Hide Forgot
Description of problem: Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1.As non admin user i launched 4 ec2 instances. QUOTA USED in overview section does not reflect the correct count. see attached screenshot. Expected results: should reflect the correct count. Additional info: [root@hp-bl480c-01 ~]# rpm -qa | grep aeolus aeolus-conductor-0.7.0-4.el6.noarch aeolus-configure-2.4.0-3.el6.noarch aeolus-conductor-doc-0.7.0-4.el6.noarch aeolus-all-0.7.0-4.el6.noarch rubygem-aeolus-cli-0.2.0-3.el6.noarch rubygem-aeolus-image-0.2.0-1.el6.noarch aeolus-conductor-daemons-0.7.0-4.el6.noarch
Created attachment 542538 [details] quota usage for nonadmin
I can't tell from the screenshot -- what is the state of the 4 instances? This will have an impact on the quota usage.
Oh one other question -- what is the user quota set to? I think I'm seeing the same thing here -- the page is showing '0 of 0', but enforcement seems to be working.
OK, we (engineering) had not come to our own conclusion about what these numbers mean. Here is that conclusion: * The "Overview" section is always specific to the user viewing it. This is the guiding principle of that area of the UI. Several things follow from that: 1. QUOTA USED should be the number of running instances owned by the user, and the QUOTA itself should be the user's personal quota. 2. POOLS should be the number of pools the user has permission to launch instances in 3. DEPLOYMENTS should be the number of deployments the user has launched 4. the numbers in the INSTANCES section are all counts of the user's various instances: Running, Pending, Failed 5. The ALERTS are the count of alerts for that user Scott let's get all these numbers set right as part of this fix.
The summary page at /pools was giving confusing (and incorrect) stats. Changes are as follows: Pools: before: number of pools the user has view permissions on after: number of pools the user has permission to launch deployments in Deployments: before: number of deployments the user has view permission on after: number of deployments owned (i.e. launched) by the user Instances: before: number of instances the user has view permissions on after: number of instances owned (i.e. launched) by the user Alerts: (we don't really have alerts yet, so this is currently just the number of failed instances) before: number of failed instances the user has view permissions on after: number of failed instances the user owns Quota Used: before: aggregated stats for environment quotas on all environments that the user has view permisions on. This was the cause of the BZ report, as by default, non-admin users have no direct permissions on the default environment -- only on the contained default pool. So for non-admin users this showed 0 of 0, since it was an aggregation over 0 environments. after: user quota stats. I've also updated the labels for the Overview to show "User Quota Usage" instead of "Quota Usage" and, for the Pool view, "Pool Quota Usage" instead of "Quota Usage" to clarify which quota measurement is being applied. Patch here: https://dhcp231-79.rdu.redhat.com/conductor/pools
oops. wrong pasted URL for the patch. Here: https://fedorahosted.org/pipermail/aeolus-devel/2011-December/007260.html
Pushed to Master.
ef2c43d7dc02e909026c730933c55400b703a86d in aeolus-conductor-0.8.0-2.el6.src.rpm
bugs in verified or on_qa moving off tracker
Correct count is displayed for non admin user Launched two instance (quota set to 5 ) Shows 40% usage ( 2 of 5 ) Stopped one instance , shows 20 % usage ( 1 of 5 ) ========================================================= Verified in rpm -qa|grep aeolus aeolus-conductor-0.8.0-8.el6.noarch aeolus-all-0.8.0-8.el6.noarch aeolus-conductor-daemons-0.8.0-8.el6.noarch rubygem-aeolus-image-0.3.0-3.el6.noarch aeolus-configure-2.5.0-5.el6.noarch rubygem-aeolus-cli-0.3.0-4.el6.noarch aeolus-conductor-doc-0.8.0-8.el6.noarch
Created attachment 556190 [details] user_quota
Created attachment 556191 [details] user_quota_with_stopped_inst