UPDATE: 17:50
Service has been restored to normal.
UPDATE: 16:00
dadup is currently still recovering. Load seems to be going down a bit, and throughput is going up again. We've ordered more hardware to further spread the load in the near future.
ETA on normal operation: ~ 30 minutes
ORIGINAL MESSAGE:
We've been upgrading dadup to a new version of Ceph during the last week. Today, one of the machines in the cluster is having a hard time which is causing some timeouts on dadup.
We have configured Ceph to try and restore normal services in a calm matter, which has stabilized the situation, but you might still encounter some timeouts while communicating with dadup.
Please note that although dadup is normally fast and very stable, dadup is not intended for high available and fast file storage, but for backups.
We will keep you updated via this notice.
