linux - High IO load during the daily raid check which lasts for hours on Debian Jessie? -
i'm experiencing load of 6 during daily raid check:
# cat /proc/mdstat personalities : [raid1] md2 : active raid1 sda3[0] sdb3[1] 2111700992 blocks super 1.2 [2/2] [uu] [=================>...] check = 87.1% (1840754048/2111700992) finish=43.6min speed=103504k/sec bitmap: 2/16 pages [8kb], 65536kb chunk md1 : active raid1 sda2[0] sdb2[1] 523712 blocks super 1.2 [2/2] [uu] resync=delayed
the suspect seems jdbc2:
total disk read : 0.00 b/s | total disk write : 433.45 k/s actual disk read: 0.00 b/s | actual disk write: 902.05 k/s tid prio user disk read disk write swapin io> command 19794 be/3 root 0.00 b 616.00 k 0.00 % 99.46 % [jbd2/loop0-8] 259 be/3 root 0.00 b 96.00 k 0.00 % 87.46 % [jbd2/md2-8] 19790 be/0 root 0.00 b 18.93 m 0.00 % 10.13 % [loop0]
the linux box debian gnu/linux 8.7 (jessie) 4.4.44-1-pve kernel.
almost instantly, when raid check finishes, load returns less one. how can figure out what's causing problem?
i'm not sure how long daily raid check should run, takes several hours, seems excessive.
the io levels drop when raid check has been finished:
total disk read : 0.00 b/s | total disk write : 8.29 m/s actual disk read: 0.00 b/s | actual disk write: 8.63 m/s tid prio user disk read disk write swapin io> command 259 be/3 root 0.00 b 188.00 k 0.00 % 28.80 % [jbd2/md2-8] 19794 be/3 root 0.00 b 720.00 k 0.00 % 28.65 % [jbd2/loop0-8]
this problem doesn't seem make sense me. further debugging useful.
Comments
Post a Comment