Theo Verelst Diary Page

Thu Jan 31 2001, 6:09 AM

I've decided after good example to write some diary pages with toughts and events.

Oh, in case anybody fails to understand, I'd like to remind them that these pages are copyrighted, and that everything found here may not be redistributed in any other way then over this direct link without my prior consent. That includes family, christianity, and other cheats. The simple reason is that it may well be that some people have been ill informed because they've spread illegal 'copies' of my materials even with modifications. Apart from my moral judgement, that is illegal, and will be treated as such by me. Make as many references to these pages as you like, make hardcopies, but only of the whole page, including the html-references, and without changing a iota or tittel...

And if not? I won't hesitate to use legal means to correct wrong that may be done otherwise. And I am serious. I usually am. I'm not sure I could get 'attempt to grave emotional assault' out of it, but infrigement on copyright rules is serious enough. And Jesus called upon us to respect the authorities of state, so christians would of course never do such a thing. Lying, imagine that.

Previous Diary Entries | List of Diary Pages | Home Page
 

Thu Jan 31 2001, 6:09 AM

I've been working on the windows 2000 / linux setup, and I've webwise read into the latest news a bit from my former university, which is worth writing about.
 

Doing the samba thing.

After a bit of hassle and a lot of simply looking around, I've come to the samba point with both the (year old) Suse linux disttribution and a recent Mandrake version. I've been looking at lots of tools and programs, some to be restricted to a capable enough system administrartor, ranging from grep and for instane 'tr', a charachter based letter translation filter type tool to graphical process browsers, and the various important input output and boot parameters and setups.

Anyhow, I was affraid last friday I could have erased all my user data and account data from a least one of the two linux machines I happen to have running, because I'd unexperiencedly used the installer to update the software. Luckily, after removing some mounts, or entries from /etc/fstab and /etc/*smb.conf  to the same effect, and rebooting, or even without, I got my original file system back, and the directory tree with users, so the problem was luckily solved. And I made a backup, to be on the safe side.

The samba thing is the server process which makes it possible to mimic a windows (it appears to be NT 4.2-4.6) network, though it uses a workgroup name, it can appear officially in some domain, in this case in my own I think it is forrest domain, next to an existing, also non-ip, intranet domain, which can be routed to another, though currently hardly are. In simply language, by using samba right, the windows 2000 or other windows machine on the same network can 'see' the linux systems' file system, the disc, or the CDROM, just as if it were a shared file system on a windows machine. So the latest mandrake system appears as an icon in the 'network environment' from windows explorer, and the files on it can be browsed, copied, changed, and deleted, just like on a windows system, depending on the permissions, of course, which can be set pretty tight and easy enough, and quick, on a linux system.

Vice versa, a windows 2000 shared disc can be viewed and used on the linux machines, for instance with some file browser, or through the command shell, for instance one machine is called 'testserver', which on a linux shell prompt can be addressed through /testserver with all the files under it. To try it out, I ran Tcl/Tk on a linux machine, taking library files from the windows 2000
machine directly, which worked fine.

Between the 2 linuxes, I used nfs file sharing, which is natural enough for unix, and now they happily share files back and forth, which made me find out that indeed different kernel versions on the same hardware may not be compatible, certain shared libraries are not present or the same, which make certain programs not run directly on another kerel version. On the other hand, cd-ing to the other linux systems file system /bin and running ./ls for instance works fine, and is an example of sharing executable files.

Using this principle for bulky program sets, combined with wise use of the 'PATH' variable is good practice for complicated and powerfull distributed system setups. Xwindows could already be exchanged between the various machines, I have used the Free 86 X root server for windows to have a windows based window in which X windows can be run, can start an xwindows based process on either linux machine or under the cygwin bash shell on the 2000 machine, and plant it on any of the attached displays, simply by adjusting the DISPLAY environment variable to for instance 192.168.200.1:0.0 for \\testserver 's Free 86 X window root, before starting up some X program.

In fact I made it worse by letting one linux machien manage the whole Free X root window, that is menu's, desktop and icons, screen switches and taskbar, over the network, and it doesn't seem to mind, it works fine. A touch slow for bulky windows over the 100baseT, but no more (probably less) then an windows terminal server/client link, and of different nature. There is hardly any inherent delay, most responses are just about immedeate, except when lots of data are transfered, for instane when drawing a 1200x800 desktop wallpaper in 24 bit/pixel, or a complicated web page running not on the local machine.

Though both suse and mandrake feature major install tools and 'wizzard' equivalents, neither ran the network in a working way just like that, not in any way. Maybe I could have tried again, but the install tools didn't get anything running in my case, only after I worked both the processes and the configuration files by 'hand' editing, I made some progress, and only by catching up on the fstab mounting syntax, the smb or samba.conf necessities from some examples, and making sure /etc/rc even ran inetd for instance, which first it didn't, leaving me without even telnet, rlogin and ftp, except my own tcl/tk based 'pcom', I got the whole machinery running. After which I still manually put in the 2000 shares though a mount, as well as the liux mutual nfs mounts. And of course the editor happens to have ended up as 'vi' (visual interactive, as followup of ex 'extended' and ed for 'editor'), which is normally not impossible but not natural to learn, but runs fine on a vt100 window, if needed even over a modem connection (I know from previous, say 15 year back, experience), and that on a white on black, major courier font 80x25 terminal xwindow. Like back in university HP workstation mode.

Those command lines and stream based tools are good for more than a little. I went trough major amounts of recorded 'ping' data of some 12 server running a intranet, and used 'grep' (context mode), wc (word count), and tr (character translate) to get clean lines of ping delay or 'timed out' requests, and them counted, to compute for instance per server or overall how many times a connection failed, or even how much. One liners of shell programming with the power of database commands with string manipulation, at major speed and ease. Someone did make all those computer ideas, they did not grow of the microsoft tree, and linux makes clear that I can pick up a book from 79 and tell no lie the the knowledge for instance about 'tr' therein is completely valid today, and that the way such a system is set up, makes sense and squeezes quite a bit out of the systems form those days. X is from somewhat later, but similar logic holds.

I did teach unix courses some 12 years ago, guess that wasn't  stupid move.
 

Electrical engineering motivations.

Maybe it was to have some of my work stolen in about that time, where I programmed stream based (still rare in our days of C++ object holy grail searching, which as anyone should know probably isn't there, and wouldn't be worth much if it was) geometrical transformation tools for rational bezier (spline like) computer graphics surfaces. That guy seems to have made to a certain kind of professorate in my former sections' remains, and definately used my idea in a paper at a graphics conference I was more or less left out of at the time, even though I had picked up the subject, me at the time not realizing that the major meat in that paper, except maybe from some general theoretical descriptions, were my ways to transform those surfaces, and I had programmed them in working practice.

This section was intended to be about something else, too, important stuff,  I'll get to that probably in a few days.

Just in case someone might have the wrong idea, even though I was 'worked out' of that section (or more or less), by someone now gone themselve, and even though at some point I did have clearly defined ambition, I don't think I want to dream going back there, because I strongly suspect (though cannot hard proof) people there for way to far have been into things I don't even like thinking about, and I don't need the risc to work with the former child abuser of girlfriends I would like, especially if on top they'd even try to be over my and their heads.

Luckily I did some share of theoretical physics, and some other things on top of subject searcher, so most probably may not have too hard a time at least making clear who is who, and that I was who I have presented myself to be, contentwise good enough. Which for people who have spare the place for some honorability is certainly essential.