Get the latest tech news
Making Machines Move
Gather round the fire to hear the tale of how we built migration for stateful apps on a global public cloud without building a SAN.
Even to a rack buddy physical server, you’ll hit a point where draining incurs minutes of interruption, especially if you’re moving lots of volumes simultaneously. To make that work, we need the target worker to see the plaintext of the source volume (so that it can do an fstrim — don’t get us started on how annoying it is to sandbox this — to read the filesystem, identify the unused block, and issue the DISCARDs where dm-clone can see them) Easy enough. This engineering work belongs to Shaun Davis, Simon Horne, Dov Alperin, Ben Ang, Dusty Hall, Saleem Rashid, and like 9 other people I’m going to have to apologize to later.
Or read this on Hacker News