eG Monitoring
 

Measures reported by OSHvVmTest

In OpenStack, virtual machines (VMs) are created and managed within the projects. The OpenStack KVM hypervisor mapped to the projects allocates the resources to the VMs in the projects. This test monitors the virtual machine (VM) instances and reports the current status of each VM. This test also reveals the CPU time, memory related statistics such as memory utilization, swap in/swap outs, memory faults that were detected etc. Using the metrics reported by this test, administrators can easily figure out which VM is waiting for a longer time to receive CPU allocation, which VM is generating the most network traffic, which VM is over-utilizing memory, and which VM is encountering more errors/packet drops while trasmission/reception.

Outputs of the test: One set of results for each virtual machine on the target hypervisor.

Descriptor: Virtual machine

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Power_state Indicates the current power state of this VM.  

The values this measure reports and their numeric equivalents are provided in the table below:

Measure Value Numeric Value
NoState 0
Running 1
Paused 3
Shutdown 4
Crashed 6
Suspended 7

Note:

By default, this measure reports the Measure Values listed in the table above. In the graph of this measure however, the power state of each VM is indicated by its corresponding numeric equivalent only.

Vm_state Indicates the current state of this VM.  

A high value for this measure indicates that the memory resources are depleting drastically. Administrators may be alerted to add additional resources before memory resources are drained completely.

Measure Value Numeric Value
Active 100
Building 90
Error 85
Paused 80
Suspended 75
Resumed 70
Resized 60
Shelved_Offloaded 50
Shelved 40
Soft_Deleted 30
Deleted 20
Stopped 10

Note:

Typically, this measure reports one of the Measure Values listed in the table above. In the graph of this measure however, the power state of each VM is indicated by its corresponding numeric equivalent only.

CPU_time Indicates the amount of time that the vCPU of this VM is waiting for the hypervisor to allocate a physical CPU to it. Seconds The value of this measure should be low.
VDA_errors Indicates the number of errors logged in the event log while performing operations on this VM via the VDA. Number Ideally, the value of this measure should be 0.
Total_memory Indicates the total amount of memory available to this VM. MB  
Free_memory Indicates the amount of free memory available for use to this VM. MB Compare the value of this measure across VMs to know which VM is running out of free memory.
Used_memory Indicates the amount of total memory that is currently being used by this VM. MB  
Memory_usage Indicates the percentage of total memory currently being utilized by this VM. Percent Compare the value of this measure across VMs to know which VM is using memory excessively.
Memory_swap_in Indicates the amount of memory swapped in to the disk on this VM from the hypervisor. MB  
Memory_swap_out Indicates the amount of memory swapped out from the disk on this VM to the hypervisor. MB  
Memory_major_fault Indicates the number of major page faults occurred on this VM. Number

A major page (memory) fault is an exception that occurs when a process attempts to access memory in a way that exceeds its permissions.

A very low or zero value is preferred for this measure.

Memory_minor_fault Indicates the number of minor page faults occurred on this VM. Number

The minor page (memory) faults mostly occur due to either a bad sector on the disk, which causes the operating system to stop reading data from the hard drive, starting over again or a corrupted file.

A low value is preferred for this measure.

VDA_read_req Indicates the number read requests received by this VM. Number  
VDA_write_req Indicates the number of write requests received by this VM. Number  
VDA_read Indicates the amount of data read from the disk on this VM. KB  
VDA_write Indicates the amount of data written on the disk on this VM. KB  
Reads Indicates the rate at which the data was received by this VM. Kbps A sudden increase or decrease in the value of this measure could be a cause of concern.
Data transmitted rate Indicates the rate at which the data was transmitted by this VM. Kbps A sudden increase or decrease in the value of this measure could be a cause of concern.
Read_packets Indicates the number of packets received by this VM. Number Indicates the number of packets received by this VM.
Write_packets Indicates the number of packets transmitted by this VM. Number  
Read_errors Indicates the number of errors occurred while data was received by this VM. Number Ideally, the value of this measure should be zero.
Write_errors Indicates the number of errors occurred while data was transmitted by this VM. Number Ideally, the value of this measure should be zero.
Read_drops Indicates the number of packets dropped by this VM during reception. Number Ideally, the value of this measure should be zero.
Write_drops Indicates the number of packets dropped by this VM during transmission. Number Ideally, the value of this measure should be zero.