venet problem

[blueice]

Hi,
before 2 days i have see the promox and seems very interest for my virtual servers.
I have make a test server (i7 920 with 8 G ram).
The server have public internet ip and the virtual servers also.

If i put venet ethernet in my vps then canot have internet. Can only ping the ip of the main server but nothing else.
If i put veth, all working fine.
Do you have any idea why?
the config is:

parnassus:~# pveversion -v
pve-manager: 1.4-6 (pve-manager/1.4/4340)
qemu-server: 1.1-4
pve-kernel: 2.6.24-15
pve-kvm: not correctly installed
pve-firmware: 1
vncterm: 0.9-2
vzctl: 3.0.23-1pve3
vzdump: 1.2-2
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1

(as you can see the pve-kvm is not available. I have try the version 1.3 and it is available (86-1)).

The vps is just plain debian 5.0 from Certified Appliances.

CPU BOGOMIPS: 45479.81
REGEX/SECOND: 726259
HD SIZE: 94.49 GB (/dev/pve/root)
BUFFERED READS: 121.84 MB/sec
AVERAGE SEEK TIME: 9.04 ms
FSYNCS/SECOND: 669.39
DNS EXT: 15.34 ms
DNS INT: 408.39 ms (test.com)

parnassus:~# cat /etc/vz/conf/101.conf
# PVE default config for 256MB RAM

ONBOOT="no"

# Primary parameters
NUMPROC="1024:1024"
NUMTCPSOCK="9223372036854775807:9223372036854775807"
NUMOTHERSOCK="9223372036854775807:9223372036854775807"
VMGUARPAGES="524288:9223372036854775807"

# Secondary parameters
KMEMSIZE="9223372036854775807:9223372036854775807"
OOMGUARPAGES="524288:9223372036854775807"
PRIVVMPAGES="524288:536788"
TCPSNDBUF="9223372036854775807:9223372036854775807"
TCPRCVBUF="9223372036854775807:9223372036854775807"
OTHERSOCKBUF="9223372036854775807:9223372036854775807"
DGRAMRCVBUF="9223372036854775807:9223372036854775807"

# Auxiliary parameters
NUMFILE="9223372036854775807:9223372036854775807"
NUMFLOCK="9223372036854775807:9223372036854775807"
NUMPTY="255:255"
NUMSIGINFO="1024:1024"
DCACHESIZE="9223372036854775807:9223372036854775807"
LOCKEDPAGES="262144:262144"
SHMPAGES="9223372036854775807:9223372036854775807"
NUMIPTENT="9223372036854775807:9223372036854775807"
PHYSPAGES="0:9223372036854775807"

# Disk quota parameters
DISKSPACE="20971520:23068672"
DISKINODES="4000000:4400000"
QUOTATIME="0"
QUOTAUGIDLIMIT="0"

# CPU fair sheduler parameter
CPUUNITS="1000"
CPUS="1"
VE_ROOT="/var/lib/vz/root/$VEID"
VE_PRIVATE="/var/lib/vz/private/$VEID"
OSTEMPLATE="debian-5.0-standard_5.0-1_i386"
ORIGIN_SAMPLE="pve.auto"
IP_ADDRESS="80.76.58.54"
HOSTNAME="blue.np.org"
NAMESERVER="217.20.116.1 217.20.115.1"
SEARCHDOMAIN="test.com"

I have try and version 1.3 but i still have the same problem.
Do you have any idea what is wrong?

 

[yatesco]

Newbie myself, but I would suggest reading: http://wiki.openvz.org/Differences_between_venet_and_veth

FYI - I followed http://www.myatus.co.uk/2009/08/31/guide-firewall-and-router-with-proxmox/ so all my VMs were private IP addressed with all the public IPs mapped to the host. Shorewall made the port forwarding of public IP/port to private IP/port trivial.

 

[dietmar]

If i put venet ethernet in my vps then canot have internet. Can only ping the ip of the main server but nothing else.

Please try to use 'tcpdump' to see whats wrong.

parnassus:~# pveversion -v
pve-manager: 1.4-6 (pve-manager/1.4/4340)
qemu-server: 1.1-4
pve-kernel: 2.6.24-15
pve-kvm: not correctly installed
pve-firmware: 1
vncterm: 0.9-2
vzctl: 3.0.23-1pve3
vzdump: 1.2-2
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1

You can safely ignore that error (known bug which does not harm).

 

[blueice]

Please try to use 'tcpdump' to see whats wrong.

ok i have see this:

17:47:31.775073 IP 78.159.98.210.internetserviceteam.com > smolikas.wt5.net: ICMP echo request, id 8963, seq 1, length 64
17:47:31.775592 IP smolikas.wt5.net > 78.159.98.210.internetserviceteam.com: ICMP echo reply, id 8963, seq 1, length 64
17:47:31.835382 00:00:00:00:00:00 (oui Ethernet) Null > 00:d0:bc:d4:08:02 (oui Unknown) Unknown DSAP 0x08 Supervisory, Receiver not Ready, rcv seq 110, Flags [Command], length 1500
17:47:31.835384 00:00:00:00:00:00 (oui Ethernet) Null > 00:d0:bc:d4:08:02 (oui Unknown) Unknown DSAP 0x08 Information, send seq 2, rcv seq 119, Flags [Poll], length 1263

the vps is the 78.159.98.210, the pve 188.72.202.67 and i have try from vps to ping the 78.159.112.18 (smolikas.wt5.net)

Do you have any idea?

 

[blueice]

hm, the vps ips are in diferent subnet from the proxmox server.
Maybe this make the problem? Have everyone do the same config?

 

[dietmar]

ok i have see this:

17:47:31.775073 IP 78.159.98.210.internetserviceteam.com > smolikas.wt5.net: ICMP echo request, id 8963, seq 1, length 64
17:47:31.775592 IP smolikas.wt5.net > 78.159.98.210.internetserviceteam.com: ICMP echo reply, id 8963, seq 1, length 64

So the ping works (there is an echo reply).

 

[blueice]

So the ping works (there is an echo reply).

hm, but never come back to the vps (in vps i still see timeout in pings).
Need to do anything if the vps have ip in diferent subnet from the main server?

 

[dietmar]

Need to do anything if the vps have ip in diferent subnet from the main server?

no. Do you have some kind of firewall installed?

 

[blueice]

no. Do you have some kind of firewall installed?

no, just plain proxmox and plain Certified Appliance debian 5.0

 

[dietmar]

Please can you try to set NEIGHBOUR_DEVS in /etc/vz.conf (see man vz.conf) (Or add an addition route for the vm subnet to venet0).

 

[blueice]

Please can you try to set NEIGHBOUR_DEVS in /etc/vz.conf (see man vz.conf) (Or add an addition route for the vm subnet to venet0).

i have change the NEIGHBOUR_DEVS to NEIGHBOUR_DEVS=eth0 in /etc/vz/vz.conf and all work fine now
thanks a lot for your help.