Latency is now back to normal and Redis is correctly serving cached data again. The Redis cluster died due to corruption of the data log and was not auto-repaired due to persistence being enabled. A patch has been applied to ensure persistence is disabled.
Posted Aug 28, 2021 - 16:50 UTC
A fix has been implemented and we are monitoring the results.
Posted Aug 28, 2021 - 16:47 UTC
Persistence should not be enabled on the cluster. We are looking at issues with the helm chart that is erroneously enforcing this.
Posted Aug 28, 2021 - 16:45 UTC
We are aware that our redis cluster ran out of memory due to a sudden spike in traffic. The site may be running slowly as we restart the cluster.
Posted Aug 28, 2021 - 16:38 UTC
We're experiencing an elevated level of API errors and are currently looking into the issue.