You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by li jerry <di...@hotmail.com> on 2019/04/23 14:00:58 UTC
答复: 答复: Cloudstack-agent gets the local IP address exception
Hi Dag Sonstebo
Because we only have three servers in total, we can't make one management node independently, so we can only downgrade and let them merge.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 21:35
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: 答复: Cloudstack-agent gets the local IP address exception
Hi Li,
Your configuration goes against best practice - you generally always keep management and hypervisors separate. To my knowledge there is nothing in the CloudStack KVM agent which will allow for the configuration you have in place, so you would have to deploy your own manual workarounds for this.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 23/04/2019, 13:59, "li jerry" <di...@hotmail.com> wrote:
HI Dag Sonstebo
We deploy management and agent on the same host, with three such nodes.
So we need to deploy keepalived on these three nodes, elect a VIP, and let all agents communicate with this VIP.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 17:08
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: Cloudstack-agent gets the local IP address exception
Li,
Why do you run with this configuration? If this is for network resilience you would run a bond/team, not keepalived.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 22/04/2019, 14:37, "li jerry" <di...@hotmail.com> wrote:
HI All
my kvm host [ip:10.226.16.11] virtualized a vip[10.226.16.10] via keepalived.
After I started the cloudstack-agent, I got vip[10.226.16.10] and I couldn't get the correct IP [10.226.16.11];
Can I have any way to get the cloudstack agent to get the correct IP address?
Cloudstack 4.11.2
CentOS 7.5
br2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether ac:1f:6b:ba:96:ea brd ff:ff:ff:ff:ff:ff
inet 10.226.16.11/24 brd 10.226.16.255 scope global noprefixroute br2
valid_lft forever preferred_lft forever
inet 10.226.16.10/24 scope global secondary br2
valid_lft forever preferred_lft forever
inet6 fe80::1456:f9ff:fe06:6228/64 scope link
valid_lft forever preferred_lft forever
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK
@shapeblue
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK
@shapeblue
答复: 答复: Cloudstack-agent gets the local IP address exception
Posted by li jerry <di...@hotmail.com>.
Hi Swen
Now our new environment is only three computing node servers, no other resources, so I can not but think of other solutions.
________________________________
发件人: Swen - swen.io <me...@swen.io>
发送时间: Tuesday, April 23, 2019 10:31:31 PM
收件人: users@cloudstack.apache.org
主题: AW: 答复: Cloudstack-agent gets the local IP address exception
Hi Li,
you can use a VM on another server for management, it does not need to be a physical server. Bit this VM needs to be not on the 3 servers which you are going to use for your Cloudstack VMs.
If you have any hypervisor outside of the 3 physical servers, than install the management server there.
cu Swen
-----Ursprüngliche Nachricht-----
Von: li jerry <di...@hotmail.com>
Gesendet: Dienstag, 23. April 2019 16:01
An: dev@cloudstack.apache.org; users@cloudstack.apache.org
Betreff: 答复: 答复: Cloudstack-agent gets the local IP address exception
Hi Dag Sonstebo
Because we only have three servers in total, we can't make one management node independently, so we can only downgrade and let them merge.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 21:35
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: 答复: Cloudstack-agent gets the local IP address exception
Hi Li,
Your configuration goes against best practice - you generally always keep management and hypervisors separate. To my knowledge there is nothing in the CloudStack KVM agent which will allow for the configuration you have in place, so you would have to deploy your own manual workarounds for this.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 23/04/2019, 13:59, "li jerry" <di...@hotmail.com> wrote:
HI Dag Sonstebo
We deploy management and agent on the same host, with three such nodes.
So we need to deploy keepalived on these three nodes, elect a VIP, and let all agents communicate with this VIP.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 17:08
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: Cloudstack-agent gets the local IP address exception
Li,
Why do you run with this configuration? If this is for network resilience you would run a bond/team, not keepalived.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 22/04/2019, 14:37, "li jerry" <di...@hotmail.com> wrote:
HI All
my kvm host [ip:10.226.16.11] virtualized a vip[10.226.16.10] via keepalived.
After I started the cloudstack-agent, I got vip[10.226.16.10] and I couldn't get the correct IP [10.226.16.11];
Can I have any way to get the cloudstack agent to get the correct IP address?
Cloudstack 4.11.2
CentOS 7.5
br2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether ac:1f:6b:ba:96:ea brd ff:ff:ff:ff:ff:ff
inet 10.226.16.11/24 brd 10.226.16.255 scope global noprefixroute br2
valid_lft forever preferred_lft forever
inet 10.226.16.10/24 scope global secondary br2
valid_lft forever preferred_lft forever
inet6 fe80::1456:f9ff:fe06:6228/64 scope link
valid_lft forever preferred_lft forever
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK
@shapeblue
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK @shapeblue
AW: 答复: Cloudstack-agent gets the local IP address exception
Posted by "Swen - swen.io" <me...@swen.io>.
Hi Li,
you can use a VM on another server for management, it does not need to be a physical server. Bit this VM needs to be not on the 3 servers which you are going to use for your Cloudstack VMs.
If you have any hypervisor outside of the 3 physical servers, than install the management server there.
cu Swen
-----Ursprüngliche Nachricht-----
Von: li jerry <di...@hotmail.com>
Gesendet: Dienstag, 23. April 2019 16:01
An: dev@cloudstack.apache.org; users@cloudstack.apache.org
Betreff: 答复: 答复: Cloudstack-agent gets the local IP address exception
Hi Dag Sonstebo
Because we only have three servers in total, we can't make one management node independently, so we can only downgrade and let them merge.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 21:35
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: 答复: Cloudstack-agent gets the local IP address exception
Hi Li,
Your configuration goes against best practice - you generally always keep management and hypervisors separate. To my knowledge there is nothing in the CloudStack KVM agent which will allow for the configuration you have in place, so you would have to deploy your own manual workarounds for this.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 23/04/2019, 13:59, "li jerry" <di...@hotmail.com> wrote:
HI Dag Sonstebo
We deploy management and agent on the same host, with three such nodes.
So we need to deploy keepalived on these three nodes, elect a VIP, and let all agents communicate with this VIP.
发件人: Dag Sonstebo<ma...@shapeblue.com>
发送时间: 2019年4月23日 17:08
收件人: users@cloudstack.apache.org<ma...@cloudstack.apache.org>; dev@cloudstack.apache.org<ma...@cloudstack.apache.org>
主题: Re: Cloudstack-agent gets the local IP address exception
Li,
Why do you run with this configuration? If this is for network resilience you would run a bond/team, not keepalived.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 22/04/2019, 14:37, "li jerry" <di...@hotmail.com> wrote:
HI All
my kvm host [ip:10.226.16.11] virtualized a vip[10.226.16.10] via keepalived.
After I started the cloudstack-agent, I got vip[10.226.16.10] and I couldn't get the correct IP [10.226.16.11];
Can I have any way to get the cloudstack agent to get the correct IP address?
Cloudstack 4.11.2
CentOS 7.5
br2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether ac:1f:6b:ba:96:ea brd ff:ff:ff:ff:ff:ff
inet 10.226.16.11/24 brd 10.226.16.255 scope global noprefixroute br2
valid_lft forever preferred_lft forever
inet 10.226.16.10/24 scope global secondary br2
valid_lft forever preferred_lft forever
inet6 fe80::1456:f9ff:fe06:6228/64 scope link
valid_lft forever preferred_lft forever
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK
@shapeblue
Dag.Sonstebo@shapeblue.com
www.shapeblue.com<http://www.shapeblue.com>
Amadeus House, Floral Street, London WC2E 9DPUK @shapeblue