Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!swrinde!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!sei.cmu.edu!fs7.ece.cmu.edu!o.gp.cs.cmu.edu!andrew.cmu.edu!jb3o+ From: jb3o+@andrew.cmu.edu (Jon Allen Boone) Newsgroups: comp.admin.policy Subject: Re: Possibly nefarious users Message-ID: Date: 26 Jun 91 13:23:26 GMT References: <2D.-_.N@cs.widener.edu> <1991Jun6.214915.18946@athena.mit.edu>, <1991Jun7.164102.672@progress.com> , <1991Jun10.053229.4305@qiclab.scn.rain.com> <4cIuopy00j69QAZE4H@andrew.cmu.edu>, <1991Jun25.061208.6009@qiclab.scn. Organization: Carnegie Mellon, Pittsburgh, PA Lines: 117 In-Reply-To: <1991Jun25.061208.6009@qiclab.scn.rain.com> leonard@qiclab.scn.rain.com (Leonard Erickson) writes: > jb3o+@andrew.cmu.edu (Jon Allen Boone) writes: > > Really? I'm amazed. We *only* have 250 ports on our main server. And we > have had to take measures to be sure that they weren't being wasted. > (things like users being logged in but just sitting at the system > prompt for *hours*) I'm not sure of the number of ports (I don't work for the data communications department) but I can call up the system any time of the day or night - and if the number I traditionally use is busy (which is exteremely rare - MAYBE once a month, if that) there are three or four other numbers I can use. End result: it's not a problem. (Note: these aren't outdial ports - these are incoming ports for loggin in.) > I'd be rather surprised if *any* system has so many outdail ports that > the use of on isn't at least a *potential problem. What do you use outdial ports for? Our system may well have NO outdial ports that are accessible to public users (I've never used them). We do have as hundreds of workstations on the internet - providing you access to almost any place that you want to go on the internet. If you want to call up a local-area bulletin board system, the you should get your own modem! > We get complaints when folks call us and say "Why can't I login?". And > when we discovered that a user had been logging in from several > machines at once to "multi-task" we had a talk with him. Never happened here, as far as I know. I can log onto as many machines as I want - there are special exceptions (some administrative machines won't let you on at all, without special permission) - if you're telnetted in, and I log in on console - you lose. But, there's not a problem with me multi-tasking on multiple machines - in fact, we used to have a system setup to let you do exactly that! > Sorry, but as long as resources are *limited* this *will* be the case. > Because the impact is *never* going to be "negligible". If your files > are scattered that widely it'll take *extra* resources just for the > "system" to find them. Extra traffic on the net looking for free space, > etc. How the hell do you get "limited"? If your site has so many users that any one of them using a socket is a problem, then there should be a serious re-evaluation of your computing systems. Also, note that this attitude is the same one that keeps universities with a surplus of computing fascilities (like mine) from sharing them more openly and regularly with systems like yours. Too bad for you. :( > You are suffereing from innumeracy. Your argument assumes that adding > up lots of little bites doesn't make one big mess. Even if *you* only > take a sector on each of a large number of machines, that doesn't mean > that the impact is negligible. What you forget is that everyone else > would be doing the same thing! And that adds up fast. Not at all. I am perfectly capable of keeping my usage down. In terms of long-term storage, I think MORE people should have smaller quotas - there should be MUCH MORE temp space which is, after all, first come - first serve. With the advent of multiple-write/multiple-read cd's, and more and more cd devices in workstations, I think that this will be acceptable. After all, when you can carry 300meg with you, what the hell do you need a large quota for? (Note: this argument looks forward -> to the day when MOST systems have these cd's - it flagrantly ignores the way things are NOW -> but if you don't look forward and just sit here in the now, things pass you by. The fact that it MIGHT not be a good idea NOW doesn't mean it will be a bad idea in 4 years or 10 years or 25 years or 100 years.) > One persons "vital files" are another person's "junk". Under *your* > system, it'd be "first come, first served". This is not practical. > When someone has purchased a system for "X" they are going to * > *justifiably* get pissed off if they can't use it for that when they > want to. As prices fall, this will become less and less common. Quite right that there is a lot of work to be done - networking bandwidth expanded, more disks, more memory, etc. But, I never said that any of that was justification for being pissed. On the contrary - if I'm using temp space, I expect you to NOT get pissed and just blow my files away - regardless of your opinion of them. > Your complaint is that the resource allocation is "less than optimal" > *for you*! Fine, obtain your *own* resources. If the costs are as > negligible as you make out, then this should not be a problem. > Otherwise, you've just proven my point. It's not a practical way to > allocate things. I have my own resources - I have unlimited disk quota - at times I've used as much as 150megs, now I bob along at about 40megs. It's not a problem. Soon, I may well have 700megs of disk space in my house - yow! But, allocating quotas of disk space, for example, is a NON-optimal way of doing things (assuming the existence of high-capacity, portable media) - for now, you can get away with it - but if they ever work the kinks out of the floptical drives, forget it. Buy a machine - buy a gig of disk space - use 250megs for systems programs and make the rest TEMP space. > For an analogy, We think that you are making less than optimal use > of your living space, so we are (all) going to borrow a little of it. > We do *mot* care that you had that empty space set aside for something > you are going to do next week. After all, you aren't using it *now*... Fine. But as soon as I AM going to use it - I'm going to throw your junk away - unless I can find you and get you to remove it first. ----------------------------------|++++++++++++++++++++++++++++++++++++++++ | "He divines remedies against injuries; | "Words are drugs." | | he knows how to turn serious accidents | -Antero Alli | | to his own advantage; whatever does not | | | kill him makes him stronger." | "Culture is for bacteria." | | - Friedrich Nietzsche | - Christopher Hyatt | -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-