Monitoring / Utilization tab remains disabled for OpenStack (Gnocchi) in Radjabov despite C&U enabled

[dhamo-ror]

Description

In ManageIQ Radjabov, we are using OpenStack with Gnocchi for metrics.
Gnocchi connectivity works from the Rails console, but the Monitoring / Utilization tab in the UI remains greyed out and no metrics or graphs are shown.

---

Environment

• ManageIQ version: Radjabov
• Provider: OpenStack
• Metrics service: Gnocchi

---

What Works

Gnocchi metrics are accessible from the Rails console:

ems = ManageIQ::Providers::Openstack::CloudManager.first
vm  = ems.vms.first

gnocchi_client = ems.connect(:service => "Metric")
gnocchi_client.list_meters([{"field" => "resource_id", "value" => vm.ems_ref}])

Valid meters (cpu, memory, vcpus, disk, etc.) are returned.

---

Issue

• Monitoring / Utilization tab is greyed out

• No metrics are shown:

  • On the OpenStack provider
  • On powered-on VMs

---

Steps Already Tried

1. Enabled Capacity & Utilization (C&U) in Region Settings
2. Restarted ManageIQ appliance
3. Checked Monitoring tab on powered-on VMs
4. Waited more than one week for rollups

Issue still persists.

---

Expected Behavior

Monitoring / Utilization tab should be enabled and display metrics when:

• C&U is enabled
• Gnocchi metrics are available

---

Actual Behavior

Monitoring / Utilization tab remains disabled and no metrics are shown.

[agrare]

@dhamo-ror bunch of initial questions:
Do you have a openstack metrics collector worker running?
If so, do you see that worker logging anything?
Do you have all 3 C&U roles enabled?
Do you have capture for all clusters and datastores enabled in the Region / C&U settings per the documentation

[agrare]

Also I would expect this to only impact cpu metrics, but this fix isn't on radjabov only spassky: #923

[dhamo-ror]

Metrics collector worker is running, I'll update the log status tomorrow, 3 C&U roles are enabled, capture for all clusters and datastores also enabled Adam

[agrare]

One last role question, do you have the scheduler role enabled on any appliances in the zone that your openstack is in?

[dhamo-ror]

Hi @agrare, I have the below configs:

1. OpenStack Metrics Collector Workers

From Rails console:

MiqWorker.where(:type => "ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker")

Result:

[#<ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker:0x00007f75e171b688
  id: 1000000002021,
  guid: "ffbb0155-b02a-4649-88e7-e2320d169286",
  status: "started",
  started_on: Fri, 23 Jan 2026 08:26:18.004680000 UTC +00:00,
  stopped_on: nil,
  last_heartbeat: Fri, 23 Jan 2026 08:26:21.444456000 UTC +00:00,
  pid: 225041,
  queue_name: "openstack",
  type: "ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker",
  percent_memory: nil,
  percent_cpu: nil,
  cpu_time: nil,
  os_priority: nil,
  memory_usage: nil,
  memory_size: nil,
  uri: nil,
  miq_server_id: 1000000000001,
  sql_spid: 225145,
  proportional_set_size: nil,
  unique_set_size: nil,
  system_uid: nil>,
 #<ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker:0x00007f75d5fe1d00
  id: 1000000002037,
  guid: "6ded26c2-1e24-485f-b92f-f224f8d389d8",
  status: "started",
  started_on: Fri, 23 Jan 2026 08:56:05.656842000 UTC +00:00,
  stopped_on: nil,
  last_heartbeat: Fri, 23 Jan 2026 08:56:08.444030000 UTC +00:00,
  pid: 225549,
  queue_name: "openstack",
  type: "ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker",
  percent_memory: nil,
  percent_cpu: nil,
  cpu_time: nil,
  os_priority: nil,
  memory_usage: nil,
  memory_size: nil,
  uri: nil,
  miq_server_id: 1000000000001,
  sql_spid: 225623,
  proportional_set_size: nil,
  unique_set_size: nil,
  system_uid: nil>]

From OS:

ps -ef | grep MetricsCollectorWorker

Result:

manageiq  225040       1  0 Jan23 ?        00:09:45 MIQ: Vmware::InfraManager::MetricsCollectorWorker id: 1000000002020, queue: vmware
manageiq  225041       1  0 Jan23 ?        00:09:41 MIQ: Openstack::CloudManager::MetricsCollectorWorker id: 1000000002021, queue: openstack
manageiq  225549       1  0 Jan23 ?        00:09:47 MIQ: Openstack::CloudManager::MetricsCollectorWorker id: 1000000002037, queue: openstack
manageiq  226146       1  0 Jan23 ?        00:09:45 MIQ: Vmware::InfraManager::MetricsCollectorWorker id: 1000000002045, queue: vmware

---

2. Metrics Worker Logs

From journald (PID 225041):

journalctl _PID=225041

Result:

[root@cmpnew ~]# journalctl _PID=225041
-- Logs begin at Sat 2025-10-25 07:37:11 EDT, end at Wed 2026-02-04 02:58:07 EST. --
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#config_out_of_date?) MIQ(ManageIQ::Providers::Openstack::CloudM>
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#heartbeat) MIQ(ManageIQ::Providers::Openstack::CloudManager::Me>
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#sync_config) ID [1000000002021], PID [225041], GUID [ffbb0155-b>
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm:
                                               ---
                                               :count: 2
                                               :cpu_request_percent: 5
                                               :cpu_threshold_percent: 100
                                               :gc_interval: 900
                                               :heartbeat_freq: 10
                                               :heartbeat_timeout: 120
                                               :memory_request: 524288000
                                               :memory_threshold: 1073741824
                                               :nice_delta: 3
                                               :parent_time_threshold: 180
                                               :poll: 3
                                               :poll_escalate_max: 30
                                               :poll_method: escalate
                                               :starting_timeout: 600
                                               :stopping_timeout: 600
                                               :dequeue_method: drb
                                               :queue_timeout: 600
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm: ---
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm:
                                               ---
                                               :guid: ffbb0155-b02a-4649-88e7-e2320d169286
Feb 03 20:12:26 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#heartbeat) MIQ(ManageIQ::Providers::Openstack::CloudManager::Me>
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#config_out_of_date?) MIQ(ManageIQ::Providers::Openstack::CloudM>
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#heartbeat) MIQ(ManageIQ::Providers::Openstack::CloudManager::Me>
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#sync_config) ID [1000000002021], PID [225041], GUID [ffbb0155-b>
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm:
                                               ---
                                               :count: 2
                                               :cpu_request_percent: 5
                                               :cpu_threshold_percent: 100
                                               :gc_interval: 900
                                               :heartbeat_freq: 10
                                               :heartbeat_timeout: 120
                                               :memory_request: 524288000
                                               :memory_threshold: 1073741824
                                               :nice_delta: 3
                                               :parent_time_threshold: 180
                                               :poll: 3
                                               :poll_escalate_max: 30
                                               :poll_method: escalate
                                               :starting_timeout: 600
                                               :stopping_timeout: 600
                                               :dequeue_method: drb
                                               :queue_timeout: 600
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm: ---
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm:
                                               ---
                                               :guid: ffbb0155-b02a-4649-88e7-e2320d169286
Feb 03 20:42:59 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#heartbeat) MIQ(ManageIQ::Providers::Openstack::CloudManager::Me>
Feb 03 21:13:01 cmp evm[225041]:  INFO -- evm: MIQ(ManageIQ::Providers::Openstack::CloudManager::MetricsCollectorWorker::Runner#config_out_of_date?) MIQ(ManageIQ::Providers::Openstack::CloudM>

---

3. C&U Roles

From Rails console:

MiqServer.first.server_roles
  .where(:name => ["ems_metrics_collector", "ems_metrics_coordinator", "ems_metrics_processor"])
  .pluck(:name, :active)

Result:

=> [["ems_metrics_collector", true], ["ems_metrics_coordinator", true], ["ems_metrics_processor", true]]

---

4. Scheduler Role

From Rails console:

MiqServer.first.server_roles.find_by(:name => "scheduler")

Result:

=>
#<ServerRole:0x00007f75d577c1d8
 id: 1000000000015,
 name: "scheduler",
 description: "Scheduler",
 created_on: Tue, 21 Oct 2025 09:22:59.235771000 UTC +00:00,
 updated_on: Tue, 21 Oct 2025 09:22:59.235771000 UTC +00:00,
 max_concurrent: 1,
 external_failover: false,
 role_scope: "region">

---

5. OpenStack Metrics Queue

From Rails console:

MiqQueue.where(:queue_name => "openstack").count

Result:

=> 0

---

6. Gnocchi Connectivity

From Rails console:

ems = ManageIQ::Providers::Openstack::CloudManager.first
vm  = ems.vms.first
gnocchi = ems.connect(:service => "Metric")
gnocchi.list_meters([{ "field" => "resource_id", "value" => vm.ems_ref }])

Result:

=> #<OpenStruct body=[{"name"=>"compute.instance.booting.time", "id"=>"0cc689ac-8249-4aa5-9349-57d451784bde"}, {"name"=>"cpu", "id"=>"51cf0c34-9986-4c2b-b936-7f1688e6c131"}, {"name"=>"disk.ephemeral.size", "id"=>"c4434127-e91c-412d-ba16-dcc772fca634"}, {"name"=>"disk.root.size", "id"=>"86d6847a-171c-44f0-b348-976251eb31e3"}, {"name"=>"memory", "id"=>"9ccd808d-614c-4d8b-a89b-4d5499ceff48"}, {"name"=>"memory.usage", "id"=>"1d162715-8451-40d9-a32a-e349d15932cd"}, {"name"=>"vcpus", "id"=>"0f1ea902-9334-4ca2-8413-10774b81d301"}]>

---

7. Performance Data

From Rails console:

MetricRollup.count

Result:

=> 32516

Vm.first.has_perf_data?

Result:

=> false

---

[dhamo-ror]

Hi @agrare, hope you are doing well. It’s been a while since we last discussed, do you have any insights or suggestions from your side?

[agrare]

Hey @dhamo-ror sorry for the delay.
I think we need to figure out if metrics captures are being scheduled at all, since from your metrics collector logs it looks like they aren't.
Do you see any logs for perf_capture_timer being queued? They will be queued by your schedule worker and dequeued by your ems_metrics_coordinator. (NOTE if you have multiple providers you'll see one for each so make sure you check that it is for your openstack specifically)