From owner-freebsd-stable Thu Oct 16 23:10:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA23063 for stable-outgoing; Thu, 16 Oct 1997 23:10:48 -0700 (PDT) (envelope-from owner-freebsd-stable) Received: from gatekeeper.tsc.tdk.com (root@gatekeeper.tsc.tdk.com [207.113.159.21]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA23057 for ; Thu, 16 Oct 1997 23:10:46 -0700 (PDT) (envelope-from gdonl@tsc.tdk.com) Received: from sunrise.gv.tsc.tdk.com (root@sunrise.gv.tsc.tdk.com [192.168.241.191]) by gatekeeper.tsc.tdk.com (8.8.4/8.8.4) with ESMTP id XAA13694 for ; Thu, 16 Oct 1997 23:10:45 -0700 (PDT) Received: from salsa.gv.tsc.tdk.com (salsa.gv.tsc.tdk.com [192.168.241.194]) by sunrise.gv.tsc.tdk.com (8.8.5/8.8.5) with ESMTP id XAA08710 for ; Thu, 16 Oct 1997 23:10:44 -0700 (PDT) Received: (from gdonl@localhost) by salsa.gv.tsc.tdk.com (8.8.5/8.8.5) id XAA28330 for freebsd-stable@freebsd.org; Thu, 16 Oct 1997 23:10:43 -0700 (PDT) Date: Thu, 16 Oct 1997 23:10:43 -0700 (PDT) From: Don Lewis Message-Id: <199710170610.XAA28330@salsa.gv.tsc.tdk.com> To: freebsd-stable@freebsd.org Subject: mfs bug in 2.1-stable? Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've found a repeatable way to hang a 2.1-stable (10/12/97) box that uses mfs for /tmp. Hardware is a Dell Pentium II 266, with an Adaptec 2940UW and an Intel Etherexpress Pro 10/100. If I tar a 65MB directory tree to a file in /tmp, rm the file, then run the tar command again, the machine wedges. It's not swap exhaustion, because I ran "pstat -s" in a loop and it shows plenty of swap left just before things hang. Device 1K-blocks Used Avail Capacity Type /dev/sd0s1b 102400 40036 62300 39% Interleaved One thing that's interesting is that at the end of the first tar command, about 32 MB of swap has been used, removing the file doesn't change the amount of swap consumed, then running the tar command again causes several MB more swap to be consumed. If I type ^T in the window that is running tar, it reports: load: 0.06 cmd: tar 366 [getblk] 0.00u 0.57s 0% 236k Something else interesting is that I had top running in another window and it continued to run after the tar command wedged, but top also froze when I tried to interrupt tar with a ^c. Once hung, the machine doesn't respond to ctrl-alt-del, so the only escape is to hit the reset button. This machine has no problem running "make world" with TMPDIR set to /tmp. --- Truck