The elevated AppGrid API response times between 16:18 - 16:45 CEST today were caused by a system configuration parameter causing an anomalous buildup of internal server logs during heavy load. This has been addressed by rotating in new machines into the cluster which are being gradually updated with the correct parameter.