Hacker News Reader

I have a small 4 node home cluster, and longhorn works great... on smaller volumes.

I have a 15TB volume for video storage, and it can't complete any replica rebuilds. It always fails at some point and then tries to restart.

View on HN

That is good to know then, I am really just using this for smaller volumes. My media is sitting at about the same size yours is and instead of using PVC's I just have it mounting a straight NFS share specifically for that to avoid any issues there.

I think I am likely keeping most of my storage just setup with a storage class that uses my NFS as storage. But longhorn will be used for the things that need to be faster like the databases. I moved jellyfin over to Longhorn and it went from being borderline unusable while metadata was grabbed to actually working well.

I can't imagine my biggest volume being more than 100gb, and even that is likely a major over estimation on my part.

Reply on news.ycombinator.com