From owner-freebsd-net Thu Feb 15 21:19:49 2001 Delivered-To: freebsd-net@freebsd.org Received: from VL-MS-MR002.sc1.videotron.ca (relais.videotron.ca [24.201.245.36]) by hub.freebsd.org (Postfix) with ESMTP id 1F72337B491 for ; Thu, 15 Feb 2001 21:19:46 -0800 (PST) Received: from jehovah ([24.202.203.190]) by VL-MS-MR002.sc1.videotron.ca (Netscape Messaging Server 4.15) with SMTP id G8U43X04.X5E; Fri, 16 Feb 2001 00:19:09 -0500 Message-ID: <002301c097d8$4b9f6000$becbca18@jehovah> From: "Bosko Milekic" To: "Dan Debertin" , "net@freebsd.org" References: Subject: Re: buffer problems with ep Date: Fri, 16 Feb 2001 00:21:19 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2919.6700 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Dan Debertin wrote: > I have a busy 4.1.1-RELEASE nfs server that stops responding at its ep > interface every hour or so. Pings to hosts on that network respond with > "sendto: no buffer space available". I've recompiled with MAXUSERS up at > 128 and NMBCLUSTERS at 32768, both of which seem plenty high. The problem > goes away, to return an hour later, after a quick 'ifconfig ep0 down; > ifconfig ep0 up'. > > Other people seem to have run into this problem as well, but > there is no sign of a fix in the archives anywhere. Can you try placing a "#define EP_LOCAL_STATS" at the _top_ of if_epvar.h and rebuilding+reinstalling the ep driver? See if you get anything printed to the console from the ep driver. If you get something that begins with "Status:" then look under it for "NOMB=" and take note of it. > Here's netstat -m when it's working: > > 194/320/131072 mbufs in use (current/peak/max): > 179 mbufs allocated to data > 1 mbufs allocated to packet headers > 14 mbufs allocated to fragment reassembly queue headers > 176/254/32768 mbuf clusters in use (current/peak/max) > 588 Kbytes allocated to network (68% in use) > 0 requests for memory denied > 0 requests for memory delayed > 0 calls to protocol drain routines > > and when it isn't: > 172/320/131072 mbufs in use (current/peak/max): > 134 mbufs allocated to data > 38 mbufs allocated to packet headers > 52/254/32768 mbuf clusters in use (current/peak/max) > 588 Kbytes allocated to network (25% in use) > 0 requests for memory denied > 0 requests for memory delayed > 0 calls to protocol drain routines > > Thanks for any help you an render, > > Dan Debertin > -- > ++ Unix is the worst operating system, except for all others. > > ++ Dan Debertin > ++ Senior Systems Administrator > ++ Bitstream Underground, LLC > ++ airboss@bitstream.net > ++ (612)321-9290 x108 > ++ GPG Fingerprint: 0BC5 F4D6 649F D0C8 D1A7 CAE4 BEF4 0A5C 300D 2387 Regards, Bosko. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message