From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Gale Subject: Re: Overruns ? Date: Thu, 19 Feb 2004 11:49:42 -0700 Sender: netfilter-admin@lists.netfilter.org Message-ID: <20040219114942.6c6ee6c9@mgalepc.utilitran.com> References: <20040219083222.7e4dc5a3@mgalepc.utilitran.com> <1077216213.1984.48.camel@ca.gts.nexusmgmt.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1077216213.1984.48.camel@ca.gts.nexusmgmt.com> Errors-To: netfilter-admin@lists.netfilter.org List-Help: List-Post: List-Subscribe: , List-Id: List-Unsubscribe: , List-Archive: Content-Type: text/plain; charset="us-ascii" To: "John A. Sullivan III" Cc: netfilter@lists.netfilter.org Hello, This network device is on a small (5 hub) network with all the other cards being on-board GB cards. This network device is a cheap Dlink 10/100 card. I am going to try changing the card to a GB card as well ... since there also seems to be a fair amount of collision. Michael. On Thu, 19 Feb 2004 13:43:34 -0500 "John A. Sullivan III" wrote: > On Thu, 2004-02-19 at 10:32, Michael Gale wrote: > > Hello, > > > > Not sure if this is the place to ask but when I do a ifconfig I see some > > overruns. I have tried google and man ifconfig but can not find what a > > "overrun" is. > > > > eth0 Link encap:Ethernet HWaddr 00:E0:81:25:A6:92 > > inet addr:10.10.10.173 Bcast:10.10.10.255 Mask:255.255.255.0 > > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > > RX packets:9639161 errors:0 dropped:0 overruns:0 frame:0 > > TX packets:9598780 errors:0 dropped:0 overruns:21 carrier:0 > > collisions:0 txqueuelen:100 > > RX bytes:2726609687 (2600.2 Mb) TX bytes:1986194437 (1894.1 Mb) > > Interrupt:10 Base address:0xb000 > Perhaps someone else can answer more authoritatively than I but I have > always assumed that it means the physical interface picked up a packet > addressed to it but had no buffer into which it could put the packet. > In other words, the system ran out of memory dedicated to packet > buffering. > > One can either increase the number of buffers allocated to buffering > packets or identify the source of the bottleneck. On a standalone > system, it might be the processor. In the case of a packet passing > device, increase the bandwidth on the egress. > > I do not know where one configures the number of cache buffers in Linux > but that would be a handy trick to know. I'm sure someone does know it > off the top of their head. Next person . . . > -- > Open Source Development Corporation > Financially Sustainable open source development > http://www.opensourcedevelopmentcorp.com > > > > > -- Michael Gale Network Administrator Utilitran Corporation