Bug 1447672

Summary: SSA fails on timeout for large images
Product: Red Hat CloudForms Management Engine Reporter: Einat Pacifici <epacific>
Component: SmartState AnalysisAssignee: Erez Freiberger <efreiber>
Status: CLOSED CURRENTRELEASE QA Contact: Einat Pacifici <epacific>
Severity: high Docs Contact:
Priority: high    
Version: 5.8.0CC: cpelland, dajohnso, dlamotta, efreiber, fsimonce, jhardy, mtayer, obarenbo, simaishi
Target Milestone: GAKeywords: Regression, TestOnly
Target Release: 5.9.0Flags: epacific: automate_bug+
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: container:smartstate
Fixed In Version: 5.9.0.1 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 1450470 (view as bug list) Environment:
Last Closed: 2018-03-06 14:34:25 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:
Bug Depends On:    
Bug Blocks: 1450470    
Attachments:
Description Flags
ssa failed task none

Description Einat Pacifici 2017-05-03 12:46:41 UTC 
Created attachment 1275912 [details]
ssa failed task

Description of problem:
CFME podified image is 1.8Gb. 
When running SSA on the image, the image download fails on timeout and as a result SSA cannot happen. 

Version-Release number of selected component (if applicable):
5.8.0.12

How reproducible:
Always

Steps to Reproduce:
1.Add Openshift 3.5. provider to CFME
2.Go to compute --> Containers --> Images
3.Find CFME image
4. Ensure Policy OpenSCAP is assigned to image
5. Select image and select Configuration --> Perform smart state analysis

Actual results:
Task always fails on timeout

Expected results:


Additional info:
Image attached.
Comment 2 Mooli Tayer 2017-05-03 14:16:16 UTC 
Erez please add details on the related docker change
Comment 3 Erez Freiberger 2017-05-03 14:41:05 UTC 
IIUC docker canceled to continued downloading after the client disconnects. I can't find it in the docs, but I see it referenced in issues. [2] is where I think it was disabled.


[1]https://docs.docker.com/engine/reference/commandline/pull/#cancel-a-pull
[2]https://github.com/moby/moby/pull/18353
[3]https://github.com/moby/moby/issues/6928
[4]https://github.com/moby/moby/issues/3136
Comment 4 Mooli Tayer 2017-05-09 11:51:06 UTC 
Some notes from email thread:
In vm scanning there are some timeout adjustment mechanisms that we can adopt[1]
The timeout there is 20 min and not 5. It should be safe to increase our timeout as well.

[1]
https://github.com/manageiq/manageiq/blob/01fb29ed1b9b3e046976bcd123ecfa8de87e649a/app/models/job.rb#L164
https://github.com/manageiq/manageiq/blob/6abb5559f81e73446712488fe27b77be0f5d4fac/app/models/vm_scan.rb#L8-L13
https://github.com/manageiq/manageiq/blob/01fb29ed1b9b3e046976bcd123ecfa8de87e649a/app/models/job.rb#L178-L186
Comment 5 Erez Freiberger 2017-05-09 13:24:04 UTC 
PR to provide setting the timeout for the scanning jobs https://github.com/ManageIQ/manageiq-providers-kubernetes/pull/14