Hey Chris,
I do not attach two ifaces to the same vswitch in vsphere. This makes you
have two ifaces in the same subnet... and there is no stp for loop-free
guarantee.
I think it is possible that when you enable the veth from HV, it sends out
the
IPv6 multicast pkts => if you have another VM wi
Hi Alex/List,
Any further ideas on troubleshooting this?
I can fix it by adding some network noise to rc.local but that seems
kinda hacky.
Is there a way to debug what frames are causing insert or updates to
fdb entries to see why this might be happening?
Regards,
Chris
___
> Sorry for delayed reply,
No problem at all.
> Do you have other iface from the same VM connected to the same esx
> switch as ens224?
Yes.
vswitch0
|
|-- ens192(base) 192.168.0.11
|
|-- ens224(base) -- OVS BR br-dmz -- veth/lxc 192.168.0.12
However, ens192 is not part of OVS:
[
Hey Chris,
Sorry for delayed reply,
I did more experiment on the netns and ovs, worked fine,
Do you have other iface from the same VM connected to the same esx
switch as ens224?
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http:
Hi Alex,
> Is there more info on this 'ens224'?
During a reboot of the container, capturing on the 'outside' ens224
port:
[root@base~]# tcpdump -nni ens224 ether src host fe:dd:3f:1e:71:f6 -e
tcpdump: WARNING: ens224: no IPv4 address assigned
tcpdump: verbose output suppressed, use -v or -vv for
Hey Chris,
On Mon, Sep 29, 2014 at 5:27 PM, Chris Bennett wrote:
> Hi Alex,
>
> > May I know what is port 1 and port 15? (via `ovs-ofctl show
> > br-inet`)
>
> The 'base' is a VM on VMWare btw, maybe that has something to do with
> the weird behaviour?
>
This should not be related.
> [root@
Hi Alex,
> May I know what is port 1 and port 15? (via `ovs-ofctl show
> br-inet`)
The 'base' is a VM on VMWare btw, maybe that has something to do with
the weird behaviour?
[root@base ~]# ovs-ofctl show br-inet
OFPT_FEATURES_REPLY (xid=0x2): dpid:005056a921dc
n_tables:254, n_buffers:256
cap
>
>
>
> Just so we are on the same page with terminology, when you say ARP
> entry are you referring to the OVS fdb entry?
>
Yes,
> Further testing and I've observed a little bit of weirdness (IMO) with
> the fdb reporting during a reboot & arping.
>
> [root@base~]# while true; do ovs-appctl
Hi Alex,
> May I know which release are you using?
>
> Did a similar test on my local setup using scapy. The arp entry is
> refreshed expectedly with Gratuitous ARP.
Thanks for the reply.
It's centos kernel & in-kernel openvswitch module, and userspace is
openvswitch 2.0 sourced from repo at
ht
Hey Chris,
May I know which release are you using?
Did a similar test on my local setup using scapy. The arp entry is
refreshed expectedly with Gratuitous ARP.
Thanks,
Alex Wang,
On Sat, Sep 27, 2014 at 9:54 PM, Chris Bennett wrote:
> Hi Ben,
>
> > If you're not using an OpenFlow controller,
Hi Ben,
> If you're not using an OpenFlow controller, then I'd expect OVS to
> learn from any packet sent by a VM.
I would have thought so too.
Here's some detailed info I'm capturing to show working/non-working:
# bit of background on setup
[root@phys ~]# ovs-vsctl get-controller br-inet
[roo
On Sun, Sep 28, 2014 at 01:36:17PM +0930, Chris Bennett wrote:
> I've setup LXC with OVS on CentOS7 yesterday and am having problems with
> OVS learning the MAC address of a container.
>
> I am only seeing ovs-appctl fdb/show reporting the correct port
> id for the container mac address when the
Hi there,
I've setup LXC with OVS on CentOS7 yesterday and am having problems with
OVS learning the MAC address of a container.
I am only seeing ovs-appctl fdb/show reporting the correct port
id for the container mac address when the container generates an ARP
reply, either in response to an ARP
13 matches
Mail list logo
