Bug 1408981

Summary:	Smartstate analysis fails against OpenShift
Product:	Red Hat CloudForms Management Engine	Reporter:	Jeff Warnica <jwarnica>
Component:	SmartState Analysis	Assignee:	Erez Freiberger <efreiber>
Status:	CLOSED DUPLICATE	QA Contact:	Einat Pacifici <epacific>
Severity:	high	Docs Contact:
Priority:	unspecified
Version:	5.7.0	CC:	bascar, brjohnso, fsimonce, jhardy, jwarnica, lavenel, obarenbo
Target Milestone:	GA
Target Release:	cfme-future
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:	container:smartstate
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2017-06-08 16:31:49 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	Container Management	Target Upstream Version:
Embargoed:

Description Jeff Warnica 2016-12-28 16:56:32 UTC

Description of problem:

SSA fails against openshift container images


Version-Release number of selected component (if applicable):
 5.7.0.13-rc3.20161129142908_1d51fd2

Smartstate analysis was failing against OpenShift docker images, with the error "cannot analyze non docker images"

On investigation into the code, this seemed reasonable, as the images were docker-pullable:// URLs

I monkey patched in the changes to  app/models/container_image.rb , from https://github.com/ManageIQ/manageiq/commit/eebf81303adf845e2cf6597824f70b402d73dd63#diff-ad3257c5ee6a799c312685f4e41d126a  but only the one file.

Reattempting to SSA an image got further, but failed with now useless errors from the Frankenstein mess. Will attempt to apply the rest of that patchset, but engineering needs to look at the right fix for 5.7.0

Comment 2 Jeff Warnica 2016-12-28 17:24:31 UTC

I did replace app/models/manageiq/providers/kubernetes/container_manager/refresh_parser.rb withg the version from that patchset. No change to the errors, samples of which are:


cannot analyze image registry.access.redhat.com/rhscl/mariadb-101-rhel7@sha256:8fb765c8210356f879f67fbf4d926f0291da5e44be838bea000f9847a27f6ad9 with id registry.acc: detected id was sha256:9e708


cannot analyze image 172.30.216.243:5000/test/cakephp-mysql-example@sha256:d4483bcb1ef23a668b53f82b8b3bab46d616863c237db435dcee250b4120f58d with id 172.30.216.2: detected id was sha256:28017

Looks to me a parsing error, but I can't quickly find the incorrect regular expression in that patchset.

Comment 3 Jeff Warnica 2017-01-04 14:02:53 UTC

Dup of https://bugzilla.redhat.com/show_bug.cgi?id=1395632 ?

Comment 4 Federico Simoncelli 2017-01-13 08:42:11 UTC

Erez can you check if this is a duplicate of bug 1395632?

Jeff did you have a chance to try the fix for bug 1395632 (of course just if it's a testing env):

https://github.com/ManageIQ/manageiq/pull/12711

Comment 5 Erez Freiberger 2017-01-15 09:38:55 UTC

Jeff,
It seems like you didn't apply all the patches of the last iteration to ssa, They are needed to get the newest image-inspector image and attempt to compare the image id with the Repository ID of the image: https://github.com/ManageIQ/manageiq/pull/12711 should fix this.

Notice that there are some other patches that followed it:
https://github.com/ManageIQ/manageiq/pull/13242
https://github.com/ManageIQ/manageiq/pull/13270

Comment 6 Erez Freiberger 2017-03-07 14:38:12 UTC

Hi Jeff,
Did you get to try updating CFME/ManageIQ or add the mentioned Patches?

Comment 7 Jeff Warnica 2017-03-07 15:11:09 UTC

I've not, as I'm long moved on from that engagement, and no longer have access to that environment.

Comment 8 Federico Simoncelli 2017-06-08 08:02:33 UTC

(In reply to Jeff Warnica from comment #7)
> I've not, as I'm long moved on from that engagement, and no longer have
> access to that environment.

Jeff should we close this?

Comment 10 Federico Simoncelli 2017-06-08 16:31:49 UTC

(In reply to Jeff Warnica from comment #7)
> I've not, as I'm long moved on from that engagement, and no longer have
> access to that environment.

Our initial assessment is that this could be a duplicate of bug 1395632. Given that we don't have any other feedback we can't confirm or deny this hypothesis.

For now closing as duplicate of bug 1395632.

*** This bug has been marked as a duplicate of bug 1395632 ***