Discord has been operating normally for more than an hour. All graphs are nominal and we understand what happened. The very brief synopsis is that we had table contention on the MongoDB instance that underlies a portion of our system. We do not believe this particular incident will recur in the near future.
Mar 17, 15:49 PDT
The Redis server is back to normal. The team is again monitoring for any ongoing issues while performance returns to normal.
Mar 17, 14:05 PDT
There is an ongoing issue with one of our Redis machines that powers certain operations that happen when messages are sent. We are working through the issue now.
Mar 17, 13:50 PDT
We've addressed the underlying issue and services are recovering. Everybody should be able to reconnect and use Discord normally, but some operations may be a little slow as the system fully recovers. The team is keeping an eye on things.
Mar 17, 13:12 PDT
We've identified that a MongoDB database is performing badly. We're taking steps to reduce load against the cluster now.
Mar 17, 12:51 PDT
We've become aware of an issue affecting connecting to Discord. We're looking into it now and will update as soon as we know more.
Mar 17, 12:34 PDT