At 10:41 AM PST our platform notified us of service instability in DC4. Our engineers began investigating.
At 10:58 AM PST we have identified high temperature in the datacenter causing servers to significantly reduce clock speed to avoid overheating. Issue was escalated to the datacenter immediately.
At 11:20 AM PST temperatures in the datacenter have been reduced by on-site staff and we are working on bringing all services back.
At 11:47 AM PST datacenter confirmed cooling system failure; and cooling was already restored.
At 12:02 PM PST we have restored all services. Remaining VMs will start automatically in the next 20 minutes.