Skip to content

Server Monitoring

Appliku continuously monitors your servers and displays key metrics on the server detail page in the dashboard.

Monitored Metrics

CPU Usage

Appliku tracks CPU utilization as a percentage. High CPU usage sustained over long periods may indicate your application needs optimization or a larger server.

Memory (RAM) Usage

Current memory usage is displayed as both an absolute value and a percentage of total available RAM.

Appliku triggers an alert when RAM usage exceeds 80% of total available memory. If you receive this alert, consider:

  • Reducing the number of applications running on the server
  • Upgrading to a server with more RAM
  • Optimizing your application's memory consumption
  • Reducing build concurrency (see below)

Disk Usage

Disk usage is monitored to ensure your server does not run out of storage. Docker images and containers can accumulate over time. Appliku performs periodic Docker cleanup to reclaim disk space, but very active servers with frequent deployments may need larger disks.

Uptime

Server uptime is tracked so you can see how long the server has been running since its last restart.

Viewing Server Metrics

  1. Open the Appliku dashboard
  2. Navigate to the Servers tab
  3. Click on a server to open its detail page
  4. Server metrics are displayed on the overview section

Server Information

The server page also shows facts about the machine, among them RAM, vCPUs (the number of CPUs), Kernel (the running Linux kernel version) and Architecture (for example x86_64 or aarch64). Appliku reads them when it fetches the server stats, so a field stays empty until the first fetch.

Refreshing Stats

Appliku fetches the stats of active servers every few minutes. To fetch them now:

  • On a server page, click Refresh Stats.
  • On the server list, click Fetch all stats. Appliku queues a refresh for all servers of the team: it asks the cloud provider about each DigitalOcean, AWS EC2 and Hetzner Cloud server (a server that was deleted at the provider is then marked as deleted), and it fetches the stats of each active server over SSH. The list reloads after some seconds.

RAM Spikes During Builds

This is normal behavior

RAM spikes during deployment builds are expected, especially for JavaScript and TypeScript projects with many dependencies. The npm install or yarn install step can temporarily consume a large amount of memory.

If you see RAM alerts only during deployments but your server is otherwise healthy, this is not a cause for concern. To reduce build-time memory pressure, you can:

  • Use a larger server -- more RAM gives builds more headroom
  • Reduce build concurrency -- if multiple applications are building at the same time, they compete for memory. Stagger deployments or limit concurrent builds
  • Use a separate build server -- in a cluster setup, you can designate a dedicated build server so builds do not affect running applications

Alerts

When a metric exceeds its threshold, Appliku displays a warning on the server detail page and in the servers list. The current alert thresholds are:

Metric Alert Threshold
RAM usage > 80%

Keeping your servers within healthy resource limits ensures reliable deployments and stable application performance.