Monitoring that someone actually reads
Plenty of self-hosted estates have monitoring. Fewer have someone whose job is to read the alert. With Pilae, every alert reaches a Pilae engineer, on every plan. Response times follow your plan, and around-the-clock incident response is part of Enterprise.
Probes run every 60 seconds against each app. They check that it answers, that it returns what it should, that it answers in time and that its certificate is valid. Machine metrics track CPU, memory, disk and network, so a disk filling up or a memory leak shows weeks before it causes an outage.
Built on open-source tools
The probes run on Gatus and machine metrics on Beszel, both open-source projects. They run in your environment, on your premises or in Pilae Cloud, and report into the console. Your team sees the same health, history and incidents as our engineers.
From alert to fix
A failing probe opens a task for the Pilae Agent. It checks recent changes, proposes a fix or a rollback, and takes a backup before anything runs. An engineer decides with you, and the whole run is recorded. Our managed operations service sets out who does what during an incident.
Reports your management can read
Each month you receive a report per app: availability against your 99.9% commitment, incidents and their causes, changes applied, and the backups and restore tests of the month. Every major incident gets its own report. The service level agreement defines availability, service credits and response times. To agree what we watch, book a scoping call.