Observability on Zenifra

Observability: logs, metrics, visitors and alerts in one console

Follow the logs of every instance, CPU and memory usage, who visits your application and its health, and get an email when something is off. All in the Zenifra Console, with no other tool to install.

Logs on every plan

The project Logs card shows each instance’s lines arriving live, in about 1 second. With Log history on, you also see the ended instances of the current version and search the logs of every project in the organization.

CPU and memory per instance

Resource usage shows how much CPU and memory each instance uses against the plan limit, refreshed every minute. Available from Premium up.

Who visits your application

Visitors shows unique visitors, page views, countries, referrers and bot traffic. From Premium up, also requests, latency, error rate, HTTP status, routes and browsers.

Health checks and email alerts

The health check calls a path of your application every 60 seconds and keeps the failures from the last 30 days. With alerts on, the organization owner gets an email when something is off.

Watch the observability of a projectConsole walkthrough: logs, resource usage, Visitors, hourly cost, health check and email alerts.
Search the logs of every projectConsole walkthrough: the organization Logs page, with filters, formats and export, and the project logs arriving live.

Organization logs

  • Turn on Log history in each project. Logs are kept in the organization Storage for 7, 30 or 90 days.
  • On the organization Logs page, search by text or request_id across the last 7 days of every project with Log history, filtering by project, instance, version and output (stdout or stderr).
  • See each line formatted, with the JSON laid out, or raw, exactly as the application wrote it.
  • Export up to 2,000 lines as CSV or JSON.
  • You pay only for the space the logs take, as Storage usage. The 5 GB free in Storage cover everything the organization keeps there.

What the Visitors page shows

  • On every paid plan: unique visitors, page views, share of bot traffic, a chart over time, countries, referrers and top pages.
  • From Premium up: requests per minute, real latency and response time (average and P95), error rate, HTTP status 2xx, 3xx, 4xx and 5xx, top routes, browsers, operating systems, bots by category and top source IPs.
  • Periods of 1 hour and 24 hours on Static and Basic; from Premium up, also 7 days.

Health check and hourly cost

  • From Premium up, turn on the health check and choose the path, such as /health. Zenifra calls that path every 60 seconds; when it fails, the instance is restarted and the failure is kept for 30 days.
  • On projects billed by the hour, the Hourly cost card shows how much each closed hour cost in compute and storage.

Email alerts

Turn on alerts in the project, on any plan. The organization owner gets an email when:

  • an instance restarts unexpectedly, including when it runs out of memory;
  • CPU usage of an instance stays at 80% or more of the plan capacity for 5 minutes;
  • memory usage of an instance stays at 80% or more of the plan capacity for 5 minutes;
  • application storage reaches 80% of capacity and, again, 95%.

At most one email per alert type every hour while the situation continues. Storage warns once at each level until usage drops.

We respect your privacy

We use essential cookies to run this site. Measurement and marketing cookies, such as Google Ads, are only enabled when you accept all cookies. Read our cookie policy.