ip cluster problem

T

testbot

New Member
Jan 25, 2012
20
0
1
sorry, networking isn't my strongest area.

i have an extra /29 and our provider says we have to bind them w/virtual nics. to make them portable between the two hosts. is it required to have portable IPs for HA to work?

they want this:
Code: 
vmbr1:0
    address  192.1.[B]233[/B].112
    netmask  255.255.0.0
    gateway  192.1.228.25
vmbr1:1
    address  192.1.[B]233[/B].113
    netmask  255.255.0.0
    gateway  192.1.228.25
vmbr1:2
    address  192.1.[B]233[/B].114
    netmask  255.255.0.0
    gateway  192.1.228.25

i tried that it the openvz containers get redirected to the proxmox management console. then i turn them off and it works but only for a short time:
Code: 
ifconfig vmbr1:0 192.168.233.112 netmask 255.255.255.248 [B]up[/B]
ifconfig vmbr1:0 192.168.233.112 netmask 255.255.255.248 [B]down[/B]

i have also tried to add routes for that range but that doesn't help. it also shouldn't matter because they use the same default gateway. one of the problems might be that the /29 is on a different subnet than the gateway because it's being routed to the vlan.

Current config:

host1
Code: 
auto lo
iface lo inet loopback

iface eth0 inet manual

iface eth1 inet manual

auto vmbr0
iface vmbr0 inet static
    address  10.10.10.130
    netmask  255.255.255.192
    bridge_ports eth0
    bridge_stp off
    bridge_fd 0
    post-up route add -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129
    pre-down route del -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129


auto vmbr1
iface vmbr1 inet static
    address  192.1.[B]228[/B].26
    netmask  255.255.0.0
    gateway  192.1.228.25
    bridge_ports eth1
    bridge_stp off
    bridge_fd 0

host2
Code: 
auto lo
iface lo inet loopback

iface eth0 inet manual

iface eth1 inet manual

auto vmbr0
iface vmbr0 inet static
    address  10.10.10.130
    netmask  255.255.255.192
    bridge_ports eth0
    bridge_stp off
    bridge_fd 0
    post-up route add -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129
    pre-down route del -net 10.0.0.0 netmask 255.0.0.0 gw 10.32.10.129

auto vmbr1
iface vmbr1 inet static
    address  192.1.[B]228[/B].27
    netmask  255.255.0.0
    gateway  192.1.228.25
    bridge_ports eth1
    bridge_stp off
    bridge_fd 0
current route
Code: 
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.1.233.116   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.117   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.118   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.133   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.112   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.113   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.114   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.115   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.124   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.126   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.140   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.127   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.120   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.121   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.122   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.123   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.128   0.0.0.0         255.255.255.192 U     0      0        0 vmbr0
192.1.0.0       0.0.0.0         255.255.0.0     U     0      0        0 vmbr1
10.0.0.0        10.10.10.129   255.0.0.0       UG    0      0        0 vmbr0
0.0.0.0         192.1.228.25    0.0.0.0         UG    0      0        0 vmbr1
 
the more research i do the more i think this has to do with arp cache. if one vm goes down in a HA cluster, does it do anything to refresh the arp cache, like arping?
 
testbot said:
sorry, networking isn't my strongest area.

i have an extra /29 and our provider says we have to bind them w/virtual nics. to make them portable between the two hosts. is it required to have portable IPs for HA to work?

they want this:
Code: 
vmbr1:0
    address  192.1.[B]233[/B].112
    netmask  255.255.0.0
    gateway  192.1.228.25
vmbr1:1
    address  192.1.[B]233[/B].113
    netmask  255.255.0.0
    gateway  192.1.228.25
vmbr1:2
    address  192.1.[B]233[/B].114
    netmask  255.255.0.0
    gateway  192.1.228.25

i tried that it the openvz containers get redirected to the proxmox management console. then i turn them off and it works but only for a short time:
Code: 
ifconfig vmbr1:0 192.168.233.112 netmask 255.255.255.248 [B]up[/B]
ifconfig vmbr1:0 192.168.233.112 netmask 255.255.255.248 [B]down[/B]

i have also tried to add routes for that range but that doesn't help. it also shouldn't matter because they use the same default gateway. one of the problems might be that the /29 is on a different subnet than the gateway because it's being routed to the vlan.

Current config:

host1
Code: 
auto lo
iface lo inet loopback

iface eth0 inet manual

iface eth1 inet manual

auto vmbr0
iface vmbr0 inet static
    address  10.10.10.130
    netmask  255.255.255.192
    bridge_ports eth0
    bridge_stp off
    bridge_fd 0
    post-up route add -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129
    pre-down route del -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129


auto vmbr1
iface vmbr1 inet static
    address  192.1.[B]228[/B].26
    netmask  255.255.0.0
    gateway  192.1.228.25
    bridge_ports eth1
    bridge_stp off
    bridge_fd 0

host2
Code: 
auto lo
iface lo inet loopback

iface eth0 inet manual

iface eth1 inet manual

auto vmbr0
iface vmbr0 inet static
    address  10.10.10.130
    netmask  255.255.255.192
    bridge_ports eth0
    bridge_stp off
    bridge_fd 0
    post-up route add -net 10.0.0.0 netmask 255.0.0.0 gw 10.10.10.129
    pre-down route del -net 10.0.0.0 netmask 255.0.0.0 gw 10.32.10.129

auto vmbr1
iface vmbr1 inet static
    address  192.1.[B]228[/B].27
    netmask  255.255.0.0
    gateway  192.1.228.25
    bridge_ports eth1
    bridge_stp off
    bridge_fd 0
current route
Code: 
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.1.233.116   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.117   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.118   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.133   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.112   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.113   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.114   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.115   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.124   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.126   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.140   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.127   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.120   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.121   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.122   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
192.1.233.123   0.0.0.0         255.255.255.255 UH    0      0        0 venet0
10.10.10.128   0.0.0.0         255.255.255.192 U     0      0        0 vmbr0
192.1.0.0       0.0.0.0         255.255.0.0     U     0      0        0 vmbr1
10.0.0.0        10.10.10.129   255.0.0.0       UG    0      0        0 vmbr0
0.0.0.0         192.1.228.25    0.0.0.0         UG    0      0        0 vmbr1
Click to expand...
Hi,
there is something wrong with your config!
Host 1 and 2 has the same IP on vmbr0: 10.10.10.130!
on host 2 you deleta an route which you don't have: gw 10.32.10.129

Why you assign "ifconfig vmbr1:0 192.168.233.112" ?? Use simply bridged network for the client and use 192.168.233.112 inside the client as network address.

Udo
 
yea, the host2 was a typo. sorry.

i might have created this thread too soon. i added post-up echo 1 > /proc/sys/net/ipv4/conf/all/proxy_arp to the interfaces and it seems to be working so far. :D i was confused because our hosting provider insisted we added vmbr0:1, vmbr0:2... for each ip. i thought it was stupid.
 
You must log in or register to reply here.
Top Bottom