Using *top *and looki at *'wa' *value can tell you I/O wait time for each CPU Dont forget to press "*1*" to expand list of CPUs
Tasks: 501 total, 4 running, 497 sleeping, 0 stopped, 0 zombie Cpu0 : 31.9%us, 52.7%sy, 0.0%ni, 15.1%id, *0.0%wa*, 0.0%hi, 0.3%si, 0.0%st Cpu1 : 29.7%us, 7.6%sy, 0.0%ni, 62.1%id, *0.0%wa,* 0.0%hi, 0.7%si, 0.0%st
Also, there is handy tool called: *iotop *which can tell how much process writes and read. We can see on our 6Gpbs SATAIII interface with SSD disks, the interfaces is being maxed out with writes at ~500MBs
At the end 7.2k disks can be easily maxed out while running a few VMs so no surprise here.
Lastly, setup some monitoring for example munin, its quite handy :
http://demo.munin-monitoring.org/disk-day.html
On Mon, Feb 8, 2016 at 1:58 PM Gokan Atmaca linux.gokan@gmail.com wrote:
Are the disk partitions properly aligned to 4k boundary on the host (and
in
the guests too) ?
There are 5 in total server. 32G ram. 2T r1 (soft) disk.
On Mon, Feb 8, 2016 at 9:41 PM, Zoltan Frombach zoltan@frombach.com wrote:
Are the disk partitions properly aligned to 4k boundary on the host (and
in
the guests too) ?
See
http://www.ibm.com/developerworks/library/l-linux-on-4kb-sector-disks/index....
and this:
http://unix.stackexchange.com/questions/247387/check-if-partitions-are-align...
On 2/8/2016 8:12 PM, Gokan Atmaca wrote:
Hello
I use KVM. In a virtual machine "jbd2 dm-0" disk I / O is very increases. It consumes up to 99%. For this reason, slowing down the other virtual machine. What should I do to solve the problem. ?
Thanks.. _______________________________________________ CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt
CentOS-virt mailing list CentOS-virt@centos.org https://lists.centos.org/mailman/listinfo/centos-virt