Date: Tue, 7 Sep 1999 23:01:38 -0400 (EDT) From: "Crist J. Clark" <cjc@cc942873-a.ewndsr1.nj.home.com> To: freebsd-questions@FreeBSD.ORG (FreeBSD Questions) Subject: Generating STABLE "Patches" Message-ID: <199909080301.XAA21882@cc942873-a.ewndsr1.nj.home.com>
next in thread | raw e-mail | index | archive | help
I have a system on which I will be running FreeBSD that is not, and never will be, connected to the Internet. I will install off of the Walnut Creek 3.2-RELEASE CDs, but I would like to incorporate bug fixes, security patches, and the other updates that get added to STABLE on a regular basis. Now, The Powers That Be are very wary about using things downloaded from the Internet. They balk at the idea of basically replacing the _whole_ OS in one swoop if we CVSup STABLE as a whole and build from a new source tree. What I am trying to do is figure out how best to keep this machine fairly up-to-date with STABLE when there is a sneakernet connection somewhere between it and cvsup*.freebsd.org. Also keep in mind that a wholesale replacement of the sourcetree each time is not an option. I basically have three ideas at this point, 1) Generate STABLE "patch sets." A machine that is on the Internet will contain two source trees. One will be the latest CVSup; the other will be a mirror of the "isolated" machine's sources. I can diff the two trees to generate a patch set that will bring the isolated machine up to STABLE. After the patch is in place, the new CVSup becomes the mirror of the isolated machine for the next CVSup. This solution has advantages, the isolated system will be exposed to the minimum number of changes and the changes are clearly recorded in the patch set. In addition, this is the solution that requires sneakernetting the least volume of media (my guess is the patches can practically fit on floppies). 2) Run both cvsup and cvsupd on the isolated system. Mirror the CVS distribution of FreeBSD. Transfer the entire CVS tree to the isolated system where it will run CVSup as a client to itself. This has some drawbacks. I need to transport the whole CVS source tree each time and figure out how to run cvsupd. 3) Figure out what the heck CTM is. I don't know if I have ever heard anyone on mailling lists talk about it. However, it sounds like it would be basically doing most of the work in idea (1) for me. Does anyone out there do anything like this? Anyone have some ideas about the best way to get something like this done? Thanks for any help. -- Crist J. Clark cjclark@home.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?199909080301.XAA21882>