an empty ebtables rule/chain is created for every LxBrNet
the ebtables rules are really only used for the WlanNode when using the basic
range model, for on/off connectivity
some quick tests suggest that this may have a negative impact on performance
for hubs/switches, where ebtables chains are not used:
*TWO PCs CONNECTED BY SWITCH*
iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[ 4] local 10.0.0.21 port 5001 connected with 10.0.0.20 port 50846
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 3.57 GBytes 3.07 Gbits/sec
[ 5] local 10.0.0.21 port 5001 connected with 10.0.0.20 port 50855
[ 5] 0.0-10.0 sec 4.27 GBytes 3.67 Gbits/sec
[ 4] local 10.0.0.21 port 5001 connected with 10.0.0.20 port 50875
[ 4] 0.0-10.0 sec 4.39 GBytes 3.77 Gbits/sec
[ 5] local 10.0.0.21 port 5001 connected with 10.0.0.20 port 50878
[ 5] 0.0-10.0 sec 4.33 GBytes 3.72 Gbits/sec
[ 4] local 10.0.0.21 port 5001 connected with 10.0.0.20 port 50879
[ 4] 0.0-10.0 sec 4.26 GBytes 3.65 Gbits/sec
*TWO PCs CONNECTED BY SWITCH* - with ebtables chain manually removed
n4.conf# iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[ 4] local 10.0.1.21 port 5001 connected with 10.0.1.20 port 42657
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 4.54 GBytes 3.90 Gbits/sec
[ 5] local 10.0.1.21 port 5001 connected with 10.0.1.20 port 42658
[ 5] 0.0-10.0 sec 4.44 GBytes 3.81 Gbits/sec
[ 4] local 10.0.1.21 port 5001 connected with 10.0.1.20 port 42662
[ 4] 0.0-10.0 sec 4.45 GBytes 3.82 Gbits/sec
[ 5] local 10.0.1.21 port 5001 connected with 10.0.1.20 port 42663
[ 5] 0.0-10.0 sec 4.39 GBytes 3.76 Gbits/sec
[ 4] local 10.0.1.21 port 5001 connected with 10.0.1.20 port 42672
[ 4] 0.0-10.0 sec 4.44 GBytes 3.81 Gbits/sec
Original issue reported on code.google.com by ahrenh...@gmail.com on 27 Aug 2013 at 5:02
Original issue reported on code.google.com by
ahrenh...@gmail.com
on 27 Aug 2013 at 5:02