From owner-freebsd-current Sat Aug 24 12:55:15 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA24961 for current-outgoing; Sat, 24 Aug 1996 12:55:15 -0700 (PDT) Received: from skynet.ctr.columbia.edu (skynet.ctr.columbia.edu [128.59.64.70]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA24956 for ; Sat, 24 Aug 1996 12:55:10 -0700 (PDT) Received: (from wpaul@localhost) by skynet.ctr.columbia.edu (8.6.12/8.6.9) id PAA04334; Sat, 24 Aug 1996 15:54:57 -0400 From: Bill Paul Message-Id: <199608241954.PAA04334@skynet.ctr.columbia.edu> Subject: Re: memory/disk error with 801-SNAP To: rsnow@lgc.com (Rob Snow) Date: Sat, 24 Aug 1996 15:54:55 -0400 (EDT) Cc: current@freebsd.org In-Reply-To: from "Rob Snow" at Aug 24, 96 11:30:29 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Of all the gin joints in all the towns in all the world, Rob Snow had to walk into mine and say: > > I started sup'ping current last night. When I looked this morning I found > this. (I'm not sure this is even related since it happened in the middle > of the sup) > > First: > Aug 24 02:51:00 rex /kernel: /usr: /optimization changed from SPACE to > TIME > > > df shows: > /dev/sd0s1f 217525 157695 42428 79% /usr This is not an error. This is a notification from the filesystem code telling you that it's changing its optimization strategy. One of the tunable filesystem parameters (man tunefs) controls whether the filesystem tries to optimize for speed or optimal disk usage. As the amount of disk fragmentation increases, it may decide to change stratgies. The tunefs man page explains this. In other words, this is normal. It just means that as your disk became fuller as the build progressed, the kernel decided to change strategies. It's not a bug. Don't file a pr for it. :) > this morning I see my make world (at'ed at 7am) died and top -n shows: > kvm_open: proc size mismatch (20196 total, 620 chunks) > top: Out of memory. > w gives me: > 11:24AM up 12:16, 5 users, load averages: 0.02, 0.02, 0.00 > USER TTY FROM LOGIN@ IDLE WHAT > w: proc size mismatch (20196 total, 620 chunks): Undefined error: 0 > > I'm going to see if I can reproduce it today. You are doing a make world which has overwritten some of your system binaries and libraries but you're still running an old kernel (or you're running a new kernel with old binaries -- same effect). Every so often someone (usually John Dyson :) makes changes to the kernel that affects the size of certain critical structures which can cause tools like ps, w or top to fail unless you recompile them to match the kernel. To avoid this problem, you have to keep your kernel and libkvm-related tools in sync. This is also not a bug. Don't file a pr for this either. :) > Now a question about sup: > > I kiced off a sup -v ports-supfile which the docs say is the "super sup" > or some such. It's supposed to get all the ports except distfiles. It > started getting all the ports into /usr/ports/ports, in other words it > starged reproducing them. Then it started transfering the distfiles, I > killed it... I then restarted it using each port collection and it > appears to have worked fine. Did I misunderstand or is something broke? Dunno about this one: I never sup the ports, only the main tree. -Bill -- ============================================================================= -Bill Paul (212) 854-6020 | System Manager, Master of Unix-Fu Work: wpaul@ctr.columbia.edu | Center for Telecommunications Research Home: wpaul@skynet.ctr.columbia.edu | Columbia University, New York City ============================================================================= "If you're ever in trouble, go to the CTR. Ask for Bill. He will help you." =============================================================================