · i: 2,147,753 (1.74%) · g: 111,715 (0.09%) · f: 317,231 (0.26%) · c: 101,020,835 (81.87%) · k: 14,395,455 (11.67%) (rows of the merged file that came from it; these partition it)
2 product-machine pairs publish no supplied count; the figures above still sum to the file's 123,391,739 rows exactly, so those pairs supplied none
6 expected, all 6 contributed
INTEGRITY sha256 a12aa8ec5ee7aabd9f67063bfd539e0b2662858ba895369784f04610b34fbfd1
the corpus checksum ledger names this file with the same digest
joined 123,415,115 to merged 123,391,739, minus 23,376 (minus 0.0189%)
boundary drops 23,376 · guard cache
SIDECARS none: this hour's manifest has no v31 key. The v3.1 sidecars are optional, and hours before them have none
MARKETS not measured (the per-hour market census runs on the capture host and its output is not published per hour)
HISTORY merged 2026-09-23T21:57:28Z as cfghik
layout union-event-sorted-v1
re-merges not measured (the merge ledger records the merge that produced this file, not the history of attempts)
supersessions not measured (supersession is recorded per corpus in the storage registry, not in the hour manifest)
CHECKS 8/8 pass
PASS all products present
PASS row counts reconcile
PASS byte ranges tile the file
PASS the merge delta is explained
PASS every expected machine contributed
PASS the checksum ledger names this file
PASS the hour is whole
PASS a boundary guard ran
UPSTREAM our incident register was checked against the venue's own status feed at 2026-09-23T20:35Z and names no event in this hour. That is a statement about what they published, not a check that every message they sent reached us.The words in this report
Each line in the CHECKS block states what was found, not what was
hoped for: a check that passes is named for the thing that is true, and a check that does
not is renamed to the thing that is wrong. So FINDING never sits beside a
sentence asserting the opposite of the one below it.
The report above is fixed-width text, so nothing in it can be clicked. The same terms are linked here instead. Its verdict is a statement about minutes and nothing else. The row it reports comes from this hour's own manifest and its witness-statistics sidecar, and the completeness line comes from the corpus attestation. The gap between the joined and merged counts is deduplication, not loss, and the boundary drops are the duplicates discarded at the seam with the hour before. Where the history line names a supersession, this hour was rebuilt to a better state after it was first published. The precision is the timestamp resolution these files carry.
Detail
The same records, per product. A product holding no rows renders its zero rather than
being dropped: dropped and empty are indistinguishable to a reader, and only one of them is
a defect. A cell reading not meas. is one nothing published, which is a third
thing again and never a zero.
The class column is the merger's own verdict for that product, and it is
keyed to CONTRIBUTION rather than presence. FULL means every machine expected to
be capturing this hour delivered rows for that product; PARTIAL means some did
and some did not; SINGLE-BY-DESIGN means only one machine was ever expected, and
SINGLE-BY-DEFECT means several were and one delivered. So PARTIAL
here is a statement about the fleet, not about the product being damaged or truncated: the
rows that are present are whole.
PRODUCT rows bytes row groups sha256 best_bid_ask 11,141,579 135,326,861 0-42 d1d2aba95b25030310eb83453287df0b63aabf96b8b630d8e752bbb8eb713c94 book 1,766,904 90,358,279 43-49 344b285f5d56a5b0fca8521bd5bd47511785d70c81187ac13479f7feea67f246 last_trade_price 65,361 3,730,098 50-50 051fcee4c490576d4891ee536413cb5188ddc9ba4c8b2a27d435cd9381218d39 market_resolved 1,414 156,754 51-51 f84effbb8b652189c494c1466703d480e864ff9ae0e4b49fae25fadf9d36f903 new_market 851 86,341 52-52 4c6cd2019016ba055a7f6abf9c9483457d357b2149396b0832d967afc47d6b16 price_change 110,413,162 1,141,267,468 53-474 599c9a32af6163733dede60f83c1d2720ceae5c3304a41954204b2e990ab5626 tick_size_change 2,468 92,598 475-475 5286c26605f4d8255326d8402619a4d23187c582c97fc033ce7eee85a916b229 EVENT MIX count share price_change 110,413,162 89.48% best_bid_ask 11,141,579 9.03% book 1,766,904 1.43% last_trade_price 65,361 0.05% tick_size_change 2,468 <0.01% market_resolved 1,414 <0.01% new_market 851 <0.01% PER PRODUCT joined merged boundary exception mkts class best_bid_ask 11,141,857 11,141,579 278 2,965 FULL book 1,787,453 1,766,904 20,549 189,720 FULL last_trade_price 65,362 65,361 1 399 FULL market_resolved 1,426 1,414 12 108 FULL new_market 851 851 0 2 FULL price_change 110,415,698 110,413,162 2,536 1,656 FULL tick_size_change 2,468 2,468 0 127 FULL ARRIVAL SKEW m rows won p50 p90 p99 max unmeasurable interpretation best_bid_ask c 10,092,322 9,283,276 5ms 14ms 89ms 4.3s 0 transport_latency_assumed best_bid_ask f 10,528,133 4,400 50ms 66ms 241ms 18.2s 0 transport_latency_assumed best_bid_ask g 10,700,888 287 80ms 106ms 448ms 21.4s 0 transport_latency_assumed best_bid_ask h 11,137,218 482,992 19ms 38ms 148ms 14.5s 0 transport_latency_assumed best_bid_ask i 11,058,414 181,890 22ms 31ms 118ms 7.8s 0 transport_latency_assumed best_bid_ask k 9,708,643 1,188,734 13ms 21ms 81ms 1.5s 0 transport_latency_assumed book c 932,041 517,450 8.1min 3.9d 41.8d 116.7d 0 book_staleness book f 612,023 256,776 4.7min 2.8d 38.9d 116.7d 0 book_staleness book g 573,098 104,498 6.4min 2.8d 38.9d 116.7d 0 book_staleness book h 267,800 60,500 1.5s 1.4d 32.1d 116.7d 0 book_staleness book i 145,504 10,704 26ms 2.8s 4.9d 98.2d 0 book_staleness book k 1,304,350 816,976 10.0min 4.0d 42.0d 116.7d 0 book_staleness last_trade_price c 59,469 54,246 7ms 20ms 128ms 1.1s 0 transport_latency_assumed last_trade_price f 61,111 38 60ms 1.8s 11.6s 18.2s 0 transport_latency_assumed last_trade_price g 59,345 8 99ms 5.2s 14.9s 21.4s 0 transport_latency_assumed last_trade_price h 65,147 2,339 21ms 132ms 823ms 6.1s 0 transport_latency_assumed last_trade_price i 64,749 1,040 25ms 173ms 919ms 4.0s 0 transport_latency_assumed last_trade_price k 57,259 7,690 14ms 38ms 181ms 1.1s 0 transport_latency_assumed market_resolved c 896 725 7ms 59.1s 5.9min 5.9min 0 transport_latency_assumed market_resolved f 927 66 52ms 55.2s 5.3min 6.5min 0 transport_latency_assumed market_resolved g 923 30 89ms 45.8s 5.5min 5.5min 0 transport_latency_assumed market_resolved h 912 231 24ms 37.7s 58.6s 1.1min 0 transport_latency_assumed market_resolved i 910 15 23ms 37.3s 59.6s 1.4min 0 transport_latency_assumed market_resolved k 1,087 347 20ms 9.5min 14.1min 15.0min 0 transport_latency_assumed new_market c 548 199 55ms 588ms 6.5min 6.5min 0 transport_latency_assumed new_market f 543 47 185ms 18.8s 52.8s 1.8h 0 transport_latency_assumed new_market g 541 32 224ms 803ms 5.2min 5.3min 0 transport_latency_assumed new_market h 541 141 69ms 18.9s 26.5s 1.2min 0 transport_latency_assumed new_market i 541 139 87ms 19.0s 52.9s 1.7min 0 transport_latency_assumed new_market k 606 293 69ms 4.7min 10.3min 10.4min 0 transport_latency_assumed price_change c 99,652,012 91,162,774 6ms 16ms 107ms 4.3s 0 transport_latency_assumed price_change f 104,375,273 55,904 50ms 100ms 3.0s 18.2s 0 transport_latency_assumed price_change g 104,964,169 6,860 81ms 183ms 7.0s 21.4s 0 transport_latency_assumed price_change h 110,302,588 4,852,510 19ms 47ms 309ms 17.4s 0 transport_latency_assumed price_change i 109,404,388 1,953,926 22ms 41ms 274ms 7.8s 0 transport_latency_assumed price_change k 96,556,839 12,381,188 13ms 24ms 109ms 1.6s 0 transport_latency_assumed tick_size_change c 2,316 2,165 6ms 12ms 69ms 153ms 0 transport_latency_assumed tick_size_change f 2,312 not meas. 50ms 60ms 134ms 349ms 0 transport_latency_assumed tick_size_change g 2,404 not meas. 79ms 99ms 212ms 428ms 0 transport_latency_assumed tick_size_change h 2,468 37 19ms 38ms 80ms 292ms 0 transport_latency_assumed tick_size_change i 2,448 39 22ms 30ms 83ms 214ms 0 transport_latency_assumed tick_size_change k 2,212 227 13ms 19ms 53ms 99ms 0 transport_latency_assumed
Captured and supplied are different numbers
Captured is how many rows each machine's own feed carried. Redundant capture means these OVERLAP, heavily and on purpose, so they do not partition the file and their sum is larger than it. Two machines each capturing nearly every event is the archive working, not double counting.
Supplied is how many rows of the merged file actually came from that machine. This one does partition: it sums to the file's row count exactly. But it is decided by the merge's tie-break, so a machine can capture almost everything and supply almost nothing. Read alone, either number tells a false story about the fleet, which is why both are printed and neither is called "the witness share".
What the skew columns mean
Arrival skew is the receipt timestamp minus the venue timestamp, in milliseconds. For six of the seven products the sidecar interprets it as transport latency, and marks that interpretation assumed: it rests on the venue timestamp being emit time, which we cannot verify from outside.
book is not latency and must not be read as latency. Its
venue timestamp is the last time the book changed, so the sidecar interprets it as book
staleness and warns in its own note that large values mean a quiet book rather than slow
transport. Its maximum runs to weeks for that reason. This is why the table above gives a
row per product per machine and never one figure for the hour: percentiles do not average,
and these seven do not measure the same quantity.
Check it yourself
The digest above is a claim. This is how you refuse to take our word for it:
curl -O https://archive.pendulumflow.com/v3/2026-09-23/15/2026-09-23T15.parquet sha256sum 2026-09-23T15.parquet # compare with the sha256 line in the report above
A range digest certifies bytes. That those bytes are the product the manifest names comes from the manifest, which is why both are published and both are checkable.