fscotto/infra
1
0
Fork 0
mirror of https://github.com/fscotto/infra.git synced 2026-09-27 19:03:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

Document validated Atlas Borg backup activation

Browse Source
This commit is contained in:
Fabio Scotto di Santolo
2026-09-18 10:09:12 +02:00
parent 7e498514dd
commit e837b0059b
2 changed files with 13 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
AGENTS.md
View File

@@ -180,13 +180,13 @@ scheduled retention prune and monthly scrub remain runtime checks.
generations, plus a monthly scrub on the first Sunday at 03:00. The timers and first hourly snapshot were generations, plus a monthly scrub on the first Sunday at 03:00. The timers and first hourly snapshot were
verified on Atlas. Still observe the first scheduled retention prune and scrub; Cockpit Scheduler is for verified on Atlas. Still observe the first scheduled retention prune and scrub; Cockpit Scheduler is for
visibility or manual operations only, and snapshot rollback is never automated. visibility or manual operations only, and snapshot rollback is never automated.
- [ ] Complete runtime activation of the encrypted offsite Borg backup to the Hetzner Storage Box. The - [x] Activate and validate the encrypted offsite Borg backup to the Hetzner Storage Box. Atlas uses the
playbook now declares the dedicated SSH identity, pinned ED25519 host key, Vault-backed `repokey` dedicated SSH identity, pinned ED25519 host key, Vault-backed `repokey` encryption, and a locked
encryption, a locked non-login `borg` account with no sudo or supplementary groups, snapshot-consistent non-login `borg` account with no sudo or supplementary groups. The initial snapshot-consistent backup,
sources prepared only by the root wrapper, Borg execution without ZFS privileges, daily retries and Borg repository check, and temporary-directory restore completed successfully; the restored `Archive`
logging, 30 daily, 8 weekly and 12 monthly archives, compaction, and monthly repository checks. Still tree matched the live data, and temporary snapshots and mounts were removed. The exported recovery key
add the Vault passphrase, install the generated public key in the sub-account, run the initial backup was copied offline. Daily backup retries and logging, 30 daily, 8 weekly and 12 monthly archives,
and check, preserve the exported recovery key offline, and validate a restore before relying on it. compaction, and monthly repository checks are enabled.
- [ ] Add the UUID-bound offline USB backup with versioned rsync, locking, capacity checks, verification, - [ ] Add the UUID-bound offline USB backup with versioned rsync, locking, capacity checks, verification,
safe unmounting and a tested restore procedure; never trigger it for an arbitrary USB disk. safe unmounting and a tested restore procedure; never trigger it for an arbitrary USB disk.
- [ ] Test restores independently from a ZFS snapshot, Borg, and the offline USB backup before relying on - [ ] Test restores independently from a ZFS snapshot, Borg, and the offline USB backup before relying on

9
README.md
View File

@@ -347,14 +347,17 @@ ANSIBLE_LOCAL_TEMP=/tmp/ansible-local \
ansible-playbook ansible/site.yml --limit atlas --tags packages,borg --check --diff ansible-playbook ansible/site.yml --limit atlas --tags packages,borg --check --diff
``` ```
Atlas runtime activation is complete: the initial backup and repository check succeeded, a full restore
to a temporary directory was validated against the live `Archive` tree, the recovery-key export was copied
to offline storage, and the temporary snapshot and bind mounts were cleaned up.
A temporary Nextcloud deployment on Atlas is also planned before Uranus: it requires separately A temporary Nextcloud deployment on Atlas is also planned before Uranus: it requires separately
declared persistent application, database, and cache storage, Vault-backed credentials, NPM-only declared persistent application, database, and cache storage, Vault-backed credentials, NPM-only
publishing through Aegis, and defined backup, upgrade, and eventual migration procedures. Do not deploy publishing through Aegis, and defined backup, upgrade, and eventual migration procedures. Do not deploy
it before the data-protection checklist is complete. it before the data-protection checklist is complete.
Runtime activation and restore validation of the Borg backup, Prometheus backup pulls, USB backup, Prometheus backup pulls, USB backup, monitoring, and disaster-recovery tests remain follow-up work. The
monitoring, and disaster-recovery tests remain follow-up work. The prioritized operational backlog is prioritized operational backlog is kept in `AGENTS.md`.
kept in `AGENTS.md`.
## How layering works ## How layering works