From owner-freebsd-stable Fri Aug 10 15:20: 4 2001 Delivered-To: freebsd-stable@freebsd.org Received: from ns.yogotech.com (ns.yogotech.com [206.127.123.66]) by hub.freebsd.org (Postfix) with ESMTP id 40DE937B401 for ; Fri, 10 Aug 2001 15:19:58 -0700 (PDT) (envelope-from nate@yogotech.com) Received: from nomad.yogotech.com (nomad.yogotech.com [206.127.123.131]) by ns.yogotech.com (8.9.3/8.9.3) with ESMTP id QAA02586; Fri, 10 Aug 2001 16:19:57 -0600 (MDT) (envelope-from nate@nomad.yogotech.com) Received: (from nate@localhost) by nomad.yogotech.com (8.8.8/8.8.8) id QAA08044; Fri, 10 Aug 2001 16:19:56 -0600 (MDT) (envelope-from nate) From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15220.24076.57998.200803@nomad.yogotech.com> Date: Fri, 10 Aug 2001 16:19:56 -0600 To: mikea Cc: stable@FreeBSD.ORG Subject: Re: Strange I/O behavior In-Reply-To: <20010809221453.A1165@mikea.ath.cx> References: <15218.65134.426946.295368@nomad.yogotech.com> <20010809221453.A1165@mikea.ath.cx> X-Mailer: VM 6.95 under 21.1 (patch 12) "Channel Islands" XEmacs Lucid Reply-To: nate@yogotech.com (Nate Williams) Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG > > I've got a dual-CPU box that's acting very strangely. At various times > > throughout the day, the box starts to crawl. > > > > In analyzing it, the only thing that jumps out at me is that the first > > fixed disk is running a huge number of I/O transactions, to the point > > that it's almost saturating the disk. > > > > However, the amount transferred is almost nil, but since this disk > > contains / and /usr (which means all the files are on it), interactive > > performance goes to the dump during these sessions. > > > > The box is an all SCSI system, with the first disk being a 17GB > > Seagate. The dmesg is below. > > > > Here's the output of iostat 2, which shows alot of transactions going > > on. However, there are no active users doing anything as far as I can > > see, and top shows the system as being mostly idle. > > > > Any ideas how I can determine which application(s) are causing the huge > > I/O loads? > > Try fstat and lsof to see what programs have what files open, > for a start. Nothing obvious. > Is there _any_ chance you've been cracked and the > cracker had a tool running that spawned a few zillion telnet > sessions? Not a chance. The box isn't/has never been on the Internet, and was just recently updated to 4.3-RELEASE. (I know, 4.4-RELEASE is almost out, but it's a production box>) > The symptoms were one CPU 97% busy with a program I didn't > recognize, and some hundreds of telnet sessions outbound. Interestingly enough, it's an I/O only problem. The CPU usage is barely creeping about 0.1%. Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message