[meta-virtualization] [PATCH] openvswitch: functionality sanity test

Joe Slater jslater at windriver.com
Thu Apr 25 09:30:49 PDT 2013

Instructions for creating a simple bridge that will take
over eth0.  You cannot be executing out of an nfs-mounted
filesystem while setting up the bridge.

Signed-off-by: Joe Slater <jslater at windriver.com>
 docs/vswitch_test.sh |   36 ++++++++++++++++++++++++++++++++++++
 1 files changed, 36 insertions(+), 0 deletions(-)
 create mode 100755 docs/vswitch_test.sh

diff --git a/docs/vswitch_test.sh b/docs/vswitch_test.sh
new file mode 100755
index 0000000..63168e3
--- /dev/null
+++ b/docs/vswitch_test.sh
@@ -0,0 +1,36 @@
+# runqemu defaults for the first "guest"
+# create a bridge
+ovs-vsctl add-br br-int
+# Add a physical interface to your virtual bridge for connectivity off box.
+# If your rootfs is nfs mounted (on eth0), you will die after attaching it
+# to the bridge.
+ovs-vsctl add-port br-int eth0
+# Zero out your eth0 interface and slap it on the bridge interface.
+ifconfig eth0 0
+ifconfig br-int ${my_inet4} netmask
+# Change your default route.  If you don't do this, nothing will
+# be reachable.
+route add default gw ${my_gw} br-int
+(You can then check that the connection works.)
+# The bridge configuration is persistant, so if something goes wrong
+# and you reboot, it will still be messed up.  Destroy the bridge,
+# then add a route, or reboot.
+ovs-vsctl del-br br-int
+ifconfig eth0 ${my_inet4}
+route add default gw ${my_gw} eth0

More information about the meta-virtualization mailing list