le-app/docs/le-app-database-migration/phase2a-host-maintenance/maintenance-worksheet.md

1.2 KiB
Raw Blame History

Phase 2A maintenance worksheet

PUBLIC_HEALTH_URLS='https://db.lasereverything.net/'
EXTRA_CRITICAL_UNITS=''
LE_PHASE2A_RECOVERY_READY=<UNSET>
LE_PHASE2A_BACKUPS_READY=<UNSET>

maintenance_window=<UNSET>
user_notification=<UNSET>
rollback_decision_owner=<UNSET>
independent_console_access=<UNSET>
independent_root_access=<UNSET>
previous_kernel_boot_procedure=<UNSET>

Verified evidence:

  • protected Directus archive manifest: PASS, return code 0;
  • Docker: 88 total, 87 running, retired directus intentionally stopped;
  • sole unhealthy running container: reviewed unrelated sonarr;
  • makearmy-app running; bgbye running and healthy;
  • le-directus-source and le-payload-pg running;
  • running restart-policy no: castopod, castopod-redis;
  • DrKonqi disposition: exact checksummed 234-unit allowlist preserved;
  • NoMachine removed; nxserver.service absent and not critical;
  • GRUB: normal, fallback, and recovery entries for linux-zen and linux;
  • Snapper snapshots 13 exist; grub-btrfs.cfg snapshot entries absent;
  • grub-btrfs-snapper.path active and watching.

The recovery and backup attestations must remain unset until the operator has personally confirmed recovery access and completed the proportional backup plan.