Network related procs hang in unusual sfpbsy state

[Gary Stanley]

New current kernel from about 3 days ago, certain processes hang after 
about a day in this state;

    80   599     1   0 -16 -2 59020 53548 sfpbsy D<    ??  1141:08.65 
/mnt/zeus/web/bin/zeus.web
    80 34021     1   0 -16 -2 38488 36892 sfpbsy D<    ??    4:24.77 
/mnt/zeus/web/bin/zeus.web
    80 35648     1   0 -16 -2 61324 57620 sfpbsy D<    ??   19:41.76 
/mnt/zeus/web/bin/zeus.web
    80 43500     1   0 -16 -2 72372 54660 sfpbsy D<    ??    6:32.22 
/mnt/zeus/web/bin/zeus.web
    80 44710     1   0 -16 -2 38288 32960 sfpbsy D<    ??    0:23.32 
/mnt/zeus/web/bin/zeus.web
    80 44818     1   0 -16 -2 57120 52936 sfpbsy D<    ??    3:04.46 
/mnt/zeus/web/bin/zeus.web
    80 45175     1   0 -16 -2 34172 32472 sfpbsy D<    ??    0:57.65 
/mnt/zeus/web/bin/zeus.web

They become unkillable (not zombies according to top/ps, but behavior would 
suggest so)

Should I tried switching schedulers from ULE to 4BSD without success. Any 
ideas?