[Openstack] [xenserver-ocata] ceilometer compute

Jianghua Wang jianghua.wang at citrix.com
Tue Sep 26 03:15:34 UTC 2017


Adhi,

I'm happy to know that you've resolved the Rrd_interface.Internal_error.

 By checking the logs you posted, the errors are all relative to measurements which are not implemented yet for XenServer. See the supporting matrix: https://docs.openstack.org/ceilometer/latest/admin/telemetry-measurements.html
 We're evaluating adding support for these measurements and put that in the development plan.
 In recent releases, upstream has been changed to log it as error for the unimplemented measurements. That doesn’t make sense indeed.

Regards,
Jianghua
------------------------------
Date: Mon, 25 Sep 2017 14:51:46 +0700
From: Adhi Priharmanto <adhi.pri at gmail.com>
To: Jianghua Wang <jianghua.wang at citrix.com>
Cc: "openstack at lists.openstack.org" <openstack at lists.openstack.org>
Subject: Re: [Openstack] FW: [xenserver-ocata] ceilometer compute
	error	Rrd_interface.Internal_error

Hi Jianghua,

Thanks for your response and attention. For the error of  "
Rrd_interface.Internal_error" , I was installed this update on my xenserver
(https://support.citrix.com/article/CTX225676) , and RRD error now was gone from ceilometer logs.

I'm still at ocata release , here is my openstack package list on my compute node

> [root at cmp1-oc-srg ceilometer]# rpm -qa |grep openstack 
> centos-release-openstack-ocata-1-1.el7.noarch
> openstack-selinux-0.7.13-2.el7.noarch
> openstack-nova-compute-15.0.0-1.el7.noarch
> openstack-neutron-common-10.0.1-1.el7.noarch
> openstack-ceilometer-polling-8.1.0-1.el7.noarch
> openstack-ceilometer-compute-8.1.0-1.el7.noarch
> openstack-neutron-openvswitch-10.0.1-1.el7.noarch
> openstack-neutron-ml2-10.0.1-1.el7.noarch
> openstack-ceilometer-common-8.1.0-1.el7.noarch
> openstack-nova-common-15.0.0-1.el7.noarch
> openstack-neutron-10.0.1-1.el7.noarch


And here's is my logs as your request
https://paste.fedoraproject.org/paste/Yv8DoGbLziE0~luY6qwbhQ

I still find error at ceilometer-compute logs like this one

2017-09-25 14:31:07.041 9048 INFO ceilometer.agent.manager [-] Polling
> pollster disk.device.iops in the context of all_pollsters
> 2017-09-25 14:31:07.042 9048 ERROR ceilometer.agent.manager [-] 
> Prevent pollster disk.device.iops from polling [<Server: prom-cl1>, <Server:
> mon-srg>, <Server: tehanget>] on source all_pollsters anymore!
> 2017-09-25 14:31:07.043 9048 INFO ceilometer.agent.manager [-] Polling 
> pollster disk.device.latency in the context of all_pollsters
> 2017-09-25 14:31:07.044 9048 ERROR ceilometer.agent.manager [-] 
> Prevent pollster disk.device.latency from polling [<Server: prom-cl1>, <Server:
> mon-srg>, <Server: tehanget>] on source all_pollsters anymore!


I think this error maybe make some of measurement won't works at gnocchi, and also some of gnocchi metric won't show the unit

[root at localhost ~]# gnocchi metric list |grep disk.device.iops
>
> +-----------------------------------+---------------------+---------------------------------+-----------+------------------------------------+
> | id                                | archive_policy/name | name
>                  | unit      | resource_id                        |
>
> +-----------------------------------+---------------------+---------------------------------+-----------+------------------------------------+
>
>> | 0062d222-2874-4a19-be34-a5bee3051d21 | low                 |
>> disk.device.iops                | None      |
>> 7b089e8e-5be7-599a-b279-15a8129f7bdd |
>
> | 3dc249e4-006d-47a3-935c-58e571ba086c | low                 |
>> disk.device.iops                | None      |
>> 0efeccca-8313-5aec-a1e2-955e091419f9 |
>
> | c2c19329-1709-4165-bdea-3bcf81c5d9d4 | low                 |
>> disk.device.iops                | None      |
>> 78aa9c1e-b199-551c-b295-fecb237639fc |
>
> | cfa1c4bf-591e-4229-bb76-7607fcba640e | low                 |
>> disk.device.iops                | None      |
>> bdf98233-442b-5fe4-aa02-ba567e487dcc |
>
>

Any suggest about the last error of ceilometer-compute  ?

On Mon, Sep 25, 2017 at 1:02 PM, Jianghua Wang <jianghua.wang at citrix.com>
wrote:

> Adhi,
>
>   Do you still have this problem?
>
>   The following errors may be caused due to this xvda disk doesn’t exist.
> Did you see any issue from the nova?
>
>         Rrd.Invalid_data_source(\\"vbd_xvda_read\\")")
>
>
>
> If you are still suffering from this issue, can you send me the 
> following log files:
>
> 1.       Log files for the nova-compute and ceilometer services located
> in nova compute node.
>
> 2.       The /var/log/xensource.log from dom0 where the compute node is
> running on.
>
>
>
> And can you confirm the release version? From the title – 
> [xenserver-ocata], it may be ocata. But by checking the code line from 
> the log, it’s probably already in pike.
>
>
>
> Regards,
>
> Jianghua



More information about the Openstack mailing list