Bug 1395911

Summary: [Scale] Sluggish User Experience with 200vm 1k Disks and continued browsing
Product: [oVirt] ovirt-engine Reporter: mlehrer
Component: Frontend.WebAdminAssignee: Alexander Wels <awels>
Status: CLOSED DUPLICATE QA Contact: Pavel Stehlik <pstehlik>
Severity: high Docs Contact:
Priority: unspecified    
Version: 4.0.5.1CC: bugs, gklein, mlehrer, oourfali, rgolan
Target Milestone: ovirt-4.1.0-alphaKeywords: Performance
Target Release: ---Flags: ykaul: needinfo?
oourfali: ovirt-4.1?
oourfali: planning_ack?
oourfali: devel_ack?
oourfali: testing_ack?
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-11-23 11:48:34 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: UX RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1388462    

Description mlehrer 2016-11-16 23:12:26 UTC 
Description of problem:

Webadmin UI sometimes became noticeably sluggish, and janky after continued usage, sometimes giving re-occuring UI general exceptions making it necessary to close the browser a few times to improve experience.  This occurred after performing multiple domain operations and checking the events tab, and browsing several VMs (each vm containing 5 disks) over time and continued usage without closing the browser.

Dataset: 
DC 1
200 VMs, 1000 disks, 2 hosts, 4 domains

Browser: Firefox 49

Firefox memory was around 625mb sometimes up to 1GB on client side

Items to note:
   Orphan Nodes around 288mb 
   Re-occurring error message in UI (Uncaught Exception)
   Dataset was 1000 Disks, 200 VMs [5 disks per VM]
   Browsing activities centered around performing common ISCSI domain management tasks over several hours and browsing disk details of VMs, start/stopping VMs.
   

Version-Release number of selected component (if applicable):
Testing performed on 4.0.5.1

How reproducible:
Requires effort to reproduce, additionally requires dataset of 1k disks, and 100 of 200 VMs 'up' state and continued browsing/usage.

Steps to Reproduce:
1. Create/Activate Storage Domains
2. Trigger long running tasks that result in many status changes like starting 5 Vms every 2-3 minutes.
3. Browse multiple VMs and respective 'disk' pane on the bottom panel.


Actual results:
Should begin to notice sluggish user experience when opening, scrolling, typing in dialog boxes of Webadmin; especially when browsing VM results and scrolling.
Sometimes browser appears temporarily unresponsive along with high client CPU usage.

Expected results:
Better user experience without locking, or sluggishness

Additional info:
When the browser did become unresponsive or sluggish generally firefox would show high CPU usage, though its unclear if this is a cause or response.

Logs, Errors found: https://drive.google.com/open?id=0B8V1DXeGhPPWbGdCaU5JQmQ3U0E
Comment 1 Yaniv Kaul 2016-11-17 07:17:33 UTC 
At least the exception, with the debuginfo, should be in the bug.

Please re-test with either 4.1 or 4.0.6, where all known leaks have been fixed.
Comment 2 Roy Golan 2016-11-17 12:17:46 UTC 
(In reply to Yaniv Kaul from comment #1)
> At least the exception, with the debuginfo, should be in the bug.
> 
> Please re-test with either 4.1 or 4.0.6, where all known leaks have been
> fixed.

master have the tooltip fix while 4.0.6 doesn't yet [1]. 

[1] https://gerrit.ovirt.org/#/c/66810/2
Comment 3 Oved Ourfali 2016-11-17 13:19:17 UTC 
Let's test this with the leak fixes and also with the gwt-rpc changes.
Comment 4 Oved Ourfali 2016-11-23 11:48:34 UTC 

*** This bug has been marked as a duplicate of bug 1368101 ***