o be honest I am a little concern that I am not sure what caused all the filesystem to go crazy. After a fsck everything seems fine but I am not going to trust it. Therefor I am planning on rebuilding the server and migrating to a new VM. You have to remember how many times we have shuffled these qcow2 to one machine to another and moved across the internet. Therefor a fresh machine just seems right to plan at lease once we get our DB 100%

@omnipotens I read somewhere RAID no longer assures correctness. That it is better to use a logical volume manager or some format with checksumming and redundancy built-in like zfs or btrfs. That RAID -- especially hardware RAID -- has been responsible for an increasing frequency of data corruption.

@nergal I dont think it was a raid issue but a qcow2 /volume group/ snapshot issue to be honest

@omnipotens oh. Wish there was a reed-solomon or such parity checksum to help verify...

Sign in to participate in the conversation

Linux Geeks doing what Linux Geeks do..