VMmanager 5 KVM Documentation

Cluster load

ISPSystem


A cluster in VMmanager is a group of physical servers where virtual machines are running.

To viewtheclusternodestatistics navigate to Statistics→ Cluster load.

Generate a report




  • Period — select a period:
    • previousday;
    • current day;
    • current week;
    • current month;
    • current year;
    • previous week
    • previous month;
    • previous year;
    • week;
    • month;
    • quarter;
    • half-a-year;
    • year;
    • any period;
    • the whole period.
  • Cluster nodes — select a cluster node to display the statistics.

You can monitor the load on each virtual machine based on the following resources:

    • incoming traffic, GiB;
    • outgoing traffic, GiB;
    • input-output operations, IOPS;
    • CPU, %;
    • RAM, %. 

Note

Per-minute statistics is kept for 5 days. Per-hour statistics is kept for one month and 3 days. Per-day statistics is kept for 2 years. When the specified periods expire, the data will be deleted.

Graphs for a random period are generated based on a per-hour statistics.

How the panel collects statistics


Raw statistics is kept in the file "/usr/local/mgr5/var/stat/raw" and is updated every 5 minutes. Once an hour the control panel processes the raw statistics and delete it from the directory. The archive of the raw statistics is kept in "/usr/local/mgr5/var/stat/mon/mainstat/".

Note

You can delete the archive of raw statistics only by means of the operating system.

Incoming and outgoing traffic

The information is collected from the /proc/net/dev file. Statistics files are copied to /var/stat/raw_bandwidth.

Receive Bytes — incoming traffic.

Transmit Bytes — outgoing traffic.

The main network interface of the cluster node is selected as the interface automatically.

Note

Network statistics are collected from physical and bonding interfaces. Statistics are not collected from virtual network interfaces.

CPU 

The information is collected from the virNodeGetCPUStats function of the libvirt library (the virsh nodecpustats command on the cluster node).

The CPU usage is calculated as follows:

(user+system) / (user+nice+system+iowait) * 100
 View details
user - percentage of CPU utilization that occurred while executing at the user level.
iowait — time spent by the CPU waiting for a IO operations to complete.
system - percentage of CPU utilization that occurred while executing at the system level.
idle - percentage of time that the CPU or CPUs were idle.

I/O operations

The information is collected from the /proc/diskstats file.

The usage is calculated as follows:

NumReads+NumWrites
 View details

NumReads — field 1 of the file /proc/diskstats(Field 1 -- # of reads issued).

NumWrites — filed 5 of the file /proc/diskstats(Field 5 -- # of writes completed)

.

The system calculates the data from all hard drives.

RAM

The information is collected from the virNodeGetMemoryStats function (the virsh nodememstats command).

The usage is calculated as follows:

MemUsed / MemTotal * 100
 View details

MemUsed = MemTotal - MemFree.

MemFree — free + cached.

MemTotal — the total field.

Thank you for your feedback!
Was the article useful?
Tell us why not:

it does not match the title

it was hard to understand

I didn't find the answer

recommendations from the article did not help

other

Tell us what you didn't like in the article: