System and diagnostics
Admin → System shows whether DataFactory is healthy and lets admins fix common problems without signing in to the server. It refreshes every 15 seconds.
What it shows
| Card | What's on it |
|---|---|
| Overview | Healthy or Attention, plus Version, Built, Platform (Single host, Kubernetes or Amazon ECS) and Licence |
| Host | Host name, OS, uptime, CPUs, memory, free disk, whether the clock is synced, and the Docker and Compose versions. Single host and appliance only. |
| Checks | Database, Cache, Prune worker and Bloom worker, and the Queued jobs |
| Services | Web, API, Prune worker, Bloom worker, Database, Cache and Database setup, with their state, health, restarts and start time |
| Diagnostics pack | Create one to send to Enov8 support |
| Recent actions | Restarts, packs and log views on this page, with who did them |
Logs
Click Logs next to a service to see its last 200, 1,000 or 5,000 lines. Download saves them as a file.
Restart a service
Click Restart next to a service and confirm. The confirmation says what users will notice: for example, restarting the API disconnects users for about a minute and stops jobs running on it.
Restarts are available on the appliance and a Linux server. On AWS, restart in the ECS console; on Kubernetes, use kubectl rollout restart.
Diagnostics pack
When you raise a support request, attach a diagnostics pack:
- Open Admin → System.
- Click Create diagnostics pack and wait for it to finish.
- Click Download and attach the file to your ticket.
| A pack holds | A pack never holds |
|---|---|
| Versions, service state and health, recent logs, settings with secrets removed, connection names and hosts, run status and errors | Passwords, keys, tokens, connection URIs, licence text, table data, Bloom plan data, generated files or user email addresses |
Every pack is checked for known secrets before it's saved. Creating and downloading one is recorded under Recent actions.
It looks after itself
On a Linux server and the appliance:
- DataFactory starts when the server starts.
- A watchdog restarts a service that stays unhealthy for about a minute. After 3 restarts in 30 minutes it stops trying and shows the service as Restart loop for you to look at. It never starts a service someone stopped.
If the web app doesn't open
On the appliance, sign in on the VM console as df-admin and use the console menu: Status, Restart Data Factory or Create diagnostics pack.
Forgot the admin password? On the appliance, use Reset admin password in the console menu. It gives admin a one-time password to change at the next sign-in, and signs the admin out everywhere.