There are two things, upgrade of system VM, and upgrade of software on system VM. Any time you start a system VM the latest scripts get copied to it. This isn't the same as using the new ipv6 template. I can understand needing to reboot the system VMS if the scripts change, but its not the same as upgrading the OS in the system VM. On Mar 27, 2013 8:59 PM, "David Nalley" <da...@gnsa.us> wrote:
> >> Deploying Vms in existing networks succeed. But Vms are not given any > ip address. > >> In router, I see that the mac address of the Vm is not populated > correctly. > >> root@r-4-VM:~# cat /etc/dhcphosts.txt > >> 02:00:42:47:00:01,set:10_1_1_17,10.1.1.17,test123,infinite > >> -4,set:10_1_1_195,10.1.1.195,-m,infinite > >> root@r-4-VM:~# > >> > >> After stopping and starting of the existing routers , Vm deployment > succeeds. > >> > >> Seems like all the routers need to be stopped and started after upgrade. > > > > Forgive the question, but I actually haven't upgraded a production > > environment yet! Is this normal to have to restart the VR's after a > > major update? I *think* I remember that this is, but I just want to > > confirm. > > > > So it's expected if we change the sysvms materially - but I would have > expected them to have continued working. > That said, I thought update of sysvms was optional and only necessary > if you wanted IPv6-enabled system VMs. > Can someone canonically answer this? Things like this will affect > install documentation. >
