Bug 1237233

Summary: alerts are no longer evaluated when one alert evaluates against a provider (Cloudforms 3.1)
Product: Red Hat CloudForms Management Engine Reporter: Felix Dewaleyne <fdewaley>
Component: ControlAssignee: Bill Wei <bilwei>
Status: CLOSED EOL QA Contact: Dmitry Misharov <dmisharo>
Severity: medium Docs Contact:
Priority: high    
Version: 5.3.0CC: cpelland, dmisharo, fdewaley, jhardy, mfalesni, obarenbo
Target Milestone: GAKeywords: ZStream
Target Release: 5.3.6   
Hardware: All   
OS: All   
Whiteboard: alert
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
: 1265323 1265325 (view as bug list) Environment:
Last Closed: 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:    
Bug Blocks: 1265323, 1265325    

Description Felix Dewaleyne 2015-06-30 15:08:45 UTC
Description of problem:
After adding an alert on a Cloudforms  3.1 appliance that triggers on "provider.clusters > 0" no longer issues alerts

Version-Release number of selected component (if applicable):
Cloudforms 3.1


How reproducible:
all the time

Steps to Reproduce:
1. setup email notifications on the zone
2. create a new policy called "reproducer" that evaluates "provider.clusters > 0" hourly
3. make it send an email to a user of the appliance

Actual results:
no email is ever sent, evm.log doesn't show ever trying to send the email

Expected results:
emails are sent as the alert is evaluated hourly and is always true

Additional info:
does not reproduce on Cloudforms 3.2

Comment 2 Bill Wei 2015-07-02 19:51:45 UTC
Please confirm this problem only occurs on CloudForm 3.1. 

Can QE loan me a 3.1 environment to debug this issue?

Comment 3 Bill Wei 2015-07-02 19:55:08 UTC
Restore overwritten attributes due to comment collision.

Comment 4 Felix Dewaleyne 2015-07-03 09:13:28 UTC
I can confirm my tests were done at the same time on 3.2 and 3.1 in lab and 3.1 failed to send emails while 3.2 did not. it seems nothing at all happened on the 3.1 instead of the alert being triggered and the email sent.

Comment 5 Bill Wei 2015-07-06 20:14:03 UTC
This has been fixed in upstream and 5.4.
https://github.com/ManageIQ/manageiq/pull/2979

It is a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=1223835.

We yet need to decide whether to back port the fix to 5.3.

Comment 7 Bill Wei 2015-09-23 13:57:46 UTC
commit 7d6463013088618014af9b733ef6e9e99ba0624a
Author: Bill Wei <bilwei>
Date:   Fri May 22 09:00:07 2015 -0400

    Call Zone's accessor methods to get models' instances in current zone
    
    Add a set of {tablename}_without_a_zone methods to get instances that are not connected to the zone
    through ext_management_system association.
    
    Add spec tests for MiqAlert.evaluate_hourly_timer
    
    https://bugzilla.redhat.com/show_bug.cgi?id=1223835

Comment 8 Chris Pelland 2015-09-29 13:36:19 UTC
*** Bug 1265323 has been marked as a duplicate of this bug. ***

Comment 9 Mike McCune 2016-03-28 22:20:08 UTC
This bug was accidentally moved from POST to MODIFIED via an error in automation, please see mmccune with any questions