Hetzner IP + /28 Subnet

Burak

New Member
Dec 14, 2015
15
0
1
34
Hi all,
I have a problem to bridge my new subnet to proxmox

My main IP is : 5.9.242.X

My Subnet is : 5.9.195.XXX /28

I cant configure my IP's to use in Proxmox anyone can help me ?
 
Hi Burak,
See my reply to another user a few threads below yours
https://forum.proxmox.com/threads/proxmox-4-hetzner-extra-subnet-configuration.25019/

This should give you enough info to start using your subet. How many bridge interfaces you end up with depends entirely on how many ip's you want to use with a single bridged interface. See for mine I ended up with one ip per bridge.

When I read up on the Hetzner routing part, it did make me wonder why they'd moved away from "managed" disparate MAC addresses that effectively puts them in complete control. I found it difficult to implement initially with Proxmox until I arrived at that config.

Hope that gets you running

Kindest regards
 
I wanna add my ips to one bridge only to use vm's but i tried ur configuration i cant access ur machine when i try it
 
OK, can you list contents of /etc/network/interfaces and anything in /etc/sysctl.conf not commented out?
Also what is the current state of your firewall ?
 
i am using debian i am not using firewall i am waiting to reboot machine because i restore my interfaces in rescue mode because when i set my ips its not working i cant access my server
 
Right, proxmox is based on debian, is this an install from the proxmox media or a manual installation?
 
Code:
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet static
        address  5.9.80.2XX
        netmask  255.255.255.224
        gateway  5.9.80.2XX
        up route add -net 5.9.80.2XX netmask 255.255.255.224 gw 5.9.80.2XX eth0

iface eth0 inet6 static
        address  2a01:4f8:161:XXXXX
        netmask  64
        gateway  fe80::1

Code:
Subnet:    5.9.195.1XX /29
Netmask:    255.255.255.248
Broadcast:    5.9.195.1XX
Usable IP addresses:

5.9.195.105 to 5.9.195.110

these are my subnet

Code:
net.ipv4.ip_forward = 1
net.ipv6.conf.all.forwarding = 1
net.ipv4.conf.default.proxy_arp = 1
net.ipv4.conf.all.rp_filter = 1
kernel.sysrq = 1
net.ipv4.conf.default.send_redirects = 1
net.ipv4.conf.all.send_redirects = 0


and my sysctl a added that
 
ok, see on mine I moved the ip address from the physical etho to bridge vmbr0 and I dont appear to have any extra routes.
In yours I don't see any extra bridge information. I wrote before, due to the Hetzner network it would appear you HAVE to sacrifice one of the additional ip's in the subnet to be able to route the rest. I suspect the only time you don't need this is if you're using pure linux containers.

Taking your config, I suspect this would work and then allow you to interface machines onto vmbr1:

Code:
iface lo inet loopback

iface eth0 inet manual

auto vmbr0
iface vmbr0 inet static
  address  5.9.80.2XX
  netmask  255.255.255.224
  gateway  5.9.80.2XX
  bridge_ports eth0
  bridge_stp off
  bridge_fd 0

auto vmbr1
iface vmbr1 inet manual
  bridge_ports none
  bridge_stp off
  bridge_fd 0

auto vmbr2
iface vmbr2 inet static
  address  5.9.195.XX
  netmask  255.255.255.224
  bridge_ports none
  bridge_stp off
  bridge_fd 0
 
so that is now correct.
You then assign your ip's in your vm's attached to vmbr1 and give it the gateway of the ip you assigned to vmbr2.
 
As I wrote, I don't believe you need any route beyond what you have. If you have created a bridge with an ip address from your subnet and you can ping that from the internet, simply use the other addresses on a vm connected to that same bridge and use the ip as gateway. try that
 
I cant connect and i wanna ask a question i have totally 8 ip in a /29 block and one is going for subnet start and one for broadcast i can use 6 ip only and if i assign an ip to vmbr i got 5 ip active, can i use the assigned ip of vmbr ? or i can use only 5 ip ?
 
Hi Burak, Hi All

I have the same questions as you do.. i am not able to use the last ip in the subnet /29. have tried multiple options... it will be good if anyone else more knowledgeable can confirm this..