Continuous monitoring with Sentinel
CPU, memory, disk, network, services and databases of every server, over encrypted connections: only SERVILINUX can read the metrics.
Back to services
Observability
From seeing something fail to understanding why
There are three steps. Monitoring tells you what is happening now; metrics keep how each server behaves day after day; and observability brings it all together to explain why it happens and what to do.
1
Monitoring: what is happening now
Sentinel reads every server continuously. When something crosses the limit, you see it in the panel and get notified.
Swipe the diagram to see all of it
- Healthy
- Warning: close to the limit
- Critical: action needed
Five servers live: api-01 is running out of memory and db-01 is filling its disk. Sentinel detects it and alerts right away, before your customers notice.
2
Metrics: how it behaves each day
Metrics are kept for weeks. That way an alert stops being an isolated scare and becomes a pattern you can see.
Swipe the diagram to see all of it
- api-01 memory
- Threshold and alerts
- Peak hours (7-10 pm)
One week of api-01 memory: every day between 7 and 10 pm it goes above 90% and the alert fires. It is not a one-off failure: it is peak time, and it repeats.
3
Observability: understand and decide
With the history in view, Eisac analyzes the behavior and proposes concrete decisions.
Swipe the diagram to see all of it
- Low to medium load (lighter means more load)
- High load
- Peak load
Server load by hour and day of the week. The backup was scheduled at 8 pm, right at peak time, competing with customers for memory. Moving it to 4 am, the quietest hour, removes the alerts without upgrading the server.
History to spot patterns
Metrics are kept for weeks. Compare days and hours, and tell a one-off failure from something that repeats.
Eisac explains and proposes
Ask about an alert and it explains what caused it with the metrics at hand, and what to change: task schedules, service settings or more resources.
The alert, in your hand
Alerts arrive instantly as notifications on your phone, and on your smartwatch if it shows your phone’s notifications. You choose which rules alert and from what value.
In SERVILINUX
How it is done
- 1Turn on Sentinel
From each server’s page, in one click.
- 2Install the app
Add SERVILINUX to your phone and turn on alerts.
- 3Tune the limits
Which rules alert and from what value, per server.
- 4Ask Eisac
“Explain this alert” and decide with data.
Start with one server and grow whenever you want
Connect your first server in minutes. The rest, from any cloud you prefer, join the same panel.