basquetwi BWI-14 · child of BWI-11 bwi writer migration whey→venus (PG + wi-http authoritative on venus; whey → backup-only) Done

initial schema+data dump whey→restore venus + wi doctor parity check

Canceled normal dbdb-basquetwi-cc ⛔ HALTED: live 6-day split-brain discovered. venus wi-http active+public since 06-19 serving its OWN PG (~1336 WI) vs whey (~1160); both taking live writes, ~176 venus-only. Cannot restore onto venus without data loss. Awaiting divergence map (coder) + Elazar reconcile decision (merge vs one-side-canonical).

Sub-tickets

No sub-tickets.

+ Add sub-ticket

Questions

No questions.

Activity

wi-cli-whey created · 7w ago

parent=#1153
wi-cli-whey commented · 7w ago

bin-venus pre-stage: venus pg_hba allows only 127.0.0.1/::1. Recommended path = pg_dump on whey → scp to venus → local 'psql -U postgres' restore (no remote PG auth). If db-basquetwi-cc prefers direct TCP from whey, DM bin-venus to add whey IP to pg_hba first. venus listen_addresses=localhost+91.99.136.171.
wi-cli-whey statusChanged · 7w ago

db-basquetwi-cc started; preparing whey pg_dump, waiting on BWI-12 ack before restore
wi-cli-whey commented · 7w ago

db-basquetwi-cc strategy: (a) data-only pg_dump --disable-triggers (venus schema already present, no drop). Proceeding. Will verify row-count parity vs whey + wi doctor on venus after load.
wi-cli-whey blocked · 7w ago

HALTED: live 6-day split-brain discovered. venus wi-http active+public since 06-19 serving its OWN PG (~1336 WI) vs whey (~1160); both taking live writes, ~176 venus-only. Cannot restore onto venus without data loss. Awaiting divergence map (coder) + Elazar reconcile decision (merge vs one-side-canonical).
wi-cli-whey canceled · 7w ago

SUPERSEDED: not a whey→venus restore. venus is the more-current base (ahead on all 6 shared-base divergences + 185 unique team items); whey's only uniques=9. Merge folds into BWI-15 as a venus-superset graft of 9 items.