Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 17 Jun 1999 17:35:46 -0500
From:      Dan Nelson <dnelson@emsphone.com>
To:        matt <matt@AIC-GW.MLINK.NET>
Cc:        mike@hyperreal.org, agifford@infowest.com, freebsd-questions@FreeBSD.ORG
Subject:   Re: ftpd processes stuck in STOR - FreeBSD 3.2
Message-ID:  <19990617173546.A7287@dan.emsphone.com>
In-Reply-To: <Pine.BSF.4.10.9906171828250.13695-100000@aic-gw.mlink.net>; from "matt" on Thu Jun 17 18:29:10 GMT 1999
References:  <19990617213110.1989.qmail@hyperreal.org> <Pine.BSF.4.10.9906171828250.13695-100000@aic-gw.mlink.net>

next in thread | previous in thread | raw e-mail | index | archive | help
In the last episode (Jun 17), matt said:
> On Thu, 17 Jun 1999 mike@hyperreal.org wrote:
> : Aaron D. Gifford wrote:
> : > I'm wondering why my system seems to collect ftpd processes stuck
> : > in the STOR state that hang around for days, weeks,
> : 
> : Don't know, but it was something I saw for years on both FreeBSD
> : and BSDI, both running old wu-ftpd betas. The solution was to start
> : using BeroFTPD.
> 
> I've never had this problem, I run wu-ftpd 2.5.0.. I've also ran the
> beta (VR 16+ series) and never had that problem either... odd.

It's because stock ftpd doesn't have an idle timeout during data
transfer, and if a dialup user disconnects during an up/down load, the
ftpd will persist until the system is rebooted.

Setting the sysctl net.inet.tcp.always_keepalive=1 works around the
problem by disconnecting the socket after two hours.

	-Dan Nelson
	dnelson@emsphone.com


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?19990617173546.A7287>