This incident has been resolved.
Nov 13, 00:24 EET
All sites are back up. We are continuing to monitor the situation.
Nov 12, 23:43 EET
We are continuing to monitor for any further issues.
Nov 12, 22:55 EET
The sites are recovering and we are currently prioritizing live sites so dev sites might take some time to recover. Thank for your patience!
Nov 12, 22:54 EET
The source of these issues was a latent bug, which quickly scaled down to 0 than back up the backends. We've reverted the faulty deploy and we are waiting for the sites to recover.
Nov 12, 22:37 EET
A set of cascading events led to backends being restarted. This in turn puts a heavy load on the git service. We are working on reducing the load on the git service so that the backends can be restarted.
Nov 12, 22:30 EET