When a server fails, surveillance cannot stop. Visylix keeps a standby server in step with the primary, and an operator brings it live in under 60 seconds, with cameras, AI and the Radha copilot all back on line.
Request HA Architecture ReviewFrom continuous copying to rehearsed recovery drills, built for control rooms that cannot go dark.
Users, settings, roles and evidence locks are copied to the standby continuously, and recordings sit on shared storage both servers can read.
When the primary fails, an operator brings the standby live with one documented action. Within a server, recording and housekeeping tasks move to a healthy worker on their own, and larger clusters move cameras off a failed streaming server.
Visylix watches CPU, memory, disk, services and camera connections, with limits you set for each environment.
Practise a cutover from the documented runbook whenever your policies require, without an outage.
Works with the network equipment and services you already use.
Recordings are written to shared storage, so footage saved before a failure is already available to the standby.
Works with common DNS services and load balancers, from Route 53 and Cloudflare to HAProxy, F5 and AWS NLB.
Primary and standby both live inside your air-gapped network, with no internet needed.
Users, audit records and evidence locks reach the standby before each change is confirmed, so the records that matter in court are never behind.
Visylix brings every camera, every analytic and Radha back on the standby in under a minute, with evidence and audit records already in sync.
On Enterprise, Visylix pairs a primary server with a standby. Users, settings, audit records and evidence locks are copied to the standby continuously, and recordings sit on shared storage that both servers can read. If the primary fails, an operator brings the standby live in under 60 seconds.
Cutover is operator-initiated and takes under 60 seconds. Video, AI and the Radha copilot all come back on the standby. Health checks on CPU, memory, disk and services warn your team early, so they can act before a problem becomes an outage.
A person confirms the switch, so a brief network blip never causes two servers to think they are in charge. The cutover itself is one documented action and completes in under 60 seconds.
Yes. Primary and standby can both sit inside your air-gapped network. Nothing needs internet access, and licensing works fully offline.
Yes. A documented drill moves work off the primary, brings the standby live and switches back once you are satisfied, so your team can practise disaster recovery as often as your policies require.
Yes. Evidence locks, SHA-256 seals, audit records, roles and case data are copied to the standby before each change is confirmed, so the chain of custody carries straight through a cutover.