Virtual instances - Los Angeles, DC2Node v7605 SSD failure
Our monitoring reported an SSD failure in a redundant RAID array. This has not affected any of the client's virtual machines, however due to the nature of this specific failure we will have to migrate all clients off of node v7605 in order to perform repairs.
Migrations have started at 16:32 PST, and will continue in the next 24 hours.
Affected VMs will not be restarted during the migration as migrations are performed without VM reboot. However, we estimate that approximately 1% of VMs will need to be rebooted during migration. More specifically, if you removed kvm_ga agent daemon then the migration will result in a VM reboot.
Update: All client's VMs have been successfully migrated away from the affected node and we will be closing this incident.