This incident was caused by an outage at our primary compute provider, GalaxyGate.
Circle hosts our critical infrastructure (database, session management, gateway [in dev], and cache management) on several other hosting providers on the East Coast of the US.
We use GalaxyGate for our main services like the bots and API, which is why this outage caused all of our services to drop offline. We were able to quickly restore Premium services by enabling a backup server at Digital Ocean, regular Circle has more moving parts and is not as easy to startup on different infrastructure on demand. By the time we were beginning to setup regular infrastructure on Digital Ocean, the GalaxyGate issue was resolved.
This was GalaxyGate’s first issue that affected us since 2022 and we do not have plans to move our infrastructure away at this time.