1

I'm testing 3 servers with RAID0 with 4 disks. During my test I noticed on all 3 servers that one of the hdd's is more %busy and has much higher averige IO wait (almost 4x higher then the other hdd's). It's always the "last" disk on the server. The disks in the servers are all the same model (number) so they should have the same results in RAID0 as far as I understand? (*to be honest I'm not sure if the servers are software raid or hardware raid)

I would like to know if that is normal in a RAID0 system? Why does does one disk have a higher IO wait? I did find a clue, see my last image where it shows the "Request size", it's 2x to 3x smaller then the other disk(s) what does that say? Maybe some setting I can change for the disk? *graphics are created by Munin.

graphics io usage

graphics io usage2

Extra information after requested:

cat /proc/mdstat

Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid0 sda5[0] sdc1[2] sdd1[3] sdb1[1]
      11713146880 blocks super 1.2 512k chunks

unused devices: none


mount

/dev/md0 on / type ext4 (rw,errors=remount-ro)
proc on /proc type proc (rw,noexec,nosuid,nodev)
sysfs on /sys type sysfs (rw,noexec,nosuid,nodev)
none on /sys/fs/cgroup type tmpfs (rw)
none on /sys/fs/fuse/connections type fusectl (rw)
none on /sys/kernel/debug type debugfs (rw)
none on /sys/kernel/security type securityfs (rw)
udev on /dev type devtmpfs (rw,mode=0755)
devpts on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620)
tmpfs on /run type tmpfs (rw,noexec,nosuid,size=10%,mode=0755)
none on /run/lock type tmpfs (rw,noexec,nosuid,nodev,size=5242880)
none on /run/shm type tmpfs (rw,nosuid,nodev)
none on /run/user type tmpfs (rw,noexec,nosuid,nodev,size=104857600,mode=0755)
none on /sys/fs/pstore type pstore (rw)
/dev/sda4 on /tmp type ext4 (rw,noexec,nosuid,nodev)
/dev/sda2 on /boot type ext2 (rw)
systemd on /sys/fs/cgroup/systemd type cgroup (rw,noexec,nosuid,nodev,none,name=systemd)



cat /proc/swaps

Filename                                Type            Size    Used    Priority
/dev/sda3                               partition       3906556 60      -1
Improve this question
9
  • You're running software RAID that's for sure. "mdX" is the common device identified mdadm uses for arrays under its control.
    – dannosaur
    Sep 28, 2014 at 19:59
  • @dannosaur thx! I was thinking, with software raid1 is the OS also installed on all the 4 drives then? If it's installed on only one drive that could be the cause of one disk being more busy then the others. (but just a wild guess)
    – klaasio
    Sep 28, 2014 at 21:24
  • No, the O/S will be on all 4 disks as well, along with the bootloader. They will be exact mirrors of each other. With software RAID, it will use 1 disk to boot, because it's not striped, each disk can operate independently (which is probably why you're seeing one disk more active than the others). Then mdadm kicks in half way through boot. The only difference is the MBR -- I can't remember if this is duplicated. It's not hard though to get grub to re-install a new MBR in the event the primary disk (that the BIOS calls to boot from) fails.
    – dannosaur
    Sep 29, 2014 at 6:42
  • @dannosaur I feel so dumb.. i said RAID1, but my system is RAID0 (it's all new to me so I got confused) as we need high write speed etc. and redundancy is less important. So the same question with RAID0 will the be installed on all the drives then?
    – klaasio
    Sep 29, 2014 at 7:13
  • Much as I hate to disagree with dannosaur, it's not certain that your OS is on all four drives; we'd need to see the output of cat /proc/mdstat, mount, and cat /proc/swaps to start to have a chance of answering that.
    – MadHatter
    Sep 29, 2014 at 7:21

0

Reset to default

You must log in to answer this question.

Browse other questions tagged .