The Gandi Community

hosting storage incident

An incident has occured on one of our EQX (Paris) storage nodes, staff is on-site to resolve the issue as soon as possible. We’ll keep you informed with this issue.

EDIT @2:00am CEST: we’re currently trying to resolve an important raid controller issue with the manufacturer support team

EDIT @2:50am CEST: no solution found with support yet, we’ll proceed to the hardware replacement of a storage array

EDIT @3:10am CEST: trying to rebuild the raid array now. 1h30 estimated time for more information on the raid status, and we’ll have to reboot the affected servers to ensure data integrity.

EDIT @5:00am CEST: rebuild is done. A read-only check was made on the disks and they all look clean. We are now starting the affected servers. Please ro check your filesystem (fsck -n), if possible do a safe full fsck, and take a fresh backup of your server in case an issue arises. We still follow this incident.

EDIT @9:00am CEST: issue is resolved and the storage array is stable. We’ll close this incident, and handle individual request at the support level.

Sorry for the unusual delays in the handling of this major issue, and thank you for your patience.