OpenVZ Forum: Support » Using veth device, why venet0 still generating?

Home » General » Support » Using veth device, why venet0 still generating?

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Using veth device, why venet0 still generating? [message #30260]

Mon, 19 May 2008 10:15

suaz
Messages: 4
Registered: May 2008

Junior Member

Hi,

first of all :
linux 2.6.18 kernel x86_64
028stab053 patch and kernel configuration
vzctl 3.0.22-1dso1
vzquota 3.0.11-1dso1
OS : debian etch
VEs : all running debian (hope i forgot nothing)

Here's my problem :

I'm working with veth device instead of venet. All my VEs have a config file who look like this :

ONBOOT="yes"

# UBC parameters (in form of barrier:limit)
# Primary parameters
NUMPROC="100:100"
NUMTCPSOCK="360:360"
NUMOTHERSOCK="360:360"
VMGUARPAGES="10240:9223372036854775807"
# Secondary parameters
KMEMSIZE="6144000:6785400"
TCPSNDBUF="1720320:2703360"
TCPRCVBUF="1720320:2703360"
OTHERSOCKBUF="1126080:2097152"
DGRAMRCVBUF="262144:262144"
OOMGUARPAGES="26112:9223372036854775807"
PRIVVMPAGES="10240:11240"
# Auxiliary parameters
LOCKEDPAGES="256:256"
SHMPAGES="21504:21504"
PHYSPAGES="0:9223372036854775807"
NUMFILE="9312:9312"
NUMFLOCK="188:206"
NUMPTY="16:16"
NUMSIGINFO="256:256"
DCACHESIZE="3409920:3624960"
#Helper values
AVNUMPROC="50:50"
NUMIPTENT="200:200"

# Disk quota parameters (in form of softlimit:hardlimit)
DISKSPACE="1048576:1153024"
DISKINODES="200000:220000"
QUOTATIME="0"

# CPU fair sheduler parameter
CPUUNITS="1000"

VE_ROOT="/var/lib/vz/root/$VEID"
VE_PRIVATE="/var/lib/vz/private/$VEID"
OSTEMPLATE="debian-4.0-test"
ORIGIN_SAMPLE="vps.basic"
NAMESERVER="91.121.126.131"
DEVICES="c:10:200:rw "
CAPABILITY="NET_ADMIN:on "
#NETIF="ifname=eth0,mac=00:0C:29:85:E5:63,host_ifname=veth101.0,host_mac=00:0C:29:E5:7E:3E"

NETIF="ifname=eth0,mac=00:0C:29:05:76:28,host_ifname=veth1005.0,host_mac=00:0C:29:5C:EB:3C"

eth0 is the interface inside the VE and vethx.0 is the one on the HN.
As you can see, there's no sign of any IP_ADDRESS field in this config file as it's supposed to be regarding to this howto : http://wiki.openvz.org/Virtual_Ethernet_device in " Making a veth-device persistent" section.

But when i start vz, all my containers are well created but with venet0 device inside and on the HN.

I forgot to say that i also patched the kernel with the patch given in this topic : http://forum.openvz.org/index.php?t=msg&goto=1993& but it doesn't have anything to do with venet device (i think...).

If i made myself clear, does anyone has any idea concerning this trouble?
Thank you very much for your answers

[Updated on: Mon, 19 May 2008 10:16]

Report message to a moderator

Re: Using veth device, why venet0 still generating? [message #30270 is a reply to message #30260]

Mon, 19 May 2008 13:01

suaz
Messages: 4
Registered: May 2008

Junior Member

Hum well,

for newbies like me who'd think it's much more complicated than it actually is, i just had to edit the /etc/network/interfaces file and remove the lines concerning venet0...

Thought i'd already done that before and posted... sorry guys.

Then just edit your template.

Report message to a moderator

Re: Using veth device, why venet0 still generating? [message #30271 is a reply to message #30260]

Mon, 19 May 2008 13:02

maratrus
Messages: 1495
Registered: August 2007
Location: Moscow

Senior Member

Hello,

Could you please clarify this statement

Quote:

But when i start vz, all my containers are well created but with venet0 device inside and on the HN.

There is a bug http://bugzilla.openvz.org/show_bug.cgi?id=478
Is this your case or you have another problem?

Report message to a moderator

Re: Using veth device, why venet0 still generating? [message #30288 is a reply to message #30271]

Tue, 20 May 2008 07:46

suaz
Messages: 4
Registered: May 2008

Junior Member

Hi,

I tried to follow the steps described in the bug report and i have the same problem.

When i create an IP address, and remove it after and then restart the VE, i have a venet interface coming up.

When i do this, there's a line in the VE conf file that remains :

IP_ADDRESS=""

and i thought this was the problem.
But then i manually removed this line and restarted the VE... and the same problem appears (venet0 coming up in the VE instead of eth0 configured with the veth as explained in my previous post).

My solution is to remove venet manually from /etc/network/interfaces file and then start on a template generated without this venet interface in this conf file.

I know this will work for me because i won't have to use the

vzctl set VEID --ipadd ipadd --save

command for any VE after but "happy" to know it's not normal behaviour.

I didn't find any patch (maybe i didn't search well on bugzilla and if yes sorry for that) and since this bug has been reported 2007-02-28 07:09, do you think it will be corrected?

Thanks for your concern Smile