New Question
0

Ceilometer-agent-compute service problem in openstack kilo

asked 2015-08-25 07:26:24 +0300

JYH gravatar image

I install nova(cloudbase kilo version) on windows 2012 r2 datacenter(Hyper-V).

But i cat get data ( network.[incoming] [outgoing].[bytes], [packets] ). And so all values

root@controller:~# ceilometer sample-list -q 'resource_id=instance-00000065-a90f05f2-b84d-47c2-95ba-43911df48046-7512ce87-ab3c-4545-9e68-edb0af28c77c'

+--------------------------------------+---------------------------------------------------------------------------------------------+-------------------------------+------------+--------+----------+---------------------+ | ID | Resource ID | Name | Type | Volume | Unit | Timestamp | +--------------------------------------+---------------------------------------------------------------------------------------------+-------------------------------+------------+--------+----------+---------------------+ | 9c0a01f1-4ad1-11e5-ae98-001d92b59ee5 | instance-00000065-a90f05f2-b84d-47c2-95ba-43911df48046-7512ce87-ab3c-4545-9e68-edb0af28c77c | network.incoming.bytes | cumulative | 0.0 | B | 2015-08-25T02:33:03 |

| 9c7a28e1-4ad1-11e5-8ae8-001d92b59ee5 | instance-00000065-a90f05f2-b84d-47c2-95ba-43911df48046-7512ce87-ab3c-4545-9e68-edb0af28c77c | network.incoming.bytes.rate | gauge | 0.0 | B/s | 2015-08-25T02:33:03 |

This value is always 0

window compute node log

---------- ceilometer-agent-compute.log -------------

2015-08-25 11:53:00.588 3992 WARNING ceilometer.agent.base [-] Continue after error from disk.allocation: 'AgentManager' object has no attribute 'inspector'

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base Traceback (most recent call last):

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\ceilometer\agent\base.py", line 177, in pollandpublish

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base resources=polling_resources

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\ceilometer\compute\pollsters\disk.py", line 709, in get_samples

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base {'inspector': manager.inspector.class.name,

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base AttributeError: 'AgentManager' object has no attribute 'inspector'

2015-08-25 11:53:00.588 3992 TRACE ceilometer.agent.base

2015-08-25 11:53:00.588 3992 INFO ceilometer.agent.base [-] Polling pollster network.outgoing.bytes in the context of meter_source

2015-08-25 11:53:02.276 3992 INFO ceilometer.agent.base [-] Polling pollster disk.write.requests in the context of meter_source

2015-08-25 11:53:02.292 3992 INFO ceilometer.agent.base [-] Polling pollster disk.device.write.requests in the context of meter_source

2015-08-25 11:53:02.308 3992 INFO ceilometer.agent.base [-] Polling pollster memory.resident in the context of meter_source

2015-08-25 11:53:02.308 3992 ERROR ceilometer.compute.pollsters.memory [-] Could not get Resident Memory Usage for a90f05f2-b84d-47c2-95ba-43911df48046: 'HyperVInspector' object has no attribute 'inspectmemoryresident'

2015-08-25 11:53:02.308 3992 TRACE ceilometer.compute.pollsters.memory Traceback (most recent call last):

2015-08-25 11:53:02.308 3992 TRACE ceilometer.compute.pollsters.memory File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\ceilometer\compute\pollsters\memory.py", line 76, in get_samples

2015-08-25 11:53:02.308 3992 TRACE ceilometer.compute.pollsters.memory memoryinfo = self.inspector.inspectmemory_resident(

2015-08-25 11:53:02.308 3992 TRACE ceilometer.compute.pollsters.memory AttributeError: 'HyperVInspector' object has no attribute 'inspectmemoryresident'

2015-08-25 11:53:02.308 3992 TRACE ceilometer.compute.pollsters.memory

2015-08-25 11:53:02.308 3992 INFO ceilometer.agent.base [-] Polling pollster disk.read.bytes in the context of meter_source

2015-08-25 11:53:02.308 3992 INFO ceilometer.agent.base [-] Polling pollster disk.device.read.requests.rate in the context of meter_source

2015-08-25 11:53:02.323 3992 INFO ceilometer.agent.base [-] Polling pollster disk.device.usage in the context of meter_source

2015-08-25 11:53 ... (more)

edit retag flag offensive close merge delete

2 answers

Sort by » oldest newest most voted
0

answered 2015-09-29 23:11:48 +0300

Claudiu Belu gravatar image

Hello,

At a first glance, it looks like metrics collection was not enabled for that instance. In order to enable metrics collection, you will have to have enable it in the configuration files:

  • for nova, for compute related metrics like cpuusage, memoryusage, diskread, diskwrite, disk_iops, etc., in nova.conf, in the hyperv section:

    [hyperv]
    enable_instance_metrics_collection=true
    
  • for neutron, for networking related metrics, in neutron_hyperv_agent.conf, in the AGENT section:

    [AGENT]
    enable_metrics_collection=true
    

By default, those config options are set to true if you have used our installer and chose to install the ceilometer-agent-compute as well. Otherwise, they are set to false.

Another important note, if the nova config option was false and you've set them to true, the nova-compute service will not enable the metrics collection for any pre-existing instance. neutron-hyperv-agent will turn on metrics collection for all bound ports.

If, for some reason, this issue persists and there are still missing metrics, there might be some errors enabling metrics collection. In this case, please include the nova-compute.log and neutron-hyperv-agent.log files.

Best regards,

Claudiu Belu

edit flag offensive delete link more
0

answered 2015-09-01 08:47:01 +0300

JYH gravatar image

I resolved.

This command "Get-VM {instance_name} | Enable-VMResourceMetering" or "Enable-VMResourceMetering *"

After I get metering data.

PS C:\Users\administrator.JYH> Get-VM instance-0000000f | Measure-VM

VMName AvgCPU(MHz) AvgRAM(M) MaxRAM(M) MinRAM(M) TotalDisk(M) NetworkInbound(M) NetworkOutbound(M) ------ ----------- --------- --------- --------- ------------ ----------------- ------------------ instance-0000000f 41809 0 0

PS C:\Users\administrator.JYH> Get-VM instance-0000000f | Enable-VMResourceMetering

PS C:\Users\administrator.JYH> Get-VM instance-0000000f | Measure-VM

VMName AvgCPU(MHz) AvgRAM(M) MaxRAM(M) MinRAM(M) TotalDisk(M) NetworkInbound(M) NetworkOutbound(M) ------ ----------- --------- --------- --------- ------------ ----------------- ------------------ instance-0000000f 5 1024 1024 1024 41809 73 4

root@controller:~# ceilometer sample-list -q "resource_id=instance-0000000f-41738d67-6ed0-444f-9253-6336a889992b-ce1c6766-87d2-41fc-9e0d-03f679ddfbce"

| 151cbe4f-504c-11e5-bd2e-001d92b59ee5 | instance-0000000f-41738d67-6ed0-444f-9253-6336a889992b-ce1c6766-87d2-41fc-9e0d-03f679ddfbce | network.outgoing.bytes | cumulative | 112197632.0 | B | 2015-09-01T01:52:20 |

| 143ee170-504c-11e5-9663-001d92b59ee5 | instance-0000000f-41738d67-6ed0-444f-9253-6336a889992b-ce1c6766-87d2-41fc-9e0d-03f679ddfbce | network.outgoing.bytes.rate | gauge | 370289.214521 | B/s | 2015-09-01T01:52:19 |

| 143a2680-504c-11e5-8b8c-001d92b59ee5 | instance-0000000f-41738d67-6ed0-444f-9253-6336a889992b-ce1c6766-87d2-41fc-9e0d-03f679ddfbce | network.incoming.bytes | cumulative | 7550795776.0 | B | 20

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

[hide preview]

Question Tools

1 follower

Stats

Asked: 2015-08-25 07:26:24 +0300

Seen: 803 times

Last updated: Sep 29 '15