Essential session storage is required to sign in. Optional analytics remain off unless you allow them.
Operate a Postgres instance like a DBA. (1) Configure WAL archiving and take a base backup (pg_basebackup), then PROVE point-in-time recovery by restoring to a moment just before a deliberate accidental DELETE/DROP. (2) Produce a logical backup with pg_dump and restore it elsewhere. (3) Tune the key memory parameters (shared_buffers, work_mem, effective_cache_size, maintenance_work_mem) for the host's RAM and justify each value. (4) Set up autovacuum monitoring and show dead-tuple/bloat before and after a vacuum. Document an RPO/RTO and that you TESTED the restore.
Working base backup + WAL archiving; PITR restore to a target time demonstrated (recovers data lost to the DROP); restore actually tested.
Memory params set with correct understanding (work_mem is per-op, effective_cache_size is a hint, shared_buffers is the cache) and justified for the RAM.
Autovacuum/bloat understood and monitored; dead-tuple reclamation shown; VACUUM vs VACUUM FULL used appropriately.
RPO/RTO stated; replication-is-not-a-backup understood; steps documented and reproducible.
Use this README structure to make the work reviewable.
DevPath fetches a read-only, size-capped snapshot. Submitted code is never executed.
# Backup, PITR, and a tuned configuration ## Goal Operate a Postgres instance like a DBA. (1) Configure WAL archiving and take a base backup (pg_basebackup), then PROVE point-in-time recovery by restoring to a moment just before a deliberate accidental DELETE/DROP. (2) Produce a logical backup with pg_dump and restore it elsewhere. (3) Tune the key memory parameters (shared_buffers, work_mem, effective_cache_size, maintenance_work_mem) for the host's RAM and justify each value. (4) Set up autovacuum monitoring and show dead-tuple/bloat before and after a vacuum. Document an RPO/RTO and that you TESTED the restore. ## Acceptance criteria - [ ] Backup & PITR: Working base backup + WAL archiving; PITR restore to a target time demonstrated (recovers data lost to the DROP); restore actually tested. - [ ] Tuning: Memory params set with correct understanding (work_mem is per-op, effective_cache_size is a hint, shared_buffers is the cache) and justified for the RAM. - [ ] Maintenance: Autovacuum/bloat understood and monitored; dead-tuple reclamation shown; VACUUM vs VACUUM FULL used appropriately. - [ ] Ops discipline: RPO/RTO stated; replication-is-not-a-backup understood; steps documented and reproducible. ## Verification - [ ] Document installation and run commands. - [ ] Record automated checks and their exact commands. - [ ] Add representative output, screenshots, or a short demo where useful. - [ ] Confirm failure paths and known constraints. ## Decision log ### Decision title - Context: - Choice: - Alternatives considered: - Tradeoffs: ## Evidence - Link each rubric criterion to the file, test, or artifact that demonstrates it. ## Known limitations - List what is intentionally out of scope and what should be improved next.