From owner-freebsd-questions  Sun Oct 27 11:12:47 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id LAA07834
          for questions-outgoing; Sun, 27 Oct 1996 11:12:47 -0800 (PST)
Received: from superior.truenorth.org (ppp013-sm2.sirius.com [205.134.231.13])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id LAA07828
          for <questions@freebsd.org>; Sun, 27 Oct 1996 11:12:43 -0800 (PST)
Received: (from jgrosch@localhost) by superior.truenorth.org (8.7.5/8.7.3) id LAA00431; Sun, 27 Oct 1996 11:11:49 -0800 (PST)
From: Josef Grosch <jgrosch@superior.truenorth.org>
Message-Id: <199610271911.LAA00431@superior.truenorth.org>
Subject: Re: tin and swap space
To: darrylb@blinx.lizard.org (Darryl Bowler)
Date: Sun, 27 Oct 1996 11:11:47 -0800 (PST)
Cc: questions@freebsd.org
Reply-To: jgrosch@sirius.com
In-Reply-To: <199610271443.OAA00608@blinx.lizard.org> from Darryl Bowler at "Oct 27, 96 02:43:16 pm"
X-Mailer: ELM [version 2.4ME+ PL13 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>Whenever I run tin it seems die. When viewing /var/log/messages I get this error
>
>Oct 27 13:35:52 blinx /kernel: pid 166 (rtin), uid 1002, was killed: out of swap
> space
>Oct 27 13:35:52 blinx /kernel: pid 166 (rtin), uid 1002, was killed: out of swap
> space                                                                         
>
>Somehow swap space hasnt been released, maybe due to the fact that someone has not exited  tin correctly.  
>
> I temporary solved the problem by using the swapon command, however it seemed
>to want a file /dev/sd0s1b, which did not exist  until I made it with MAKEDEV.
>(swapon -a)
>
>Tin now works this way, but is this a valid solution to this problem,
>how I view the swap space been used and by what process (vmstat??? maybe?)?
>
>How do I release swap space?
>
>
>Regards Darryl.
>

Tin is a memory pig. If I remember right, for each newsgroup tin builds an
in memory index of the articles that are available. Since the number of
newsgroups is now hovering around 13,000 it doesn't take tin very long to
eat all the swap space. This assumes you are subscribing to a very large
number of newsgroups, have a small swap space, and/or have a "small" amount
of memory.  

I bumped into this problem several months ago. I have 16 meg of ram and 100
meg of swap space (I plan on upgrading to 32 meg of ram very soon) and
still I was running out of swap. I trimmed the number of newsgroups I was
subscribing to. At the time I was subscribing to ALL the newsgroups so
paring it down to a couple of hundred solved the problem.

Use the swapinfo command [man pstat(8)] to see how much swap you have and
how much is being used. The death of a process should release the swap
space being used by that process.  

Josef

-- 
Josef Grosch       | Laugh while you can, monkey boy ! |    FreeBSD 2.1.5
jgrosch@sirius.com |          - John Warfin -          | UNIX for the masses