0

I'm experiencing a 99% IO load caused by [md0_raid1] process every 20-30 seconds. That 99% IO is very short-term, about 1-2 seconds, but it causes lag for other processes due to io wait (it's very harmful for gaming server). Is this normal behaviour? How can I avoid these high IO lags? I would be grateful for any advice.

I have SW RAID 1 (Ext4) on new Linux Debian 9.3 server, I run apache/mysql webserver and several Minecraft servers.

iotop:

Total DISK READ :       0.00 B/s | Total DISK WRITE :      15.30 K/s
Actual DISK READ:       0.00 B/s | Actual DISK WRITE:        0.0 B/s
  TID  PRIO  USER     DISK READ  DISK WRITE  SWAPIN     IO>    COMMAND
  199 be/4 root        0.00 B/s    0.00 B/s  0.00 % 99.99 % [md0_raid1]
45302 be/4 root        0.00 B/s    3.82 K/s  0.00 %  0.00 % java -Xms~spigot.jar
51303 be/4 root        0.00 B/s    3.82 K/s  0.00 %  0.00 % java -Xms~spigot.jar
51306 be/4 root        0.00 B/s    3.82 K/s  0.00 %  0.00 % java -Xms~spigot.jar
51200 be/4 root        0.00 B/s    0.00 B/s  0.00 %  0.00 % java -Xmx~ee-private
    1 be/4 root        0.00 B/s    0.00 B/s  0.00 %  0.00 % init
    2 be/4 root        0.00 B/s    0.00 B/s  0.00 %  0.00 % [kthreadd]
    3 be/4 root        0.00 B/s    0.00 B/s  0.00 %  0.00 % [ksoftirqd/0]
    5 be/0 root        0.00 B/s    0.00 B/s  0.00 %  0.00 % [kworker/0:0H]

mdstat:

md1 : active raid1 sda5[0] sdb5[1]
    33501184 blocks super 1.2 [2/2] [UU]

md0 : active raid1 sda1[0] sdb1[1]
    943111168 blocks super 1.2 [2/2] [UU]
    bitmap: 5/8 pages [20KB], 65536KB chunk

HW specs:

Intel Xeon 1231v3 (3.4 Ghz) 4C/8T 8MB
32 GB DDR3 ECC 1600 MHz
2x SATA 6G 1TB 7.2k (SW RAID 1)

lspci specs:

https://gist.github.com/FreeWall/98f4bc8029387dbe4d47feacdd37aa85

2
  • What filesystem are you using on that RAID?
    – Jan
    Commented Jun 12, 2018 at 11:40
  • i'm using Ext4 filesystem
    – FreeWall
    Commented Jun 12, 2018 at 11:50

1 Answer 1

0

I'd suggest to check for system messages by using command:

dmesg

If it shows nothing the post more verbose information regarding hardware by using command:

lspci

Note that this may be also sign of hardware issue, if this is not rack server but ordinary PC then I'd suggest also to replace sata wires for disks having this issue.

2
  • Thank's for you reply. I already checked the logs, didn't find anything relevant. There is output from "lspci -v", don't know if it's useful. link
    – FreeWall
    Commented Jun 17, 2018 at 16:02
  • I'd suggest to check SATA cables and verify that these were made for 6Gb/s. 3Gb/s may work as well as connection is exactly same, however key aspect is cable's quality (shielding etc). I've seen many issues (both performance issues and not detected disks at all) caused by deformed connector on Sata cables. These cables comes without latches on connectors and connectors may bent over time by heat. It all may sound silly but I've seen performance issues caused by this.
    – Mr.Holub
    Commented Jun 23, 2018 at 9:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .