Bug 1367583

Summary: Memory report for containers does not show expected values
Product: Red Hat CloudForms Management Engine Reporter: ldomb
Component: C&U Capacity and UtilizationAssignee: Ari Zellner <azellner>
Status: CLOSED NOTABUG QA Contact: Dave Johnson <dajohnso>
Severity: medium Docs Contact:
Priority: medium    
Version: 5.6.0CC: azellner, jhardy, ldomb, obarenbo, simaishi
Target Milestone: GA   
Target Release: 5.7.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: container
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-10-10 16:07:29 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
chargeback none

Description ldomb 2016-08-16 20:54:20 UTC 
Description of problem:
When creating a report for container chargeback with the following values 

Memory Used
Memory Used Cost  

the memory total for example for openshift-infra shows 184.4 GB. 

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


5.6.0.13.20160624114606_13a9153


How reproducible:


Steps to Reproduce:
1. Create a report with the values 
Memory Used
Memory Used Cost 

Actual results:
Memory values do not seem to be based on container usage. 

Expected results:
Correct value for memory consumption

Additional info:
Comment 2 ldomb 2016-08-16 21:07:37 UTC 
Created attachment 1191391 [details]
chargeback

this attachment is a pdf
Comment 3 Federico Simoncelli 2016-08-17 21:26:00 UTC 
(In reply to ldomb from comment #0)
> the memory total for example for openshift-infra shows 184.4 GB. 
> 
> ...
> 
> Actual results:
> Memory values do not seem to be based on container usage. 

Laurent the memory reported (184.4 GB) is based on the summation of all the pods in the project per each hour.

So for example if the relevant project has multiple pods that are using 7 Gb of memory (total) for 24 hour, then the we would be charging the use of (7 * 24) 184 Gb.

Anyway I'll leave you to investigate this further with Ari.
Comment 4 Ari Zellner 2016-08-23 14:16:04 UTC 
180 GB is plausible.
Possible scenario: openshift-infra had 3 pods running for 24 hours 
Each pod uses 2.5GB of memory an hour.

The chargeback report for that day would show 24 * 3 * 2.5GB = 180gb in mem used.
Comment 5 Ari Zellner 2016-09-19 12:47:00 UTC 
Can I move this to POST?
Comment 6 ldomb 2016-09-27 10:53:19 UTC 
Yes please move post. Thanks for the explanation.