Bug 2063223

Description of problem, and how to reproduce it:

With reference to https://bugzilla.redhat.com/show_bug.cgi?id=1999089#c10 :

The files virt-who-config-*.conf all report into the same Satellite organization (owner=Example_Org) but authenticate to Satellite as different users (rhsm_username=virt_who_reporter_2 , rhsm_username=virt_who_reporter_3, and so on), they generate separate connection requests to report on their generated host-to-guest mappings. This is when the customer experiences the issue with missing hypervisors in the mappings.

Changing 'rhsm_username' in all virt-who-config-*.conf files to 'rhsm_username=virt_who_reporter_2', and restarting the virt-who service, forced all virt-who-config-*.conf files to submit one single report, and the customer no longer encounters the issue.

Also observed that when each virt-who config is configured to use its own virt-who reporter, there are multiple Hypervisors tasks with very short (1 or 0 seconds) duration which although show as being successful in Satellite. But when you examine them in the dynflow console, they have actually failed:

For example:
~~~
Id: a4e6c9e0-9209-490c-bd59-799a65212f93
Label: Actions::Katello::Host::Hypervisors
Status: stopped
Result: success
Started at: 2021-12-05 14:14:39 UTC
Ended at: 2021-12-05 14:14:39 UTC


3: Actions::Candlepin::AsyncHypervisors (success) [ 0.10s / 0.10s ]
Queue: hosts_queue

Started at: 2021-12-05 14:14:39 UTC
Ended at: 2021-12-05 14:14:39 UTC
Real time: 0.10s
Execution time (excluding suspended state): 0.10s

Input:

---
task_id: 8a3188c67d8ae15d017d8af25d390035
session_id: 
remote_user: admin
remote_cp_user: admin
current_request_id: 
current_timezone: UTC
current_user_id: 17
current_organization_id: 
current_location_id: 
Output:

---
hypervisors: []
task:
  created: '2021-12-05T14:14:39+0000'
  updated: '2021-12-05T14:14:39+0000'
  id: 8a3188c67d8ae15d017d8af25d390035
  name: Hypervisor Update
  group: 
  origin: satellite.example.com
  executor: 
  principal: foreman_admin
  state: FAILED				<<<<<----- reporting 'FAILED' here ----->>>>>
  previousState: CREATED
  startTime: 
  endTime: 
  attempts: 0
  maxAttempts: 1
  statusPath: "/jobs/8a3188c67d8ae15d017d8af25d390035"
  key: HypervisorUpdateJob


5: Actions::Katello::Host::HypervisorsUpdate (success) [ 0.08s / 0.08s ]
Queue: hosts_queue

Started at: 2021-12-05 14:14:39 UTC
Ended at: 2021-12-05 14:14:39 UTC
Real time: 0.08s
Execution time (excluding suspended state): 0.08s

Input:

---
hypervisors: []
remote_user: admin
remote_cp_user: admin
current_request_id: 
current_timezone: UTC
current_user_id: 17
current_organization_id: 
current_location_id: 
Output:

---
results: 


7: Actions::ForemanVirtWhoConfigure::Config::Report (success) [ 0.02s / 0.02s ]
Queue: default

Started at: 2021-12-05 14:14:39 UTC
Ended at: 2021-12-05 14:14:39 UTC
Real time: 0.02s
Execution time (excluding suspended state): 0.02s

Input:

---
hypervisors:
  hypervisors: []
current_request_id: 
current_timezone: UTC
current_user_id: 17
current_organization_id: 
current_location_id: 
Output:

--- {}
~~~



Actual results:

Multiple 'Actions::Katello::Host::Hypervisors' tasks with very short (1 or 0 seconds) duration which although show as being successful in Satellite. But when you examine them in the dynflow console, they have actually failed.


Expected results:

When a similar condition occurs, 'Actions::Katello::Host::Hypervisors' tasks which failed as per the dynflow console are also reported as failed in Satellite.