[SAM] [www.ietf.org/rt #50961] The IETF rsync server is down again

"Glen via RT" <ietf-action@ietf.org> Wed, 26 September 2012 17:03 UTC

Return-Path: <wwwrun@ietfa.amsl.com>
X-Original-To: sam
Delivered-To: sam@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 30) id 0F97F21F85A1; Wed, 26 Sep 2012 10:03:05 -0700 (PDT)
From: Glen via RT <ietf-action@ietf.org>
In-Reply-To: <rt-3.6.5-9790-1348677577-1953.50961-7-0@www.ietf.org/rt>
References: <RT-Ticket-50961@www.ietf.org/rt> <34E66D8E-3849-4260-BDB7-1A5D7813F3E2@vpnc.org> <rt-3.6.5-19112-1348676806-713.50961-6-0@www.ietf.org/rt> <1A0FA67E-5A04-4A87-81F2-1EA0DD29B1E3@vpnc.org> <rt-3.6.5-9790-1348677577-1953.50961-7-0@www.ietf.org/rt>
Message-ID: <rt-3.6.5-30703-1348678984-1201.50961-7-0@www.ietf.org/rt>
Precedence: bulk
X-RT-Loop-Prevention: www.ietf.org/rt
RT-Ticket: www.ietf.org/rt #50961
Managed-by: RT 3.6.5 (http://www.bestpractical.com/rt/)
RT-Originator: glen@amsl.com
To: "OtherRecipients of www.ietf.org/rt Ticket #50961":;
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
X-RT-Original-Encoding: utf-8
Date: Wed, 26 Sep 2012 10:03:04 -0700
X-Mailman-Approved-At: Wed, 26 Sep 2012 13:38:11 -0700
Subject: [SAM] [www.ietf.org/rt #50961] The IETF rsync server is down again
X-BeenThere: sam@irtf.org
X-Mailman-Version: 2.1.12
Reply-To: ietf-action@ietf.org
List-Id: "For use by members of the Scalable Adaptive Multicast \(SAM\) RG" <sam.irtf.org>
List-Unsubscribe: <http://www.irtf.org/mailman/options/sam>, <mailto:sam-request@irtf.org?subject=unsubscribe>
List-Archive: <http://www.irtf.org/mail-archive/web/sam>
List-Post: <mailto:sam@irtf.org>
List-Help: <mailto:sam-request@irtf.org?subject=help>
List-Subscribe: <http://www.irtf.org/mailman/listinfo/sam>, <mailto:sam-request@irtf.org?subject=subscribe>
X-List-Received-Date: Wed, 26 Sep 2012 17:03:05 -0000

On Wed Sep 26 09:39:37 2012, paul.hoffman@vpnc.org wrote:
> > The IETF rsync server was not, in fact, down.  All available
> connections were being used, and that
> > situation has been rectified.
> 
> Thanks. Out of curiosity, why did you set "max connections" in
> rsyncd.conf?
> 
> --Paul Hoffman

Hi Paul -

Glen here.  This has been escalated to me.

I always try to set sane limits on all processes, to prevent one runaway
from consuming the entire server.  It seems like a responsible way to
proceed; and, in this case, my guess is that the offending external site
could have used up a much larger number of processes, potentially
impacting the rest of the server, had those limits not been set.

I hope this information is helpful, let me know if you need more.

Glen
Glen Barney
IT Director
AMS (IETF Secretariat)