Jeff Gentry <jgentry@jimmy.harvard.edu> writes: > > I can't find any instructions for anonymous cvs access to the bio > > conductor cvs repository. The faq say to use rsync which does not seem > > to be set up yet. > > Its not, and as far as anon cvs I don't think that such a beast is set up > either right now. rsync is now set up on rsync.r-project.org (a.k.a. franz.stat.wisc.edu) $ cd /var/tmp bates@bates3:/var/tmp$ rsync rsync.r-project.org:: r-release R-1.5.1 sources (current released version - approx 25 MB) r-patched R sources (patched released version - approx 25 MB) r-devel R sources (devel version, less stable - approx 25 MB) r-manuals Development sources for manuals for R r-recommended Sources for recommended R packages xlispstat xlispstat sources (development version) CVS tree (approx 12 MB) CRAN Complete CRAN ftp area Bioc-release Bioconductor sources (current release version) Bioc-devel Bioconductor sources (devel version, less stable) omega-cvs Omega Project for Statistical Computing CVS tree r-project-web Web pages for www.r-project.org and mirrors Omegahat Complete ftp archive for Omegahat I created these views by cvs -d /home/madman/CVS-ARCHIVE co -P -r RELEASE_1_0 madman and cvs -d /home/madman/CVS-ARCHIVE co -P madman Please check it and tell me if that is the view of the CVS tree that was intended. If you want to add rsync.bioconductor.org as a CNAME, the machine that runs the rsync server is franz.stat.wisc.edu (128.105.174.95). AFAIK rsync doesn't have an idea of a virtual server so connection to the rsync server under any CNAME will give access to all the modules.

>>>>> "robert" == Robert Gentleman <rgentlem@jimmy.harvard.edu> writes: robert> Either the web cvs down load or rsync are robert> possible. anonymous cvs is not generally viewed as a good robert> idea. R has now dropped it (I beleive). I would disagree partially. anon cvs is an excellent idea, far easier than rsync for maintaining local versions of code; it's just a potential security hole (hence a headache). Never the less, Bioconductor will NOT be running one on franz (we could set one up at UW, using the existing facilities there, but that is another discussion). It does look like R's (and hence, franz's) anoncvs server is dead, though. It wasn't, a few (many? time goes by fast these days) weeks ago. best, -tony -- A.J. Rossini Rsrch. Asst. Prof. of Biostatistics U. of Washington Biostatistics rossini@u.washington.edu FHCRC/SCHARP/HIV Vaccine Trials Net rossini@scharp.org -------------- http://software.biostat.washington.edu/ ---------------- FHCRC: M: 206-667-7025 (fax=4812)|Voicemail is pretty sketchy/use Email UW: Th: 206-543-1044 (fax=3286)|Change last 4 digits of phone to FAX (my tuesday/wednesday/friday locations are completely unpredictable.)

Robert Gentleman <rgentlem@jimmy.harvard.edu> writes: > Either the web cvs down load or rsync are possible. anonymous cvs is > not generally viewed as a good idea. R has now dropped it (I > believe). As Tony remarked anoncvs it has been shut down although not intentionally. (Franz's /var filesystem filled up a few days ago because mailman log files for one list were gargantuan.) I am going to take this opportunity to kill anoncvs. We now have rsync as I said in my other message.