From root@www3.shellnet.co.uk Fri Apr 30 08:35:13 1999
Return-Path: <root@www3.shellnet.co.uk>
Received: from www3.shellnet.co.uk (www3.shellnet.co.uk [194.129.209.10])
	by hub.freebsd.org (Postfix) with ESMTP id B670C14F69
	for <FreeBSD-gnats-submit@freebsd.org>; Fri, 30 Apr 1999 08:35:03 -0700 (PDT)
	(envelope-from root@www3.shellnet.co.uk)
Received: (from root@localhost)
	by www3.shellnet.co.uk (8.9.1/8.9.1) id QAA04547;
	Fri, 30 Apr 1999 16:35:02 +0100 (BST)
	(envelope-from root)
Message-Id: <199904301535.QAA04547@www3.shellnet.co.uk>
Date: Fri, 30 Apr 1999 16:35:02 +0100 (BST)
From: steven@shellnet.co.uk
Sender: root@www3.shellnet.co.uk
Reply-To: steven@shellnet.co.il
To: FreeBSD-gnats-submit@freebsd.org
Subject: ps: bad namelist & etc.
X-Send-Pr-Version: 3.2

>Number:         11405
>Category:       kern
>Synopsis:       pwd_mkdb with no tmp space leads to kernel problems
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Apr 30 08:40:00 PDT 1999
>Closed-Date:    Fri Jul 20 08:53:24 PDT 2001
>Last-Modified:  Fri Jul 20 08:57:26 PDT 2001
>Originator:     Steven Fletcher
>Release:        FreeBSD 3.0-RELEASE i386
>Organization:
Shellnet Limited
>Environment:

	Apache webserver with 200 user accounts and not much spare diskspace. :)

>Description:

	After accidentally filling up the / partition in the morning (to 107%), I ran adduser later
in the day to add a new user to the server. When pwd_mkdb came to run as part of the adduser script, the following happened:

Apr 30 15:14:19 www3 /kernel: pid 14922 (pwd_mkdb), uid 0 on /: file system full

And adduser failed. Then proceeding to run "uptime/w" and "ps" (Did not try top) gave errors. ps returned:

ps: bad namelist

w/uptime returned (something like, don't have a log):

/dev//umount: /cdrom: not currently mounted
umount: No such file or directory
/dev//: /proc: not currently mounted
/var:: No such file or directory
/dev// not currently mounted
/usr: not cur: No such file or directory
/dev//rently mounted
/home: not currently : No such file or directory
/dev//mounted
: No such file or directory
4:20PM  up  41 mins, 1 user, load averages: 0.02, 0.05, 0.06


>How-To-Repeat:

	I haven't tried this, I don't have a spare box to risk it at the moment. I would suggest don't give pwd_mkdb
enough space in /tmp and try to run it.

>Fix:

Fix for the ps problem:
	
	cd /dev
	./MAKEDEV all

Fix for the uptime/w problem:

	cd /var/run
	mv utmp /tmp
	touch utmp

I've since linked /tmp to a directory on another harddisk with many gigs free, hopefully this shouldn't happen for me again.


>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed 
State-Changed-By: mike 
State-Changed-When: Fri Jul 20 08:53:24 PDT 2001 
State-Changed-Why:  

The system will always return errors like these when it runs out of 
disk space.  A better partitioning scheme would likely solve these 
problems, if you anticipate running out of disk space again. 

http://www.FreeBSD.org/cgi/query-pr.cgi?pr=11405 
>Unformatted:
