Bug 796695

Summary: Provider quota not enforced for multi-instance deployment.
Product: [Retired] CloudForms Cloud Engine Reporter: Rehana <redakkan>
Component: aeolus-conductorAssignee: Tzu-Mainn Chen <tzumainn>
Status: CLOSED ERRATA QA Contact: wes hayutin <whayutin>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 1.0.0CC: akarol, deltacloud-maint, hbrock, ssachdev
Target Milestone: beta4Keywords: Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: v0.8.0-40 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-05-15 22:40:18 UTC Type: ---
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
Erron on UI
none
provider quota
none
multiple err none

Description Rehana 2012-02-23 14:07:29 UTC
Description of problem:


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


How reproducible:


Steps to Reproduce:
1. configured the cloud provider quota to 1
2. selected a comp. outline which is pushed only to rhevm provider
3. provision multi instance(two)
  
Actual results:
Since the image was pushed to only one provider the gave the error on UI (Error on UI.png)

but the instance went to "New" status and the application in "Pending" state

Expected results:
Should do a provider quota check before launching multi-instances


Additional info:

[root@hp-xw6600-02 test]# rpm -qa | grep aeolus
aeolus-conductor-daemons-0.8.0-33.el6.noarch
aeolus-conductor-0.8.0-33.el6.noarch
rubygem-aeolus-image-0.3.0-8.el6.noarch
aeolus-conductor-doc-0.8.0-33.el6.noarch
aeolus-configure-2.5.0-14.el6.noarch
aeolus-all-0.8.0-33.el6.noarch
rubygem-aeolus-cli-0.3.0-9.el6.noarch

Comment 1 Rehana 2012-02-23 14:08:15 UTC
Created attachment 565308 [details]
Erron on UI

Comment 2 Rehana 2012-02-23 14:08:46 UTC
Created attachment 565309 [details]
provider quota

Comment 3 Hugh Brock 2012-02-27 16:41:06 UTC
The error message doesn't make a ton of sense, although the last one is accurate. We should not report that we couldn't start on the other providers, especially since we didn't try. It would also be nice if the formatting for multiple errors was a bit better (newlines, anyone?).

Comment 4 Tzu-Mainn Chen 2012-02-29 16:46:26 UTC
Patch created:

https://fedorahosted.org/pipermail/aeolus-devel/2012-February/009278.html

It actually checks on the launch_time_params page, ensuring that the deployment will not get created if there are no appropriate providers.

Mainn

Comment 5 Tzu-Mainn Chen 2012-03-01 20:34:57 UTC
Patched pushed to master:

commit dfb61f73df0e02f86d73c1323f958561c22cf51e
BZ 796695 enforce provider quota for multi instance deployments

Comment 6 Aziza Karol 2012-03-05 06:21:39 UTC
Provider quota is now enforced for multi-instance deployment.

However, it would be nice if the formatting for
multiple errors was a bit better. I agree with Hugh Brock from comment 3.

see attached screenshot.Multiple err messages get displayed.

rpm -qa | grep aeolus
aeolus-configure-2.5.0-17.el6.noarch
aeolus-conductor-0.8.0-40.el6.noarch
aeolus-conductor-doc-0.8.0-40.el6.noarch
aeolus-all-0.8.0-40.el6.noarch
rubygem-aeolus-cli-0.3.0-12.el6.noarch
aeolus-conductor-daemons-0.8.0-40.el6.noarch
rubygem-aeolus-image-0.3.0-12.el6.noarch

Comment 7 Aziza Karol 2012-03-05 06:22:36 UTC
Created attachment 567481 [details]
multiple err

Comment 8 Tzu-Mainn Chen 2012-03-05 15:05:42 UTC
The formatting looks fine to me - wasn't the issue that the errors weren't on multiple lines?

The error is also technically accurate; I'm guessing you had a deployable that was supposed to launch three instances.  The code iterates through the provider accounts, trying to find one that can handle that.  In one case (mock), it can't because the quota is too low and the three component outlines aren't pushed.  In the other case (aziza_vsphere), it can't because the three component outlines aren't pushed.

The problem with compressing the error messages is that the error creation is tied up with the matching code, which iterates one by one through the component outlines and the provider accounts.  It won't be easy to extricate the two without doing some fairly large code surgery.

Can this be done for a future release, and with a new ticket (as this one is no longer accurate)?

Comment 9 Aziza Karol 2012-03-06 05:51:23 UTC
Provider quota is enforced for multi-instance deployment.

Have logged https://bugzilla.redhat.com/show_bug.cgi?id=800246 to track error handling messages.

verified:
rpm -qa | grep aeolus
aeolus-configure-2.5.0-17.el6.noarch
aeolus-conductor-0.8.0-40.el6.noarch
aeolus-conductor-doc-0.8.0-40.el6.noarch
aeolus-all-0.8.0-40.el6.noarch
rubygem-aeolus-cli-0.3.0-12.el6.noarch
aeolus-conductor-daemons-0.8.0-40.el6.noarch
rubygem-aeolus-image-0.3.0-12.el6.noarch

Comment 10 errata-xmlrpc 2012-05-15 22:40:18 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHEA-2012-0583.html