Tony Sarendal said: > Now about netstat on your openbsd box ? > netstat -I <interface> -w10
results: (netstat -I em1 -w1) em1 in em1 out total in total out packets errs packets errs colls packets errs packets errs colls 45461 0 23878 0 0 138684 0 138680 0 0 48678 0 25173 0 0 147717 0 147720 0 0 46782 0 24444 0 0 142449 0 142439 0 0 43420 0 22977 0 0 132808 0 132806 0 0 43880 0 23109 0 0 133964 0 133961 0 0 47932 0 24928 0 0 145733 0 145731 0 0 48065 0 24938 0 0 146007 0 146003 0 0 44539 0 22644 0 0 134363 0 134365 0 0 I tried one more thing, changing the bridges so they all land on the same IRQs, I thought the dual and quads would have 1 irq per card but doesn't seem like the case: em0 at pci3 dev 4 function 0 "Intel PRO/1000MF QP (82546EB)" rev 0x01: irq 5, address: 00:04:23:45:d9:20 em1 at pci3 dev 4 function 1 "Intel PRO/1000MF QP (82546EB)" rev 0x01: irq 10, address: 00:04:23:45:d9:21 em2 at pci3 dev 6 function 0 "Intel PRO/1000MF QP (82546EB)" rev 0x01: irq 3, address: 00:04:23:45:d9:22 em3 at pci3 dev 6 function 1 "Intel PRO/1000MF QP (82546EB)" rev 0x01: irq 11, address: 00:04:23:45:d9:23 em4 at pci4 dev 2 function 0 "Intel PRO/1000MT DP (82546EB)" rev 0x03: irq 3, address: 00:30:48:74:e0:86 em5 at pci4 dev 2 function 1 "Intel PRO/1000MT DP (82546EB)" rev 0x03: irq 11, address: 00:30:48:74:e0:87 em6 at pci7 dev 1 function 0 "Intel PRO/1000MT DP (82546EB)" rev 0x03: irq 5, address: 00:04:23:b3:d6:8e em7 at pci7 dev 1 function 1 "Intel PRO/1000MT DP (82546EB)" rev 0x03: irq 10, address: 00:04:23:b3:d6:8f doing this had no noticable impact on throughput or cpu time spend servicing interrupts. thanks nate
