Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!um-math!hyc From: hyc@math.lsa.umich.edu (Howard Chu) Newsgroups: comp.sys.atari.st Subject: Re: pc7 anonymous ftp? Keywords: anonymous, ftp, clio, pc7 Message-ID: <532@stag.math.lsa.umich.edu> Date: 2 Jan 89 03:14:21 GMT References: <909@ccncsu.ColoState.EDU> <2227@imagine.PAWL.RPI.EDU> <5560@boulder.Colorado.EDU> Sender: usenet@math.lsa.umich.edu Reply-To: hyc@math.lsa.umich.edu (Howard Chu) Organization: University of Michigan Math Dept., Ann Arbor Lines: 65 UUCP-Path: {mailrus,umix}!um-math!hyc In article <5560@boulder.Colorado.EDU> walkerb@tramp.Colorado.EDU (Brian Walker) writes: >OK, the first item of business is that the pc7 and clio archives are separate >and exist on separate machine half way across the nation from each other. >There no longer is an archive at clio. Sorry to disagree... You *are* correct that there no longer is an archive site at clio. The two archives were on separate machines, but they are separated by about two miles, not half a continent. I am system/network administrator of the network clio resides on, and a systems programmer/mail guru for the machine the PC7 archives are stored on. > >The second item of business is that him1 (pc7 archive site) probably still >accepts logins. The problem appears to be worse because the machine I reach >at ftp 35.1.1.43 is um.cc.umich.edu. A quick look at a local hosts file >indicates that um.cc.umich.edu is another hosts in the same domain which should >have the address 35.1.1.81. Furthermore, ftp 35.1.1.81 gets me to yet another >host (umix.cc.umich.edu). So, either the umich.edu domain has been scrambled >or part of the connection between here and there has been scrambled. There is a mixup, yes, but it shouldn't affect FTP users. If you want all the gory details, keep reading. Otherwise, just hit 'n', safe in the knowledge that the PC7 archive is accessible at him1.cc.umich.edu... Here's the scoop: him1.cc.umich.edu is a network interface for an IBM 3090, running MTS (the Michigan Terminal System). The particular host him1 talks to is um.cc.umich.edu, also known as umichum.bitnet, or by local users, just "UM" (as opposed to "UB," another MTS host in another partition of the 3090). MTS really doesn't support TCP/IP very well, so it relies on front ends like the HIM (Host Interface Machine, a PDP-11/70 with custom software & hardware) to talk on the networks for it. The HIM is a relatively new machine; the TCP/FTP service on MTS has only been usable for a few months. The MTS hosts are used by literally thousands of people, for quite a large variety of applications. Probably the most significant being electronic mail. Of course, since MTS doesn't deal with TCP/IP very well, it can't deal with Internet SMTP mail either. So, a long time ago, (two years, at least) our resident Unix expert kludged up a way for a Vax running BSD 4.3 Unix to "mooch" for the MTS host. That is, the vax answers any incoming connections headed to um and ub.cc.umich.edu. While the primary reason was to intercept SMTPs, it catches everything, which means incoming telnets, ftps, etc... (This was in the days before everyone ran mailers that used name servers, i.e., before the age of the MX record...) But, none of this is relevant to accessing the PC7 archives. So - the PC7 archives physically reside on the um.cc.umich.edu machine, but must be accessed through its internet network interface, at him1.cc.umich.edu. Perhaps sometime in the future, when this HIM stuff evolves past the experimental stage, the um.cc.umich.edu address will point to the correct machine. But for now, it doesn't, and that's about it. Are you sorry you asked? }-) > > >Brian Walker, University of Colorado at Boulder >walkerb@tramp.colorado.edu ...!{ncar,nbires}!boulder!tramp!walkerb >Q: What do you have when you have a lawyer buried up to his neck in sand? >A: Not enough sand. -- / /_ , ,_. Howard Chu / /(_/(__ University of Michigan / Computing Center College of LS&A ' Unix Project Information Systems