hello there,
Its time to update one of servers to an SSD, websites are too demanding these days.. so the long pending move from proxmox to cloudmin would happen.. i would be going for the ubuntu 16 lts version as the base OS. Have one cloudmin running on ubuntu 14 LTS, so Q1. is 16 supported and work fine for KVM? LXC for example? Q2: i have had no luck making snapshot function work with LVM to work on the 14 LTS.. apparently the OS did not play well with the installation, i did consult with you guys but we could not make it work.. so basically LVM and snapshot did not work-- do care to comment on this with 16 LTS? Q3: i will have 480 GB of SDD on soft raid, what could be advantage of LVM as against the normal ext3/4 installation. Q4: i do plan to have 4-5 VPS running on this.. with virtualmin, as special recomendation for the cloudmin installation? I would like to get the backup/restore/snapshot functions working.. so just checking?
Comments
Submitted by andreychek on Wed, 12/27/2017 - 11:48 Comment #1
Ubuntu 16.04 should work great with KVM.
KVM is going to be one of the simplest Virtual Machine types to get working, and most distros have built-in support for it.
As far as LVM snapshots go -- Jamie, are there any distro/version requirements for LVM snapshots?
Submitted by JamieCameron on Wed, 12/27/2017 - 12:27 Comment #2
No, snapshots should work on all recent and reasonably recent Linux distros, as long as the kernel is new enough.
Submitted by masterg0g0 on Wed, 12/27/2017 - 16:33 Pro Licensee Comment #3
ok good, i confirm snapshots are working fine..
I was able to get the bridge setup for hertzner root servers.. using my old documentation https://re-views.tech/cloudmin-installation-hetzner-root-server/ .. but i am having trouble with the LXC networking to work.. i simply installed LXC packages
https://www.itzgeek.com/how-tos/linux/ubuntu-how-tos/setup-linux-contain... and tried to configure the host.. when i create the LXC guest, it does work..
also LXC config seems to be complaining about it not able to control/limit memory, which is weird.. will post some details..
Submitted by andreychek on Thu, 12/28/2017 - 09:33 Comment #4
Okay, we'll wait to see what those details are.
Note that we've experienced a number of issues with networking at Hetzner. Unfortunately, we've often had to tell folks that we can only help if they switch away from Hetzner.
When we've been able to help folks, it's often been with the help of your guide :-)
But we'll certainly take a look at the issues you're seeing and see what we can do to help!
Submitted by masterg0g0 on Mon, 01/01/2018 - 14:58 Pro Licensee Comment #5
Hi,
That good to hear that my guide has helped people, if you want to you can re-direct people to me..i can help them further..
Coming to the issue above, on my server 1(ubuntu 14.04) with the below network config.. lxc works fine..
Hetzner Online GmbH - installimage
Loopback device:auto lo br0 eth0 iface lo inet loopback iface lo inet6 loopback
device: eth0iface eth0 inet static address 144.76.8.226 netmask 255.255.255.255 pointopoint 144.76.8.225 gateway 144.76.8.225
iface br0 inet static address 144.76.8.226 netmask 255.255.255.255 bridge_ports none bridge_stp off bridge_fd 0 up ip route add 144.76.8.244/32 dev br0 up ip route add 144.76.8.249/32 dev br0 up ip route add 144.76.8.250/32 dev br0
on the server2(ubuntu 16.04), with the below network config.. things dont work..
Hetzner Online GmbH installimage
source /etc/network/interfaces.d/*
auto lo br0 eth0 iface lo inet loopback iface lo inet6 loopback
iface eth0 inet static address 94.130.220.55 netmask 255.255.255.255 gateway 94.130.220.1 pointopoint 94.130.220.1 post-up iptables-restore < /etc/iptables.up.rules
iface eth0 inet6 static
pre-up /sbin/modprobe -q ipv6 ; /bin/trueiface br0 inet static address 94.130.220.55 netmask 255.255.255.192
bridge_ports none bridge_stp off bridge_hello 2 bridge_maxage 12 bridge_fd 0 up ip route add 94.130.220.35/32 dev br0 up ip route add 94.130.220.36/32 dev br0 up ip route add 94.130.220.37/32 dev br0
iface br0 inet6 static address 2a01:4f8:13b:1807::2 netmask 64 gateway fe80::1
both servers are on hetzner root baremetals.. i guess difference is the ipv6 config(which i dont understand i guess anyway)
any hints or directions, i could go ahead with ?
Submitted by masterg0g0 on Mon, 01/01/2018 - 15:33 Pro Licensee Comment #6
ok got the LXC networking to also work.. apparently we still have that bug where the networking setting for the LXC does not take the gateway settings from the ip address from the pool.. let me say it works partially..
the networking file inside of the container should have only the ip address and the gateway(which should be the host) this is what seems to work with the hertzner installations.. i had to go inside of the container and then remove all the others parameters.. is there way to somehow generate a custom networking file for the LXC containers, via the cloudmin settings/gui?
Submitted by JamieCameron on Mon, 01/01/2018 - 21:32 Comment #7
You could define a post-VM-creation script that does custom configuration to setup network config files.
These can be setup at Cloudmin Settings -> Cloudmin Configuration -> Pre and post-modification scripts.
Submitted by masterg0g0 on Tue, 01/02/2018 - 08:18 Pro Licensee Comment #8
ok i might need some help to get that to work... but this is not priority for me for now.. i wanted to ensure that nothing was broken on the cloudmin or hetzner side..
Can i ask you if upgrading the server 1 host from ubuntu to 14 to 16 LTS pose any risk? i could potentially keep a backup of the guest images...
Submitted by andreychek on Tue, 01/02/2018 - 09:07 Comment #9
There's always a risk when doing an in-place upgrade like that, and that impact can be bigger when it's a host that runs multiple Virtual Machine guests.
We unfortunately haven't done testing of in-place upgrades on a Cloudmin host.
That said, my gut tells me you'd likely be fine moving from Ubuntu 14.04 to 16.04, but I'd definitely put together a backup and/or contingency plan :-)