Automated monitoring noted that our marketing site was returning errors at 03:05 UTC, and systems engineers were paged to investigate. It turned out that the marketing errors were actually due to timeouts talking to the Core service, and the errors had resolved themselves in the ~10 minutes it took us to come to that conclusion. With further digging we ascertained that our upstream provider had seen some routes dropped and it took about 10 minutes for the routing to reconverge on their side.
This appears to have been an intermittent routing failure, so it is not guaranteed that you will have seen any affect in your interactions with Spreedly. However, if you did get timeouts during that period, the routing issue will have been the cause.
Posted about 2 years ago. Jul 22, 2017 - 23:57 EDT