[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openbsd-tech
Subject:    memory buffers de0 (was re: ping no buffer space)
From:       happy emory <emory () hellyeah ! com>
Date:       1998-07-22 2:40:44
[Download RAW message or body]


okay thanks to a kind soul that pointed me to the magic of netstat -m,
i've been able to get some help on this issue.

this machine typically has about 12 people logged into it, does webservice
and a bit of mail.  i was reading the manpage and it says that the network
manages a private pool of memory buffers per network device.  

i'm really concerned:

bitchslap# netstat -m
63 mbufs in use:
        34 mbufs allocated to data
        28 mbufs allocated to packet headers
        1 mbuf allocated to socket names and addresses
32/60 mapped pages in use
135 Kbytes allocated to network (58% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines
bitchslap# netstat -m
151 mbufs in use:
        57 mbufs allocated to data
        93 mbufs allocated to packet headers
        1 mbuf allocated to socket names and addresses
32/60 mapped pages in use
157 Kbytes allocated to network (64% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

it fluctuates in the 50+ to 60+ range when i'm the only person on here and
traffic is low since it just came back up :P

can i tweak this in the OS at all to allocate more resources TO the device
de0?  how can i jack this up a bit?  my p2 system with an fxp0 (intel)
card is around 30% and has more available to it.

is this *soley* in the hardware?

tips tricks and whatnot are very much appreciated!  

again: dec alpha udb/multia 233 openbsd 2.3



[youdeserveabreakfromcounterfeitsfraudsandfakes]
[getittogethergetittogetherget]			s u b l i m i n a l
					http://gravity.hellyeah.com
pgp key via finger, baybee.		  mailto:emory@hellyeah.com

[prev in list] [next in list] [prev in thread] [next in thread]