Hello,
I have working for 8 months my new storagenode. A someday I saw that I have high disk usage and high I/O wait. Can you help me fix this heavy load on my system? Does it affects to my node responsibility? My Synology DS2413+ is only for sunday pc backups. Rest of it is for Storj docker container.
This setup is 3x 4TB HDD’s with SHR Raid.
Please look on below graphs:
I would recommend disabling atime.
Hello,
Thank for your reply. What is atime?
EDIT: Looked fast on web. This thing is associated with filesystem. I have never seen this before.
Can I change this without backup? Or I need backup changing this option in my filesystem on volume2?
This is my etc/fstab:
/dev/root / ext4 defaults 1 1
/dev/mapper/cachedev_0 /volume2 btrfs auto_reclaim_space,ssd,synoacl,relatime,nodev 0 0
/dev/mapper/cachedev_1 /volume1 btrfs auto_reclaim_space,ssd,synoacl,relatime,nodev 0 0Due to lack of community interest (possibly the problem is rare) I decided to take matters into my own hands and changed the file system from BTRFS to EXT4.
Now I/O wait is 0% much of the time. Only disk usage jumps to 100% for 20 minutes for someday but that’s not a problem.
I think this is the right move. I hadn’t seen your post before and would have suggested this. Btrfs tries to do too much and isn’t useful for storage nodes. It just end up being slow.