> > Also, run a netstat -m and look at peak usage. If that number is > > reaching max, increase kern.maxclusters with sysctl. > > $ netstat -m > 1543 mbufs in use: > 1539 mbufs allocated to data > 1 mbuf allocated to packet headers > 3 mbufs allocated to socket names and addresses > 1538/6152/6144 mbuf clusters in use (current/peak/max) > 12732 Kbytes allocated to network (27% in use) > 0 requests for memory denied > 0 requests for memory delayed > 4896 calls to protocol drain routines
I thought so. The patch, I believe, is in 3.8-STABLE which allocates mbufs on the fly as they are needed. Try a -current snapshot and see if that helps. Keep the kern.maxclusters at the default. When you hit the limit, it will print a warnings on the console, but will not panic. I guess you can safely double the default value and you should be fine. Don't set it to a ridiculously high number. If the double is not enough for you, then the real problem with network data buffering is somewhere else.