The platform: highly available, isolated and open to your sign-in
How Rapitt keeps running and keeps customers apart: instances, a database standby, a second site in another country, backups, per-customer keys, sign-in options and an Arabic interface.
Built to keep running, built to be isolated
A monitoring platform that goes down when you need it most is worse than none. Rapitt runs several instances per workspace, keeps a database standby ready to take over, replicates everything to a second site with a different provider in a different country, and backs each customer up on its own.
And every customer is separate: its own process, database, encryption key, files and tunnel network.
Reliability
What exists, stated plainly.
Several instances per workspace
Background jobs are held by one instance at a time and move to a surviving one if it stops. Restarts and updates roll through instances so one keeps serving. In a test, an instance was killed mid-traffic: 142 requests, none failed, every job held again in 9 seconds.
Database standby with failover
A PostgreSQL standby takes over automatically. A planned switchover to the standby completes in under 10 seconds, and the programs reconnect to whichever node is primary without configuration changes.
A disaster-recovery site
A second site with another provider in another country holds a streaming replica of every database and a mirror of the files the platform needs. Data loss is measured in seconds, and a restore drill runs weekly.
Backups and point-in-time recovery
Nightly per-customer database backups, point-in-time recovery, off-site copies and a weekly restore drill. Restores can be tested, copied beside the live database, or replace it with an automatic rollback.
Isolation, access and reach
Who can get in, and what each customer shares with nobody.
Isolated per customer
Own process, database, encryption key, files and tunnel network. Large customers can have a dedicated node: their whole stack on a virtual machine of their own.
Sign-in your way
Microsoft, Google, Apple, an e-mailed link that shows a number, username and password, or Active Directory. Each switched on per workspace. Two-step sign-in with an authenticator app, recovery codes and the option to require it.
Roles, permissions, audit
Super Admin and read-only roles, per-section permissions, 24-hour sessions you can end, and an audit trail of every change with who, what and from where.
Retention you control
Raw data plus hourly and daily rollups, kept per data class. Capacity history can run for years.
English and Arabic
A full Arabic interface, right to left, chosen per browser.
Mobile app
The Rapitt app for iPhone and iPad on the App Store, and on Huawei AppGallery. See the badges below.
Zero-touch install
One command detects whether a machine is a server or a workstation and installs the right agent.
Licensing that never stops monitoring
Per device kind; clients are never counted. A lapsed term or a reached cap refuses only the next add.
The platform questions
Is there an uptime guarantee?
Rapitt makes no uptime promise on this site. What it publishes is how the platform is built to keep running, described on this page.
Can a customer run on a server of its own?
Yes. A dedicated node runs one customer’s whole stack on a virtual machine of its own, including its own tunnel network, while updates and licensing stay managed centrally.
Which sign-in methods are supported?
Microsoft, Google, Apple, an e-mailed link that shows a number, username and password, and Active Directory. Each can be switched on or off per workspace, and two-step sign-in with an authenticator app can be required.
How long is data kept?
You set retention per class of data. Raw readings roll up to hourly and daily summaries, so capacity history can be kept for years while raw detail ages out.
See it on your own systems
A walkthrough runs on your own estate. Reserve your workspace address while you book.