2

I have two 3TB drive that are formated as btrfs and mounted on /media/andy/btrfs as one drive.

I run btrfs balance start -dconvert=raid1 -mconvert=raid1 /media/andy/btrfs/ which takes forever.

It outputs ERROR: error during balancing '/media/andy/btrfs/' - Input/output error There may be more info in syslog - try dmesg | tail

There's an output there, it's very long.

Sure enough, I try btrfs filesystem df /media/andy/btrfs

I get

Data, RAID1: total=1.14TiB, used=1.13TiB Data, single: total=106.01GiB, used=94.40GiB System, DUP: total=8.00MiB, used=184.00KiB System, single: total=4.00MiB, used=0.00B Metadata, RAID1: total=1.00GiB, used=1017.65MiB Metadata, DUP: total=1.00GiB, used=441.45MiB Metadata, single: total=8.00MiB, used=0.00B GlobalReserve, single: total=488.00MiB, used=0.00B

So sure enough I have 106 GiB that's not in raid. Anyone has an idea how to fix this? I'm running Debian testing. Kernel 3.16.7

  • 1
    Can you run sudo btrfs fi usage /media/andy/btrfs/ and add the output to your question? Sorry I just checked the date, I'm quite late... – Falk May 19 '17 at 21:14
  • Time flies. The problem was most likely caused by my kernel being 3.16.7 at the time. BTRFS has changed a lot since. Might even become stable for raid-1 in the next couple decades. Seriously, it's super fast when it works, but it rarely works for more than a few months before catastrophically self-destroying itself. Great for live, but have a real backup elsewhere. On a basic array of mechanical HDDs, I regularly have 2-3 GB/s write speeds. Amazing! Sure beats GoogleFS write speeds of 10 MB/s per array. But GoogleFS doesn't crash. BTRFS does. – Andy Chow Jan 12 '18 at 7:47
  • You are right, btrfs has changed a lot. I'm using a raid1 on my laptop for months and it works just fine. I'm waiting for raid56, but this doesn't seem to change. – Falk Jan 12 '18 at 7:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.