OK, when I do a netstat -k ce1, i get the stats, but I also need to change them.
The command,
ndd -set /dev/ce instance 1
works ok, but
ndd -get /dev/ce link_speed
does not, I get `operation failed, invalid argument`
I have tried this with other options that I get out of the netstat -k command. I look through the /var/adm/messages file also, and see this error:
Sep 3 09:40:00 ebzees01 genunix: [ID 408822 kern.info] NOTICE: ce1: no fault external to device; service available
Sep 3 09:40:00 ebzees01 genunix: [ID 611667 kern.info] NOTICE: ce1: xcvr addr:0x01 - link up 100 Mbps half duplex
Then I get the same for ce2, which we don't use yet, but it is plugged in.
This started because of performance. The longer it is up, the slower it gets. If I go in via an HP Secure web console, it performs fine. I can ping it and the maching next to it, and get 70ms to the other one, but in excess of 1300ms to this machine. The port on the switch is hard set to 100M full, but the interface shows as up at half in the messages file. I can't figure out how to change it to full on the ce card.