On Thu, Sep 29, 2005 at 08:37:23AM -0600, Herb Peyerl wrote:
> My machine way up above looks like it's starting again... If I  
> understand the math here...
> 
> Name        Size Requests Fail Releases Pgreq Pgrel Npage Hiwat Minpg  
> Maxpg Idle
> mbpl         256    44130    0    37383   464    41   423   423      
> 1   inf    1
> mclpl       2048    10215    0     6484  2444   574  1870  1870     4  
> 32768    4
> 
> saab-web-1# netstat -m
> 6153 mbufs in use:
>         4307 mbufs allocated to data
>         1845 mbufs allocated to packet headers
>         1 mbufs allocated to socket names and addresses
> 164 calls to protocol drain routines

44130 - 37383 = 6747
10215 - 6484 = 3731

There's probably something wrong, but this time there is a high number of
mbufs in use too, so the clusters are probably not orphaned.

does netstat -a show large send/receive queues ?

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 26 ans d'experience feront toujours la difference
--