On Jul 29, 2009, at 5:57 AM, Mr liu wrote:
[b]Unfortunately however, even with the TCP tweaks enabled, I still
only get a little over 1 Gb/s through the two aggregated Gig ports.
It seems as though the aggregation is only using one port,[/b]
though MRTG graphs of the two switch ports do in fact show that they
are both being utilized equally, essentially splitting the 1 Gb/s
speed between
the two ports.
Problem with the server? switch? Aggregation software? All the
above? At any rate, I seem to be missing something.. Any help
regarding this issue would be greatly appreciated!
There's no problem at all, actually. You're seeing the intended
behavior of LCAP as a protocol.
In order to prevent out-of-order ethernet frames that can be caused by
blindly splitting traffic across 2 (or more) LCAP group links, the
source/destination IP address is hashed and traffic to/from those
hosts are sent down one link to ensure o-o-o frames are kept to a
minimum. This is why your tests show one of your two links as busy,
and not a 50/50 split of the traffic load.
Again, this is a property of LCAP, not of Solaris's implementation of
it. With LCAP, you're likely never to get near a 50/50 balance of
traffic across two links.
/dale
_______________________________________________
networking-discuss mailing list
[email protected]
