The only thing that seems to cure this problem is doing several: > rmmod ixgbe.ko > modprobe dca > insmod ixgbe.ko IntMode=3 > rmmod pf_ring.ko > insmod pf_ring.ko transparent_mode=2 quick_mode=1
Eventually everything plays nice together. As this is intermittent, I'm thinking that maybe there's some uninitialized variable in either the ixgbe.ko (tnapi) module or the pf_ring.ko module. Regards, Karl On 6/23/11 8:59 AM, "Henderson, Karl" <[email protected]> wrote: >Yes, I still got a errno 22 on the bind function. > >On 6/23/11 8:15 AM, "Luca Deri" <[email protected]> wrote: > >>Karl >>I don't see a failure. Has this test been run while you open failed? >> >>Thanks Luca >> >>On Jun 22, 2011, at 7:33 PM, Henderson, Karl wrote: >> >>> Hi Luca, >>> >>> Here's the output: >>> >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_create() >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_insert() >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] Added >>> /proc/net/pf_ring/7796-none.0 >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_create(): created >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] --> >>>SO_RING_BUCKET_LEN=200 >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_bind() called >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] called >>>ring_release(none) >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] Removing >>> /proc/net/pf_ring/7796-none.0 >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] Removed >>> /proc/net/pf_ring/7796-none.0 >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_remove() >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] Found socket to remove >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] leaving ring_remove() >>> Jun 22 06:34:35 hydradevva-nr1 kernel: [PF_RING] ring_release: done >>> >>> Cheers, >>> >>> Karl >>> >>> On 6/22/11 12:36 PM, "Luca Deri" <[email protected]> wrote: >>> >>>> insmod ./pf_ring.ko enable_debug=1 >>> >>> _______________________________________________ >>> Ntop-dev mailing list >>> [email protected] >>> http://listgateway.unipi.it/mailman/listinfo/ntop-dev >> >>--- >>If you can not measure it, you can not improve it - Lord Kelvin >> >>_______________________________________________ >>Ntop-dev mailing list >>[email protected] >>http://listgateway.unipi.it/mailman/listinfo/ntop-dev > _______________________________________________ Ntop-dev mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop-dev
