Created attachment 1123536 [details]
screenshot of the Events list with Hosts page

Description of problem
======================

When I run Accept Node task for all 6 nodes available and a failure happens
during the process, I see mismatch in Events, Host state and Tasks.

This BZ is concerned about the fact that this mismatch happened and not about
the fact that accepting of some node(s) failed.

Version-Release number of selected component
============================================

# rpm -qa 'rhscon*'
rhscon-ceph-0.0.6-2.el7.x86_64
rhscon-ui-0.0.11-1.el7.noarch
rhscon-core-0.0.8-2.el7.x86_64

How reproducible
================

no data available so far, I may add stats later if needed

Steps to Reproduce
==================

1. Install skyring on server and prepare few hosts for cluster setup
2. Accept all nodes

I assume that Initialize Node task fails for one node here.

Actual results
==============

Tasks page reports that "Initialize Node" task for node1 is in Failed state.

On the other hand, there is a mismatch in reporting of this event in other
places:

There are 6 nodes in total (shown on the Hosts page), one of them is in
a failed state (small red icon with a cross).

There are 4 events in Events popup list (for node5, node2, node3 and node4),
all of the events report a success: "host is accesable now on"
(sic, BZ 1307053).

Expected results
================

When failure of "Initialize Node" taks happens for a single node, I would
expect to that:

 * 5 success events are shown in Events popup list
 * 1 failure event is shown in Events popup list

Note that this is just an example and other solutions are possible as well, but
information provided by usm should be consistent in every situation.