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
http://repos.fedorapeople.org/repos/openstack/openstack-icehouse/epel-7

  [root@base /]# cat /etc/redhat-release
  CentOS Linux release 7.0.1406 (Core)

  [root@base /]# uname -r
  3.10.0-123.8.1.el7.x86_64

  [root@base /]# modinfo /lib/modules/`uname 
-r`/kernel/net/openvswitch/openvswitch.ko
  filename:       
/lib/modules/3.10.0-123.8.1.el7.x86_64/kernel/net/openvswitch/openvswitch.ko
  license:        GPL
  description:    Open vSwitch switching datapath
  srcversion:     1241855A733802E089FD201
  depends:        libcrc32c,vxlan,gre
  intree:         Y
  vermagic:       3.10.0-123.8.1.el7.x86_64 SMP mod_unload modversions
  signer:         CentOS Linux kernel signing key
  sig_key:        B4:0B:A1:CF:51:04:A1:14:69:97:51:5E:90:84:0A:8B:2C:9F:B1:44
  sig_hashalgo:   sha256

  [root@base ~]# rpm -q openvswitch
  openvswitch-2.0.0-6.el7.x86_64

Just so we are on the same page with terminology, when you say ARP
entry are you referring to the OVS fdb entry?

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 fdb/show br-inet | grep 
fe:dd:3f:1e:71:f6; sleep 1; done
  (working)
     14     0  fe:dd:3f:1e:71:f6    1
     14     0  fe:dd:3f:1e:71:f6    1
     14     0  fe:dd:3f:1e:71:f6    1
     14     0  fe:dd:3f:1e:71:f6    1
     14     0  fe:dd:3f:1e:71:f6    0
     14     0  fe:dd:3f:1e:71:f6    0
     14     0  fe:dd:3f:1e:71:f6    0
      1     0  fe:dd:3f:1e:71:f6    0 - rebooted lxc container
      1     0  fe:dd:3f:1e:71:f6    1
     15     0  fe:dd:3f:1e:71:f6    0 - fdb temporarily shows *correct* port
      1     0  fe:dd:3f:1e:71:f6    1 - fdb reverts back to port 1 for some 
reason
      1     0  fe:dd:3f:1e:71:f6    0
      1     0  fe:dd:3f:1e:71:f6    1
      1     0  fe:dd:3f:1e:71:f6    2
      1     0  fe:dd:3f:1e:71:f6    3
      1     0  fe:dd:3f:1e:71:f6    4
      1     0  fe:dd:3f:1e:71:f6    5
      1     0  fe:dd:3f:1e:71:f6    6
      1     0  fe:dd:3f:1e:71:f6    7
      1     0  fe:dd:3f:1e:71:f6    9
      1     0  fe:dd:3f:1e:71:f6   10
      1     0  fe:dd:3f:1e:71:f6   11
      1     0  fe:dd:3f:1e:71:f6   12
      1     0  fe:dd:3f:1e:71:f6   13
      1     0  fe:dd:3f:1e:71:f6   14
      1     0  fe:dd:3f:1e:71:f6   15
      1     0  fe:dd:3f:1e:71:f6   16
      1     0  fe:dd:3f:1e:71:f6   17
      1     0  fe:dd:3f:1e:71:f6   18
      1     0  fe:dd:3f:1e:71:f6    1 - ran 
                                        /sbin/arping  -U -A -c 5 -I inet 
192.168.0.12
                                        from container
     15     0  fe:dd:3f:1e:71:f6    1 - port id updated
      1     0  fe:dd:3f:1e:71:f6    1 - port id reset back to port 1
     15     0  fe:dd:3f:1e:71:f6    1 - port id back to correct value
     15     0  fe:dd:3f:1e:71:f6    1 - port id remains at correct
                                        value for lifetime of
                                        container, until next reboot
     15     0  fe:dd:3f:1e:71:f6    1
     15     0  fe:dd:3f:1e:71:f6    1

I'm not heavily experienced with OVS to know whether the above is
'normal' but it seems strange :)

Thanks,

Chris
_______________________________________________
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss

Reply via email to