Subject: port-alpha/24366: sa going apparently going crazy
To: None <gnats-bugs@gnats.netbsd.org>
From: None <diro@nixsys.bz>
List: netbsd-bugs
Date: 02/08/2004 19:00:10
>Number:         24366
>Category:       port-alpha
>Synopsis:       sa going apparently going crazy
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-alpha-maintainer
>State:          open
>Class:          support
>Submitter-Id:   net
>Arrival-Date:   Sun Feb 08 19:01:00 UTC 2004
>Closed-Date:
>Last-Modified:
>Originator:     
>Release:        NetBSD 1.6.1
>Organization:
>Environment:
System: NetBSD nixsys 1.6.1 NetBSD 1.6.1 (NIXSYS) #9: Fri Jan 30 23:41:47 EST 2004 root@nixsys:/usr/src/sys/arch/alpha/compile/NIXSYS alpha
Architecture: alpha
Machine: alpha
>Description:

I'm not sure if this is a port specific problem. I've never seen anything
like this before and am unsure why it's happening.

I couldn't find anything online describing this problem. First:

Date: Sun,  8 Feb 2004 04:58:04 +0000 (UTC)
From: Cron Daemon <root@nixsys.bz>
To: "root@nixsys.bz" <root@nixsys.bz>
Subject: Cron <root@nixsys> /bin/sh /etc/daily 2>&1 | tee
    /var/log/daily.out | sendmail -t

    postdrop: warning: uid=0: File too large
    sendmail: fatal: root(0): Message file too big

Well, that's interesting, so, in /var/log/daily.out:

Purging accounting records:
sa: add key 0 to user accounting stats: No space left on device

There's _many_ of those lines.

diro@nixsys% du -k /var/log/daily.out
88776   /var/log/daily.out

In my /var/log/conslog:

Feb  8 03:22:55 nixsys /netbsd: uid 0 comm sa on /tmp: file system full
Feb  8 03:23:26 nixsys last message repeated 34586 times
Feb  8 03:25:27 nixsys last message repeated 151277 times
Feb  8 03:35:28 nixsys last message repeated 740074 times
Feb  8 03:42:02 nixsys last message repeated 491001 times
Feb  8 04:58:03 nixsys postfix/sendmail[18489]: fatal: root(0): Message file too big

So:

diro@nixsys% df
Filesystem  512-blocks     Used     Avail Capacity  Mounted on
/dev/sd0a       253806    55928    185186    23%    /
/dev/sd3a      7860020  1033994   6433024    13%    /var
/dev/sd0d      6596898  4846532   1420520    77%    /usr
/dev/sd1a      5894158   355208   5244242     6%    /home
kernfs               2        2         0   100%    /kern
procfs              16       16         0   100%    /proc
/dev/sd2a      6845056  4303496   2199306    66%    /usr/src/sys
/dev/sd2d       510322      286    484518     0%    /tmp
/dev/sd2e       510322       40    484764     0%    /var/tmp
/dev/sd1d      1967866    94340   1775132     5%    /www

Which is strange, because /tmp is 0%

Since I couldn't find anything online about this, I'm thinking it could possibly
be a new bug that no one's seen before.

>How-To-Repeat:
	Not sure how it happened.
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: