@if ($recovered)
Database backups on {{ $host }} are healthy again.

{{ $reason }}
@else
The database backup on {{ $host }} did not produce a usable dump.

{{ $reason }}

Until this is fixed there is no fresh restore point. The offsite copy is an
rsync mirror with --delete, so it does not compensate — it reflects whatever
is on this server, including the absence of a new dump.
@endif
@if ($context !== [])

Detail:
@foreach ($context as $label => $value)
- {{ str_pad($label, 14) }} {{ $value }}
@endforeach
@endif

Where to look:
- storage/logs/db-backup.log        output of both scheduled backup commands
- php artisan backup:database       run it by hand; it prints the reason
- df -h /var/www                    a full disk truncates dumps mid-write
- sudo crontab -u www-data -l       no schedule:run entry means nothing runs
                                    at all. Reinstall it from the
                                    ansible-thinkpi repo:
                                      ansible-playbook \n                                        playbooks/setup-kulibri-backup.yml \n                                        --ask-vault-pass -e target_group=prod

Retention is {{ config('backup.retention_days') }} days, so a gap longer than that leaves no dump
from before it. Copy anything you still need out of {{ config('backup.path') }}
before restarting the scheduler: backup:cleanup deletes on age, not on count.

--
See Hosting.md for the full backup setup.