Bug 803096

Summary: Scalability Testing : Logout times impacted by number of users logged in concurrently
Product: [Retired] CloudForms Cloud Engine Reporter: Ronelle Landy <rlandy>
Component: aeolus-conductorAssignee: Jiri Tomasek <jtomasek>
Status: CLOSED NOTABUG QA Contact: wes hayutin <whayutin>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 1.0.0CC: akarol, athomas, dajohnso, deltacloud-maint, hbrock, ssachdev
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-08-10 14:53:03 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Ronelle Landy 2012-03-13 23:53:28 UTC
Description of problem:

With 100+ users logged in to one conductor instance, it can take up to 4 minutes for a user to log out.

The following lines from a jmeter test results table show that it took twice as long for a user to log out than it did for that same user (or an equivalent user) to access the conductor/pools page or conductor/users page (which has to load 300+ users)

738	19:31:10.973	Add User Group 1-44	/conductor/pools	107822	Success	15784
739	19:31:10.965	Add User Group 1-99	/conductor/users	117398	Success	101797
740	19:28:52.172	Add User Group 1-8	/conductor/logout	256192	Success	5313

Other logout times from the same test are copied below:

757	19:29:07.830	Add User Group 1-10	/conductor/logout	244794	Success	5308
758	19:31:34.728	Add User Group 1-49	/conductor/pools	108223	Success	15779
759	19:29:07.830	Add User Group 1-7	/conductor/logout	255122	Success	5308
760	19:29:07.835	Add User Group 1-6	/conductor/logout	255121	Success	5318
761	19:29:24.146	Add User Group 1-4	/conductor/logout	238815	Success	5313
762	19:29:24.156	Add User Group 1-5	/conductor/logout	238806	Success	5308

Note that when the load is reduced (other users are logged out already), the remaining user can log out in +/- 3.5 seconds:

899	19:33:59.499	Add User Group 1-100	/conductor/logout	3635	Success	5323
900	19:33:59.717	Add User Group 1-99	/conductor/logout	3438	Success	5308


Steps to Reproduce:
1. Create 100+ users 
2. Make sure those users are logged into the same conductor instance concurrently
3. Have one of the users log out
  
rpms tested:

rpm -qa |grep aeolus
rubygem-aeolus-cli-0.3.0-14.el6.noarch
aeolus-configure-2.5.0-18.el6.noarch
aeolus-all-0.8.0-41.el6.noarch
rubygem-aeolus-image-0.3.0-12.el6.noarch
aeolus-conductor-0.8.0-41.el6.noarch
aeolus-conductor-daemons-0.8.0-41.el6.noarch
aeolus-conductor-doc-0.8.0-41.el6.noarch

Comment 1 Hugh Brock 2012-03-19 17:22:54 UTC
Similar to 802571. Moving to 1.0.z.

Comment 2 Dave Johnson 2012-08-10 14:53:03 UTC
Closing this out... at this point we are not necessarily concerned about a large number of users and all scalability concerns are around the number of instances.