Ubuntu
heat package

cpu metric not getting created for resource instance | Openstack wallaby | Centos

Bug #1989584 reported by swogat pradhan
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
heat (Ubuntu)
New
Undecided
Unassigned

Bug Description

Hi,
I am trying out the heat autoscalling and i am facing an issue where the instance created doesn't have cpu metric (enabled in ceilometer polling.yaml file)
(overcloud) [stack@hkg2director ~]$ gnocchi resource show 130a0dfd-4e43-4373-ad0c-b5ffd4fcca95
+-----------------------+---------------------------------------------------------------------+
| Field | Value |
+-----------------------+---------------------------------------------------------------------+
| created_by_project_id | 7c9659572662449da4906cc77b33f21b |
| created_by_user_id | 4cb066d9701440dea941c5157186d06d |
| creator | 4cb066d9701440dea941c5157186d06d:7c9659572662449da4906cc77b33f21b |
| ended_at | None |
| id | 130a0dfd-4e43-4373-ad0c-b5ffd4fcca95 |
| metrics | compute.instance.booting.time: c0f77048-a85d-48a3-92cd-cdca50b341e8 |
| | disk.ephemeral.size: 46d77e5d-aba1-4b93-92eb-3a97f9512a3c |
| | disk.root.size: 3f60cb9f-9c06-4ccc-b47d-d2ca899fe5ea |
| | memory: 97f796ed-c26c-4625-8670-64cf004a7b26 |
| | vcpus: 513908f9-1bba-429e-8106-3bf933f04a76 |
| original_resource_id | 130a0dfd-4e43-4373-ad0c-b5ffd4fcca95 |
| project_id | 5d922243077045c48fe4b075e386551b |
| revision_end | None |
| revision_start | 2022-09-14T13:00:53.216445+00:00 |
| started_at | 2022-09-14T12:57:21.846635+00:00 |
| type | instance |
| user_id | c1905e3c5a374924980e851840e7f028 |
+-----------------------+---------------------------------------------------------------------+

For other instances also i do not see any cpu metric.
Now the issue is as cpu metric is not getting created the alarm is not getting any data from the VM in terms of cpu usage and my autoscaling feature is failing.

Please suggest how to fix this issue.

Revision history for this message
lokendra (lokendrarathour-b) wrote :

Hi Swogat,
By default as the instance is created, I also see this as missing, but I saw this coming up as soon as I trigger some load on the VM/Instance.

Example: yes > /dev/null &
and then run below:

(overcloud) [stack@undercloud ~]$ gnocchi resource show --type instance e6890d6e-3e5f-4e8a-9399-fc262f57017e

+-----------------------+---------------------------------------------------------------------+
| Field | Value |
 +-----------------------+---------------------------------------------------------------------+
| availability_zone | nova |
| created_at | 2022-09-15T09:27:09+00:00 |
| created_by_project_id | 7abe36ce16574b53ad1de77e03e91906 |
| created_by_user_id | f52fb8c4d8da471abc16d92da4ff15f4 |
| creator | f52fb8c4d8da471abc16d92da4ff15f4:7abe36ce16574b53ad1de77e03e91906 |
| deleted_at | None |
| display_name | ab-qhje-o7tpsf4sbjsc-3yyzws2exifc-server-etgudmvrriqv |
| ended_at | None |
| flavor_id | af93b39a-c8a6-463d-afbd-bac5f6860f8c |
| flavor_name | m1.small |
| host | overcloud-novacompute-0.mydomain.com |
| id | e6890d6e-3e5f-4e8a-9399-fc262f57017e |
| image_ref | 56643348-40dc-4e21-8305-4b7f2563c888 |
| launched_at | 2022-09-15T09:27:17.725687+00:00 |
| metrics | compute.instance.booting.time: 9007c507-951a-4b70-8d08-13da5fffeca3 |
| | cpu: 84bb36a4-dc5f-4790-af7b-ac33b0132bcd |
| | disk.ephemeral.size: 0d3a2cbb-eea0-4943-9be0-3c10f98634b7 |
| | disk.root.size: 73710230-85a8-47f9-a1a3-272ac936ec7c |
| | memory.usage: fbdd7a3f-8f28-443f-ab24-53975f54f8c8 |
| | memory: f89f76b6-2a34-4b97-a036-b8f71ecf33c1 |
| | vcpus: 6a7b4161-9f02-4b20-805d-480eedeba4c3

Revision history for this message
swogat pradhan (swogat) wrote :

Hi lokendra
Thanks for the response.
In my case it worked after i restarted the gnocchi and ceilometer services on the controller nodes.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.