Bug 1591272

Summary: [Regression] GCE provider refresh fails in CFME 5.9
Product: Red Hat CloudForms Management Engine Reporter: Nandini Chandra <nachandr>
Component: ProvidersAssignee: Tomas Coufal <tcoufal>
Status: CLOSED CURRENTRELEASE QA Contact: Nandini Chandra <nachandr>
Severity: high Docs Contact:
Priority: high    
Version: 5.9.0CC: cpelland, gblomqui, ikaur, jfrey, jhardy, obarenbo, simaishi, tpapaioa
Target Milestone: GAKeywords: Regression, TestOnly, ZStream
Target Release: 5.10.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: 5.10.0.1 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 1592504 1596336 (view as bug list) Environment:
Last Closed: 2019-02-11 14:06:57 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: CFME Core Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1592504, 1596336    

Description Nandini Chandra 2018-06-14 12:42:55 UTC
Description of problem:
-----------------------
I'm running into this error in 5.9.z(latest build).

I thought the Fog google upgrade to 1.3.3 made through the following PR would take care of this issue.

https://github.com/ManageIQ/manageiq-providers-google/pull/54

Also, see https://bugzilla.redhat.com/show_bug.cgi?id=1585821


Snippet from evm.log

2018-06-13 06:46:01,981 [E] Cannot validate the provider. Error occured: accessNotConfigured: Access Not Configured. Compute Engine API has not been used in project 25xxxxxxx before or it is disabled. Enable it by visiting https://console.developers.google.com/apis/api/compute.googleapis.com/overview?project=25xxxxxx then retry. If you enabled this API recently, wait a few minutes for the action to propagate to our systems and retry. 


Version-Release number of selected component (if applicable):
------------------------------------------------------
5.9.3.1


How reproducible:
-----------------
Always


Steps to Reproduce:
-------------------
1.Attempt to validate GCE provider in CFME


Actual results:
--------------
Unable to validate GCE provider 


Expected results:
----------------
It should be possible to validate GCE provider


Additional info:
----------------

Comment 3 Tomas Coufal 2018-06-14 16:21:51 UTC
For the record:

After meeting with Nandini, we discovered this issue is present only when the provider is being added via API - QE uses an automated test execution, I'll try to reproduce this behavior locally and if that proves to be the point of failure we'll forward this to the API people.

Comment 5 Tomas Coufal 2018-06-15 11:25:36 UTC
Hi, I've located the issue. It's caused by Google. I've prepared a PR fixing it upstream [1] but we might need a quick fix for MIQ. I'm looking into that right now.

[1] https://github.com/fog/fog-google/pull/356

Comment 6 Tomas Coufal 2018-06-15 11:45:11 UTC
MIQ hotfix is ready to review https://github.com/ManageIQ/manageiq-providers-google/pull/63

Comment 7 Tomas Coufal 2018-06-15 12:03:23 UTC
Also for the record the full story:

The issue is not caused by adding a provider but it's rather a result of initial refresh. When adding a provider, the provider entity in MIQ database is created just fine. Also credentials validation passes. Both via UI and via API. The error appears after initial refresh is processed. The error message appears as an result of the refresh. It's caused because Google removed a project from public images listing, so our refresh fails when we try to pull images from this nonexistent project.

Provider addition works just fine. The refresh is failing.

Comment 8 Chris Pelland 2018-06-18 13:39:16 UTC
*** Bug 1592216 has been marked as a duplicate of this bug. ***

Comment 9 Tomas Coufal 2018-06-18 14:46:57 UTC
Fixed by: https://github.com/ManageIQ/manageiq-providers-google/pull/64
Merged, requires backporting

Comment 13 Nandini Chandra 2018-07-17 19:45:33 UTC
verified in 5.10.0.4