From owner-freebsd-current Sat May 31 14:09:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA12889 for current-outgoing; Sat, 31 May 1997 14:09:55 -0700 (PDT) Received: from sliphost37.uni-trier.de (blank@sliphost37.uni-trier.de [136.199.240.37]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA12879 for ; Sat, 31 May 1997 14:09:49 -0700 (PDT) Received: (from blank@localhost) by sliphost37.uni-trier.de (8.8.5/8.8.5) id XAA02312 for current@freebsd.org; Sat, 31 May 1997 23:09:41 +0200 (CEST) From: Sascha Blank Message-Id: <199705312109.XAA02312@sliphost37.uni-trier.de> Subject: Re: ctm In-Reply-To: <19970531180718.HX14681@uriah.heep.sax.de> from J Wunsch at "May 31, 97 06:07:18 pm" To: joerg_wunsch@uriah.heep.sax.de Date: Sat, 31 May 1997 23:03:21 +0200 (CEST) Reply-To: blank@fox.uni-trier.de (Sascha Blank) X-System: FreeBSD 2.2-STABLE X-Mailer: ELM [version 2.4ME+ PL31H (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk J Wunsch has written recently: > The only reliable method is to CVSup from the CTM master site again, > before switching back to CTM. This would update your .ctm_status file > to the exact status of the master site. > > I once tried to convince Richard into providing a CVSup server (and be > it only for registered people, exactly for this purpose), but it seems > he feared the effort. (Well, he probably doesn't know it takes as few > as 15 minutes to get it up & running... ;-) If people are disciplined > enough to not abuse this service, but only use it for the final > catchup before switching back to CTM, this shouldn't cause any notable > load on Richard's machine. You can neglect the time it takes to unpack and apply the deltas. As soon as the filter gets hold of a delta, ctm_rmail takes over and does the rest. I'm already doing this on the fly to update my ports-cur and src-cur repository on our ftp server, and even though the server is usually heavily loaded and the 9GB disks aren't the fastest ones, it often takes less than a minute or two. > An alternative could it be if Sascha's machine would provide this > service, although the latency is a little higher then (since he has to > extract the CTM deltas first). The idea does not sound bad. We're already serving as a (unoffical) sup server (mostly for NetBSD), so setting up a cvsup server shouldn't be to complicated. In case I can find an entirely statically linked version of CVSup 15 that will work on BSDI 2.1 and 3.0, I would be willing to do a little testing (I once fought against the M3 compiler and do not want to do it again, especially on a machine that has no ports collection). But you shouldn't forget even with such a CVSup server I'm still relying entirely on Richard's server. The moment _his_ server fails for any reason _my_ server will be useless as well. The past incident made this very clear. -- Sascha Blank - mailto:blank@fox.uni-trier.de Student and System Administrator at the University of Trier, Germany Finger my account to receive my Public PGP key I don't speak for my employers, they don't pay me enough for that.