personal_infra/ansible/roles/backup_source/tasks/main.yml
counterweight c2de6dbbd9
backups: monitor both the dump and the pull, per source
Twelve endpoints in two groups, because they answer different questions and
fail for different reasons:

  backup-dump_<svc>       pushed by the SOURCE right after its dump runs
  backup-store_<svc>      pushed by the BOX at 05:30, per source
  backup-store_pull-job   pushed by the BOX, about the box itself

The store alone could catch almost everything, because the artefact filename
carries the source's dump timestamp - a source whose timer died still pulls
"ok" forever, but the timestamp gives it away. What the source side adds is
LATENCY and DIAGNOSIS: the store only learns at the next 04:00 pull, and it
cannot tell you whether the dump broke or the pull did.

pull-job is separate from the per-source checks because it is a LEADING
indicator where those are lagging ones. A disabled pull timer, a failed pull
job, or a filling disk are all visible immediately, while the per-source checks
only fire once an artefact is >26h stale - about a day later. Disable the timer
at 10:00 and every source stays green until tomorrow; pull-job goes red this
morning and names the cause instead of showing six stale sources.

Frequencies: dumps 02:00-02:30 staggered, pull 04:00, verification 05:30, all
daily and Persistent. 26h staleness decides red; a 30h Gatus heartbeat catches
the verification itself having stopped, so a dead check-backups.timer cannot
hide a stale backup.

arbret has no dump endpoint: prd-arbret is in [arbret], which `managed`
deliberately excludes, so nothing of ours runs there. Store-checked only.

check-backups.sh was manual-only; it now runs on a timer and reports per source
rather than only printing. The human-readable report is unchanged.

A SERIOUS bug introduced and fixed in this change, recorded because the shape
is easy to repeat: the reporting hook was added to backup.sh as a second
`trap ... EXIT`. Bash REPLACES the EXIT handler rather than adding to it, so
that silently deleted the trap which restarts the stopped service - the one the
script's own comment calls "the point", and the bug the role was written to
eliminate. Every backup then stopped its service and left it stopped. It took
forgejo, lnbits, headscale and memos down for several minutes each, and nothing
caught it: the dumps exit 0, the artefacts are correct, the deploy reports
failed=0, and liveness only proves the HOST is up. There is now ONE EXIT
handler doing both jobs, armed BEFORE the stop so a failure during the stop
still restarts. Verified by rendering both variants, asserting exactly one EXIT
trap in each, and simulating a mid-way failure to confirm the restart fires.

Verified: all 12 endpoints UP; six sources pulled cleanly (arbret 31M,
headscale 198K, memos 7.8M, vaultwarden 2.1M, lnbits 30M, forgejo 2.6G), store
16% full, "RESULT: all checks passed".

Known gaps, deliberately not closed here:
  * `yell` warnings - disk 75-90%, an artefact under half the previous size,
    retention not pruning - never reach Gatus, because a push is binary.
  * Nothing verifies a backed-up service came back UP. That is the gap that let
    the trap bug run unnoticed, and it is what the next change addresses.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-14 09:22:12 +02:00

110 lines
4.2 KiB
YAML

---
- name: Assert backup_source parameters are sane
ansible.builtin.assert:
that:
- backup_source_name | length > 0
- backup_source_description | length > 0
- backup_source_dump_command | length > 0
- backup_source_recipient | length > 0
- backup_source_recipient is match('^age1[0-9a-z]{58}$')
- not (backup_source_stop_command | length > 0 and backup_source_start_command | length == 0)
fail_msg: >-
backup_source: '{{ backup_source_name | default("<unnamed>") }}' needs a name,
description, dump command and a valid age recipient (age1... 62 chars).
backup_source_stop_command must be paired with backup_source_start_command.
quiet: true
# Declared here rather than assumed. Stage 1 installed it by hand; this is what
# makes a rebuilt host get it too.
# Cache refresh is best-effort on purpose. An unrelated third-party repo with a
# bad signing key (spacey had two: an expired Caddy subkey and a SHA1 nodesource
# key) makes `apt-get update` return warnings, which the apt module treats as a
# hard failure — and that must not stop backups being configured. Installing the
# package is NOT best-effort: if age is genuinely unavailable, the next task fails.
- name: Refresh the apt cache (best effort)
ansible.builtin.apt:
update_cache: yes
cache_valid_time: 3600
failed_when: false
changed_when: false
- name: Ensure age is installed
ansible.builtin.apt:
name:
- age
# curl is needed only when backup_source_push_url is set, but installing it
# unconditionally keeps the task idempotent and it is present on every
# Debian host here anyway.
- curl
state: present
# The pull account: unprivileged, no sudo, exists only so small-backups-box can
# read the dump directory. Trust points one way — the box can read backups, and
# can do nothing else on this host.
- name: "Ensure the {{ backup_source_pull_user }} account exists"
ansible.builtin.user:
name: "{{ backup_source_pull_user }}"
system: yes
shell: /bin/sh # rsync-over-ssh needs a shell; nologin breaks it
home: "/var/lib/{{ backup_source_pull_user }}"
create_home: yes
password: '!' # no password login, ever
when: backup_source_pull_user | length > 0
- name: "Authorise the backup box's key for {{ backup_source_pull_user }}"
ansible.posix.authorized_key:
user: "{{ backup_source_pull_user }}"
key: "{{ backup_source_pull_key }}"
key_options: "restrict" # no pty, no forwarding, no user rc
exclusive: yes
state: present
when: backup_source_pull_user | length > 0
# The shared container above the per-service directories. It must be traversable
# or the pull account cannot reach its own directory. The script's `mkdir -p`
# runs under `umask 077` and would otherwise create this 0700.
- name: "Ensure {{ backup_source_dir | dirname }} is traversable"
ansible.builtin.file:
path: "{{ backup_source_dir | dirname }}"
state: directory
owner: root
group: root
mode: '0755'
- name: "Ensure {{ backup_source_dir }} exists"
ansible.builtin.file:
path: "{{ backup_source_dir }}"
state: directory
owner: root
group: "{{ backup_source_pull_user | default('root', true) }}"
mode: '0750'
- name: "Install the {{ backup_source_name }} backup script"
ansible.builtin.template:
src: backup.sh.j2
dest: "/usr/local/bin/{{ backup_source_name }}-backup.sh"
owner: root
group: root
mode: '0750'
validate: "bash -n %s"
# The .service carries the push token in an Environment= line, so it is 0600.
# The .timer holds nothing secret and stays world-readable.
- name: "Install the {{ backup_source_name }}-backup systemd units"
ansible.builtin.template:
src: "backup.{{ item.unit }}.j2"
dest: "/etc/systemd/system/{{ backup_source_name }}-backup.{{ item.unit }}"
owner: root
group: root
mode: "{{ item.mode }}"
loop:
- {unit: service, mode: "0600"}
- {unit: timer, mode: "0644"}
notify: Reload systemd for backup units
- name: "Enable the {{ backup_source_name }}-backup timer"
ansible.builtin.systemd:
name: "{{ backup_source_name }}-backup.timer"
enabled: yes
state: started
daemon_reload: yes