"Peter N. M. Hansteen" <[EMAIL PROTECTED]> wrote: > > "Florian Fuessl" <[EMAIL PROTECTED]> writes: > > > Increasing "kern.maxclusters" has solved the problem here. :) > > It would be interesting to hear a bit more about which values you have > tried and to what effect.
Doubling the default (6144) should be far enough; I did setup the value 128000 which is obviously exaggerated but seems to works fine here: ~ # netstat -m 2766 mbufs in use: 2246 mbufs allocated to data 515 mbufs allocated to packet headers 5 mbufs allocated to socket names and addresses 1675/6152/128000 mbuf clusters in use (current/peak/max) 14460 Kbytes allocated to network (27% in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines