Actual ring memory is rounded up to pow of 2, instead /proc/net/pf_ring/info contains the original value.
Alfredo On Jun 4, 2013, at 5:38 PM, frwa onto <[email protected]> wrote: > Dear All, > We just cat into this file /proc/net/pf_ring/1774-eth0.1 below is > the results. > > Bound Device(s) : eth0 > Active : 1 > Breed : Non-DNA > Sampling Rate : 1 > Capture Direction : RX+TX > Socket Mode : RX only > Appl. Name : pfcount > IP Defragment : No > BPF Filtering : Disabled > # Sw Filt. Rules : 0 > # Hw Filt. Rules : 0 > Poll Pkt Watermark : 1 > Num Poll Calls : 447484 > Channel Id Mask : 0xFFFFFFFF > Cluster Id : 0 > Slot Version : 15 [5.5.3] > Min Num Slots : 6368 > Bucket Len : 128 > Slot Len : 328 [bucket+header] > Tot Memory : 2097152 > Tot Packets : 305059 > Tot Pkt Lost : 0 > Tot Insert : 305059 > Tot Read : 305058 > Insert Offset : 1672888 > Remove Offset : 1672576 > TX: Send Ok : 0 > TX: Send Errors : 0 > Reflect: Fwd Ok : 0 > Reflect: Fwd Errors: 0 > Num Free Slots : 6367 > > What confuses is that the minimum ring show in the another command cat > /proc/net/pf_ring/info is 4096 but here it shows as 6368? Which is the right > values to look and how does this effect the overall packet capture efficiency? > > Regards, > Frwa. > > > _______________________________________________ > Ntop-misc mailing list > [email protected] > http://listgateway.unipi.it/mailman/listinfo/ntop-misc
_______________________________________________ Ntop-misc mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop-misc
