I'm getting very high cpu usage and IO wait when I try to download to my cache drive. system load skyrockets and download speed drops from 100mb/s to 5-10k/s.
I have my dockers and download folder on a BTRFS cluster of 2 cache drives.
I originally thought it was the sonarr process of moving the files to the array that was killing things, but this is happening well before the download completes.
Prior to this, I have had no issues downloading at full speed.
load shows:
15:14:41 up 16 days, 1:20, 1 user, load average: 9.05, 6.95, 3.65
USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
root pts/0 10.0.1.154 14:57 1.00s 0.01s 0.00s w
I've seen load hit 25+ over the past few days when I download larger files
iotop is showing me:
Total DISK READ : 0.00 B/s | Total DISK WRITE : 94.77 K/s
Actual DISK READ: 0.00 B/s | Actual DISK WRITE: 54.18 M/s
TID PRIO USER DISK READ DISK WRITE SWAPIN IO> COMMAND
17449 be/4 root 0.00 B/s 0.00 B/s 0.00 % 99.99 % shfs /mnt/user -disks 1023 -o noatime,allow_other -o remember=330
15101 be/4 root 0.00 B/s 0.00 B/s 0.00 % 86.57 % [kworker/u32:10+btrfs-worker]
19429 be/4 root 0.00 B/s 14.58 K/s 0.00 % 6.41 % shfs /mnt/user -disks 1023 -o noatime,allow_other -o remember=330
26760 be/4 root 0.00 B/s 72.90 K/s 0.00 % 0.00 % [kworker/u32:5-bond0]
18835 be/4 root 0.00 B/s 7.29 K/s 0.00 % 0.00 % shfs /mnt/user -disks 1023 -o noatime,allow_other -o remember=330
Was going to try and remove a cache drive and switch to a single ssd as a next step?
homer-diagnostics-20220109-1513.zip