In article <199704250521.IAA03420@pyy.jmp.fi>,
	Jukka Marin <jmarin@pyy.jmp.fi> writes:
> Was there an mbuf leak in 1.2?

Not on my machines, both running NetBSD-i386:

tron@colwyn:~>uptime
 9:34AM  up 34 days, 19:30, 1 user, load averages: 0.10, 0.12, 0.08
tron@colwyn:~>netstat -m
43 mbufs in use:
        22 mbufs allocated to data
        15 mbufs allocated to packet headers
        6 mbufs allocated to socket names and addresses
2/120 mapped pages in use
245 Kbytes allocated to network (3% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

tron@jehova:~>uptime
 9:35AM  up 16 days, 10:39, 1 user, load averages: 0.21, 0.19, 0.12
tron@jehova:~>netstat -m
16 mbufs in use:
        3 mbufs allocated to data
        11 mbufs allocated to packet headers
        2 mbufs allocated to socket names and addresses
3/282 mapped pages in use
566 Kbytes allocated to network (1% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

On what kind of system do you experience that problem? What networking
hardware? It might be the interface driver which loses mbufs.

> Is this some kind of special kernel memory which I will soon run out of?

Yes, but the system will be able to allocat thousands of mbufs before you
get into trouble.

-- 
Matthias Scheler                                http://home.pages.de/~tron/