Bug 1035443

Summary: Summary Last Metrics values show different values than Chart Last Metric values
Product: [JBoss] JBoss Operations Network Reporter: Mike Thompson <mithomps>
Component: UIAssignee: Mike Thompson <mithomps>
Status: CLOSED CURRENTRELEASE QA Contact: Mike Foley <mfoley>
Severity: high Docs Contact:
Priority: unspecified    
Version: JON 3.2CC: ahovsepy, hrupp, loleary, skondkar
Target Milestone: CR01   
Target Release: JON 3.2.1   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: 1034991 Environment:
Last Closed: 2014-01-10 15:57:45 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:
Embargoed:
Bug Depends On: 1034991    
Bug Blocks: 1034852    
Attachments:
Description Flags
metric_dummary_chart
none
free-space-chart-summary-values none

Description Mike Thompson 2013-11-27 19:27:21 UTC 
+++ This bug was initially created as a clone of Bug #1034991 +++

Description of problem:
This is one that @ahovsepy and @skondkar brought up with the summary metric last value sometimes not matching the value in the graphs.

Version-Release number of selected component (if applicable):
RHQ 4.1.0 and JON 3.2.0 ER7


How reproducible:
Depends on time range.


Steps to Reproduce:
1. Navigate to the monitor metrics tab
2. Navigate to the EAP RHQ Server
3. Observer the summary metrics value for 'Maximum Request Time'
4. click on the chart link for 'Maximum Request Time' and hover over last bar. Observer the Avg. value.

Actual results:
Depending on various time ranges these last values may or may not match.


Expected results:
The summary metric last value and chart last value (avg value on aggregate bar)should match.

--- Additional comment from Mike Thompson on 2013-11-26 15:22:08 EST ---

I have changed the code to use exactly the same date range and metric api. It was possible to use different metrics apis for the different time range types. I eliminated the variation, so that the summary last values are pulling from the same single metrics api as the charts.


Committed to master: a532e90
Comment 1 Mike Thompson 2013-11-27 19:29:32 UTC 
Cloning for JON 3.2.0 (CR01) as this needs to be taken as part of BZ 1034852.
Comment 2 Mike Thompson 2013-11-27 19:35:27 UTC 
Committed to release/jon3.2.x as: 393d350
Comment 3 Simeon Pinder 2013-12-03 23:19:44 UTC 
Moving to ON_QA for testing in latest(CR1) brew build.
Comment 4 Armine Hovsepyan 2013-12-04 10:29:30 UTC 
Created attachment 832548 [details]
metric_dummary_chart
Comment 5 Armine Hovsepyan 2013-12-04 10:32:10 UTC 
Created attachment 832560 [details]
free-space-chart-summary-values
Comment 6 Armine Hovsepyan 2013-12-04 10:33:20 UTC 
re-assigned
issue is still visible in CR1 - screen-shots attached
Comment 8 Heiko W. Rupp 2013-12-04 11:26:53 UTC 
The charts use different precision which end up in slightly different values displayed - rounding 2.1G and 1.9G to GB boundaries both result in 2 G -- this is probably not nice, but it is not way off.
Comment 9 Mike Thompson 2013-12-04 15:43:33 UTC 
The original issue here is fixed and a new issue was found. This bug was referring to the summary metrics not being refreshed when the time range was changed and this was a bug and it fixed with this commit. Now, the issue is with round/precision through javascript instead of java. This kind of sophisticated rounding requires knowledge of units (Bytes, seconds) and groupings  of precision(Terabyte, Gigabyte, Kilobyte, byte).

On the summary metrics detail graph, the averages box at the right and the hovers  has rounding/precision errors. That should be the new BZ. Much less severe than the blocker that was fixed.
Comment 10 Mike Thompson 2013-12-05 14:53:51 UTC 
I fixed the significant/digits precision issue with graph legend.

committed to master: ec67bac
Comment 14 Mike Foley 2014-01-10 15:57:45 UTC 
per bz triage ... closing this as VERIFIED CLOSED cURRENT RELEASE 


and

opening a new bug for comment #9  (i will document new BZ in comment #15)
Comment 15 Mike Foley 2014-01-10 16:01:20 UTC 
new BZ to capture loose end in comment #9 

https://bugzilla.redhat.com/show_bug.cgi?id=1051583