Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

How much did you deal with snapshots? Because I've had crippling performance problems stemming from them. Even something as simple as having 90 daily snapshots and deleting the oldest few can cause trouble, where the filesystem does not respond to any requests for multiple seconds. And that's on an SSD. I don't remember if it was deleting snapshots or running balance, but I've had Btrfs on a hard drive not respond to I/O requests for two minutes. A light-use server that had snapper running for a few months fragmented so badly that it regularly hitched up even after snapshots were paused. I had to migrate the entire filesystem.

I'm still using Btrfs on my backup system, but that's only because I like the dedup enough to overlook the brief hangs.



Not only that but in conjunction with journald you can achieve amazing disk space leaks that cannot be repaired easily without losing data.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: