Bug 1512703

Summary: Long time to refresh network provider on OpenStack
Product: Red Hat CloudForms Management Engine Reporter: Andrea Perotti <aperotti>
Component: ProvidersAssignee: Sam Lucidi <slucidi>
Status: CLOSED CURRENTRELEASE QA Contact: Jad Haj Yahya <jhajyahy>
Severity: high Docs Contact:
Priority: high    
Version: 5.8.0CC: aperotti, gblomqui, jfrey, jhajyahy, jhardy, maufart, mlele, obarenbo, simaishi, slucidi
Target Milestone: GAKeywords: TestOnly, ZStream
Target Release: 5.10.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: 5.10.0.0 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 1554541 1554543 (view as bug list) Environment:
Last Closed: 2018-08-01 02:48:08 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: Openstack Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1468726, 1554541, 1554543    

Description Andrea Perotti 2017-11-13 21:37:43 UTC 
Description of problem:
cloudforms refresh an openstack network manager it took more than 5 minutes, when refresh a cloud manager o storage manager, it took about one minute.

Look at the example below:

Worker PID:             2930
Message ID:             1000000461622
Message fetch time:     2017-10-30T10:17:48.535652
Message time in queue:  11.466717693 seconds
Provider:               Openstack::CloudManager
EMS Name:               OSP9
Refresh type:           full
Refresh start time:     2017-10-30T10:17:48.623460
Refresh timings:
  collect_inventory_for_targets:       0.002257 seconds
  parse_legacy_inventory:              25.098824 seconds
  parse_targeted_inventory:            25.099295 seconds
  save_inventory:                      3.148565 seconds
  ems_refresh:                         28.250349 seconds
Refresh end time:       2017-10-30T10:18:16.874024
Message delivered time: 2017-10-30T10:18:17.001680
Message state:          ok
Message delivered in:   28.465715544 seconds


---
Worker PID:             2940
Message ID:             1000000461781
Message fetch time:     2017-10-30T10:22:11.088119
Message time in queue:  234.240138968 seconds
Provider:               Openstack::NetworkManager
EMS Name:               OSP9 Network Manager
Refresh type:           full
Refresh start time:     2017-10-30T10:22:11.092803
Refresh timings:
  collect_inventory_for_targets:       0.002506 seconds
  parse_legacy_inventory:              241.966221 seconds
  parse_targeted_inventory:            241.966239 seconds
  save_inventory:                      5.755467 seconds
  ems_refresh:                         247.724421 seconds
Refresh end time:       2017-10-30T10:26:18.817420
Message delivered time: 2017-10-30T10:26:18.829892
Message state:          ok
Message delivered in:   247.741669532 seconds

Version-Release number of selected component (if applicable):
CFME 4.5

How reproducible:
Always
Comment 3 Sam Lucidi 2017-11-14 18:01:49 UTC 
There is a pull request waiting to be backported that should resolve the slow refresh of the network manager. https://github.com/ManageIQ/manageiq/pull/16427
Comment 6 Sam Lucidi 2017-11-28 19:23:25 UTC 
The PR for Fine has been merged, moving to POST.
Comment 19 Jad Haj Yahya 2018-01-25 13:38:59 UTC 
Verified on RHOS 10, CFME 5.8.3.1

1. Created 100 cloud tenants from OSP
2. Added admin as member to all Tenants 
3. Perform refresh
4. measure the time till all tenants appear on cloud tenant page


Measured time was 4 min
Comment 20 Jad Haj Yahya 2018-01-25 15:26:17 UTC 
Also set is_admin to true at settings.yaml:

:ems:
  :ems_openstack:
    :excon:
      :omit_default_port: true
      :read_timeout: 60
    :refresh:
      :is_admin: true
Comment 21 Sam Lucidi 2018-03-08 17:29:46 UTC 
For master: https://github.com/ManageIQ/manageiq-providers-openstack/pull/216 (already merged)

Fine backport: https://github.com/ManageIQ/manageiq/pull/16695