Laravel VPS Hosting: A Production Setup and Sizing Guide
In this article8 sections
Laravel VPS hosting makes sense when your application needs reliable background workers, Redis, custom services, or control over PHP and the web server. It also gives you more responsibility: someone must maintain the operating system, secure access, monitor disk and memory, apply updates, and test recovery. The best Laravel server is not the largest VPS you can afford. It is the smallest configuration that supports the app’s real traffic and jobs, with a clear way to grow.
This guide helps you choose a VPS size from the actual Laravel workload: web requests, queue workers, database, and scheduled jobs. Siteharbour’s free initial setup installs and hardens the server stack, then hands over root access; the VPS is self-managed after that. If your app needs workers or Redis, review the Laravel VPS plans and start from the plan recommendations below.
Quick recommendation: VPS-1 (2 vCores, 4 GB RAM, 40 GB NVMe) fits development, staging, a small app, or a low-traffic API. VPS-2 (4 vCores, 8 GB RAM, 75 GB NVMe) is the sensible starting point for most production SaaS apps and is the plan to choose when queues or Horizon run beside the database. Consider VPS-3 (6 vCores, 12 GB RAM, 100 GB NVMe) for heavy jobs or sustained traffic. Choose VPS-2 for a production Laravel app or compare the Laravel setup and plans.
When Laravel needs a VPS
A VPS is a strong fit for production SaaS products, APIs with steady traffic, applications with queue workers, scheduled processing, Redis, or integrations that need custom system packages. It is also useful when you need root or SSH-level control, want predictable resources, or need to choose exactly how PHP-FPM and the web server are configured.
Shared cPanel hosting can work for a small Laravel site if the plan supports the app’s PHP version, extensions, document root, command-line access, and cron needs. A scheduler command that starts and exits is different from a queue worker that must stay alive. If your app depends on Horizon, long-running workers, WebSockets, or a custom daemon, verify the hosting model explicitly; a VPS is usually more suitable.
Do not move to a VPS solely because Laravel is in the framework name. A brochure site, prototype, or low-use internal tool may be well served by shared hosting while it remains within the account limits. The Laravel on cPanel deployment guide covers that route and its constraints.
For a focused comparison of what cPanel plans can and cannot handle, read our companion guide to Laravel shared hosting.
Check the framework’s current requirements
Use the documentation for the exact Laravel version your application runs. Requirements change by release. Check PHP version and extensions, database compatibility, Composer version, and any extra needs introduced by packages. Keep the CLI PHP version used by Composer and Artisan aligned with the web runtime, otherwise commands may install or execute against a different set of extensions.
Laravel’s production deployment guidance also highlights configuration caching, route and view caching, secure environment values, and serving requests from the application’s public directory. Keep .env, source code, and private storage outside the web document root. The web server should route requests through public/index.php, rather than expose the project root.
Size the VPS around the work it will perform
Estimate CPU and memory from concurrent requests, queue volume, database activity, scheduled jobs, and deployment processes. A test or low-traffic app needs less than a production service that generates reports, imports data, or processes user uploads. Also account for the operating system, web server, database, Redis, and monitoring. A server with 4 GB of RAM does not have 4 GB available to PHP.
- Development, staging, or a small app: a modest VPS can be sufficient if the database and queue load are light.
- Production SaaS or business app: allow room for several PHP workers, the database, and any queue or cache service. Monitor memory pressure during normal peaks.
- Data-heavy or job-heavy app: load test representative requests, observe slow queries and queue time, and consider separating database or worker resources when evidence shows contention.
Use this plan matrix to choose a starting point. It reflects Siteharbour’s published Laravel guidance; it does not replace measuring your own app. A queue worker can consume more memory than an ordinary request, and a report or import can keep several workers busy at once.
| Laravel workload | Starting plan | Why | Order |
|---|---|---|---|
| Development, staging, or a small production site | VPS-1 · 2 vCores · 4 GB RAM · 40 GB NVMe | Room for a modest PHP-FPM pool, database and Redis when the app is lightly used. Keep worker count conservative and monitor memory. | Choose VPS-1 |
| Production SaaS or business application | VPS-2 · 4 vCores · 8 GB RAM · 75 GB NVMe | Recommended starting point for several queue workers, Horizon, a larger PHP-FPM pool and the database on one server. | Choose VPS-2 |
| Busy app, imports, reports, or heavy queues | VPS-3 · 6 vCores · 12 GB RAM · 100 GB NVMe | More CPU for parallel jobs and more memory for the database and application processes to coexist. | Choose VPS-3 |
| High traffic, large database, or several production apps | VPS-4 · 8 vCores · 24 GB RAM · 200 GB NVMe | Use when monitoring and load tests show the VPS-3 tier is consistently constrained or multiple apps need to share resources. | Choose VPS-4 |
Our default recommendation is VPS-2 for production. Start smaller only when the app is genuinely light or still in staging. Move up when CPU, memory, queue wait time, or database latency stays high after application tuning. The plans upgrade in place; check the current price, location allowance, and tax at the Laravel VPS page or in checkout.
Plan the application stack
A typical deployment includes a supported Linux release, NGINX or Apache, PHP-FPM, a compatible database, HTTPS, and a process manager for queue workers. Redis can support cache, sessions, or queues where the app is designed to use it. Supervisor or a systemd service can keep workers running and restart them after a failure. The Laravel scheduler normally needs a cron entry that invokes schedule:run each minute.
Keep each service’s role clear. Use PHP-FPM for web requests, and run queue workers separately with appropriate timeouts and retry settings. A job that handles a large file may need different memory and timeout limits from an email notification. Do not treat a queue as a magic speed switch: slow jobs still consume CPU, and a backed-up queue needs investigation.
Set production environment values securely. Turn off debug output, use a strong application key, restrict database privileges, and do not commit secrets into the repository. Use a separate database and credentials for staging when possible. Protect administrative endpoints and restrict firewall access to services that do not need to be public.
Deploy safely and make rollback possible
Use a repeatable deployment process rather than editing production files by hand. A common flow fetches a tagged release, installs Composer dependencies without development packages, applies reviewed migrations, clears or rebuilds Laravel caches, and gracefully restarts workers so they load the new code. Test migrations against a recent copy of production data where practical; a destructive migration can turn a successful code deploy into an outage.
Keep previous releases available until the new version is healthy. Check the homepage, login, API endpoints, scheduled tasks, queue status, and error logs after deployment. If a release fails, know how to switch code back and whether database changes are reversible. A backup is important, but rollback should also have a documented sequence.
Security, backups, and monitoring
Use SSH keys, remove unused accounts, apply security updates, and allow only required network ports. Enforce HTTPS and renew certificates automatically. Give the web process only the write permissions it needs, commonly for Laravel’s storage and cache directories. Avoid broad recursive permissions such as making the whole application world-writable.
Back up the database and uploaded files on a schedule that matches how much data you can afford to lose. Store a copy away from the VPS and test restoring it. Monitor uptime, disk space, CPU, memory, PHP-FPM saturation, database errors, queue depth, failed jobs, and certificate expiry. Logs should be useful, rotated, and protected from public access; remove or redact sensitive data where appropriate.
Common Laravel VPS problems
- 502 or 504 responses: inspect PHP-FPM service health, worker capacity, timeouts, and application logs.
- Scheduled work never runs: verify the cron entry uses the correct project path, user, and PHP binary.
- Queue jobs remain pending: check whether the worker is active, its queue name matches, failed-job records are growing, or Redis/database connections are failing.
- Uploads fail: confirm disk permissions, storage links, available disk space, and application-level upload limits.
- Works in shell but not browser: compare PHP versions, environment variables, permissions, and web server document root.
Choose the right hosting path
Choose shared hosting if the app is small, stateless enough for its constraints, and the plan provides the required PHP runtime, extensions, SSH or terminal access, and cron. Choose a VPS when your Laravel application needs persistent workers, Redis, custom packages, more predictable capacity, or control over deployment and services. For projects moving from a control panel, the existing deployment walkthrough helps identify which shared-hosting assumptions will need to change.
Ready to deploy? Most production Laravel apps should start by comparing VPS-2. Use VPS-1 for a small app or staging and VPS-3 for heavier queues or sustained load. Siteharbour’s one-time setup covers the server stack and hardening, not deploying or maintaining your Laravel application. Review the service scope and current specifications before checkout, and decide who will own updates, backups, monitoring, and incident response after handover.