Hi Dedhi, Could you please let us know the device id of the e1000g NIC on your system? You can get it with the following command:
% prtconf -vpPD | grep "pciclass,0200" Thanks, Ted Dedhi Sujatmiko > James Carlson wrote: > >> I suggest contacting the driver-discuss list. What you're describing >> sounds a little like CR 6727837, and trying the workaround would be a >> good experiment, but that CR is supposed to have been fixed in build >> 103. > > Thanks for the pointer James, > > tried the LSO disable but the problem still persist. > > >> Another thing to try would be to plug that e1000g port into the >> 100Mbps switch and repeat the test, and vice-versa. It's at least >> possible that one of the switches is behaving badly, and finding the >> common denominator for the problem will help. (In other words, you've >> got several variables here: driver, wires, and switch are all >> separate.) > > Replace the : > a. NIC (got spare each) > b. wires with Cat 6 > > Problem still persist. > > Replace the switch with Fast Ethernet switch. The problem vanished, but > it can be because the e1000g running on 100BaseT is more stable. Need to > purchase Gigabit Ethernet switch but from different vendor, or I will > test the NIC to different OS. > >> I'm seeing a lot of retransmits and no errors, which suggests that >> something is trashing the packets such that they don't get through at >> all. > > Thanks for the commentary. > > Will test the above possibilities. > > best regards, > > Dedhi > _______________________________________________ > networking-discuss mailing list > [email protected] _______________________________________________ networking-discuss mailing list [email protected]
