[
https://issues.apache.org/jira/browse/CLOUDSTACK-1771?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13613309#comment-13613309
]
Alena Prokharchyk commented on CLOUDSTACK-1771:
-----------------------------------------------
The reason this bug was never caught before - we never allowed VR destroy and
networkRestart with cleanup=true for Shared networks. Then somebody added
support for that, but never assumed that on each VR recreation, it gets a new
ip address, and all the user vms in the network won't be able to send DHCP
requests to the VR.
The fix I'm putting in, will ensure that the IP address assigned to the very
first VR instance in the Shared network, will remain with this network for its
entire lifecycle, and will be re-used by all the VRs ever recreated in the
network. This is a big change for Shared network, will involve the DB upgrade
as well. I'll update the bug with instructions for QA once the fix is committed.
> IPv6 - Network restart for a dual network , results in the ipv4 address of
> the router to be changed. After netwpork restart , name resolution of the Vms
> fail.
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CLOUDSTACK-1771
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-1771
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Management Server
> Affects Versions: 4.1.0
> Environment: Build from 4.1
> Reporter: Sangeetha Hariharan
> Assignee: Alena Prokharchyk
> Fix For: 4.1.0
>
>
> IPv6 - Network restart for a dual network , results in the ipv4 address of
> the router to be changed. After netwpork restart , name resolution of the Vms
> fail
> Steps to reproduce the problem:
> Pre Req:
> 1.Create Dual Stack Shared Network by passing ipv6 params and ipv4 params
> 2. Deploy few Vms in this network.
> Steps:
> 1. Restart Network.
> Expected Behavior:
> After network restart , we should be able to access all the Vms within the
> network using its name.
> Actual Result:
> After network restart , we are not able to access all the Vms within the
> network using its name.
> This is because of the ipv4 address of the router being changed after
> network restart.
> Router had ip address - 10.223.136.69
> After reboot , Router now has ip address - 10.223.136.69
> But the guest Vms still point to the routers old ipaddress - 10.223.136.69
> for providing name resolution.
> sangeetha@test123:~$ cat /etc/resolv.conf
> # Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
> # DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
> nameserver 10.223.136.69
> nameserver 72.52.126.11
> nameserver 72.52.126.12
> search hello1361
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
