Bug 1434681

Summary: CloudForms 4.2 is not displaying vm network metric info from OSP10 provider
Product: Red Hat CloudForms Management Engine Reporter: tachoi
Component: ProvidersAssignee: Richard Su <rwsu>
Status: CLOSED CURRENTRELEASE QA Contact: Ido Ovadia <iovadia>
Severity: high Docs Contact:
Priority: unspecified    
Version: 5.7.0CC: cpelland, iovadia, jfrey, jhardy, obarenbo, rwsu, simaishi, tzumainn
Target Milestone: GAKeywords: TestOnly
Target Release: 5.9.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: openstack
Fixed In Version: 5.9.0.1 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 1442900 (view as bug list) Environment:
Last Closed: 2018-03-06 15:12:50 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: Openstack Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1442900    
Attachments:
Description Flags
ceilometer meter-list,gnocchi metric list output
none
screeshot from the performance view none

Description tachoi 2017-03-22 06:37:42 UTC 
Description of problem:
- RHOS 10 is deployed with Gnocchi/ Ceilometer for metering (which is being ingested by CloudForms 4.2)
- CFME 5.7.1.3
- CFME is showing metric info for cpu/mem/disk not network of OSP instance

Version-Release number of selected component (if applicable):
CFME 5.7.1.3
OSP10

How reproducible:
Connect CFME with OSP10.
Enable gnocchi for metric publisher(ceilometer.conf,pipeline.yaml)
Restart ceilometer service
Monitoring performance from CFME instance page

Steps to Reproduce:
1.
2.
3.

Actual results:
Not showing network IO

Expected results:
Showing network IO info

Additional info:
Comment 3 tachoi 2017-03-22 06:43:19 UTC 
Created attachment 1265311 [details]
ceilometer meter-list,gnocchi metric list output
Comment 4 tachoi 2017-03-22 06:46:46 UTC 
Created attachment 1265312 [details]
screeshot from the performance view
Comment 10 Richard Su 2017-03-28 20:20:52 UTC 
Hi Taeho,

1. You are correct, I am also seeing null values for net_usage_rate_average in the database.

vmdb_development=# select count(*) from metrics where net_usage_rate_average is null ;
 count 
-------
  2090
(1 row)

2. There is network usage data in gnocchi.

gnocchi resource list -f yaml
Find network interface of instance1.

- ended_at: null
  id: f5c79575-dbd2-5926-8f54-b1f67474529d
  original_resource_id: instance-00000001-78cd5fe0-e218-441c-919e-b598258539fe-tap83a5fffa-b9
  project_id: 5f92fa2ed5d84d7587e860ece0913ef6
  revision_end: null
  revision_start: '2017-03-27T21:47:58.563989+00:00'
  started_at: '2017-03-27T21:47:58.563968+00:00'
  type: instance_network_interface
  user_id: 92ffaaf9aff34317a79609db962c3dac

[stack@instack ~]$ gnocchi metric list | grep f5c79575-dbd2-5926-8f54-b1f67474529d
| 1479050e-7c52-464b-9765-8e868d6f08e5 | low                 | network.incoming.bytes.rate     | None      | f5c79575-dbd2-5926-8f54-b1f67474529d |
| 335a2887-6504-4526-8e45-55681977acb5 | low                 | network.incoming.packets.rate   | None      | f5c79575-dbd2-5926-8f54-b1f67474529d |
| 379658e5-c05d-41fc-8f84-70d2581570e5 | low                 | network.incoming.packets        | packet    | f5c79575-dbd2-5926-8f54-b1f67474529d |
| 9ba1e09c-ef80-43f1-b117-47c73a0f4ff9 | low                 | network.outgoing.bytes.rate     | B/s       | f5c79575-dbd2-5926-8f54-b1f67474529d |
| a9f69e74-a877-4644-9aae-23dcf49bda69 | low                 | network.outgoing.packets        | packet    | f5c79575-dbd2-5926-8f54-b1f67474529d |
| c1e72ec6-3e9d-4055-ae65-f438554229aa | low                 | network.outgoing.packets.rate   | None      | f5c79575-dbd2-5926-8f54-b1f67474529d |
| dcb02696-7fca-4285-92c3-6e3af3ea2af6 | low                 | network.incoming.bytes          | B         | f5c79575-dbd2-5926-8f54-b1f67474529d |
| fae412a9-b8f0-407c-9588-2284e695c595 | low                 | network.outgoing.bytes          | B         | f5c79575-dbd2-5926-8f54-b1f67474529d |

gnocchi measures show dcb02696-7fca-4285-92c3-6e3af3ea2af6
+---------------------------+-------------+--------------+
| timestamp                 | granularity |        value |
+---------------------------+-------------+--------------+
| 2017-03-28T19:35:00+00:00 |       300.0 |        684.0 |
| 2017-03-28T19:45:00+00:00 |       300.0 |        684.0 |
+---------------------------+-------------+--------------+

gnocchi measures show fae412a9-b8f0-407c-9588-2284e695c59
+---------------------------+-------------+--------+
| timestamp                 | granularity |  value |
+---------------------------+-------------+--------+
| 2017-03-28T19:35:00+00:00 |       300.0 | 2928.0 |
| 2017-03-28T19:45:00+00:00 |       300.0 | 2928.0 |
+---------------------------+-------------+--------+

When I add debugging output to the metrics_capture code, I do not see the net_usage_rate_average counter being updated, so that may explain why the values are null in the database. 

I see the disk_usage_rate_average counter being updated. For  disk_usage_rate_average the meters' values are also coming in as 0. There maybe a sourcing problem here and with net_usage_rate_average.
Comment 13 Richard Su 2017-03-30 22:24:08 UTC 
Hi Taeho,

Network io metrics are missing because the measures in Gnocchi are associated not with the instance but with the instance's network interface. CloudForms, however, only fetches the metrics using the instance's resource id. We will need to make a modification to also fetch metrics using network interface's resource id.

I will post another update when the fix is ready. 

Thanks,

Richard
Comment 17 Richard Su 2017-04-11 06:26:18 UTC 
Hi Taeho,

The fix is in progress. I will post an link to the PR when available.

Thanks,

Richard
Comment 18 Richard Su 2017-04-13 01:32:35 UTC 
Fix posted for review: https://github.com/ManageIQ/manageiq-providers-openstack/pull/12
Comment 21 Ido Ovadia 2018-02-08 16:21:00 UTC 
Verified
========
5.9.0.20