Virtual machine that hosts master of caching service has crashed due to unidentified reason.
This virtual machine was also affected in previous outage, it seems that secondary node couldn't host it.
This resulted in a brief downtime (3-5 minutes). Some of background services were degraded until all slaves synced (lasted for about hour and a half):
- background tasks of all sorts
- fetching data from 3rd party services (eg: payment)
- sending emails
All services are fully operational.
Timetable:
- 10:40 CET
- 04:40 EST
- 09:40 UTC