Bug 1723872

Description of problem:

* Manager was updated from 4.2 to 4.3

* After the post upgrade reboot, there are ovirt_engine_history proceses running for a long time that consume 98% of CPU. 

Eg: 

~~~
14424 12967 postgres postgres 14424    1 08:08 postmaster      10:devices:/system.slice/rh-postgresql10-postgresql.service,1:name=systemd:/system.slice/rh-postgresql10-postgresql.service
~~~

When I look at the query output provided  I see:

~~~
 14424 | ovirt_engine_history | ovirt_engine_history | SELECT name, AVG(total) AS total, AVG(used) AS used, MAX(previous_used) AS pre
vious_used, MAX(previous_total) AS previous_total, MAX(vm_disk_usage_percent) AS usage_percent FROM (SELECT history_datetime, vms.vm_
name AS name, SUM(vm_disk_size_mb) AS total, SUM(vm_disk_actual_size_mb) AS used, COALESCE(MAX(previous_actual_size), 0) AS previous_
used, COALESCE(MAX(previous_total_size), 0) AS previous_total, CASE WHEN MAX(vm_disk_size_mb) = 0 THEN 0 ELSE (CAST(MAX(vm_disk_actua
l_size_mb) AS DECIMAL)/ MAX(vm_disk_size_mb) * 100) END AS vm_disk_usage_percent FROM v4_2_statistics_vms_disks_resources_usage_sampl
es samples INNER JOIN v4_2_latest_configuration_vms_devices devices ON devices.device_id = vm_disk_id INNER JOIN v4_2_latest_configur
ation_vms vms ON vms.vm_id = devices.vm_id INNER JOIN v4_2_latest_configuration_vms_disks disks ON disks.vm_disk_id = samples.vm_disk
_id LEFT OUTER JOIN (SELECT vm_name, AVG(previous_actual_size) AS previous_actual_size, AVG(previous_total_size) AS previous_total_si
ze FROM (SELEC
~~~

* The process 14424 terminated, but there is another one with the same behavior: 

~~~
postgres  2808 93.7  0.0 269576 45560 ?        Rs   13:48 142:14 postgres: ovirt_engine_history ovirt_engine_history 127.0.0.1(38350) SELECT
~~~

Version-Release number of selected component (if applicable):

~~~
rhvm-4.3.3.7-0.1.el7.noarch                                 Wed Jun 12 15:35:33 2019

ovirt-engine-dwh-4.3.0-1.el7ev.noarch                       Wed Jun 12 15:36:02 2019
~~~

How reproducible:

Don't know

Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:

No dwh processes using 98% CPU. 

Additional info: