Functional reference for main settings

Interface and global notifications

• Application name, subtitle, letter/logo and login tagline.
• Timezone: either the server timezone or a custom one.
• Dashboard refresh interval.
• Default statistics period.
• Mail and SMTP configuration.

Security, WebAuth, audit and roles

• Absolute session lifetime, inactivity timeout, warning delay and database sync frequency.
• Strict session fingerprinting.
• Maximum re-authentication age and pending second-factor TTL.
• Maximum number of sessions per user.
• Enforced 2FA for administrators and login notifications.
• RP name, RP ID, timeouts, user verification, resident keys and attestation for WebAuthn.
• Audit retention for activity, restores, cron, logins, rate limits, queue and archives.
• Editing of the role hierarchy and role badges.

Backup, restore and default behaviors

• Global repository alert threshold.
• Backup job scheduling defaults: enabled or not, time, days, failure notification.
• Retention defaults: enable, keep_last/daily/weekly/monthly/yearly and prune.
• Global reference retry policy.
• Default paths for local, remote and partial restores.
• Page size for restore history.

Disk Space Monitoring

The DiskSpaceMonitor service continuously monitors available disk space on the mountpoints involved in your backups. It runs as a global scheduled task and periodically probes the configured volumes.

• What is monitored: local Restic repository mountpoints, and the root / of each remote SSH host.
• Remote probing: performed via the SSH command df -Pk executed on the target host.
• Data stored: point-in-time history in the disk_space_checks table; latest known state in disk_space_runtime_status.
• Forecast model: uses 30 days of history (requires at least 3 daily data points) to project the number of days until the disk is full.
• Preflight checks: before each job (backup, copy, restore), disk space is checked — if the critical threshold is exceeded, the job is blocked with an explicit error.
• Severity levels: ok / warning / critical — thresholds are configurable as usage percentages. error indicates a probe failure itself (SSH host unreachable, command error). unknown indicates no data has been collected yet.
• Manual probe: can be triggered from a repository or host page, or via the probe_disk_space.php script.
• UI surfaces: dashboard widget, stats page, individual repository and host pages.

Scheduler, central cron and worker

• Central cron is CLI-only (legacy HTTP entry cron.php?token=... is disabled).
• Execution modes (manual, diagnostic, quick) controlled by FULGURITE_CRON_MODE.
• Weekly report, integrity check and vacuum maintenance with days, times and notifications.
• Default worker name, sleep interval, jobs per loop, stale minutes and heartbeat stale seconds.

Reconfiguration wizard (Resetup)

The Reconfigure system button in the Automation tab opens the reconfiguration wizard. It lets you change the PHP-FPM user, file permissions and worker configuration without reinstalling the application.

• 7-step flow: preflight → user/group selection → permission validation → sudo password (if needed) → apply → audit trail → completion.
• Sudo handling: if a sudo password is required, it is held in session only for the duration of the wizard and is never stored. Each sudo command is individually validated (defense in depth).
• Audit trail: all actions are logged with a timestamp, the command executed and the actor.
• Docker detection: if the application is running inside a Docker container, systemd-related steps are automatically skipped.
• No-sudo fallback: in environments without sudo, the wizard outputs the manual commands to run directly on the server.
• Session lock: only one reconfiguration session can be active at a time. Navigating away from the wizard resets its state.
• Rate limiting: maximum 3 reconfiguration attempts per hour.

Internal data, caches and thresholds

• Explorer view cache TTL and exploration page size.
• Maximum number of search results and maximum explorable file size.
• Warm index batch size, adhoc index retention and number of recently indexed snapshots.
• Restic cache TTLs: snapshots, ls, stats, search and tree.
• Slow-query thresholds for HTTP, SQL, Restic and logged commands.
• Auto-refresh and cache settings for the performance page.

External integrations and public API

• Discord, Slack, Telegram, ntfy, webhook, Teams, Gotify, in-app and web push.
• Internal notification retention.
• secret://-based secret handling with agent provider by default and encrypted local fallback.
• Infisical configuration: URL, project, environment and secrets path.
• API enable/disable, default token lifetime, rate limit, API log retention, idempotency retention and allowed CORS origins.
• Rate limits per endpoint families, including restores, scheduler, worker, explore, search and WebAuthn.