App Hosting That Grows With Your Traffic

App Hosting is billed hourly from a prepaid balance. You set your reserved baseline and burst limit, and autoscaling stays within those limits.

App Hosting gives you prebuilt stack templates and scaling controls, so you do not have to assemble and operate each layer yourself. If you need full OS control, Cloud VPS is the better fit.

A cloudlet is one unit of compute: 128 MiB RAM + 400 MHz CPU. You set baseline and burst scaling in cloudlets.

Scaling adds capacity within the limits you set. Whether that happens without downtime depends on your architecture and application. Start with a calculator preset, or chat with an engineer about the right setup.

Database Clusters covers MySQL, MariaDB, and PostgreSQL with HA options.

Kubernetes Hosting is available, with architecture presets to get started.

Many customers run production on App Hosting. Use the trial for hands-on evaluation, and talk to an engineer to design your production setup.

We can map a target setup and migration plan. Portability & Migration has the details.

Togglebox manages the platform and infrastructure, including the runtime, networking, and base OS layers. Your team manages the application code, dependencies, and runtime upgrades.

Your environment can use resources up to the maximum you set. Togglebox meters actual CPU and RAM usage in cloudlet steps, so billing follows measured use within that range.

In a clustered setup, the load balancer routes traffic to the remaining healthy app nodes. For databases, replication modes such as primary/replica and Galera help keep data available during a node failure.

RAM, CPU, and traffic charges stop immediately. Charges for retained storage and reserved resources, such as public IPs and SSL certificates, continue. That lets you stop idle dev or staging environments while keeping data and network resources in place.

Deploy with application archives (WAR, JAR, ZIP, or EAR), Git or SVN repositories, URL-based archive pulls, or build-node workflows. You can also add pre- and post-deploy hooks for smoke tests, migrations, or cache warming.

Protect platform access with 2FA using time-based codes from an authenticator app. Recovery codes provide backup access.

Add Memcached for distributed in-memory caching to reduce database load and improve response times. It also supports session storage for PHP and Java apps in clustered setups to maintain session continuity during failover.

If your team runs DevOps, the building blocks are here: Git and CI/CD deploys, pre- and post-deploy hooks, environment cloning for staging and release rehearsal, one-dashboard metrics and logs, and web SSH to any node. Togglebox runs the platform layer so your pipeline is the thing you maintain, not the servers under it.

Open the topology, select the runtime node, and pick the new version. The platform redeploys the runtime layer; your app code redeploys with it. No rebuild of the environment is required.

Yes. Vertical and horizontal scaling triggers on CPU, RAM, network, disk I/O, or disk IOPS thresholds you set. The same cloudlet model (128 MiB RAM + 400 MHz CPU bundle) applies to PHP, Node.js, Python, Ruby, Java, Go, .NET, Docker, and Kubernetes node groups.

Bind a custom domain via CNAME (shared load balancer) or DNS A record (public IP). Automatic SSL covers public endpoints. For zero-downtime cutovers, use domain swapping between staging and production environments.

Pre-deploy, post-deploy, and rollback hooks work for every native runtime and Docker. Use them for database migrations, cache warm-up, smoke checks, asset precompile, or readiness probes before traffic switches.

Stateless apps balance evenly behind the load balancer. Session-bound apps use either sticky sessions or a shared session store (Memcached, Redis, or a database). Tomcat, TomEE, GlassFish, WildFly, and Payara also support session replication on multi-node Java setups.

Choose Cloud VPS for OS-level control, custom kernel modules, root-owned long-running daemons, GPU/CUDA workloads, or Windows / IIS / Active Directory / RDP / COM interop / .NET Framework workloads. App Hosting runs .NET on Linux + Kestrel only.

Bring code, lockfile, environment variables, and a data export. Most stacks land in a native runtime; the rest run as Docker. Portability and migration covers the process; an engineer can review your topology before cutover.

Each runtime exposes the standard build entrypoint (composer install, npm ci, pip install, bundle install, mvn package, go build, dotnet publish). Add pre-deploy hooks for additional steps. For full custom build control, use Docker.

Add MySQL, MariaDB, PostgreSQL, Redis, OpenSearch, or other managed services as nodes inside the same environment. Apps reach them over the internal network using standard TCP. Database hosting and Database clusters cover engine and topology choice.

Yes. Any runtime not in the native list runs as Docker, with the same scaling, deploy, and load balancer integration. If a runtime is on the native list, prefer the native path because you get OS-level patching and version updates without maintaining a Dockerfile.

Pick Kubernetes when orchestration control is the buying requirement: multi-service deployments, custom pod scheduling, Helm-driven release pipelines, HPA/VPA tuning, or platform integrations that depend on raw kubectl access.

Run workers as a separate node group, isolated from the web tier. Sidekiq, Resque, Celery, BullMQ, and similar workers scale independently of web traffic. Job state typically lives in a managed Redis or database node.

Yes. Clone an environment, validate the change, then swap domains for an instant cutover. That gives you a blue-green-style deploy. Portability and migration covers the promotion workflow.

Per-node CPU, RAM, disk, and network metrics ship by default. Attach APM agents (New Relic, Datadog, AppDynamics) at the runtime layer; the JVM also runs a built-in GC optimization agent. The App Hosting feature detail covers the available signals.

Start with the App Hosting calculator. Presets cover starter, production, data-backed, and custom topologies. Or chat with an engineer to map your stack to runtime, scaling profile, and database choice before launch.

Pull an image from Docker Hub, GitHub Container Registry, GitLab Container Registry, AWS ECR, Google Artifact Registry, or any OCI-compliant private registry, or build from a Dockerfile in a connected Git repo. Configure registry credentials per environment. The platform handles image layers, networking, SSL termination, and container startup automatically.

Image-based custom Docker runs third-party images with platform-managed networking, scaling, and lifecycle (deploy via dashboard or API). Docker Engine CE gives you the native Docker CLI on managed infrastructure. You control the daemon directly. For privileged containers, GPU passthrough, Windows images, or full kernel control, use Cloud VPS.

Native Compose-file orchestration is not supported on a single host, but the topology wizard handles multi-container setups with the same result. Translate each Compose service to a separate platform node connected over the private environment network; scale, deploy, and observe each service independently. Multi-host Compose orchestration belongs on Cloud VPS.

The platform manages internal DNS, port routing, and SSL termination. Containers in the same environment communicate over a private network. External traffic routes through the managed load balancer with automatic SSL via Let’s Encrypt or your own certificates, so the image itself stays lean and does not need to terminate TLS.

Any CI/CD tool. Git auto-deploy, CI artifact upload, the Deployment API, and the CLI all run the same five-stage flow with the same atomic-swap guarantees. GitHub Actions, GitLab CI, Bitbucket Pipelines, Jenkins, and CircleCI authenticate with scoped, expiring personal access tokens.

Yes. The new build is prepared on platform build nodes, then traffic switches with an atomic swap. In-flight requests complete on the previous release while new requests route to the new one. Pre-deploy hooks (migrations, smoke tests) run before the swap; non-zero exit halts the release and the previous build keeps serving.

Yes. The Deployment Manager keeps a versioned inventory of every release with its artifact, hooks, environment-variable snapshot, and probe result. Pick any prior release from the dashboard or call the rollback API endpoint. The load balancer drains the current node only after the rollback target passes its health probe.

Code rollback is platform-managed via the Deployment Manager. Database schema rollback is application-managed: pair every forward migration with a reversible down-migration (Liquibase, Flyway, Active Record) before merging. The platform does not roll back schema for you.

Add a platform-generated webhook URL to your GitHub, GitLab, or Bitbucket repository. Every push to the mapped branch fires the webhook and queues a deploy. You can also configure scheduled repo checks that pull and deploy whenever the source changes, with no webhook required.

Personal access tokens can be restricted to specific API methods (for example, deploy-only) and given a configurable expiry. Issue a separate token per workflow or team member. No need to share account credentials or grant permissions beyond what the task requires.

Yes. Hooks run inside the container before traffic switches. Use them for database migrations, smoke tests, or cache warming. Default timeout is 60 seconds; adjust in environment settings if the migration or warm-up step is longer. Non-zero exit halts the release.

Each environment has a maximum cloudlet limit per node group. The platform never adds capacity beyond that ceiling, even under sustained traffic. Pair the maximum with a prepaid balance or low-balance alert to set a hard upper bound on spend.

Scale-in waits for active requests to finish on the cloudlet being released. The load balancer stops routing new traffic to it, the cloudlet drains, and only then is it removed. Long-running connections are not severed mid-flight.

Billing is per cloudlet-hour, prorated to the hour. A cloudlet that is active for any portion of an hour bills for that hour. Reserved cloudlets bill while the environment is running, regardless of utilization. Burst cloudlets only bill for the hours they are actually allocated.

Vertical scaling (more cloudlets per node) is automatic on every tier. Horizontal scaling (more nodes) is automatic for stateless app tiers and load balancers. Database tiers scale horizontally through topology changes (primary/replica, Galera) that you choose explicitly, not as an automatic reaction to load.

The application keeps running on the maximum cloudlets you set. Scaling stops there; the platform does not silently exceed the limit. You receive a scaling-limit-reached notification and can raise the maximum from the dashboard or API.

Stateless tiers typically begin serving traffic within approximately 30 seconds of a scale-out trigger: the new node clones from the stack template and registers with the load balancer. Database tiers take longer because data must sync before queries are accepted, and adding a node is an operator-approved topology change.

No. CPU, RAM, disk, network, and request-rate metrics come from the platform itself, with no in-container agent to install or maintain. Application-level metrics (custom counters, traces, error rates) still need their own instrumentation via OpenTelemetry, Datadog, New Relic, or similar.

Every environment ships with low-balance, scaling-limit-reached, node-restart, and SSL-certificate-expiry alerts active from first deploy. Add custom resource alerts on CPU, RAM, disk, inodes, and network thresholds per node role from the dashboard.

Metrics and access logs are kept long enough to investigate recent incidents and trend recent traffic. For longer retention, export logs and metrics to your own collector through the platform API or via syslog/HTTP shipping. Application logs are retained on the environment alongside the runtime; you control rotation.

The metrics view and the hourly invoice read from the same usage stream. Each hour, the platform takes whichever is higher (peak RAM or average CPU) and bills against that line. Dashboard usage equals invoice usage; there is no separate billing telemetry to reconcile.

Yes. Tail app-server, database, load-balancer, and build-node logs from a single browser tab. Each node role exposes its own log set; download any log through SFTP or the dashboard. SSH into any node directly from the dashboard for deeper diagnostics.

Yes. Collaborators with the right role receive the same scaling, balance, and incident notifications as the primary account holder. Configure notification preferences (email, webhook) per user.

Yes. Export logs and metrics to your own collector through the platform API or via syslog/HTTP shipping. The platform provides the infrastructure signal at 1-minute resolution; the application signal (traces, custom counters, error rates) is yours to instrument with OpenTelemetry, Datadog, or New Relic.

Add Imunify360 or a managed WAF as a security add-on to filter common attack patterns (SQLi, XSS, file-inclusion, bad bots) before traffic reaches your application. Without an add-on, traffic still flows through platform-level network filtering, but request-content inspection requires the WAF.

By default, only the public ports your environment exposes (HTTP/HTTPS on app nodes, configured database ports between tiers) accept traffic. Internal node-to-node traffic stays on the private network. Open additional ports explicitly per environment from the topology view.

Invite collaborators with explicit roles (admin, developer, billing, read-only). Roles control which environments and which actions (deploy, restart, billing changes) the user can perform. All actions are logged in the account audit trail.

Yes. Generate an SSH key in the dashboard and connect to any node in your environment over the platform SSH gateway. Access is keyed to your account; collaborators get their own keys, scoped by role.

Application code, dependency updates, secret management inside the app, and database-level access controls remain with your team. Togglebox manages the host OS, runtime base images, network perimeter, and platform updates. Statement of Support has the full split.