简体   繁体   English

是否可以使用神经节监控系统测量开放堆负载?

[英]Is it possible to measure the openstack load with ganglia monitoring system?

I am trying to figure out if it is possible to measure load on openstack say for ex CPU usage measurement on slave/child openstack VMs with Ganglia monitoring system? 我正在尝试找出是否可以测量Openstack上的负载,例如使用Ganglia监视系统对从/子OpenStack VM上的CPU使用率进行测量? I could not find any proper answers for this. 我找不到任何适当的答案。 Here I am using separate servers for measuring the load and for running VMs. 在这里,我使用单独的服务器来测量负载和运行VM。

Thanks for any help! 谢谢你的帮助!

technically, yes - although using it easily and "out of the box" with no twaeking might be troublesome. 从技术上讲,是的-尽管使用起来很方便,并且不需开箱即用,但可能会很麻烦。 Adding a ganglia agent to each of your openstack compute nodes is where you'd want to start, and from there its very dependent on what openstack network configuration youre using and if multicast is directly viable between all your nodes. 要在每个OpenStack计算节点上添加神经节代理,从那里开始,它取决于您使用的OpenStack网络配置以及组播是否在所有节点之间直接可行。

That will give you the hypervisor host details. 这将为您提供虚拟机监控程序主机的详细信息。 If you want the details of the VMs themselves, you're better using a plugin to get data from Libvirt (im assuming a KVM based openstack here, not the other hypervisors) for best/most honest numbers, although you could also run ganglia agents inside your VMs as well. 如果您需要虚拟机本身的详细信息,则最好使用插件从Libvirt获取数据(假设这里是基于KVM的openstack,而不是其他虚拟机管理程序),以获取最佳/最诚实的数字,尽管您也可以运行ganglia代理在您的虚拟机中。 Again, how they communicate (ie is multicast available) will detemine how easy/hard it is to get functioning. 同样,他们如何通信(即是否可以使用多播)将确定其运行的难易程度。

You can also explore possibility of Nagios for monitoring. 您还可以探索Nagios进行监视的可能性。 There is a plugin for ceilometer in nagios. nagios中有一个用于云高仪的插件。 ref: http://blog.zhaw.ch/icclab/nagios-ceilometer-integration-new-plugin-available/ 参考: http : //blog.zhaw.ch/icclab/nagios-ceilometer-integration-new-plugin-available/

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM