Hi Jesse,
Sorry to bother you with so many questions. But my question from the
beginning has been how to configure gre tunnel for vm-to-vm communication
using openvswitch. I have got it to work for vms in the same subnet, but am
not able to for vms in different subnets. It would be very helpful to
I think at this point these questions don't particularly relate to
Open vSwitch, so perhaps some background reading on general network
principles would be appropriate first.
On Fri, Sep 9, 2011 at 8:21 PM, Aishwarya wrote:
> Hi Jesse,
>
> I was doing the following:
>
> ifconfig eth0 0.0.0.0 up fr
Hi Jesse,
I was doing the following:
ifconfig eth0 0.0.0.0 up from teh terminal.
Regarding gre tunnel, if one vm has an ip address of 192.168.0.1/24 and the
other has an ipaddress of 192.168.10.1/24, how do i configure gre to make
these 2 communicate. Right now, i have created a bridge br0 on bo
On Fri, Sep 9, 2011 at 8:08 PM, Aishwarya wrote:
> Hi Jesse,
>
> Ok.. this solution does not seem to work for me. The problem is that as soon
> as I take the physical interface eth0 off the network(i.e. assign it an ip
> of 0), all networking stops for me and the machine hangs. This happens every
Hi Jesse,
Ok.. this solution does not seem to work for me. The problem is that as soon
as I take the physical interface eth0 off the network(i.e. assign it an ip
of 0), all networking stops for me and the machine hangs. This happens every
time.
I tried the gre tunnel and that seems to work for me.
Logically, bridge should be assigned the IP address of physical interface,
not VM's.
--
Usman
On Sat, Sep 10, 2011 at 7:30 AM, Aishwarya wrote:
> Hi Usman,
>
> I kind of understand what you mean now:
>
> Actually I have created a tap0 interface(that is associated with the vm)
> that I have adde
That's correct but please don't drop on the mailing list in your responses.
On Fri, Sep 9, 2011 at 7:48 PM, Aishwarya wrote:
> Hi Jesse,
>
> Let me know if the following is correct according to what you explained:
>
>
> Initially, host1 has physical ip address 10.0.0.1(eth0 on the physical
> host
On Fri, Sep 9, 2011 at 7:08 PM, Aishwarya wrote:
> Hi Jesse,
> I dont understand... do i need to add eth0 to the bridge or not? Also,
> ultimately I want these vms to talk over the network... i.e. vm1 located on
> host1 should be able to talk to vm2 located on host2 where host1 and host2
> are co
Hi Jesse,
I dont understand... do i need to add eth0 to the bridge or not? Also,
ultimately I want these vms to talk over the network... i.e. vm1 located on
host1 should be able to talk to vm2 located on host2 where host1 and host2
are connected over a network to each other... Could you elaborate a
\On Fri, Sep 9, 2011 at 6:23 PM, Aishwarya wrote:
> HI Jesse,
>
> Yes, eth0 is my physical ethernet interface thats connected to the external
> network.
You need to move the IP address off of eth0 onto the bridge port.
Experimenting with networking changes while connected over the network
is also
HI Jesse,
Yes, eth0 is my physical ethernet interface thats connected to the external
network.
Thanks,
Aish
On Fri, Sep 9, 2011 at 6:20 PM, Jesse Gross wrote:
> On Fri, Sep 9, 2011 at 6:03 PM, Ben Pfaff wrote:
> > On Fri, Sep 09, 2011 at 05:06:44PM -0700, Aishwarya wrote:
> >> Added a bridge
On Fri, Sep 9, 2011 at 6:03 PM, Ben Pfaff wrote:
> On Fri, Sep 09, 2011 at 05:06:44PM -0700, Aishwarya wrote:
>> Added a bridge on both host 1 and host2. Also added a tap interface for the
>> vms on both hosts. Next, I added the tap interfaces to the bridge on both
>> hosts. Then I added eth0(the
On Fri, Sep 09, 2011 at 05:06:44PM -0700, Aishwarya wrote:
> Added a bridge on both host 1 and host2. Also added a tap interface for the
> vms on both hosts. Next, I added the tap interfaces to the bridge on both
> hosts. Then I added eth0(the physical interfaces of both hosts) to the
> bridge. As
Hi Ben,
So I tried what you suggested:
Added a bridge on both host 1 and host2. Also added a tap interface for the
vms on both hosts. Next, I added the tap interfaces to the bridge on both
hosts. Then I added eth0(the physical interfaces of both hosts) to the
bridge. As soon as I do this, the mac
On Tue, Aug 30, 2011 at 06:51:45PM -0500, Tyler Coumbes wrote:
> If you add a tap to a fake bridge(br101) and delete the tap without
> removing it from the bridge first ovs does not delete the port.
> If you add a tap to a regular bridge(br0) and delete the tap without
> removing it from the bridge
It's simplest not to use VLANs or tunnels.
VLANs are useful for partitioning an L2 network into multiple virtual
networks, but they require extra coordination.
Tunnels are useful for partitioning an L3 network into multiple
virtual networks, but they require even more coordination.
I suggest usi
On Fri, Sep 9, 2011 at 15:15, Ben Pfaff wrote:
> If you want to implement it and (this is important) test it, and the
> implementation is clean enough that it's clear it doesn't break
> anything else, then it seems reasonable to accept it.
>
I will certainly take a look at it.
--
Nick
_
The difference is that this doesn't use a GRE tunnel.
(Why should I have guessed that you wanted to configure a GRE tunnel?)
On Fri, Sep 09, 2011 at 12:14:11PM -0700, Aishwarya wrote:
> What is the difference between this and using the gre tunnel?
>
> Thanks,
> Aish
>
> On Fri, Sep 9, 2011 at 1
On Fri, Sep 09, 2011 at 03:11:57PM -0400, Nicholas Bastin wrote:
> On Fri, Sep 9, 2011 at 15:05, Ben Pfaff wrote:
>
> > It seems reasonable to me, if it solves David's problem.
>
> Well, it solves other problems too.. :-) (It mirrors an in-band mode that
> indigo has, so if you already are tagg
On Fri, Sep 09, 2011 at 12:09:51PM -0700, Aishwarya wrote:
> Lets say I have 2 hosts with ip address 10.0.0.1 and 10.0.0.2, both on the
> same subnet.
> Now, i have a vm1 with ip address 192.168.0.1 on host 1 and vm2 with ip
> address 192.168.0.2 on host 2. I have openvswitch installed on both host
On Fri, Sep 9, 2011 at 15:05, Ben Pfaff wrote:
> It seems reasonable to me, if it solves David's problem.
>
Well, it solves other problems too.. :-) (It mirrors an in-band mode that
indigo has, so if you already are tagging your in-band control traffic, it
makes sense to match that with more sp
On Fri, Sep 09, 2011 at 11:39:23AM -0700, David Erickson wrote:
> On 9/9/2011 9:54 AM, Ben Pfaff wrote:
> >On Fri, Sep 09, 2011 at 09:36:25AM -0700, David Erickson wrote:
> >>On 9/9/2011 9:16 AM, Ben Pfaff wrote:
> >>>On Thu, Sep 08, 2011 at 06:11:39PM -0700, David Erickson wrote:
> Hi All-
> >
On 9/9/2011 12:05 PM, Ben Pfaff wrote:
On Fri, Sep 09, 2011 at 02:55:18PM -0400, Nicholas Bastin wrote:
Would you accept a patch to OVS which allowed a user to configure a single
VLAN to which the in-band rules applied? Basically, instead of wildcarding
the VLAN tag on all the in-band rules, it
On Fri, Sep 09, 2011 at 12:04:35PM -0700, Aishwarya wrote:
> I am trying to get virtual machines located on 2 different physical hosts to
> communicate using openvswitch. Has an yone had any luck with this?
Yes, this works fine.
___
discuss mailing list
On Fri, Sep 09, 2011 at 02:55:18PM -0400, Nicholas Bastin wrote:
> Would you accept a patch to OVS which allowed a user to configure a single
> VLAN to which the in-band rules applied? Basically, instead of wildcarding
> the VLAN tag on all the in-band rules, it would optionally set them all to
>
HI All,
I am trying to get virtual machines located on 2 different physical hosts to
communicate using openvswitch. Has an yone had any luck with this?
Thanks,
Aishwarya
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/lis
I'm still parsing David's reply, but I have a suggestion that I floated by
him offline, which may solve some problems...maybe.. :-)
Would you accept a patch to OVS which allowed a user to configure a single
VLAN to which the in-band rules applied? Basically, instead of wildcarding
the VLAN tag on
On 9/9/2011 9:54 AM, Ben Pfaff wrote:
On Fri, Sep 09, 2011 at 09:36:25AM -0700, David Erickson wrote:
On 9/9/2011 9:16 AM, Ben Pfaff wrote:
On Thu, Sep 08, 2011 at 06:11:39PM -0700, David Erickson wrote:
Hi All-
I'm seeing some confusing behavior happening relating to ARPs and
OVS 1.2.1 (also
On 09.09.2011 20:09, Ben Pfaff wrote:
On Fri, Sep 09, 2011 at 08:05:00PM +0200, S?bastien Riccio wrote:
Okay thanks it's clear. I'm trying to find a way to be nearly sure
that on a xen host if a customer vm gets hacked and starts flooding
the network like hell, it doesn't render the whole host u
On Fri, Sep 09, 2011 at 08:05:00PM +0200, S?bastien Riccio wrote:
> Okay thanks it's clear. I'm trying to find a way to be nearly sure
> that on a xen host if a customer vm gets hacked and starts flooding
> the network like hell, it doesn't render the whole host unreachable
> (That's what happened
On 09.09.2011 19:54, Ben Pfaff wrote:
On Fri, Sep 09, 2011 at 07:47:52PM +0200, S?bastien Riccio wrote:
That's expected behavior. When new flows constantly pop up, it takes
CPU time to decide what to do with them, and eventually you run out of
CPU time. This will be true of any kind of smart s
On Fri, Sep 09, 2011 at 07:47:52PM +0200, S?bastien Riccio wrote:
> It's more a cpu usage problem than memory, last test i did:
>
> hping3 -q -S -L 0 -p 80 -i u50 192.168.50.2
>
> PID USER PR NI VIRT RES SHR S*%CPU* %MEMTIME+ COMMAND
> 2211 root 10 -10 7660 3816 1076 R *93
On 09.09.2011 19:34, Ben Pfaff wrote:
On Wed, Sep 07, 2011 at 08:53:14PM +0200, S?bastien Riccio wrote:
For the details about the versions:
root@xen-blade13:~# ovs-vswitchd --version
ovs-vswitchd (Open vSwitch) 1.2.1+build0
Compiled Sep 6 2011 01:01:15
OpenFlow versions 0x1:0x1
It's the one f
On Wed, Sep 07, 2011 at 08:53:14PM +0200, S?bastien Riccio wrote:
> For the details about the versions:
>
> root@xen-blade13:~# ovs-vswitchd --version
> ovs-vswitchd (Open vSwitch) 1.2.1+build0
> Compiled Sep 6 2011 01:01:15
> OpenFlow versions 0x1:0x1
>
> It's the one from de debian unstable re
On Fri, Sep 09, 2011 at 09:36:25AM -0700, David Erickson wrote:
> On 9/9/2011 9:16 AM, Ben Pfaff wrote:
> >On Thu, Sep 08, 2011 at 06:11:39PM -0700, David Erickson wrote:
> >>Hi All-
> >>I'm seeing some confusing behavior happening relating to ARPs and
> >>OVS 1.2.1 (also on 1.1.1 and probably 1.0.
Thanks Nick for the reference that was useful, I had seen it before but
forgot where it lived.
On 9/9/2011 9:16 AM, Ben Pfaff wrote:
On Thu, Sep 08, 2011 at 06:11:39PM -0700, David Erickson wrote:
Hi All-
I'm seeing some confusing behavior happening relating to ARPs and
OVS 1.2.1 (also on 1.1.
On Thu, Sep 08, 2011 at 06:11:39PM -0700, David Erickson wrote:
> Hi All-
> I'm seeing some confusing behavior happening relating to ARPs and
> OVS 1.2.1 (also on 1.1.1 and probably 1.0.1). This is a XS machine
> with 1 ethernet port, so in-band connection from OVS to the
> controller. Specifically
On Fri, Sep 09, 2011 at 07:28:46AM -0700, David Nguyen wrote:
> punix:/usr/local/var/run/openvswitch/db.sock: listen failed: No such file or
> directory
Does /usr/local/var/run/openvswitch exist? Create it.
___
discuss mailing list
discuss@openvswitch.
How can I do this step ?
create a /etc/init.d/local and run it ? it is not successful
% ovsdb-server --remote=punix:/usr/local/var/run/openvswitch/db.sock \
--remote=db:Open_vSwitch,manager_options \
--private-key=db:SSL,private_key \
On Fri, Sep 9, 2011 at 10:05, David Nguyen wrote:
> ovs-openflowd: Command not found.
> why got this errors ?
>
The standard ovs-vswitchd is a fully functional openflow datapath - use this
instead of ovs-openflowd.
> how can I create "start up" step in INSTALL.linux. db.sock allway not found
ovs-openflowd: Command not found.
why got this errors ?
how can I create "start up" step in INSTALL.linux. db.sock allway not found ??
can you help ?___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
On Thu, Sep 8, 2011 at 21:11, David Erickson wrote:
> Also is there a document that specifies OVSs behavior in in-band
> environments such as XS with one ethernet port? I would expect after the
> controller is connected that all packets coming in/out other than the
> controller connection (and p
On 9/9/2011 12:10 AM, Justin Pettit wrote:
On Sep 8, 2011, at 1:22 AM, David Erickson wrote:
On 9/7/2011 7:10 PM, Justin Pettit wrote:
On Sep 7, 2011, at 5:29 PM, David Erickson wrote:
Is there a specific procedure for moving between OVS versions in XenServer?
I've done this a few times, a
On Sep 8, 2011, at 1:22 AM, David Erickson wrote:
> On 9/7/2011 7:10 PM, Justin Pettit wrote:
>> On Sep 7, 2011, at 5:29 PM, David Erickson wrote:
>>
>>> Is there a specific procedure for moving between OVS versions in XenServer?
>>> I've done this a few times, and frequently I find myself in
44 matches
Mail list logo
