Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 6 Jun 2000 10:30:42 +0200
From:      Stefan `Sec` Zehl <sec@42.org>
To:        Jeff Gray <jwg2@adsl-63-201-55-220.dsl.snfc21.pacbell.net>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: How not to reboot, was part of df - du leakage
Message-ID:  <20000606103042.A23681@matrix.42.org>
In-Reply-To: <Pine.BSF.4.10.10006050905360.53838-100000@adsl-63-201-55-220.dsl.snfc21.pacbell.net>; from jwg2@adsl-63-201-55-220.dsl.snfc21.pacbell.net on Mon, Jun 05, 2000 at 09:12:51AM -0700
References:  <Pine.BSF.4.10.10006050905360.53838-100000@adsl-63-201-55-220.dsl.snfc21.pacbell.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jun 05, 2000 at 09:12:51AM -0700, Jeff Gray wrote:
> As part of the earlier thread it was suggested that the way to resolve the
> problem of a growing /var fs [as shown by df] and a stable /var [as shown
> by du] was to reboot into single user mode and run fsck.

I never had a problem of this kind. So far it's always been a running
process holding a file open.

If you already checked the usual suspects (syslog/apache), you can use
lsof (port sysutils/lsof) which can list all open files. With a bit of
digging you should be able to find the Process.

You can then check if that process has a way of closing and reopening
the logfiles without terminating the process (usually kill -HUP)

> Running 3.3 R
> /var is growing at 50K an hour.  No log files being rotated, that I am
> aware of.

Did you add any file to /etc/newsyslog.conf ?

CU,
    Sec
-- 
One of the main causes of the fall of the Roman Empire
was that, lacking zero, they had no way to indicate
successful termination of their C Programs.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000606103042.A23681>