January 2, 20242 yr For more than six months I had been experiencing frequent crashes of my unRAID server which I was unable to resolve. Every 1-3 days the server would lock up. I completed a lot of troubleshooting, including testing the memory, wiping the cache and reinitialising (multiple times), combing through all the logs I could possibly find, scrubs; even a docker elimination test where I tried turning them on one by one. This eventually led to corruption in my application databases (which has been difficult to correct). Most recently everything shit the bed so hard I had to spend multiple days repairing corrupt databases by hand (the automated SQL repairs did not work). So I wiped the cache again and this time formatted with ZFS. We're on day seven now without any crashes. The system is responsive and I'm not detecting any more file system errors in logs. I have no idea why ZFS is working but BTRFS did not. Perhaps it's more resilient? I'm too tired tired to keep fighting. It works and I'm happy with that. I'm writing this because I've read dozens of other reports of users experiencing the same issues as I was. If so, ZFS on cache could resolve your issue. I'm using mirror mode (two cache SSDs mirrored).
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.