The database restart is now complete and errors are now resolved.

We're experiencing increased write errors in one of our main DBs. Replicase failover worked properly and read replicase are working as expected. We're monitoring the issue to provide resolution, and our engineering team will continue working on improving failover for write operations as well.