cluster（如果是基于zone-wide的primary storage，是可以跨cluster的）, affinity group,
capacity
在 2014年7月11日 上午5:35，虚无缥缈 83675...@qq.com写道：
一共10台实体机，vm迁移有个实体机显示not suitable，有点实体机显示suitable，请问显示not
suitable状态的原因，怎么查找，实体机负载都很低，刚刚建立，vm一共20左右。存储使用sharemountpoint.
大家好，这是我最近的测试数据。经测试发现KVM下新建的虚机与局域网内其他机器交互的速度仍然不正常。
我们公司经过了多伦测试，每次装好的CS4.2.1+KVM环境，都存在这个问题。
各位在KVM环境下都没遇到过这个性能问题（使用KVM作虚级化平台，KVM上的虚拟机与外界交换数据的速度很慢）？
我感觉这不是我们环境的问题，是CS自身的问题。
1.从FTP服务器上下载文件到被测试的KVM主机：
平均速度63M/S
ftp get centos.iso
local: centos.iso remote: centos.iso
227 Entering Passive Mode
你用的是advanced zone? 还是basic或者advanced zone with security group?
如果是后两个，流量不经过virtual router，所以可能跟cs没关系。
如果是第一个，流量都要经过virtual router，建议到virtual router上测试一下。
另外，看看virtual router和vm的配置（virsh
dumpxml），看看里面的bandwidth设置，inbound和outbound的单位都是KBps。
global
哦，对，即使是basic或者advanced-sg，vm.network.throttling.rate也会影响vm
在 2014年7月11日 上午8:48，Wei ZHOU ustcweiz...@gmail.com写道：
你用的是advanced zone? 还是basic或者advanced zone with security group?
如果是后两个，流量不经过virtual router，所以可能跟cs没关系。
如果是第一个，流量都要经过virtual router，建议到virtual router上测试一下。
另外，看看virtual
basic 和 advanced zone我们都测试过，I/O速度差不多，都很慢。
virtual router和vm的bandwidth配置都是：
inbound average='25600' peak='25600'/
outbound average='25600' peak='25600'/
global
setting里面的network.throttling.rate和vm.network.throttling.rate的默认值都是200
好像没有问题啊
张雷(Evan)
中航信华东数据中心有限责任公司
这么看可能是跟限速有关系。
200不够，你改大点，比如2000，2都可以。然后重启vm和router
在 2014年7月11日 上午9:11，evanitsharp evanitsh...@gmail.com写道：
basic 和 advanced zone我们都测试过，I/O速度差不多，都很慢。
virtual router和vm的bandwidth配置都是：
inbound average='25600' peak='25600'/
outbound average='25600' peak='25600'/
global
不应该是限速的问题，现在的软件，哪儿还会默认设置成几K的速度？
还是GOOGLE一下，看是不是哪儿有BUG
---对网络这儿也熟（以前没系统学过）-
说一下我们这儿用的情况，希望有帮助 ：
1. qemu-kvm版本：qemu-kvm-0.12.1.2-2.415.el6_5.10.x86_64
2. VM版本 Debian7
系统VM模板http://cloudstack.apt-get.eu/systemvm/systemvm64template-2014-01-14-master-kvm.qcow2.bz2
网络设置，eth0DHCP
你现在的这个问题和我去年的11月发现的问题几乎一样，寻遍了所有可能的解答，
始终没有找到原因，后来研究了近一个月，做了几十次不同的安装组 合，终于绕
过去了，由于这个问题，差点让我们放弃再使用CS了，不过我认为这个问题与CS的
版本有关，在4.1.1上不会出现这个问题，只有 4.2以后会有。我碰到这个问题是
在Ubuntu做CS安装环境下出现的。后来发现在CentOS下就没有了。 说这些供你参
考吧。
于 2014/7/11 15:11, evanitsharp 写道:
basic 和 advanced zone我们都测试过，I/O速度差不多，都很慢。
virtual
这么说就不是virtual router的问题了。
vm的配置文件是用的virtio网卡么？还是e1000？
vm里面ethtool显示的speed是千兆的么
在 2014年7月11日 上午10:28，evanitsharp evanitsh...@gmail.com写道：
你说改成2我试过了，还是老样子。
200其实够了，同样的配置，在XenServer和VMware平台下，速度就很快
张雷(Evan)
中航信华东数据中心有限责任公司
Mail:evanitsh...@gmail.com
Hi,
Due to power blackout i have to recreate ssvm cpvm, i am able to deploy
new vms also create new template but on System Capacity, General Alerts
area, latest alerts are not populating.
Also ssvm cpvm status is 'running' but Agent state is not displayed in
Cloudstack UI.
I have production
If i recreate ssvm cpvm, guest vm will be affected?
On Fri, Jul 11, 2014 at 11:50 AM, Tejas Gadaria refond.g...@gmail.com
wrote:
Hi,
Due to power blackout i have to recreate ssvm cpvm, i am able to deploy
new vms also create new template but on System Capacity, General Alerts
area,
NO it won't impact guet vms .destroy the ssvm and cpvm , it will recreate
again and during that time you can't create new guest vms.
--sadhu
-Original Message-
From: Tejas Gadaria [mailto:refond.g...@gmail.com]
Sent: 11 July 2014 11:53
To: users@cloudstack.apache.org;
Before destroying the cpvm and ssvm ,please analyse the logs. May be agent is
not up.
Also check below
https://cwiki.apache.org/confluence/display/CLOUDSTACK/SSVM%2C+templates%2C+Secondary+storage+troubleshooting
regards
sadhu
-Original Message-
From: Suresh Sadhu
Sent: 11 July 2014
So, I'm wondering whether the guest not having the interfaces configured
correctly (i.e. not having an IP) is just a symptom of more generally
broken networking - my interfaces file for the KVM host is below, does
anyone spot any issues?
auto lo
iface lo inet loopback
auto eth0
iface eth0
Where can I find a tutorial to modify systemVM image or create my own
systemVM template ?
--
Regards,
Tomasz Zięba
Twitter: @TZieba
LinkedIn: pl.linkedin.com/pub/tomasz-zięba-ph-d/3b/7a8/ab6/
http://pl.linkedin.com/pub/tomasz-zi%C4%99ba-ph-d/3b/7a8/ab6/
Hi Elliot,
Try recreating router (destroy the router and deploy new vm, router get
recreated).
After recreation if the problem still exists, check the router logs to see why
the interfaces are brought up.
Thanks,
jayapal
On 11-Jul-2014, at 1:38 PM, Elliot Berg elliot.b...@avcosystems.com
Rohit has put up a how-to here http://bhaisaab.org/logs/building-systemvms/
-abhi
On 11/07/14 2:27 pm, Tomasz Zięba t.a.zi...@gmail.com wrote:
Where can I find a tutorial to modify systemVM image or create my own
systemVM template ?
--
Regards,
Tomasz Zięba
Twitter: @TZieba
LinkedIn:
I have all vms HA enabled on ACS 4.3 on xenserver 6.2 .
I also want to enable auto start functionality in xen 6.2 for all vms.
but i found this
http://support.citrix.com/article/CTX133910
So will it be safe to enable auto start on xen6.2 with command line?
Regards,
Tejas
Hi,
I want to provide HA to VMs, currently i have created service offering with
'HA enable'
option. Hence VM is showing 'HA Enable'. but Host is hot HA enable.
Also for host If I change global setting 'ha.tag' = 'label' apply 'label'
to host tag it shows 'HA Enable'.
but while deploying vm
On Fri, Jul 11, 2014 at 12:01 AM, Harikrishna Patnala
harikrishna.patn...@citrix.com wrote:
Download vhd-util @
http://download.cloud.com.s3.amazonaws.com/tools/vhd-util, copy to
/opt/xensource/bin and /opt/cloud/bin on XenServer and give executable
permissions(a+x)
thanks a lot.
it works
Hi, All:
My CloudStack environment is 4.3.0 running on rhel 6.5 and kvm hypervisor also
running rhel 6.5.
When I create my first VM from an imported rhel 6.5 ISO image, the root
volume is identified as /dev/sda. Is there a way to force the root volume to
be identified as /dev/vda device ?
What do you mean by ³profile² ?
I am just doing interactive installation from ISO, using a 64 bit RHEL 6.5
server iso which I just uploaded.
Thanks,
Yiping
On 7/11/14, 12:02 PM, Nux! n...@li.nux.ro wrote:
Hi,
What profile did you choose for the ISO? It should be CentOS/RHEL, that
should
When you upload or edit the template iso using the GUI it brings up a drop down
for operating system. Which did you select ?
On Jul 11, 2014, at 12:10 PM, Yiping Zhang yzh...@marketo.com wrote:
What do you mean by ³profile² ?
I am just doing interactive installation from ISO, using a 64
Hm, I had it listed as “Other (64bit)”, since the list does not have rhel
6.5 (64bit), perhaps I need to select “RHEL 6.4(64bit)” ?
Yiping
On 7/11/14, 12:15 PM, Bret Mette bret.me...@dbihosting.com wrote:
When you upload or edit the template iso using the GUI it brings up a
drop down for
Thanks, Kirk, that make it to work for manual installation !
Now the next logic question is how to make this work with kickstart ? In
our environment, we normally use kickstart to (re)build all nodes, so it
would be extremely helpful to get kickstart working for VM¹s as well.
Yiping
On
If you set the global setting 'ha.tag' and tag the hosts then those hosts
become dedicated HA hosts as you observed.
If you want to use all your hosts for regular VM deployments, then you should
not ha tag them.
Just using 'ha.enabled' service offering while deploying the VMs should be
26 matches
Mail list logo
