Platform resilience · plain-language overview
Everything runs on one main server. Behind the scenes, copies of the data flow out continuously to a second server and to outside storage, so the whole thing can be brought back after almost any kind of failure.
Runs every website, the chat platform, and the live database that holds all customer and lead data.
production-01 · Ashburn, VirginiaNobody has to remember to do any of this. The main server keeps four separate safety nets fed on their own schedules:
A second server keeps a near-instant copy of the database, ready to take over. Website files and security certificates copy across every half hour.
cwad-standby-01 · Hillsboro, OregonA full, scrambled copy of the data and every website is sent nightly to an unrelated storage company. Kept for months, so you can go back in time.
Backblaze cloud storageA handful of encryption keys can never be recreated — lose them and that data is gone forever. They are copied off-server automatically each day.
iCloud + password managerA monitor checks that the standby copy is keeping up and that services are healthy. If anything slips, it emails an alert right away.
Automated health checksThe main server goes downHardware dies, data center outage, anything that takes it fully offline.
Run one failover command on the standbyIt promotes its database copy, starts up all the apps, and switches web traffic over — automatically.
Data is lost or corruptedA bad change or corruption that the live copy would just mirror — the server itself is fine.
Restore from the nightly cloud backupPull back a clean copy from before the problem, and restore the encryption keys from their off-server copy.