@rhtyd I can't say how it should be. I noticed that behavior has been
changed...
For example, before upgrade I had such dhcp leases on vms:
~~~
lease {
interface "eth0";
fixed-address 172.16.0.25;
option subnet-mask 255.255.255.240;
option routers 172.16.0.17;
option dhcp-lease-time 2692800;
option dhcp-message-type 5;
option domain-name-servers 172.16.0.17,8.8.8.8; 10.9.x.x; <---------
172.16.0.17 is gateway on VR, 8.8.8.8 and 10.9.x.x - is public dns ips which
are set in ZONE settings: public dns1, public dns2
option dhcp-server-identifier 172.16.0.17;
option dhcp-renewal-time 1346400;
option broadcast-address 172.16.0.31;
option dhcp-rebinding-time 2356200;
option host-name "front-web-vm1";
option domain-name "cs16cloud.internal";
renew 2 2018/08/14 02:14:34;
rebind 0 2018/08/26 20:54:45;
expire 4 2018/08/30 18:24:45;
}
~~~
now I have
~~~
lease {
interface "eth0";
fixed-address 172.16.0.25;
option subnet-mask 255.255.255.240;
option routers 172.16.0.17;
option dhcp-lease-time 2692800;
option dhcp-message-type 5;
option domain-name-servers 172.16.0.17<--------- only VR ip stayed
option dhcp-server-identifier 172.16.0.17;
option dhcp-renewal-time 1346400;
option broadcast-address 172.16.0.31;
option dhcp-rebinding-time 2356200;
option host-name "front-web-vm1";
option domain-name "cs16cloud.internal";
renew 2 2018/08/14 02:14:34;
rebind 0 2018/08/26 20:54:45;
expire 4 2018/08/30 18:24:45;
}
~~~
So, from my point of view the behavior has been changed
[ Full content available at: https://github.com/apache/cloudstack/issues/2856 ]
This message was relayed via gitbox.apache.org for [email protected]
