Need help with cluster configuration

T

twister666

New Member
May 11, 2011
14
0
1
Montreal, Quebec, Canada
Hello everyone,

I just started playing with PVE and love its capabilities but having trouble with cluster configuration. Hope someone could help me out.

This is my configuration:
vmsrv0:
vmrb0 (eth1) IP:10.0.0.2/24 GW:10.0.0.1/24 DNS:10.0.0.1/24 [100mbps]
vmbr1 (eth0) IP:192.168.0.1/24 [1gbps]

vmsrv1:
vmbr0 (eth1) IP:10.0.0.3/24 GW:10.0.0.1/24 DNS:10.0.0.1/24 [100mbps]
vmbr1 (eth0) IP:192.168.0.2/24 [1gbps]

*Two nodes are connected with crossover cable [ vmsrv0(eth0) - vmsrv1(eth1) ]
*vmsrv0(eth1) connects to router, vmsrv1(eth1) connects to router.

The cluster synchronizes fine and each host can ping each other on 192.168.0.0/24 and the 10.0.0.0/24 network

Problems:
1. When I try to move running virtual machine (kvm) with live migration from one node to the other I get an error saying thay I cannot do live migration because I am running local storage. There is no way of running live migration without san?
2. When I try to move a virtual machine (kvm) that is stopped everything works fine but the speed is too slow (arround 11mbps) and from chacking adapter's activity to migrate machines it uses 10.0.0.0/24 network instead of 192.168.0.0/24. Is there a way to make the servers use the faster connection if available and slower connection for backup?

vmsrv0:
pve-manager: 1.8-15 (pve-manager/1.8/5754)
running kernel: 2.6.32-4-pve
proxmox-ve-2.6.32: 1.8-32
pve-kernel-2.6.32-4-pve: 2.6.32-32
qemu-server: 1.1-30
pve-firmware: 1.0-11
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-11
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.14.0-2
ksm-control-daemon: 1.0-5

vmsrv1:
pve-manager: 1.8-15 (pve-manager/1.8/5754)
running kernel: 2.6.32-4-pve
proxmox-ve-2.6.32: 1.8-32
pve-kernel-2.6.32-4-pve: 2.6.32-32
qemu-server: 1.1-30
pve-firmware: 1.0-11
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-11
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.14.0-2
ksm-control-daemon: 1.0-5



I will appreciate any help! Thanks
 
Last edited:
twister666 said:
1. When I try to move running virtual machine (kvm) with live migration from one node to the other I get an error saying thay I cannot do live migration because I am running local storage. There is no way of running live migration without san?
Click to expand...

Only work with OpenVZ containers. KVM needs shared storage. (or DRBD)

twister666 said:
2. When I try to move a virtual machine (kvm) that is stopped everything works fine but the speed is too slow (arround 11mbps) and from chacking adapter's activity to migrate machines it uses 10.0.0.0/24 network instead of 192.168.0.0/24. Is there a way to make the servers use the faster connection if available and slower connection for backup?
Click to expand...

Proxmox always use the 'vmbr0' network. So you should simply assign the fast network to vmbr0.
 
Last edited by a moderator:
Thank you dietmar for clarifying. I will setup NFS on a third computer to use as central storage for the cluster. Concerning the network speed I am still confused. The vmbr0 is the only one that has ip/subnet/gateway, the consequent bridge devices only have ip/subnet fields. How can I fix that problem? Any suggestions? Thanks
 
twister666 said:
Thank you dietmar for clarifying. I will setup NFS on a third computer to use as central storage for the cluster. Concerning the network speed I am still confused. The vmbr0 is the only one that has ip/subnet/gateway, the consequent bridge devices only have ip/subnet fields. How can I fix that problem? Any suggestions? Thanks
Click to expand...
Hi,
have a look at "/etc/network/interfaces". You can swap there vmbr0 with vmbr1 (reboot after that).
See also the ip used in /etc/hosts.
If you swap vmbr0/1 you must update your vms - simply change "vlan0: e1000=xx:xx:xx:xx:xx:xx" to "vlan1: e1000=xx:xx:xx:xx:xx:xx" in the config-files (below /etc/qemu-server/).

Udo
 
Hello,
Thanks for instractions. I switched vmbr0 with vmbr1 like you said and now I have 192.168.0.0/24 network set for vmbr0 but it seems like the speed didn't increase.

/usr/sbin/qmigrate 10.0.0.3 101
May 24 18:53:03 starting migration of VM 101 to host '10.0.0.3'
May 24 18:53:03 copying disk images
vm-101-disk-1.raw

rsync status: 4194762752 13% 11.14MB/s 0:40:55


my /etc/network/interfaces
vmsrv0:
auto vmbr0
iface vmbr0 inet static
address 192.168.0.1
netmask 255.255.255.0
bridge_ports eth0
bridge_stp off
bridge_fd 0
auto vmbr1
iface vmbr1 inet static
address 10.0.0.2
netmask 255.255.255.0
gateway 10.0.0.1
bridge_ports eth1
bridge_stp off
bridge_fd 0

on vmsrv1:
auto vmbr0
iface vmbr0 inet static
address 192.168.0.2
netmask 255.255.255.0
bridge_ports eth0
bridge_stp off
bridge_fd 0
auto vmbr1
iface vmbr1 inet static
address 10.0.0.3
netmask 255.255.255.0
gateway 10.0.0.1
bridge_ports eth1
bridge_stp off
bridge_fd 0

Is there anything else I should do to force it to use that 192.168.0.0/24 network?

I didn't know what to change on /etc/hosts (it seemed to be ok) and my /etc/qemu-server is empty.
 
Last edited:
Hello once again,
Thanks dietmar for suggestion. I verified /usr/sbin/qmigrate, however, didn't find any statement with IP address inside. I am noob in linux, if you could please give me some steps i will greatly appreciate.

After configuring NFS I can do live migration of KVM machines but still takes long time to transfering.
 
I don't know how to use other IP address. I did the migration with web interface. Should I do it instead with cli. If yes, could you please tell me the right way of doing it. Thanks so much in advance and sorry for asking so many questions.
 
twister666 said:
I don't know how to use other IP address. I did the migration with web interface.
Click to expand...

That is confusing. You posted a command lin, so I assumed you executed that from CLI. So where did you get that command line (which should use the other IP address)?
 
You must log in or register to reply here.
Top Bottom