mtu significantly affects ingress throughput

Linux Advanced Routing and Traffic Control

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi all,

I'm using tc to limit ingress throughput per IP dimension. How I do
this is installing some tc rules on a bond interface, as the
followings:

tc qdisc add dev bond1 handle ffff: ingress
tc filter add dev bond1 parent ffff: protocol ip prio 1 u32 match ip
dst 1.1.1.1/32 police rate 1000mbit burst 100m drop flowid :1

After this setup, iperf shows max bandwidth less than 100Mbps, which
is unexpected. Then I've tried to modify some parameters, eventually I
find out that changing mtu parameter's value from default 2k to 100k
solves this problem.

tc filter add dev bond1 parent ffff: protocol ip prio 1 u32 match ip
dst 1.1.1.1/32 police rate 1000mbit burst 100m mtu 100k drop flowid :1

Although I can reach 1Gbps finally, I still cannot understand why mtu
matters here? AFAIK, mtu limits max packet size travels through tc, so
any value greater than 1500 makes no sense since mtu on physical
interface is 1500. This is really confusing, hope to get advice from
you.

Regards.
--
To unsubscribe from this list: send the line "unsubscribe lartc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [LARTC Home Page]     [Netfilter]     [Netfilter Development]     [Network Development]     [Bugtraq]     [GCC Help]     [Yosemite News]     [Linux Kernel]     [Fedora Users]
  Powered by Linux