From owner-freebsd-questions Thu Apr 3 18:35:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA08394 for questions-outgoing; Thu, 3 Apr 1997 18:35:41 -0800 (PST) Received: from alliance.workgrp.com (Marty.Bluebird.COM [204.29.238.72]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id SAA08389 for ; Thu, 3 Apr 1997 18:35:38 -0800 (PST) Received: from happy.workgrp.com (happy [207.207.207.120]) by alliance.workgrp.com (8.6.11/8.6.9) with SMTP id SAA07617 for ; Thu, 3 Apr 1997 18:36:00 -0800 Message-Id: <2.2.32.19970404023817.00690ba4@207.207.207.90> X-Sender: freebsd@207.207.207.90 X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 03 Apr 1997 18:38:17 -0800 To: freebsd-questions@freebsd.org From: FreeBSD Administrator Subject: Where is the 'procfs' filesystem? Sender: owner-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Greetings fellow BSDers, I've inherited a small mail server that has recently started spewing out errors about full file systems and temporary write errors that has me baffled. The machine is running this version of FreeBSD, on a 486 system with a small IDE disk: bonkers: /kernel: FreeBSD 2.0.5-RELEASE #0: Wed Jul 5 20:22:26 PDT 1995 All seems well, ie, mail flows in and out, but the console keeps displaying this message: Mar 31 19:22:16 bonkers /kernel: uid 0 on /: file system full Trying to process the mail queue gives: bonkers: {78} sendmail -q mail.local: temporary file write error The only thing I can see is that the "procfs" file system is 100% full. Where the heck is this thing defined? Could this be the source of the problem? Any suggestions on how to fix this? bonkers: {79} df Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/wd0a 24815 16201 6628 71% / /dev/wd0s1e 206323 123258 66559 65% /usr procfs 4 4 0 100% /proc I don't see anything in the disklabel that would indicate that "procfs" is real filesystem and the docs talk about it's function as related to tracking processes but not how or where it's defined, and how to "enlarge" it. extracted from /var/log/messages: -------------------------------------------------------------- Mar 31 19:22:16 bonkers /kernel: uid 0 on /: file system full Mar 31 19:28:51 bonkers /kernel: uid 0 on /: file system full Mar 31 19:28:51 bonkers syslogd: /var/log/maillog: No space left on device Mar 31 19:28:51 bonkers /kernel: uid 0 on /: file system full I'm in the process of building a new, larger machine but given everything else I have to do, that won't happen anytime soon, so I'm hoping this is a 'fixable' problem. All space on this tiny disk is allocated, so another idea I has would be to backup both filesystems with cpio, carve up a new disk and restore the existing system into a larger home. Is there any equivalent to SCO's emergency boot & root floppies, or do I have to try to cook up my own? Any and all comments, tips, suggestions would be appreciated. Thanks, Marty Gordon marty@workgrp.com