2.7. Statistics History Views

Statistics data is available in hourly, daily, and samples views.

To query a statistics view, run SELECT * FROM view_name_[hourly|daily|samples];. For example: 

# SELECT * FROM v4_3_statistics_hosts_resources_usage_daily;

To list all available views, run: 

# \dv

2.7.1. Enabling Debug Mode

You can enable debug mode to record log sampling, hourly, and daily job times in the /var/log/ovirt-engine-dwh/ovirt-engine-dwhd.log file. This is useful for checking the ETL process. Debug mode is disabled by default.

  1. Log in to the Manager machine and create a configuration file (for example, /etc/ovirt-engine-dwh/ovirt-engine-dwhd.conf.d/logging.conf).

  2. Add the following line to the configuration file: 

    DWH_AGGREGATION_DEBUG=true

  3. Restart the ovirt-engine-dwhd service: 

    # systemctl restart ovirt-engine-dwhd.service
    To disable debug mode, delete the configuration file and restart the service.

2.7.2. Storage Domain Statistics Views

Table 2.2. Historical Statistics for Each Storage Domain in the System

NameTypeDescriptionIndexed

history_id

bigint

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history row (rounded to minute, hour, day as per the aggregation level).

Yes

storage_domain_id

uuid

Unique ID of the storage domain in the system.

Yes

storage_domain_status

smallint

The storage domain status.

No

seconds_in_status

integer

The total number of seconds that the storage domain was in the status shown state as shown in the status column for the aggregation period. For example, if a storage domain was "Active" for 55 seconds and "Inactive" for 5 seconds within a minute, two rows will be reported in the table for the same minute. One row will have a status of Active with seconds_in_status of 55, the other will have a status of Inactive and seconds_in_status of 5.

No

minutes_in_status

numeric(7,2)

The total number of minutes that the storage domain was in the status shown state as shown in the status column for the aggregation period. For example, if a storage domain was "Active" for 55 minutes and "Inactive" for 5 minutes within an hour, two rows will be reported in the table for the same hour. One row will have a status of Active with minutes_in_status of 55, the other will have a status of Inactive and minutes_in_status of 5.

No

available_disk_size_gb

integer

The total available (unused) capacity on the disk, expressed in gigabytes (GB).

No

used_disk_size_gb

integer

The total used capacity on the disk, expressed in gigabytes (GB).

No

storage_configuration_version

integer

The storage domain configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_storage_domains view and it can be used to join them.

Yes

2.7.3. Host Statistics Views

Table 2.3. Historical Statistics for Each Host in the System

NameTypeDescriptionIndexed

history_id

bigint

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history row (rounded to minute, hour, day as per the aggregation level).

Yes

host_id

uuid

Unique ID of the host in the system.

Yes

host_status

smallint

  • -1 - Unknown Status (used only to indicate a problem with the ETL. Please notify Red Hat Support. )
  • 1 - Up
  • 2 - Maintenance
  • 3 - Problematic

No

seconds_in_status

integer

The total number of seconds that the host was in the status shown in the status column for the aggregation period. For example, if a host was up for 55 seconds and down for 5 seconds during a minute, two rows will show for this minute. One will have a status of Up and seconds_in_status of 55, the other will have a status of Down and a seconds_in_status of 5.

No

minutes_in_status

numeric(7,2)

The total number of minutes that the host was in the status shown in the status column for the aggregation period. For example, if a host was up for 55 minutes and down for 5 minutes during an hour, two rows will show for this hour. One will have a status of Up and minutes_in_status of 55, the other will have a status of Down and a minutes_in_status of 5.

No

memory_usage_percent

smallint

Percentage of used memory on the host.

No

max_memory_usage

smallint

The maximum memory usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

ksm_shared_memory_mb

bigint

The Kernel Shared Memory size, in megabytes (MB), that the host is using.

No

max_ksm_shared_memory_mb

bigint

The maximum KSM memory usage for the aggregation period expressed in megabytes (MB). For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

cpu_usage_percent

smallint

Used CPU percentage on the host.

No

max_cpu_usage

smallint

The maximum CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

ksm_cpu_percent

smallint

CPU percentage ksm on the host is using.

No

max_ksm_cpu_percent

smallint

The maximum KSM usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

active_vms

smallint

The average number of active virtual machines for this aggregation.

No

max_active_vms

smallint

The maximum active number of virtual machines for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

total_vms

smallint

The average number of all virtual machines on the host for this aggregation.

No

max_total_vms

smallint

The maximum total number of virtual machines for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

total_vms_vcpus

integer

Total number of vCPUs allocated to the host.

No

max_total_vms_vcpus

integer

The maximum total virtual machine vCPU number for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

cpu_load

integer

The CPU load of the host.

No

max_cpu_load

integer

The maximum CPU load for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

system_cpu_usage_percent

smallint

Used CPU percentage on the host.

No

max_system_cpu_usage_percent

smallint

The maximum system CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

user_cpu_usage_percent

smallint

Used user CPU percentage on the host.

No

max_user_cpu_usage_percent

smallint

The maximum user CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

swap_used_mb

integer

Used swap size usage of the host in megabytes (MB).

No

max_swap_used_mb

integer

The maximum user swap size usage of the host for the aggregation period in megabytes (MB), expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

host_configuration_version

integer

The host configuration version at the time of sample. The host configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_hosts view and it can be used to join them.

Yes

2.7.4. Host Interface Statistics Views

Table 2.4. Historical Statistics for Each Host Network Interface in the System

NameTypeDescriptionIndexed

history_id

bigint

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history view (rounded to minute, hour, day as per the aggregation level).

Yes

host_interface_id

uuid

Unique identifier of the interface in the system.

Yes

receive_rate_percent

smallint

Used receive rate percentage on the host.

No

max_receive_rate_percent

smallint

The maximum receive rate for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

transmit_rate_percent

smallint

Used transmit rate percentage on the host.

No

max_transmit_rate_percent

smallint

The maximum transmit rate for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

received_total_byte

bigint

The total number of bytes received by the host.

No

transmitted_total_byte

bigint

The total number of bytes transmitted from the host.

No

host_interface_configuration_version

integer

The host interface configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_hosts_interfaces view and it can be used to join them.

Yes

2.7.5. Virtual Machine Statistics Views

Table 2.5. Historical Statistics for Each Virtual Machine in the System

NameTypeDescriptionIndexed

history_id

bigint

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history row (rounded to minute, hour, day as per the aggregation level).

Yes

vm_id

uuid

Unique ID of the virtual machine in the system.

Yes

vm_status

smallint

  • -1 - Unknown Status (used only to indicate problems with the ETL. Please notify Red Hat Support. )
  • 0 - Down
  • 1 - Up
  • 2 - Paused
  • 3 - Problematic

No

seconds_in_status

integer

The total number of seconds that the virtual machine was in the status shown in the status column for the aggregation period. For example, if a virtual machine was up for 55 seconds and down for 5 seconds during a minute, two rows will show for this minute. One will have a status of Up and seconds_in_status, the other will have a status of Down and a seconds_in_status of 5.

No

minutes_in_status

numeric(7,2)

The total number of minutes that the virtual machine was in the status shown in the status column for the aggregation period. For example, if a virtual machine was up for 55 minutes and down for 5 minutes during an hour, two rows will show for this hour. One will have a status of Up and minutes_in_status, the other will have a status of Down and a minutes_in_status of 5.

No

cpu_usage_percent

smallint

The percentage of the CPU in use by the virtual machine.

No

max_cpu_usage

smallint

The maximum CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

memory_usage_percent

smallint

Percentage of used memory in the virtual machine. The guest tools must be installed on the virtual machine for memory usage to be recorded.

No

max_memory_usage

smallint

The maximum memory usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value. The guest tools must be installed on the virtual machine for memory usage to be recorded.

No

user_cpu_usage_percent

smallint

Used user CPU percentage on the host.

No

max_user_cpu_usage_percent

smallint

The maximum user CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregation, it is the maximum hourly average value.

No

system_cpu_usage_percent

smallint

Used system CPU percentage on the host.

No

max_system_cpu_usage_percent

smallint

The maximum system CPU usage for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

vm_ip

text

The IP address of the first NIC. Only shown if the guest agent is installed.

No

currently_running_on_host

uuid

The unique ID of the host the virtual machine is running on.

No

current_user_id

uuid

The unique ID of the user logged into the virtual machine console, if the guest agent is installed.

No

disks_usage

text

The disk description. File systems type, mount point, total size, and used size.

No

vm_configuration_version

integer

The virtual machine configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_vms view.

Yes

current_host_configuration_version

integer

The host configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_hosts view and it can be used to join them.

Yes

memory_buffered_kb

bigint

The amount of buffered memory on the virtual machine, in kilobytes (KB).

No

memory_cached_kb

bigint

The amount of cached memory on the virtual machine, in kilobytes (KB).

No

max_memory_buffered_kb

bigint

The maximum buffered memory for the aggregation period, in kilobytes (KB). For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

max_memory_cached_kb

bigint

The maximum cached memory for the aggregation period, in kilobytes (KB). For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

2.7.6. Virtual Machine Interface Statistics Views

Table 2.6. Historical Statistics for the Virtual Machine Network Interfaces in the System

NameTypeDescriptionIndexed

history_id

integer

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history row (rounded to minute, hour, day as per the aggregation level).

Yes

vm_interface_id

uuid

Unique ID of the interface in the system.

Yes

receive_rate_percent

smallint

Used receive rate percentage on the host.

No

max_receive_rate_percent

smallint

The maximum receive rate for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

transmit_rate_percent

smallint

Used transmit rate percentage on the host.

No

max_transmit_rate_percent

smallint

The maximum transmit rate for the aggregation period, expressed as a percentage. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average rate.

No

received_total_byte

bigint

The total number of bytes received by the virtual machine.

No

transmitted_total_byte

bigint

The total number of bytes transmitted from the virtual machine.

No

vm_interface_configuration_version

integer

The virtual machine interface configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_vms_interfaces view and it can be used to join them.

Yes

2.7.7. Virtual Disk Statistics Views

Table 2.7. Historical Statistics for the Virtual Disks in the System

NameTypeDescriptionIndexed

history_id

bigint

The unique ID of this row in the table.

No

history_datetime

date

The timestamp of this history row (rounded to minute, hour, day as per the aggregation level).

Yes

vm_disk_id

uuid

Unique ID of the disk in the system.

Yes

vm_disk_status

smallint

  • 0 - Unassigned
  • 1 - OK
  • 2 - Locked
  • 3 - Invalid
  • 4 - Illegal

No

seconds_in_status

integer

The total number of seconds that the virtual disk was in the status shown in the status column for the aggregation period. For example, if a virtual disk was locked for 55 seconds and OK for 5 seconds during a minute, two rows will show for this minute. One will have a status of Locked and seconds_in_status of 55, the other will have a status of OK and a seconds_in_status of 5.

No

minutes_in_status

numeric(7,2)

The total number of minutes that the virtual disk was in the status shown in the status column for the aggregation period. For example, if a virtual disk was locked for 55 minutes and OK for 5 minutes during an hour, two rows will show for this hour. One will have a status of Locked and minutes_in_status of 55, the other will have a status of OK and a minutes_in_status of 5.

No

vm_disk_actual_size_mb

integer

The actual size allocated to the disk.

No

read_rate_bytes_per_second

integer

Read rate to disk in bytes per second.

No

max_read_rate_bytes_per_second

integer

The maximum read rate for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

read_latency_seconds

numeric(18,9)

The virtual disk read latency measured in seconds.

No

write_rate_bytes_per_second

integer

Write rate to disk in bytes per second.

No

max_read_latency_seconds

numeric(18,9)

The maximum read latency for the aggregation period, measured in seconds. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

max_write_rate_bytes_per_second

integer

The maximum write rate for the aggregation period. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

write_latency_seconds

numeric(18,9)

The virtual disk write latency measured in seconds.

No

max_write_latency_seconds

numeric(18,9)

The maximum write latency for the aggregation period, measured in seconds. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

flush_latency_seconds

numeric(18,9)

The virtual disk flush latency measured in seconds.

No

max_flush_latency_seconds

numeric(18,9)

The maximum flush latency for the aggregation period, measured in seconds. For hourly aggregations, this is the maximum collected sample value. For daily aggregations, it is the maximum hourly average value.

No

vm_disk_configuration_version

integer

The virtual disk configuration version at the time of sample. This is identical to the value of history_id in the v4_2_configuration_history_vms_disks view and it can be used to join them.

Yes