The cloud provider changed the limits of its free plan and stopped, without warning, the machine running Axis (then called Argos), ONYX and the website. It was not an attack or a bug.
About an hour and a half without the bot in Discord, and about eight hours without web verification, appeals or the panel. Anyone who joined a server during that time could not verify. Afterwards the bot was slower for a few hours. No data was lost.
We recovered the data from the disk, set up a new machine within the new limits, restarted the bot and ONYX, and moved the website to a new address.
There is now a second backup node that takes over if the main one goes down. And this page records every outage on its own, even when nobody is watching.
Everything is working and links now point to the new address. No data was lost.
Web verification, appeals and the panel are back with a new address. We are updating the links.
The new machine has fewer resources and the bot is responding slower than usual.
ONYX is working again. Still working on the website.
The bot is back online in Discord from a new machine. All data is intact. The website is still down while we move it.
The main server stopped responding: the provider stopped the machine Axis runs on. The bot, ONYX, web verification, appeals and the panel are down.
Every incident that lasts 15 minutes or more, or is serious, gets a public report within 48 hours.