elevated latency/errors for a subset of customers

Final update

This incident has been resolved.

1. Resolved · Oct 31, 07:52 UTC
   This incident has been resolved.
2. Monitoring · Oct 31, 06:00 UTC
   We are continuing to monitor for any further issues.
3. Monitoring · Oct 31, 05:59 UTC
   There will be sporadic availability issues for a small subset of non-pro customers. We apologize for this impact. There was an issue with a downstream provider which resulted in a database cluster running with an inefficient table format. We're in the process of rewriting this data format but have had to shed load to provide headroom while reoptimizing the database.
4. Monitoring · Oct 31, 04:24 UTC
   All deployments should be restored. We are monitoring the database to ensure it remains stable.
5. Identified · Oct 31, 03:08 UTC
   We're continuing to work on a fix for the issue. Expect another update in ~90 minutes.
6. Identified · Oct 31, 00:25 UTC
   we've brought back a subset of free deployments. Continuing to remediate.
7. Identified · Oct 30, 23:44 UTC
   We've restored affected pro deployments. Working on bringing back affected free deployments.
8. Identified · Oct 30, 23:43 UTC
   We're investigating an issue with a database cluster leading to degraded performance and higher error rates for affected deployments.