The Shindig (english translation)
On Fri, 2 Aug 2002, Official Flamer/Cabal NON-Leader wrote: There was a shindig today. It was attended by a community of misfits and elderly geeks, some pushing the wrong side of 35, obviously well past their use by date. there was a gathering today. It was attended by a group of over 100 unrelated inDUHviduals of all ages (14-45 would be my guess). It was opened by a weirdly coiffure'd person speaking about Elvis, of all things. Elvis, obviously, has significant timing problems, since his frequency has been administratively increased from 100 appearances per alien abductee to 1024. Which is, nevertheless, progress. it was opened by gilad ben yosef (the next generation) who found some obscure excuse to involve Elvis with embedded linux systems - in an attempt to persuade his listeners that embedded linux is gapping the strangled desktop PC business from every direction - in an increasing rate. according to the applause in the end - everybody agreed with that claim. Then a person with a very variegated taste in shirts spoke to the assembled about parasitic insects' victims which send each other letters. Thus, Host A send Host B something (and, sometimes, vice versa) which Host C, for some ineffable reason, wants to hijack. While highly unlikely, since rarely do hosts hijackanything, unless they are hostesses, of course, in which case they are likely to catch something and not hijack, barring Muslim gentlemen with rudimentary flight training, in which case they are hijacked and do no hijack themselves which, basically, means tha Never mind. then aviram jenik took over, and tried showing people how insecure their secure connections realy are - this attracted quite a few questions from the crowd, ranging from i know the server is protected by a firewall, and i connect from my remote PC - is there software to protect my PC from being broken into (the answer, surprisingly, was use a firewall on the PC), to i telnet to a machine. is it realy that insecure?. some questions were more advanced - and there were quite a few of those. There were no loo breaks so all suffered distended bladders. due to hickups in the organization, time flew by (time goes faster, when you move slow ;) ), so the proper solution was chosen - skip the break! a short protest, and some recalculation, suddenly brought a 10-minute break into existance. the 2nd break was shortened to 10 minutes (from the originally planned 20), and thus bad infections were prevented. Then came a film in which a whole collection of aging communist hippies lashed out at Microsoft, denigrating the glorious company's contribution to world culture, technology and, incidentally, to quite a few Congressmen and Senators. These representatives of the drug taking, stuttering, unwashing, oversexed, under-stratified and otherwise anti- American elements lauded a damned foreigner's method of playing Solitaire! The attack on our culture is intensifying, mein komaraden! then came a film, in which about 70-80% of the time there were 'talking heads'. Some heads looked quite familiar, such as the linux head, the RMS head, the clown head... After the film ended, there was a signing of some form of Satanic contracts in the cafeteria and finally a drunken binge! after the film ended, a bunch of the said group of inDUHviduals took over the cinematheque's cafeteria, in order to hold a PGP signing ritual. after that, people started spreading around, many went home, and many took over a near-by restaurant, fighting with the waiters in order to connect tables in a large 'L' shape - well, this _is_ a Linux users group... I, of course, can say nothing of the stubby monkey whichclimbed the stage and, instead of singing the Anthem, declaimed something in an unhuman language, waves his arms a bit and climbed off. Who the fuck was he? ah - but i can. one of the attending inDUHviduals was asked, 15 minutes before the event begun, to carry the opening 5-minutes keynote speech. To this challenge he responded by speaking in a language no one(*) understood - and which sounded like a mix between german and latin. That inDUHvidual looked suspiciously similar to the former reporter. Your Reporter this last phrase is in english... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: FW: Blank screen when moving from X to console
On Sun, 25 Aug 2002, Tal Achituv wrote: Here you go: mov ax,7 int 10h this is a bios call - does it work under linux? won't it conflict with what linux thinks is the terminal's mode? its also not in a format understood by GNU assembler - they use the ATT format, not the intel format. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: X problems
On Sun, 25 Aug 2002 [EMAIL PROTECTED] wrote: when i 'startx' i get a black screen and sometimes (when running startx again) i get to see the desktop for a variable amount of time (0.5 seconds to 3 secs aprox.) which program did you try to configure X with? there are several such programs - one of them might work. not sure which comes with mandrake - you should have mandrake's own X config utility (drakex?), or Xconfigurator (is it redhat specific?) or xf86setup, or also - you get lots of messages when you run X, on the console from which you ran 'startx'. can you copy them _all_ into a file, and send them over? specifically the last few might tell exactly what the problem is - or they might not ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: X problems
On Sun, 25 Aug 2002, Aviv wrote: i attached the file to this mail just btw, now when i ran it i only saw the kde screen for half a sec when switching between VCs oki doki. you have the following error, which repeats for a few applications: /opt/kde/bin/artsd: error while loading shared libraries: libaudiofile.so.0: cannot open shared object file: No such file or directory as you can see - you are missing the 'audiofile' library. this is a low-level library, used by most of the KDE applications. you need to find how you managed to miss installing it - and find which package it belongs to. since i'm not familiar with slackware - soemone else might be able to tell how to find which package this library belongs to. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
re: X problems
On Mon, 26 Aug 2002 [EMAIL PROTECTED] wrote: oki doki. you havethe following error, which repeats for a few applications: /opt/kde/bin/artsd: error while loading shared libraries: libaudiofile.so.0: cannot open shared object file: No such file or directory as you can see - you are missing the 'audiofile' library. this is a low-level library, used by most of the KDE applications. you need to find how you managed to miss installing it - and find which package it belongs to. since i'm not familiar with slackware - soemone else might be able to tell how to find which package this library belongs to. well, i get an error when starting kde on another slack 8.1 box saying i dont havea sound card configured or something like that, i just click ok and everything works smoothly i guess thats the same error and i doubt i didnt install that package because both systems are installed from the same cd and i chose the same kde packages for them so i either installed it or the package isnt necessary for running kde this library _must be installed_ if you want thoose underlying applications to work - when an application is dynamically linked with several libraries, and one of those libraries are not found - the application does not run. this still possibly does not have anything to do with your problem - but it is nontheless a problem that you should fix. waving with your hands and muttering holly matramonies - won't make your problems go away ;) kindly do seek a fix to this problem by finding which package was not installed - this has nothing to do with a lack of soundcard, and thus it is NOT the same problem. why do you think its the same - just because it has the word 'audio' in the error message? can't you see that the error message is different? you also ought to remember that trying to identify a problem, while other problems clutter your view - makes it much harder to identify the problem. so fix this one, and if it does not solve your major problem - then you'll at least have a clearer view of what is going on. and if you're lucky - this _is_ what causes your problem. good luck, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Ivrix site
On Tue, 27 Aug 2002, Nadav Har'El wrote: In particular I was wondering what it would take to host the ivrix.org.il domain on the iglu.org.il server. Who makes such a decision? the purpose of iglu, is to host linux-related information and services (mostly the mirror service) for use by the israeli community. as such - ivrix.org.il itself seems to fit the bill (the other domains probably don't). assuming this domain does not require a lot of disk space or network resources, there should be no problem hosting it on linux (asside from administrative overhead of setting things up). this discussion, however, apparently belongs on 'iglu-web' - where matters regarding administration of the iglu machine are discussed. if you want to pursue this further - please post on iglu-web. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: sh question
On 12 Sep 2002 [EMAIL PROTECTED] wrote: i want to redirect both strdout and stderror into a pipe in tcsh i do: process1 | process2 how do i do it in sh ? in bash ? ( is that the same ) process1 21 | process2 -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: DMZ
On Thu, 12 Sep 2002, Guy Cohen wrote: You couldn't be more right. The art of mailing list is slowly dying. people who spend 5 and more years investigating unix want to get payed and are sick and tired of not finding a job because a potential employer could just send a question to the mailing list and get the answer he *should* pay for. so you are in favor of free software, but not of free support? what about free documentation? and you're one of those who seem to answer people's questions here. tsk tsk ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Women and Linux
On Mon, 23 Sep 2002, Nadav Har'El wrote: On Mon, Sep 23, 2002, Orna Agmon wrote about Re: Women and Linux: Another imporant aspect is the pass making. Taking the woman who is interested in Linux first as a woman, and only later as a linux person. Why would you want to be treated as a linux person before of a woman? This completely contradicts the IGLU meetings should be more friendly idea. Friendly means taking you as a person (man or woman) first and as a walking-talking-Linux-manual only second. this is the issue - as a person. not as a walking target. The first time I went to Haifux, the only person who talked to me was trying to hit on me. The Doesn't that happen to you in other places where men and women meet?? Will you stop going to the supermarket because loudmouth macho-types try to hit on you by the vegetable stands? and the fact that it happens in other places is your ustification for accepting this here? whats more, you'd expect more from a civilized group of people. If not coming from an obnoxious person, then you should take the advance in the friendly manner it was offered, and decline. If the advance was obnoxious, rude or harassing, then feel free reprimand the guy sternly. wait until you become a woman (in some future life) and get hit on too often, and then see if you manage to 'accept it with grace'. the fact is that you, as a male, are quite safe from these kinds of harrasements, and can attend a meeting with strange people without being afraid that someone will address you in a way that'll require you to 'refute'. the reason why you migth think differently, is that being a male, you get too few women hit on you (i'm talking statistically here, ofcourse). I don't think that being afraid of being hit on (to the point that another girl mentioned her BOYFRIEND in capital letters in a previous post) should be part of you girls' Linux experience. Being hit on happens everywhere, and it is not always unwelcome; You, of all people, shouldn't outlaw it altogether, seeing where you met your own boyfriend :) i'd say this is a blow below the waist. obviusly, once you know people for a while, you get to know some of them and get new friends. but this is a process that takes time, not a 'hit and run' thing like what was mentioned earlier. And you know what, you girls can feel free to hit on us guys (and me in particular :)) once in a while, whether it is in a Linux meetings or any other mixed-gender crowd. see, this is where your hitting on a person is never a bad thing perception comes from. The point is that there are rude and disgusting people in both genders, and if you let that bother you too much to the point that you stop doing activities you like, you'll miss many opportunities. Obviously this lesson isn't specific to Linux. and obviously, if people are more aware of how annoying this situatin is, less people will do those things, making our lives better. will you sign the 'no hitting on first sight in LUG meetings' manifest? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Women and Linux
On 24 Sep 2002, Oleg Goldshmidt wrote: Orna Agmon [EMAIL PROTECTED] writes: the only person who talked to me was trying to hit on me. Orna, he just saw more in you than just a Linux user... ;-) oleg, it is precisely these kinds of comments which drive people crazy. obviously, you did not get the clue... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: linux 2.4.20-pre?-ac? kernels
dear people, i think there's some great confusion here about the usa of the term 'stable'. ariel (at least as far as i know) runs systems that bear a rather heavy load over network connections. most people run machines that do not bear that heavy load. thus, your definition of 'stable' is quite different. and its a rather known fact, that for realy heavy network, memory and processing load on linux, most of the 2.4 kernel series was rather flowed, at least until 2.4.18 came out. a company i worked with encountered lots of woes in this area (running a machine with redhat's kernel or any vanilla 2.4.X kernel, about 8-12 month ago) year back, perhaps 8 month back) in the lab and bombing it with traffic while running their application on it (which used a lot of memory), caused networking to stop working after several hours or half a day, in a repeatable manner. only when then eventually installed a variant of the 2.4 kernel with the 'aa' patches - did they manage to have a machine running steadily - and that certainly wasn't a tested kernel, back then, not to mention QA. you can argue that your experience was different as much as you will - but the fact was that the VM code caused problems on a machine that was hammered heavily the same problem existed with pre 2.4.10 kernels, that is 2.4.7, 2.4.9, etc). and as far as i know, not all of arcanageli's patches went into 2.4.19 (muli - please correct me if i'm wrong regarding this merge). this VM fiasco apparently did not have a parelel during the 2.2 kernel release cycle, or with earlier versions - it did happen with 2.4 . btw, at the same time, another company that used much earlier 2.4 kernels with an appliction that performs network processing - did not encounter _any_ problem. ofcoruse, they handled traffic flowing at only 10 mega-bit per second - and that did not use as much memory. and they were using very early 2.4 kernels (including some 2.4-test kernels) - and had no problems. so from their point of view, 2.4 was stable since quite long ago. so, i'd say again - stability depends on what you're doing with the system. and QA alone cannot fix problems stemming from major bugs in the underlying system. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: linux 2.4.20-pre?-ac? kernels
On Thu, 26 Sep 2002, voguemaster wrote: True, but then how would you explain the article by Moshe ? He tested several 2.4 kernels and the RH 7.2 kernel, while a bit slower, was stable. different usage patterns. btw, i didn't read the article. thought it does not matter - the fact that it worked for one test can NOT be used to refute the case that it did NOT work for another test. and by telling ariel you are talking bullshit because it works for us - you are not making a good case. fact is, that it didn't work. and as far as i know ariel, i don't think he would go saying something did not work without doing some proper testing, and without realy seeing it not testing. on the other hand, i don't think its redhat's fault at all - its the fault of the 2.4 kernel developers, and their decision making. the good thing, is that we can see a path of it stabilizing - but it took quite long to stabilize. now, here is a question for those 'in the know' - how far is alan cox involved in choosing what patches go into redhat's kernel version, since he begun working for them? or, who in redhat decides on that? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Options (Re: VMware and competing products)
On Tue, 1 Oct 2002, Omer Zak wrote: According to a brief Web surfing session, the options are: Bochs - a Free emulator - not suitable for my needs, because it is emulator and not virtualizer (i.e. very slow). plex86 - a Free virtualizer - suitable for people, who want to develop it rather than use it as a reliable tool for their own work. For me to use it, would be like using a 2.5.* kernel for my regular development work. VMware - Workstation 3.2 costs $299 per license, and 30-day free evaluation is available. Some people recommend it. Can use Linux based host. VirtualPC (http://www.connectix.com) - Virtual PC5 for Windows costs $229, There is no information about a version running under Linux host, but I sent them an E-mail message asking about this. use vmware, and forget about anything else. for 300$, you get something that works, that gets further developed, that works quite well, and with a near-current CPU will work fast enough for your needs (any p-III 700Mhz or above would make it almost invisible). the support for 'undo' of file system changes is valuable with software that might corrupt the system, as well as with testing installation scripts. saving on those 300$ is a simple waste of time, and other products don't tend to come near what VMWare does. at my former workplace, VMWare was used to test installatin procedures - they had several operating system copies with it, and it saved a lot of time for developers. btw, just to add to the confusion, you always have the option of using norton ghost to save a copy of partitions, and install and delete them when needed - it takes about 5 minutes to recover a typical windows 2000 installation (10 minutes if done over a network). but this is quite inferior relative to the time it takes booting an OS under VMWare. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RE: what's going on with the list ?
On Thu, 3 Oct 2002, Adir Abraham wrote: On Thu, 3 Oct 2002, Aviram Jenik wrote: If instead of arguing you would have actually tried to find the cause of the problem, you would have seen that both emails that continually bounce are sent via the technion's mail gateway mailgw2.technion.ac.il. Oh. That's an interesting fact. I would have seen that the emails are bounced via the Technion's mailgw2.technion.ac.il. I have nothing to do with the Technion's mailing gateway systems, so it's interesting that you got to that conclusion. [.. snip ..[ people, people, what's the matter with y'all? shouting at each other for no good reason. such a misunderstanding - as far as i saw in the headers of the bounces (i only looked at one of them), there was no information there that could point towards any bosmat servers. indeed, i have no idea where nadav came with that email address from - i erased all the redundant copies (and i see the bouncing has stopped by now) so i might have missed something. now, if i may add constructive suggestions for next time (please don't get mad - i'm suggesting something practical - not fair). adir - if people get to such wonderous conclusions in the future, just state the fact that you looked at the mail headers, found nothing regarding those email addresses in them, and wonder where the original sender got this info from. you _almost_ did that, except for trying to explain why nadav's conclusion seemed to clash with the reality you see. think it over a bit - you might see what i mean. and aviram - there's no need to jump to conclusions about what adir's role is. nothing in the letters he wrote implied he's in charge of the technion's mail gateway - he only responded to the email address issue that nadav brought up. true - adir's response _seemed_ to imply a fwe things - but mostly if it was read out of context, and still it probably did not warrant such a harsh response. and i thought the heat wave is over by now ;) i'd suggest that in the next occasion, we all try to read email letters in full, before we send such harsh messages to the list. will help to keep everyone's calm. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Why jobs are wanted (was Re: job wanted also - NOT - YES!)
On Sun, 6 Oct 2002 [EMAIL PROTECTED] wrote: Quoting [EMAIL PROTECTED]: In addition, I would also like to know what is going on because it seems that many people are loosing their job lately. I consider this information on topic as well. Ah, well, there are several reasons: The quarter has just ended, and with it, dire financial results for many companies. Also, many companies have deferred the cutbacks till after the holidays. and there are still people that switch jobs - even if they are a minority. do not immediatly assume that if person X stopped working during an economic depression - then they were fired or the company was closed. cheer up - there are still those that look for new people. you need to have something extra to give them, to get there. if you don't have that something extra - then use the time you're at home to learn and practice that something extra - its better then just sitting at home and feeling bad about your situation. (i;m not talking specifically about you - i've no idea regarding your skills and knowledge - i'm talking in general about job seeking when you seem to be unable to find a job). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RE: A disturbing article...
On Mon, 14 Oct 2002, Eytan Heidingsfeld wrote: I don't think you understood my about my ADSL problem (BTW) I tried installing it and I still can't get it to work. I have read the how-to BEZEQ_ADSL_LINUX document but can't get the stupid thing to work! my memory fails me here - did you write about this before to the list and asked for some help? with the relevant details that could allow people to help you (relevant error messages, relevant logs from pppd's debug mode, exact system setup? pinging the adsl modem works?) btw, this is not a reply to your letter per ce - i'm not saying that refering to a mailing list for assitance should be the normal route to getting ADSL working - just saying that it _can_ be a route to go, and if you wish to solve the problem, and haven't mailed the list before with these details - you might want to do that. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
typing Vs. car mecahnics (was: Re: A disturbing article...)
On Mon, 14 Oct 2002, Guy Cohen wrote: On Sun, Oct 13, 2002 at 09:44:48PM +0200, Oron Peled wrote: So they didn't waste a month or so fighting their keyboards. But the poor guys still use their keyboards *for years* at a fraction of the speed I regularly type. Thats a good example. I'm starting to learn auto mechanics tomorrow. It will take me a while to learn, but next time my car brakes, I won't have to spend all that money at the shop. not quite the same kind of analogy. if someone never uses a computer, and lets some typist do all the 'heavy' typing for them - then your analogy has something to it. but there are those people who simply have to type various documents from time to time, spend quite a while on that, and don't figure they could either pay a typist to type for them, or learn how to type faster. i've seen some programmers with this typing slowness - you would agree they'd want to learn how to type faster, right? btw, i did see people who type with 2 fingers in high speed - 10 fingers is not the only way to type fast. its all in the 'long' brain. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Reverse ADSL?
On Tue, 15 Oct 2002, Oleg Kobets wrote: No, there is what is called Synchronous ADSL. Both streams can go as far as 12Mbps (I think), but Bezeq using the Asynchronous one. i think you're refering to Synchronous DSL (SDSL). the 'A' in ADSL stands for Asynchronous. there are also other types of DSL protocols (VDSL, for example) which are not yet used in israel - some of these are probably in use abroad (SDSL is in use in the united states, as a replacement for frame-relay and fractional T1 technologies). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Reverse ADSL?
On Thu, 17 Oct 2002, Barak Kaufman wrote: the A stands for asymmetric oops. ofcourse. stupid me :0 -- guy sorry, no world domination today. come back - one year = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RedHat 8.0 on IGLU - finally
we finally got RedHat Linux 8.0 ISO images on iglu's server. it can be found on the ftp server with the rest of RedHat's files. note that for installation you only need the 1st 3 ISOs - the other 2 contain SRPM (source code) files. the ISO images were taken from tel-aviv university's redhat mirror - the MD5 signatures taken from redhat's site and compared with md5sum of the images. the 8.0 updates are not yet on iglu's site - hopefully they'll start flowing later tonight, or over the weekened. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: increasing the keyboard's responsiveness
On Mon, 28 Oct 2002, Arie Folger wrote: When typing in kde apps, I feel that typing has definitely become more sluggish lately (as opposed to earlier versions). Does this mean that it's time to upgrade (need a new job, first) or is this tweakable? what kind of hardware do you run this KDE on? myself, i use an AMD k6-2 with 256MB of RAM, and since KDE 2.X and beyond runs to slow, i switched to gnome 1.2 (RH 7.3). KDE became too bloated to run on this hardware. [please don't start flaming - i liked KDE's look better ;) ] i might upgrade my hardware soon for other reasons - but i shadder to think i'll do it for the sake of a desktop environment - that's one of the reasons why i don't use windows... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: CHANGING the routing for established connections.
On 29 Oct 2002, Meir Michanie wrote: with nat table and DNAT target you can redirect new connections. but how can you break all ready established connections and redirect to a certain ip? mangle support established connections but do not support DNAT. i think at least in the vanilla kernel, this is not supported. normally, you can not redirect an already established connection. this makes sense - how would the new target machine know how to synchronize in on an already established connection, that has an established state, and established sequence numbers, etc? can you explain why you find the need to redirect alerady established connections? perhaps this can be remedied... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: dvd player*s* show twice same half screen
On Fri, 8 Nov 2002, Arie Folger wrote: OK, after having decided that RH8.0's xv extensions are the probable source of trouble, I moved my /usr/local partition's content to /usr/local on the /usr partition, and installed a barebones RH7.3 on the freed partition (1GB). Xine now plays the dvds beautifully. Of course, dual booting is not the world's best way of playing dvds (I could even have used windows' player, if dual booting was an ok performance -- forget it, I have dvds from outside my region), so I'd like to confirm the source of trouble and either fix it or file a bug report. mplayer has an option that specifies how it will send images (the '-vo' option). you can run it and tell it to use something other then 'xv', and see if the problem goes away. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: 4 tars in a row
On 13 Nov 2002, Gilad Ben-Yossef wrote: On Wed, 2002-11-13 at 18:18, Ben-Nes Michael wrote: I noticed my web server is extremely busy so I ran 'top' and sow 4 tar process where running in the background under 'nobody' I killed them and start to look around in my logs but vile :( any ideas ? ( its not cron ) Yes - congratulations! your box has most likely been rooted. Your log files have been edited. Your 'ps', 'netstat' 'login', 'less' and zillion other binaries have been replaced with trojaned copies programed to not show you exactly that which you are looking for. Your best bet is to disconnect the computer from the network, boot using a floppy and take whatever *data* files you might need. *DO NOT* copy or run any executable. Format HD and reinstall OS, and make sure to apply all the latest patches etc... hold your horses before jumping to such a conclusion - one should verify that indeed their system was broken into. by killing those processes - michael lost invaluabale information - e.g. what files were handled by these tar commands - which process was their parent process, etc. however, if there was a break-in, one could find it e.g. by looking for system binaries whose checksum is different then that of a 'safe' system. or checking last update times (ls -c, not just ls) on various system files. and then doing some thinking... too many times i had situations that alerted me to think oh my god, someone broke into my system - when some checking showed there was no break-in - some application broke, or some other valid user with valid 'root' access made changes to the system and did not tell me about them, etc. ofcourse, the matter should not be overlooked - but also not automatically assumed to be a 'break-in' just by using circumstential evidence. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Netvision accounts price raise
On Thu, 14 Nov 2002, Oleg Kobets wrote: Umm, I am paying ~39$ = 225NIS (vat included) for 1.5m at Actcom. I also got static IP free of charge :-) you can also pay for a year in advance, and get a discount (the discounted price is around 33$ a month, on average, for the said account). the discount is higher then the interest rate you can get on most types of savings (around 18%) - so its better to pay in advance (if you have the money, ofcourse) then to put the money in the bank during that time - unless you expect one of: 1. an average drop in the price of the dollar of arond 15% for that year. 2. a drop in the price of similar services over this year in a similar ammount. 3. a drop in the actually conection speed during this year, that'll make you want to switch providers. i've paid this yearly price for the 2nd time, a few month back. just for general info - i'm using actcom's services not just because on economical reasons. i used to work there several years back. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SNMP question
On Wed, 13 Nov 2002, Michael Sternberg wrote: I'm using UCD-SNMP version 4.2.3 over Linux. I've compiled the ucd tutorial mib module into an *.so library, which SNMPD loads. The following lines the access section of snmpd.conf: [.. snip ..] The only response I get is for the system MIB. On call: snmpget MyLinuxMachine public enterprises.2021.13.4242.1.1.1 I get the response: Error in packet Reason: (noSuchName) There is no such variable name in this MIB. Failed object: enterprises.ucdavis.ucdExperimental.ucdSnmpTutorialMIB.ustMIB Objects.ustScalarSet.ustSSSimpleString try to 'snmpwalk' over this machine (from the 'root' node) and see if you get nodes that appear to belong to the tutorial's MIB. perhaps the path you're using is not the same as is actually defined in the MIB or used by the system at run-time. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Ideas about presentation needed!
On Wed, 13 Nov 2002, Boulgakov Andrei wrote: Hi, everybody! I call to your experience in presentations. At my work I've got the task: make presentation about Internet technologies. On my question what does it mean Internet technologies i didn't get an answer. Do you have any ideas or materials about Internet technologies. Audience - UNIX programmers and QA testers, so I guess, HTTP header structure is not good subject :) what exactly do those UNIX programmers do (what kind of programms do they write)? you should check what the _purpose_ of this presentation (i.e. is it to enrich peple about hot buzzwords not related to what they do at work? is it to allow people to think if they should use internet technologies in their products? is it in order to show people how to use the internet as internet users?) - this is something that the people who asked you to give the presentation - should be capable of answering. without this - you'll have to talk directly to the people that will be present in this lecture, and ask _them_ what they want to get from this presentation. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: longhorn's winFS, what about linux ?
On Sun, 24 Nov 2002, Amir Tal wrote: Longhorn will include a database-like file system called Windows Future Storage (WinFS), which is based on technology from SQL Server 2003 (code-named Yukon). This file system will abstract physical file locations from the user and allow for the sorts of complex data searching that are impossible today. For example, today, your email messages, contacts, Word documents, and music files are all completely separate. That won't be the case in Longhorn. i was wondering if there's any intention to try and implement something similar in linux, and if there are any advantages to this technology that justifies the effort. I'd like to review this issue in whatsup.org.il, and your thoughts\comments\opinions are highly welcomed here. look for information about 'universal databases'. they aimed at turning everything into queriable database material. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: cdrecord, mount complain there is no disk
On 24 Nov 2002, Oleg Goldshmidt wrote: Geoffrey S. Mendelson [EMAIL PROTECTED] writes: Any one will do. Put it in your drive and try to mount it. Does not matter if it does or not. Obviously, it didn't. Why should it find this particular CD if it does not find any other? Nor does the disk spin with any CD player app I tried (it's supposed to be a music CD). No change in status, and I have just wasted NIS 35, I think... oh, not at all. you bought some knowledge with those 35 NIS. until now - you had a _hunch_. why don't you step over to the technician to check the CD? it could be that it got damaged beyond cleaning. its a rule commonly known by peple who worked in customer support: if it stopped working 'all of a sudden', do NOT exclude hardware malfunction from the list of potential problems to check. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: USB 2.0 operating at 1.1 speeds
On Tue, 26 Nov 2002, Miki Lewinger wrote: Hi ppl. Has anybody seen that a USB 2.0 controller operates at USB 1.1 speeds on RH or Mandrake ? It seems that the full speed mode is not enabled for some reason. I googled and mandkraked (sic) the net, maybe my query was not well formulated, but I came short of answers. Maybe some of you has already solved/experienced this ? some guesswork: look at your boot messages, for those dealing with recognition of the various chipsets. i just moved my redhat 7.3 hard disk to a new machine, and the VIA chipset that handles IDE (in my case) was not recognized. i downloaded the latest (as of yesternight) kernel from redhat's updates, and it had this chip (and several others) in its lists. the symptom that lead me down that road, was getting an 'operation not permitted' error when trying to enable DMA for the CD-ROM. your boot messages are located somewhere in /var/log/messages (and possibly also in /var/log/boot.log, if you look there right after the system boots - not sure about this one, though). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Ever wanted to have a lawyer at your mercy?
On Thu, 28 Nov 2002, Guy Baruch wrote: 2) regarding apps attractive to lawyers: gnucash (though I have problems with it on RH 8.0 ). gnupg and encryption. I suggest demonstrating : email signing email encryption. whole file-system encryption. perhaps gnu-keyring ? i think perhaps a very interesting 'app' for them will be the various licenses of free source (GPL, LGPL, BSD, X11 and so on). if only a few will start checking these licenses out, there's a chance that in time we'll have more lawyers capable of answering questions related to using free software in commercial places (i'm talking from a project development point of view - of using free software inside company products, not from a sys admin's point of view). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Alternatives to Mozilla
On Thu, 28 Nov 2002, Alexander Maryanovsky wrote: When do you know you're talking to an open source supporter? When he's looking for alternatives to *Mozilla* :-) the fact that something is open-source, doesn't have to mean its a resource hog. and mozilla is a great resource hog, and so is KDE. and unlike various movie playing software - they don't _have_ to be such resource hogs. just that nobody cares enough to make them less 'hoggish'. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Alternatives to Mozilla
On Thu, 28 Nov 2002, Shlomi Fish wrote: On Thu, 28 Nov 2002, guy keren wrote: On Thu, 28 Nov 2002, Alexander Maryanovsky wrote: When do you know you're talking to an open source supporter? When he's looking for alternatives to *Mozilla* :-) the fact that something is open-source, doesn't have to mean its a resource hog. and mozilla is a great resource hog, and so is KDE. and unlike various movie playing software - they don't _have_ to be such resource hogs. just that nobody cares enough to make them less 'hoggish'. I beg to differ. Mozilla has to support a lot of things: all the HTML versions (2.0, 3.2, 4.0, 4.01, XHTML ), broken HTML, CSS, images, the XUL portable GUI library, java and Flash applets, JavaScript, many protocols (all versions of HTTP, FTP, gopher, etc), XML and XSL and the other W3C inventions, and possibly other things I forgot. It needs to be bloated if it wishes to support all of those things, and with the advancement of W3C standards, the situation is only getting worse. when i say something is bloated, i mean it takes too many resources. as you quite well know, two programs doing the same things can use different ammounts of resources. in fact, the same program can do the same thing in two manners - one optimized and one not. the problem with bloatware is not the fact they try to do too much - but rather that they don't give a time to make sure they don't waste resources. i'll call upon your own pet to show that - you modified your algorithms and data structures several times, not to gain new funcitonality, but rather to make it run faster. you did it because you cared about its resource use (a CPU is a resource, too). the kde folks went over board with things, without caring if it runs on older hardware - hence, the bloat. when you have a new PC that runs very fast, you can loose awareness to how bloated your code is. i just bought a new computer a week ago, and suddenly things run fat, that i don't feel the bloat on every spot - suddenly netscape 6 launches quickly. suddenly galeon does not slag behind. so you see - if i was developing on this new PC, it would hardly run on older hardware, cause i wouldn't _feel_ the bloat. only if i care about it, or try it on older hardware, will i notice this bloat properly, and be reminded to keep my code optimized. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Alternatives to Mozilla
On Sun, 1 Dec 2002, voguemaster wrote: when i say something is bloated, i mean it takes too many resources. as you quite well know, two programs doing the same things can use different ammounts of resources. in fact, the same program can do the same thing in two manners - one optimized and one not. And what would you say were the minimum system requirements for Mozilla to run with (not too sluggish) ??? i don't know what are the minimum requirements, as i didn't run it on computers other then those that i have. on a 366Mhz AMD k6-2 it was much too slow. from what i understood from people's posts to this list, if you have a 600Mhz p-III, mozilla still doesn't run very fast - but at least its useable. I can tell you exactly how the computers in the CC at TAU are behaving with IE (the computers in the basement). The are all P2's I think, and most of them act sluggish as well with IE, so I wouldn't just come to Mozilla with complaints. there you go. the fact that IE runs slow on some hardware, does not mean that every other browser has to run as slow. if we looked up to misrosoft for comparing stuff, linux would have crashed every few days and we'd be filling fine. is that the kind of standard you're looking for? the fact that other browsers do manage to run much better then mozilla on a given hardware, suggest that mozilla contains bloat that it does not necessarily have to carry. this is bloat demonstrated in its purest form. when mozilla runs as fast as opera (for example), i won't say it has bloat. and don't sell me the argument that 'mozilla does much more'. when i view a given web site, the other features of mozilla do not 'run', so they should not slow it down. it runs slower, because it wasn't designed and implemented as good as opera was. plus, often you can add more functionality without paying anything in speed - if you do it right. if this was not the case, there would have been no meaning to the word 'optimization'. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Hack sought
On 3 Dec 2002, Oleg Goldshmidt wrote: [..snip..] 1) You'll go to userland for each packet, paying in performance. I don't see how you can send only packets with bad checksum across the border: if you could, you would have a simple solution for your problem, I guess. 2) Libpcap grabs a copy of the packet, so you need to configure your firewall (ipchains, I guess) to drop everything you grab, otherwise the original packet will live on regardless of what you do to the copy in userland. If you can rely upon the kernel to drop corrupted packets (can you?) you can send only those you have corrected back, ignoring the rest. actualy, netlink can be used to do both sides of the task. one can write a netlink program that will see all packets, find those with the broken CS field, fix only this field, and tell the kernel to move the packet onwards. all this - wihtout copying the full packet to user space - only its header. i don't remember the full details, but as far as i saw, it is do-able, and is not _that_ hard. this has an advantage of making sure the packets don't go out-of-order - while if you grab them using pcap, the packets _may_ go out-of-order - it shouldn't hurt functionality (after all, proper UDP-based applications need to be able ot handle out-of-order packets) - but it may hurt performance. using this solution, btw, will easily handle a sustained 10MBps link on a 500MHz p-3, even with full packet copying - and is likely to sustain a much higher bandwidth (especially that you don't copy all pakcets to user-space in this way). note: it could be that pcap uses BSD filters to filter the packets already inside the kernel, which might make it give better performance. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: makefile question
On Wed, 4 Dec 2002, Michael Sternberg wrote: Oleg Goldshmidt wrote: Use recursive make: ppc: $(MAKE) ARCH=PPC It does not work. Here is a simple makefile: ppc: @ $(MAKE) ARCH=PPC obviously, you did not read what oleg wrote - and modified his suggestion. the above rule should instead be changed to: ppc: @ $(MAKE) ARCH=PPC debug and the makefile should be invoked using: make ppc the other alternative (i forgot who wrote it) would also work (i.e. setting the variable on the command line when running make, rather then in the makefile). Makefiles should not be medled with. if you try to do something complicated - most likely you'll fail, cause the language that gmake (and the commercial 'make' variants) supports - are very very limited. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: makefile question
On Thu, 5 Dec 2002, Tzafrir Cohen wrote: On Thu, 5 Dec 2002, guy keren wrote: Makefiles should not be medled with. if you try to do something complicated - most likely you'll fail, cause the language that gmake (and the commercial 'make' variants) supports - are very very limited. A chalange, then, ha? not realy - gnu make's language realy _is_ very limited. every once in a while i try to enhance the Makefile of the project i work on (been using gnu make for such things since around 97 - so i thought i should be able to do what i want with it) - and eventually fall back to using cut and paste, external aliases and scripts and the like - because of the limits of its language, and the lack of a proper debugger (or proper error messages). (reading the gnu-make manual) so did i yet again before i wrote my previous post ;) Well, the best I could find was: target-spesific values There is one more special feature of target-specific variables: when you define a target-specific variable, that variable value is also in effect for all prerequisites of this target (unless those prerequisites override it with their own target-specific variable value). So, for example, a statement like this: prog : CFLAGS = -g prog : prog.o foo.o bar.o will set `CFLAGS' to `-g' in the command script for `prog', but it will also set `CFLAGS' to `-g' in the command scripts that create `prog.o', `foo.o', and `bar.o', and any command scripts which create their prerequisites. So how about: ppc: ARC=PPC ppc: release release: ifeq ($(ARCH),PPC) (is that the right syntax?) tzafrir, if you want a challenge - then _try_ what you write. it does not work and will not work, because of a simple reason - the given variable is being defined ONLY inside the commands of the rule. this 'ifeq' is NOT part of the commands for the rule - since it is a gnu make command. the 'commands of the rule', as i perceive the term now, are only the _shell_ commands of the rule. The problem is that 'release' has to be a pre-requirement of ppc. the problem is - this does not work as you think it does ;) But then-again, recursive makes are not much nicer. but at least they can be understood, and they do work. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
'per-process' DNS setup?
i'm trying to 'fake' DNS settings for a _single_ process on a linux box (i.e. when this process will query for the address of a given host, it will get a reply i set a-priory, while other processes will get the proper repely). i noticed there are various environment variables that allow overriding various settings for the resolver library, but they don't seem to allow enough control to actually override such settings, without changing the settings for the whole box. i've been searching on google, and looking for info in the o'rreily DNS and bind book - to no avail. if anyone has an idea if such a feature is available - i'd be glad for any pointer. it sounds odd for such a feature not to be already available via the default libraries. thanks, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
interoperable data formats (was: Re: sizeof(bool))
On Wed, 11 Dec 2002, Muli Ben-Yehuda wrote: And if you can, use XDR (External Data Representation). Standard is always good. http://www.faqs.org/rfcs/rfc1014.html. the problem with this specification is it is wasteful in network resources, because it uses a lot of padding for small data types. so one should consider whether they should go with the standard, or they have a reason not to use it (as as the neet to pass lots of data over a slow medium). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Blessed Religious Wars [Was: Mandrake 9.0 is fantastic]
On Fri, 13 Dec 2002, Eli Marmor wrote: PC Magazine, for example, has done a great job for years, with 22 annual comparisons, per year. One of those comparisons, repeated any year, compared all the printers that were announced that year (more than 100 PER YEAR). These comparisons covered almost anything you can imagine. They were very objective, although some of the competitors advertised in PC-Magazine. and because this kind of thing takes a lot of effort - they only did it for a more important products in their field. and they sell the magazine for money, to a lot of subscribers. perhaps then you are approaching the wrong crowd - you should approach the different linux magazines with this suggestion - hopeing they begin it now, and be able to grow the ammount of products tested, as their circulation (as they call it) increases. If it is a strategic decision for you and/or your company, maybe it will pay to do the research yourself. You have a very good head start cynic Let's go forward: if it's strategic, maybe it will pay to develop it from scratch! /cynic then don't be a perfectionist. there are only a _few_ stategic decisions a company needs to make. most other decisoins are simply tactical decisions, and can be corrected - _if_ the need arises. already. You know of a wide range of products that fit the general description. This is much more than most people start with. List your criteria, search for information about each product and try to grade them based on each criterion. Such comparative tables do help. If nothing else, it will clarify quite a few things for you, and will help you make informed decisions. You can't just harvest details from the Internet and build a check list; You should try all of the choices in order to get a decision. This is the only way to decide which of them is really the easiet. And which of them is really the fastest. But this forces you to download and install all of them learn all of them, etc. well, you don't really have to download and test all of them, not with a full-scale test. when i need something form the open source world, i first try what i know about - if it is good enough, i take it. for things i'm not aware of a 'category killer' for, i google, read about a few, cancel most on the bases of their APIs or config file formats, or so, try one or two, and then decide if one is good enough, or i'll write it on my own. i also sometimes just read about tools randomally, assuming that knowing them would probably come handy in the future. you might argue that with a better tool, i could do what i wanted in less time. but this time is time i already saved by only doing a rather shallow comparison. so there's a trade off here. if you waste too much time on the comparisons, you eventually lose time on doing what you actually want to do. and eventually, what you're measured with is what you bring to the customer in, at the end of the day. -- guy, who taught himself that perfectionism will only make him miserable. For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Compaq Evo N1015v
On 15 Dec 2002, Gilad Ben-Yossef wrote: Thank you very much to all that replied. I was already aware of the possibility to install from HD , floppies or network. Unfortunately, the RedHaT install mailing list post I quoted indicates that all of these have been tried and failed because the problem seems to lie with the IDE chip set - e.g. no HD, floppy etc will work. I was more hoping for someone that might have gotten the IDE chip set to work but it looks I will have to do it myself. question: which kernel exactly is used during the installation process? i presume this is NOT the kernel found in redhat's updates? cause the updates redhat add seem to also include support for newer hardware. i guess it should be possible to somehow take their installation floppy, replace its kernel with the updated kernel, and try again. ofcourse, checking if the updated kernel supports the IDE chipset beforehand would be a good idea ;) but then again - you probably already thought about trying this route... and btw, its your fault that you didn't state you already tried those ideas ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: CD music
On Mon, 16 Dec 2002, shlomo solomon wrote: I played around for a while with the KsCD setup, but nothing helped (/dev/cdrom, dev/scd1, dev/cdrom2, etc). tell it to work directly with the relevant /dev/hd* device (/dev/hdc? /dev/hdd?). when you run the ide-scsi emulation, the software that sets things up tends to also replace the links of /dev/cdrom* to point to the matching scsi device - while for audio CD playing you need to accedd the IDE device. happened to me too, while doing the same thing. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: strange file permission problem
On Thu, 19 Dec 2002, Arie Folger wrote: On Thursday 19 December 2002 11:19, shlomo solomon wrote: I have a strange problem. Actually, I've solved it, but I don't like the solution and I don't like not knowing what's causing it. So maybe someone can help. snip The problem is that every so often (I don't know when it happens), the permission becomes 600 and non-root users can no longer read the file. There are also some gz files in the /var/log/mylogs directory (created by logrotate). The same thing happens to their permissions too. My solution was simple - run a cron job to reset the permissions for all files in the directory to 644. But, although that works, it seems strange that **something** is changing the permissions back to 600. PAM does that. Read the various settings under /etc/pam.d I had a similar problem with certain device files, until I discovered that the settings were reset (as part of devfs, IIRC) and would best be set using pam.d settings. pam changes permissions for device files - NOT for log files - at least not the way it is ocnfigured on redhat systems - and it has no reason to touch any permissions related to log files at all. it messes with device files, in order to allow the user that logged on via the console, to gain access to certain 'console-based' hardware (e.g. sound device via /etc/dsp). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: counting emails and pine questions
On Sun, 29 Dec 2002, Ishai Parasol wrote: Another question. Is it possible to run pine with another user email file (/var/spool/mail/USER) ? for example, I'm logged as root and want to read USER2 emails (ofcourse, USER2 can't login, he's only a mail account and not a full user). If it's impossible with pine, is there a client (not for X) that can do it ? pine -f ../../../../../var/spool/mail/USER2 pine's -f flag interprets its argument relative to the user's 'Mail' (or 'mail') directory - hence the need to use these '../' things. note that if you use '..' from the '/' directory, you remain in the root directory - so you can always use more '../' things, without getting errors. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: URL with memory management explained
On Wed, 25 Dec 2002, Michael Sternberg wrote: I'm looking for a good URL that explains memory management in modern Linux kernel. I.e. what happens when I'm typing malloc, how many memory is assigned to a process, what is virtual memory, how swapping works, etc... ok. this was the last push i needed ;) a new tutorial, titled Unix and C/C++ runtime memory management for programmers tutorial was just released on the LUPG site (http://www.actcom.co.il/~choo/lupg/). the tutorial actually tries to explain what happens when you're using malloc(), what is virtual memory, etc. took me over a year to eventually complete it. comments and fixes are welcome, as usual. the tutorial's direct url is http://www.actcom.co.il/~choo/lupg/tutorials/unix-memory/unix-memory.html hope this helps, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
kernel versions (was: Re: Kernel Build Problem - Solved)
(note: i'm going to repeat here things that were discussed several times in the past - just to avoid another similar loop in the next few weeks (no doubt it'll rise again in a few month, as it usually does ;) ):) read below: On 30 Dec 2002, Aharon Schkolnik wrote: Oleg == Oleg Goldshmidt [EMAIL PROTECTED] writes: Oleg Aharon Schkolnik [EMAIL PROTECTED] writes: Not only Oleg that, I could not even find anything related in the code Oleg (2.4.20 though). Out of curiosity, where did you find it? Google search Oleg Oh, it's a patch... ;-) Wouldn't it be more correct to say that it WAS a patch which has since been incorporated into the mainstream kernel code ? i think people keep forgetting (or don't know) somehting fundamental about kernels and distributions. there is the 'vanilla' kernel, which linus handles (for the 2.5, i.e. current development version) or marcelo what's-his-name keeps (the 2.4 version - linus moves maintenance of stable non-development kernels ot other people. i'd imagine alan cox still maintains the 2.2 kernel version). now, the major distributions usually take some semi-stable vanilla kernel, and add 10s of patches to it, that float around the internet, and that were not incorporated into the vanilla kernel. in redhat specifically, if you open the RPM of the kernel sources, you'll see they took soem vanilla kernel, and added more then a hundread patches to it - patches they write, or they collect from other sources. why those patches are not in the vanilla kernel? this could be either due to policy, or due to timing (you can add too many patches to a given kernel revision, and expect it to remain stable without a given cycle of 'testing' - and vanilla kernels don't go through format testing). thus, when you have a problem with a kernel from your distibution (and usually, something with 2.4.18-19.8.7 - i.e. a set of 3 numbers after hte regular set of numbers, implies a distribution-specific kernel version) - you check your distribution for a newer kernel. also note that bugs occur. and bugs exist. and bugs get added. don't be surprised to stumble upon bugs, especially not in software that has thousands of different configurations in which it may be compiled. the kernel dependencies system is not complete and considered a half-hack (at least the one that comes with the 2.4 kernels - no idea what eventually happened with the 2.5 kernel source tree) - and thus it cannot properly catch all dependencies in the kernel. btw, in your specific case - you might have been able ot ignore the errors you got. they only suggested that 2 of your modules oculd not be loaded. this would matter only if you actually needed those modules. if you never need to use them, you don't realy care that they try to use a feature that's missing from the kernel. i had systems with many similar problems (e.g. because i mistakenly compiled in PCMCIA support on a desktop computer, and i didn't include some other feature that was missing) - and the machine had no problem working (just that the error messages were annoying). If it has been incorporated into the kernel, shouldn't it be documented ? If it doesn't work as a module, shouldn't it be impossible to build it as a module ? yes, except that the kenel build system is not perfect, and it appears that the kernel grew a bit large for it to handle properly. Seems to me that someone slipped up here. there are many such slips all the time. don't be so surprised that you stumbled over one of them. eventually, if there were no bugs in software, we all would be millionairs (or loose our jobs). and remember, its not wierd - its a bug. report it, or work around it - just don't sound so surprised. and don't think but i didn't do anything special - beause you did - you just tried out a configuration for this specific kernel, that redhat's people haven't tried before ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RE: Strange cout behaviour
On Wed, 8 Jan 2003, Boulgakov Andrei wrote: It doesn't matter __FILE__ : when I change it to Hello, World it doesn't print on speedy machine and print on slow machine! Something with time slice? someone said here something about the some shells 'eating' out the last printed line, if it has no new-line at its end. try running this program under different shells. for example, run 'tcsh', and then run the program, and see what you get. do the same with 'bash'. see what you come up with. also, your prompt might cause the shell to overwrite the string, if it contains a control character that makes the shell print backspaces (or a 'move to line's front' symbol). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Cable-Modem Mini-Howto for Israeli Linux Users
On Thu, 9 Jan 2003, Alex Chudnovsky wrote: On Thursday 09 January 2003 00:55, Amit Margalit wrote: Can you tell me what the 'VPN server' is? Probably something like matav.inter.net.il or something? I'll add this to the cable-modem mini-howto. Do you mind if I answer instead? VPN server is a server you run PPTP connection against. For example : pptp cable.netvision.net.il - cable.netvision.net.il is a VPN server. i think you were answering someone else's question. amit didn't ask what is a 'VPN server'. he asked what is the specific name of the 'VPN server' used when connecting to matav. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: 99.6% idle 5.16 load
On Sun, 12 Jan 2003, Nadav Har'El wrote: If you have load 5 but no process is using CPU time, it is most likely that you have 5 processed in the D (uninterruptable sleep) state. Run ps aux and look for a D in the STAT column to confirm this hunch. this sounds odd - a process in the 'D' state is in an uninterruptible wait on a resource. thus, it is not supposed to be in the 'ready' (or 'run') queue, and hence shouldn't be counted inside the 'load average' (which counts the number of processes in the 'ready' queue or the 'running' state, in the last X minutes). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: NEWS: Cables configuration
On Sat, 18 Jan 2003, Official Flamer/Cabal NON-Leader wrote: Quoth Eran Levy: Yes I have DNS resolution problem. Im not uses my own DNS server. Im using Netvision's DNS server only. here are the files: /etc/resolv.conf: nameserver 194.90.1.5 add a domain foo.com or search bar.net to this file. doesn't sound relevant to the problem. the 'search' or 'domain' entries are used just to optimize the queries - not to make them 'correct'. this addition simply allows partial domain names lookup to work (i.e. searching for 'foo', which is under the 'bar.net' domain, will work if the 'search bar.net' entry is found in resolv.conf). thus, i presume this is not going to solve the problem (unless the user was trying to look up such partial names, and expected this to work without the 'domain' or 'search' entries). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: NEWS: Cables configuration
On Sun, 19 Jan 2003, Dani Arbel wrote: Hi! The setup looks ok. I'd remove nis from nsswitch.conf , since I believe Eran does not have a Linux farm ... so I'd change the hosts line in that file to be: hosts: files dns Anyway, I think it is not the problem. Start ethereal to capture traffic from eth0 and then try to do some kind of name resolution (nslookup, host etc) could it be that he has a firewall setup that blocks his DNS requests? they go over UDP connect5ions, and if one blocks all UDP ports, one gets stuck. eran? any firewall set up there? you can also try to make DNS queries directly to the DNS server, using the 'host' command. something like: host ftp.netvision.net.il 194.90.1.5 'host' will query the given name server (194.90.1.5) directly for the details of 'ftp.netvision.net.il'. if this works, it means no firewall problems. if this does not work - i'd suggest making sure the firewall is (temporarily) disabled (if indeed it is running at all) by 'iptables -F', and then running this check again. please note that in current distributions, one chooses the firewalling level during system installation, and you _might_ have accidentally set it up, and a mode that blocks your DNS queries (or the DNS replies). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: /usr/bin/top for kernel processes
On Mon, 20 Jan 2003, Michael Sternberg wrote: Hello. From time to time I notice in output of /usr/bin/top that my system CPU time jumps to nearly 90%. How can I know which task kernel performs ? I'm looking for some kind of top for kernel.. Maybe some file in /proc directory ? please note that 'system' time will include the time spent by user-space processes, running in kernel mode. for example, if a process does a lot of disk I/O, then you'll see it is spending a lot of time in 'system' mode, and thus 'top' will show the 'system' time as non-zero. the same goes for other types of heavy I/O (networking, swapping, interrupt handling). without too much hasteling, you could use a few standard utilities ot help you debug problems. there is 'vmstat' (run it as 'vmstat 3' to refresh every 3 seconds, and ignore the first line of output. read the man page for more info). there is 'iostat' (similar, but concentrates on device I/O, like disks). you might also have 'sar'. in order to actually decipher the output of these programs, you should look at what they show you wihle the system is in 'silent' mode, and then compare that to what they show under various types of activities. then compare to other systems - otherwise, you'll have no idea what 'high ammount of context switches' is, or what ammount of interrupts per second makes the mahcine too slow... you need to know the capacity of your machine, in order to know if its saturated regarding activity of a given resource, or not (i.e. if its getting so many interrupts per second, that this comes on account of its ability to run user-space programs, etc). i don't suppose this will actually help you in the short run, but perhaps it will in the long run. if you manage to find a resource that gives 'proper numbers' for various types of hardware - let me know, please ;) will help me quite alot. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: People n' computers interest groups - help wanted
On Sat, 25 Jan 2003, Shachar Shemesh wrote: Second, I am looking for content for the programming interest group. I have a suggestion for a lecture about Perl, and another about kernel programming. I'm looking for more people interested in sharing their insights. i presume these interest groups are going to have their meetings instead of the non-existing meetings of linux-il (not including various 'special events')? by the way - where are these meetings going to take place (i assume somewhere in gush dan)? at what intervals (every other week? every month? every two month?) this info is important if people from far areas are going to get there 'regularly'. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: OT: Net Loss of $100000000000 to AOL for 2002
On 30 Jan 2003, Oleg Goldshmidt wrote: Eli Marmor [EMAIL PROTECTED] writes: Not yet slashdotted: AOL removed $45.5 billion in the last quarter. Total loss in 2002 was $99 billion. A new world record. Well, IIRC in September 2000 Intel lost $91 billion in one day... you surely realize these are not losses 'in one day'. the stock value of the company (or of its holdings) dropped over a period of time, and in their reports for a given quarter, they decided to adjust their 'reported money worth' to (part of) this drop in stock prices. thus, this is loss just 'on paper', much like it was a gain just 'on paper' when the stock prices shot to the skies. ofcourse, companies often use this opportunity to hide their real losses (most likely a few hundreads of millions of dollars, or a few 10s of millions) - but these numbers are quite small relative to the 'net worth' loss of the company, so nobody looks at them when they can shout 'hei, they lost billions of dollars the last quarter!'. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: OT: Net Loss of $100000000000 to AOL for 2002
(this will be my last email to the mailing list on the subject, due to its off-topicness.) On 30 Jan 2003, Oleg Goldshmidt wrote: guy keren [EMAIL PROTECTED] writes: On 30 Jan 2003, Oleg Goldshmidt wrote: Well, IIRC in September 2000 Intel lost $91 billion in one day... you surely realize these are not losses 'in one day'. the stock value of the company (or of its holdings) dropped over a period of time, and in their reports for a given quarter, they decided to adjust their 'reported money worth' to (part of) this drop in stock prices. thus, this is loss just 'on paper', much like it was a gain just 'on paper' when the stock prices shot to the skies. It was a loss in one trading day. If you were to buy (the whole of) Intel in the morning and sell it in the evening you would have lost 91 billion dollars (wrong, of course - you would be a market maker for Intel then). Think what you may of stock prices, they reflect what the markets think of how much a company is worth, which is just about the only half-reasonable notion of value there is. I did not quite understand the they decided in what you wrote - they do not decide what their stock is worth. obviously, we're talking about two different things. you're talking about 'market price' of a company. when this price drops, the company does not loose money - its share holders do. i was talking about the 'value assesment' of a company's holdings, as written in its books. this is something that is in control by the management of the company - not by the stock market. if company A holds part of company B, and company B's price in the stock market dropped - then company A does not have to change the value it attaches to its share of company B, if they think this change is 'temporary'. and when the price of a stock falls or jumps on any given day, this is usually deemed as temporary by most managements. furthermore, when a company is sold or merged, usually a 'value assesment' is being ordered from a financial consultant by the buyers and by the sellers - and it often does not match the stock price. Actually, IIRC, in that particular case (Intel again) there were no real losses in your definition. On that day (or maybe the day before) Intel reported that their European sales grew by a few per cent in the previous quarter. The problem was, the world was expecting higher growth. That expectation was of course reflected in INTC stock price (if you look at the graph, there was a sharp rise before that drop). This is identical to the situation with AOL/TW, in principle, but on a much shorter time scale. when i refer to losses - i refer to what the company writes in its quarterly and yearly reports to the stock market - the figures there got little to do with the stock price of the company. for example, when lucent lost almost 50$ billion in one quarter during 2000, it was mostly because they decided to 'write off' a lot of 'virtual incomes' (i.e. sales that were written on paper, for which they got no money, and which they financed with their own money). the loss happened a while before that already - its just then when they finally decided to 'write it off' in their formal report. its not that they shelled out money to someone else, and its also not that their stock price droped by that much - they only droped by that much _after_ they sent out these reports. so as you see - we're talking about different issues. btw, in israel, there is an attempt now to pass a decision that will force public copanies to value their holdings in other companies based on the stock price (its called Teken 15 by the people dealing with this) in order to avoid situations such as when Kur wrote E.C.I's value in its books, as being about 4 times their value in the stock market, after the change was not realy temporary. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Internet Gold Tech Support - IMAP
On Fri, 31 Jan 2003, Ilya Konstantinov wrote: On Friday 31 January 2003 16:22, you wrote: How can they? In pop3 you can * get a message * delete a message keep mail on server basically means that the client does not order the sever to delete those messages. The server can delete the messages the user fetched anyway, even if not directed to by a DELETE command. It might violate some RFC, but real life (the large number of users who simply mark 'Keep mail on server' without thinking of the consequences) requires it this will cause user support problems from the other side - a user connected, started downloading, got disconnected, and lost all their email. (i.e. it is not transaction-based, and handling mail in a non-transactional manner is a bad idea). you might thus say 'we'll delete all messages only if we got the 'quit' command from the client' - but this is not a good idea as well - what if the client only downloaded some of the messages? so you need to keep track of which messages from the mailbox were actually downloaded by the user. -- otherwise, lots of users' mailboxes would quickly grow to enormous sizes. The ISP can enforce mailbox size limits to solve this, but this way they'll hurt legitimate users (who expect a one-time large delivery or leave for a 2 week vacation) and increase complaints from the users who lightheartedly enabled the 'Keep mail on server' option. actually, i don't remember ever seeing a problem with using 'keep mail on server' to-date - except for configuration problems on the users' side. can you name an ISP that added this odd feature to their pop server? or rather, a pop server that supports this feature? i did see 'mail box over quota' messages several times. most ISPs limit the mail box to a given number. if a user goes on vacation, then ask to enlarge their quota (for a fee) for the duration. some ISPs give you different 'soft' and 'hard' quotas, so you can have a larger mail box temporarily. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: KCrorn
On Sat, 8 Feb 2003, Amichai Rotman wrote: Hi Clan, I am trying to run a cron job to use the record command as follows: KDEDIR/bin/konsole -e /usr/bin/record -cv -t 00:15 why are you trying to run an X application from within cron? just tell it to run the 'record' application directly, without the konsole, and see what it does. ofcourse, make sure 'record' records into a file, and a different file at each run (unless you want each run to override the previous run). note that yo need to make sure that the user running 'record' has access to the /dev/dsp device (the sound device) no matter who is logged into the console, and even if no one is logged into the console. by default, distributions tend to set the ownership of /dev/dsp to the user that's logged into the console, or to 'root' when no one is logged into the console. you'll probably have to either change the group of the device and add the user with the cron job to this group, or something else. note also that you'll have problems with this because KDE 'takes over' the /dev/dsp device, when it is running, and at that time, 'record' will probably fail opening /dev/dsp - and will need to be run using the kde program that simulates /dev/dsp for non-kde applications (i forgot this program's name). note that if your purpose is more limited, you might be able to get away with doing only a small part of this setup - just remember that the environment of a process running under 'crond', is not the same as that of processes you run from within your console. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: samsung PPPoE instructions?
On Sun, 16 Feb 2003, Hetz Ben Hamo wrote: You can also find instructions there. After the package is installed (along with pppd of course), run: adsl-setup And answer the questions. To connect do: adsl-start OK. The modem's IP address is 10.0.0.2. And the NIC IP is? I didn't manage to get it connected. I tried with @OBezeq, with a real user/pass - but the authentication always failed. I tried with different NIC (should the NIC have any IP or just configure it to connect to a some DHCP?) as i wrote a few days ago - my NIC has no IP address (i disabled it from being initialized during boot, accoridng to the instructions on the roaring penguin's site) and the connection works. when i tried setting the connections, things started getting messy, so i re-disabled it. note that i didn't use their 'adsl-setup' script - i followed their manual configuration instructions, and i disabled their software from modifying /etc/resolv.conf - i like it static and manually configured. you might want to follow that path as well and see where it leads you to. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: policy routing question
On Fri, 14 Feb 2003, Stanislav Malyshev wrote: I have rather strange problem with routing on Linux. The host in question is 2.2.19. It is connected to Frame Relay and ADSL (eth0 and ppp0 interfaces, accordingly). The intranet is on eth1, all connections outside are masqueraded. What I want to do is to make requests to port 80 go to ADSL and all other things got to FR. i did something similar once, with a 2.2.X kernel, and it took several hours to get it done (a more complex setup, unfortionately, which involved 2 linux machines). there is no sane way to debug this - you must be very persistent, and stare at the code as your only debugging method ;) Manual suggest following setup: ipchains rule along the lines: ipchains -A input -s 10.0.0.0/8 80 -i eth1 -m 1 10.0.0.0/8 and eth1 being the intranet addresses and interface, and then iproute setup: ip ru add fwmark 1 lookup adsl ip ro add default via PPP-host dev ppp0 and where is the definition of your 'adsl' routing table? there is no 'adsl routing table' by default. perhaps you forgot to add 'adsl' somewhere in the 'ip ro add' command? something like: ip ro add default table adsl via PPP-host dev ppp0 ?? The problem is that the setup doesn't work, and in a very weird way. The packet from inside gets out through ppp0, as intended, ofcourse it does - after all, your default route for the default routing table is set to go via ppp0 - thus, _All_ your traffic will be going out via that interface. this, while the incoming data is defined to go according to routing table 'adsl', which was not defined. hope this helps, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: top level project directory
On Wed, 26 Feb 2003, Muli Ben-Yehuda wrote: Let's say that I have a complex project, using many Makefiles. In some of them, I'd like to refer to other directories of the project. So far, I've been using various combinations on ../, ../../, etc, but that's awfully brittle when you start moving directories around (not to mention potentially dangerous). Is there a way, which does NOT involve setting an environment variable, to refer to the top level project directory? Specifically, I have a Rules.make in that directory, which is included from the other Makefiles. If I could make the Rules.make contain the top level directory in a variable, it would solve most of my problems. you cannot have that, unless you force all 'make' invocations to be run from that top-level directory. here are a few things to try - non seems to be too satisfactory: 1. make your project 'shallow' (i.e. all source directories are located under a single 'src' directory, no hierarchies). that's what i use in all my projects (small and large alike), together with '-I..' and using directoy names in includes (i.e. #include dir/file.h). this kind of structure also helps avoid most of the directory moving - only files moving between directories need more work. ofcourse, if your project inherits from the linux kernel itself, you won't dare go via that path. 2. having a symlink from each directory to the top-level directory (i.e., in directory 'top_dir/subsys1/unit1/': ln -s ../../.. top_dir_link then use this link to access the top-level rules file. this won't help alot with directory renaming, though. 3. avoid moving directories. think a bit further before you create a directory - in case you have a reason to beleive it might turn out more 'generic' then you first think it is - make it into that generic directory, and put it in the right place from the start. Now that I think of it, another way would be to change the build structure from recursive builds to a centralized build - all directories are built from the top level directory. Let's say that I do want to use recrusive build, though. What trick am I missing? none. all possible tricks have problems of various sorts. btw, why are you so against using an environment variable? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: webcam driver
On Sat, 22 Feb 2003, Alon Altman wrote: Hi, Anyone know of a linux driver for a USB webcam called twinklecam? Windows identifies it as CIF USB CAMERA 2110. Google search returns nothing of use. when i bought a camera that wasn't supported under linux (eventually i found an alpha-grade driver that was just in the beginning of being written, and did nothing) - i passed the camera to a windows machine, and used what i learned form the attempt of installing a non-working camera - to buy one that works. (the non-working was Creative Webcam PD1001 [the PD1001 is written in small letters on a sticker attached to the camera - nowhere in the documentation]. the working was Creative Webcam-5, which i found in atid machshevim in chorev center - althought in order to use its more advanced features (apprently a larger image size capture) require installing a binary-only module, which i didn't do yet because it does not match my current kernel). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SOLVED: Slow Linux response during disk operations (was: Testingon various computers needed)
On Mon, 16 Jun 2003, Yedidyah Bar-David wrote: Sorry for not reading it (apparently wasn't here, maybe hackers-il, but google doesn't find it), but what I usually do is kill -STOP all of them, and only then kill -SOMETHINGTERMINAL. Unless the user intended to abuse the system, and wrote a signal handler for STOP, it would work well (and does in practice, with all the OS course students I have here). a user cannot write a signal handler for the 'STOP' signal. this one, and the 'KILL' signal, are un-stopable in a user application. by the way, you forgot to mention that you kill -STOP the process group, rather then the processes directly one by one. i still remember the note regarding this that you (or someone else?) placed in bar ilan university's terminals farm, some 10+ years ago ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Reverse ftp over ssh does send files
On 17 Jul 2003, Micha Feigin wrote: I try to connect to a remote computer using ssh and then tunnel the ftp connection back to by computer using ssh -R 1234:local machine:21 ... why do you expect to be able to tunnet 'ftp' like that? ftp sends only commands via port 21. data is sent via a seperate connection (data is both the output of 'ls', and files you transfer with 'get' or 'put'). it looks like you _might_ be able to do what you wanted, _if_ your could force the 'data' port to always be the same port on the remote machine, and then tunnel that port too via ssh. if this is possible, perhaps someone on the list can show us how to do that. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Dial up (ppp) connecion to ActCom fails to get out to theInternet.
On Sat, 9 Aug 2003, Shaul Karl wrote: Can you advise what to do with the following? $ ping -c1 192.115.23.205 PING 192.115.23.205 (192.115.23.205) 56(84) bytes of data. never check connectivity without the '-n' flag to ping (this disables reverse name lookup). otherwise, you're likely confusing connectivity problems with name lookup problems. and second - what is customer support for? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: gnome vs. kde wars not over?
On Fri, 15 Aug 2003, Diego Iastrubni wrote: Distrbing or what? what disturbes me, is that you bothered sending this to the mailing list. this is rumors pushing at it's best. that guy writing this is not acting like a journalist in any manner - so taking what he writes as a possible truth, is quite like throwing a dice. This article says that companies are feeding anti-KDE articles to news sites from supposedly neutral sources. http://www.pclinuxonline.com/modules.php?name=Newsfile=articlesid=7488 if you read the links at the bottom, and the replies and replies to replies, you will probably see what this guy did - someone told him something, he went on spreading the rumor, and then he wrote that he was all wrong. but he did more - he wrote 2 things in the same letter. one was denied. for the other - i think he didn't even contact anyone (redhat? sun?) to ask for their stand. so pull yourself together, diego - don't spread rumors and drive ;) anyway, if it'll make you happy, i could write a post on linux-il, claiming that a reliable source 'close to a leading linux vendor' claims that this vendor spreads anti-gnome articles to known websites, claiming these articles come from 'neutral' sources. would that make you feel the situation is back to normal, with everyone bashing each other equaly? :) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kernel-image
On Sun, 17 Aug 2003, Amir Tal wrote: debian sid, dell inspiron 4000 laptop. installed kernel-image-2.6.0-test2-1 with apt-get. installation went fine. took a look at lilo.conf after installing, looks ok to me. when booting the new kernel i get a kernel panic, and a message that tells meto append a correct root= option in lilo.conf. running lilo returnes to errors, and the old image (who uses the same is that 'to' or 'not'? if it's 'to' (which is odd, anyway) - what are the errors? parameter from lilo, root=/dev/hda3, boots just fine. images are in the right place in /boot with the correct names, and links from / are ok and matching entries in lilo.conf. (note: the following is written in a didactic tone - not in a scorning, angry or you don't know anything tone - so please read it that way) here you are, hiding important information from the public, and yet expect ideas to come. someone here asked about debugging lately - the first rule of debugging is - never make assumptions about what works. the second rule of debugging is - never make assumptions about what works. the third rule of debugging is - hmmm... well, i think i shouldn't mention that third rule - you wouldn't like it anyway. anyway, you should not tell us where there are no problems. if you gave up on finding the problem yourself and you refer to the list for help - you supply _all_ the information, and you don't force your own assumptions on us - one of them is _probably_ wrong, or you would have solved the problem already. so, mind writing again, and this time, without information hiding and data encapsulation? we want the private members of amir's problem to be made public ;) here are some things you've hidden: - what does it mean that 'installation went fine'? does it mean that you saw no errors, or that the installation also has a 'test' optoin, that performs post-installation tests which report that the installation meets all post-install conditions? - what is the contents in lilo.conf that looks ok to you ? - _what_ kind of kernel panic message do you get? - what is the _exact_ message? - what are the 'ok' paths and links? - did you read the post shaul karl sent here about installing kernel 2.6? what did you do about it? (by the way, i'm not expecting a reply to the first item - but i do expect a reply regarding the rest of them). waiting for your next post, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Cost of qt development license
On Sun, 31 Aug 2003, Aviram Jenik wrote: We are currently debating on what GUI infrastructure to use for one of our products, and the main downside of qt seems to be its constraining license. Can anyone shed more light on this subject? How much does the qt license cost to develop a non-GPL product that uses qt library? Does it also apply for tools like kdevelop? Are there any free (read: completely free, with no we're-free-but fine print) alternatives other than gtk and WxWindows? you didn't state what kind of application you need to develope - which, i think, is quite relevant to the choice of toolkit. i had the opportunity to develope some linux GUI application for some company, which needed to show both simple windows (canvas display, dialog-boxes, tables) and some time-based graphs - and was supposed to run on linux, with no real portability requirements. i went for gtk+ (programming in C, mind you) because i could use it for free, and i could find the proper extentions to show the time-based graphs for it. eventually, we managed to come up with a GUI that looked mostly decent (it wasn't an application meant for end-users, althought it was used in a trade show in the USA, to demonstrate the hardware that company developed). the down-sides i stumbled on were: 1. documentation. gtk+'s reference manual (that was 1-1.5 years ago, and we used gtk1.2, when gtk2.0 was just released) was very 'empty' (i.e. each function had an enrty, but often with no text attached to it). i used some experience i had with gtk+ previously (on a hobbist level) and lots of looking in the header files, and guessing, to overcome this issue. 2. the time-based graph widget i used contained a fundamental bug. ofcourse, since it was open-source, i was able to debug it and fix the bug. yet, i had to spend about 2-2.5 days on decyphering the code ;) 3. the timers of glib/gtk do not account for drifting, which caused the time-based graphs to come out with shifts. i had to re-write the timers code several times, eventually implementing my own delta-list on top of gtk's basic timer, to overcome the problem, until i managed to get the graphs to display properly, with a normal deviation of about 50ms per dot (we drew 1 point per second). 4. gtk's GUI builder, glade, has two modes. one is 'code generation' (un-useable in a proper project, since it re-writes the same code you modify). the other is using libglade (less convinient due to the way it handles callbacks - but useable). glade itself has some quirks which require getting used to, but if you don't try to play with it too much, and use it systematically, it is useable. the libglade mode also made it easy to write code that dynamically loads windows on-the-fly - which is a feature we needed for part of the application. overall, the time we spent on the overcoming gtk+'s shortcomings, had cost (in a rough estimation) less (money-wise) then buying a Qt license (and Qt did not have a time-based plotting widget - i found one freely available such widget on the net, that i did not like) - and i don't think that using Qt would make the development time shorter then it was - when you're using some 3rd-party library - any library - for a less-then-trivial project, you're bound to hit the library's rough points. in general, it _looks_ like the greatest trade-off between open-source and closed-source code, is having the source and not having to pay royaltees/licenses, against having good documentation and having no source (well, with Qt you have the full source, which is useful for debugging your code - that alleviates a lot of the problem). i find the same trade-off exists with other libraries (e.g. ACE Vs. Roguewave's commercial libraries) - and which of those suites you, depends alot on your personality. after developing using commercial tools for ~4.5 years, and free software for 1.5 years - i know which side i want to be on - but that's me. if you'll have an intention of using gtk+ (and judging by your letter - you currently don't) and would like some more details - feel free to mail me. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Cost of qt development license
On Mon, 1 Sep 2003, Aviram Jenik wrote: The application is very simple in terms of GUI, and heavy on internal logic - so the GUI is just a few buttons/edit boxes/progress bars, etc. Nothing too complicated. Cross platform is also not an issue: this GUI will be Linux only. in that case, and assuming this application is not for internal use, i'll avoid using Qt, and just use the toolkit that looks most appealing, visually-wise. by the way, if your application might be multi-threaded, you should take that into account. assuming the GUI library is not thread-safe, you'll need to make sure you perform all GUI operations from within one thread only - this requires some 'thread-to-thread' delegation mechanism - not hard to implement, but requires _some_ time. by the way, gilad - did you personally use fltk for a real (non-trivial) application? if so, can you comment on that? technically-wise, it looks like i should dump gtk+ in favor of fltk - if i find the widgets visually appealing ;) - but i want to know, form someone who actually used it, how their experience with it was - and without information hiding :) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: VI(M) - Changing Merging 2 lines?
On Mon, 1 Sep 2003, Subba Rao wrote: Hello, I have a large text file where 2 lines (fields) form a record.Now I want to merge the 2 lines into one line seperated by a comma. Line1 Line2 The fields should be seperated by a comma. Line1,Line2 How can I define a keystrokethat will, add a comma at the end of Line1 perform a JOIN of Line1 and Line2 move the cursor to the next record? Any help appreciated. if this is a large file - isn't it more appealing to do this with a 2 lines perl/python/whatever script? or does the file contain more data then just the lines you described here? --guy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Cost of qt development license
On Wed, 3 Sep 2003 [EMAIL PROTECTED] wrote: guy keren wrote: that into account. assuming the GUI library is not thread-safe, you'll need to make sure you perform all GUI operations from within one thread only - this requires some 'thread-to-thread' delegation mechanism - not hard to implement, but requires _some_ time. Why is 'thread-to-thread' delegation required? What's wrong with Mutexes at the right spots, so only one thread gets to run the critical sections at any point in time? because normally, the GUI library runs inside a 'main loop' that waits for events and handles them. so you'll dedicate one thread to run this 'main loop'. and since this thread might wake up whenever, in order to handle a new event, and this thread cannot wait on a mutex (cause no one told it to) - so you'll not want to start messing with the GUI library from within another thread. what you'll normally do (again, all this - iff the GUI library isn't thread safe) is that other threads that want to update the GUI will delegate the operatoin to the main thread (e.g. writing a 'command' on a GUI commands queue, and waking this thread by writing a byte into a pipe, and having the GUI thread add this pipe's file descriptor to the set of file descriptors it select(2)-s on. (all GUI library i saw have a method to ask the main loop to add file descriptors to its select()/poll() loop). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [OT] Digital Cameras and Linux
On 11 Sep 2003, Oleg Goldshmidt wrote: Geoffrey S. Mendelson [EMAIL PROTECTED] writes: I found that the best way to use a digital camera is to treat the memory cards as film. You buy several of them acording to your needs and replace one when it gets full. What's wrong withdumping files onto a hard disk from time to time? You can re-use the memory, can't you? i think geoff is talking about a completely different scale of cameras/prices, so most of his advice is probably not relevant for you. Canon just anounced a $1200 (US list price) EOS-300D. which has lots of features but a CMOS or CCD sensor (I think Canon prefers CMOS to CCD, but I'm not sure). I would very much prefer something a few times cheaper... how much cheaper? there are cameras for any price-range you'll define. the one i bought, i did via wallashop's group sale for an HP 850 (which has an optical zoom of X8, which seems to be missing from most cheap cameras), and it cost 1900 NIS including shipping, about a month ago. this includes a (very small) 16MB SD memory card. i had to buy rechargable batteries for it, and will hopefully buy a larger card, too (for few hundeads NIS). this camera supports both usb-storage (which works simply under linux - just mount it) and some photo exchange protocol that _appears_ to be something standard, thought i didn't try using it. when you buy a camera, do note: 1. how large is the _optical_ zoom (a digital zoom, as far as i understood, is generaly quite pointless). 2. what kind of lens it has (if you know anything about lens types - i don't ;) ). 3. how much control you have over taking pictures (things like shutter speed, and the rest of the buzz-words photographers use) 4. how much control you have about the number of pixels it uses (e.g. in the HP 850, you can either do 4M pixels, or 1M pixels - not a number in between, which is quite annoying). and indeed stick to USB, and forget about serial - i think it'll also mean a faster transfer of pictures from the camera to the computer. not to mention the fact that a USB camera can be easily used as a 'floppy' (making the use of a seperate disk-on-key redundant). well, that's about all i know about digital cameras ;) -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux Real Time OS.
On Tue, 16 Sep 2003, Josh Roden wrote: Can anyone advise us on which Linux Real Time distro to use + any tools etc... that could help us. Here at the Hadassah College our Virtual Reality department wants to give its students the opportunity to do their final project in the RT field. To this end we will have to build a RT computer environment. since no one else responded here yet, i'll write what little i know about this subject. the only 'real-time' linux system i have put my hands on, was RT-Linux (from FSM labs - www.fsmlabs.com, and more specifically - http://www.fsmlabs.com/products/openrtlinux/). it used to come as a patch to the kernel (it required very specific kernels back then) and a set of libraries. basically it runs linux as the lowest-level task of the rt-linux kernel, and your write the real-time stuff using their API (which is a subset of POSIX) which runs in non-protected mode... it took less then a day to install it on the first time and getting their example programs running. and indeed, they seem to respond to interrupts in the time-frame they claim (tested on a non-loaded machine, though). there are other real-time linux versions, such as hard-hat linux, which you might want to take a look at. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: vi question
On Thu, 18 Sep 2003, Erez Doron wrote: i am using a version control software i want vi to automatically checkout a file whenver i try to edit it while it is read-only. this is not a good idea, since a lot of times, you open a file with vi ust to view it. this includes running a command line, then reloading the file (because the date changes when i checkout the file), then let me edit it. i want all to be automatically ( e.g. when i change the file) ah, this makes more sense (i.e. the change operation causes the file to be checked out) but this ignores the fact that often one accidentally writes in a file without meaning to do so, and then immediately un-do the write operation). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: linux.org.il - help wanted
On Mon, 29 Sep 2003, Shachar Shemesh wrote: That said, I think I'll take Ilya's design. You can view it at http://furr.shiny.co.il/linux.org.il/. Does anyone have any objections? no. after some text editing (i could do that if needed, once the text is ready), some fixes to the 'what is linux' page (on galeon 1.2 that comes with redhat 7.3, the images are shown on the left, on top of part of the text), and completing the site, it could do well. the only question to ask is - are we (iglu) letting go of the linux.org.il domain in favor of such a neutral entry page? this is the question that should have been asked - lets not do this change like thieves in the night. and my vote would be 'a painfull ok' :0 -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kernel q
On Mon, 20 Oct 2003, Erez Doron wrote: i am using copy_from_user and copy_to_user functions. how do i make that physical memory non cached ? ( it seems to write when it wants and read adresses i didnt ask it to read) can you explain what do you mean by 'reads when it wants to'? after you 'copy_from_user', your kernel-space buffer does not contain the data that exists in the user-space buffer? if this is the case - you seem to have some bug in your code. the same should be for the writing part, i.e. after 'copy_to_user', the user-space buffer should contain the desired text. why do you think caching is your problem, anyway? also, in what context do you use these functions? a system call? an interrupt handler? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: GPL Licensing Question
On Sun, 26 Oct 2003, Tal, Shachar wrote: Hi all, I have a GPL licensing question that came from a customer of mine: That customer is currently developing a distributed client-server, where the communication protocols between clients and servers are non-standard (i.e. not HTTP or likes of it). The customer wishes to include somewhat-modified GPLed software components in its client software (e.g. python, GTK or LAM/MPICH), while keeping his server implementation, protocol implementation and part of client code proprietary, in order to actually make money off the software. Does that conform to the GNU Public License? How much of the code does the customer have to release under the GPL? you might want to read the FSF's GPL FAQ (too many TLAs there ;) ). for example, regarding 'mere aggregation' Vs. 'combining': http://www.gnu.org/licenses/gpl-faq.html#TOCMereAggregation or regarding linking a GPL-ed program with non-free libraries: http://www.gnu.org/licenses/gpl-faq.html#TOCMoneyGuzzlerInc -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: compress communication between two servers
On Fri, 21 Nov 2003, redbaron wrote: I any of you know expand device an his way of work I'm looking for a solution like the expand device gives... The expand learn patterns of communication and find the best way to compass them, SSH is a stupid compass (he wont learn and wont improve any more then the first time I will start the service). SSH can't compress binary files so wall as the expand (based on experience of SCP whit and without compression). i'm not aware of a _free software_ solution that such adaptive compression (except for a free software solution that works on the X protocol only - but that's not what you want - is it?). there are commercial things that do that on linux, but i guess that's not what you're looking for. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On Mon, 15 Dec 2003, Shachar Shemesh wrote: I have a (couple of) client(s) that have performance problems across linux distirbutions. That is - they try an operation on a given platform with a given (redhat both cases) distribution, and get a certain performance. Then they try the same operation on the same hardware with a newer distribution, and get considerably lower performance. In one case, the fast distro is RedHat 7.2, and the slow one is 9. In another I'm not sure what the fast one is, but the slow on is AS 3. I suspect that RedHat screwed something up with their newer kernels (perhaps something incorrectly backported from 2.6?). people have such short memories: http://www.mail-archive.com/[EMAIL PROTECTED]/msg30749.html (summary - nadav harel, checks why hspell on redhat 9 runs several times slower then on redhat 7.something, and finds that changing the LD_LIBRARY_PATH to use the non-tls libraries solves the problem). -- guy good consultants are tested by their speed of memory recall operations ;) For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On 16 Dec 2003, Oleg Goldshmidt wrote: guy keren [EMAIL PROTECTED] writes: people have such short memories: http://www.mail-archive.com/[EMAIL PROTECTED]/msg30749.html From memory (thanks for reminding, Guy), someone (Mulix?) found this (I just found it on my own, but it does ring a bell) http://kerneltrap.org/comment/reply/1574 Note that Ulrich Drepper says there that Fedora Core 1 and RHEL3 should not have the problem. that's not what he was writing. he said that the benchmark should be perfrmed on one of those systems. he didn't say these systems contained the NPTL code from CVS head. get yourself one of those systems, and test. Shachar says that RHAS3 is slow - question is, whether or not that is the same kernel as RHEL3 uses, and which version Ulrich actually meant (he raises the point of kernel version numbers, then provides benchmarks without any indication of versions...). So in the 7.2 vs 9 case, can the client grab the kernel RPM from Fedora Core 1, rpm -ivh it reboot, and compare again? it's not a kernel issue - it's a libraries issue. do read that post again, to see the light ;) and please DON'T suggest that people take binaries frm one distro and install on another - some people on this list will not notice your question mark, thinkthis is the way to go, and find their systems unable to boot any longer. people tend to read what they want see :0 -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On 16 Dec 2003, Oleg Goldshmidt wrote: Oleg Goldshmidt [EMAIL PROTECTED] writes: http://kerneltrap.org/comment/reply/1574 Note that Ulrich Drepper says there that Fedora Core 1 and RHEL3 should not have the problem. Shachar says that RHAS3 is slow - question is, whether or not that is the same kernel as RHEL3 uses, and which version Ulrich actually meant (he raises the point of kernel version numbers, then provides benchmarks without any indication of versions...). So in the 7.2 vs 9 case, can the client grab the kernel RPM from Fedora Core 1, rpm -ivh it reboot, and compare again? Eh, it might be more involved. Look at the last posting from Jakub Jelinek in the URL above: Ulrich meant glibc CVS HEAD. Makes sense: NPTL is split between the kernel and glibc. Jakub is, IIRC, the glibc maintainer at Red Hat. The posting is from Nov 5, the latest glibc erratum is from Nov 13, https://rhn.redhat.com/errata/RHSA-2003-325.html It is not clear from the description if this includes the locking patch. I don't know how dangerous it is to upgrade glibc on RH9 to Fedora's version. I would not do it on a mission-critical machine. and i would not do it on _any_ system - unless i am realy bored, and want to see what if at work ;) it's fix-able - but what will the user gain from this excersize? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On Tue, 16 Dec 2003, Aaron wrote: I in fact noticed in RH 9.0 hspell and aspell were locking up my system. I was running a lowlatency kernel but found that jack and ardour were slow. I am now running Fedora, in kde and gnome window minimize slowly and maximize slowly. Programs lockup and apt-get takes over to the point where my mouse pointer dissapears for a time. I am wondering if this is just a RH issue or Mandrake and friends also have this problem?? did you _read_ the mentioned post(s)? they give you a direction regarding what to try doing. try running the problematic program using: LD_ASSUME_KERNEL=2.4.1 program parameters... read the posts for the exact command - i might be typing it wrong here, and i have no RH9 to test it on. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On Tue, 16 Dec 2003 [EMAIL PROTECTED] wrote: guy keren wrote: did you _read_ the mentioned post(s)? they give you a direction regarding what to try doing. try running the problematic program using: LD_ASSUME_KERNEL=2.4.1 program parameters... read the posts for the exact command - i might be typing it wrong here, and i have no RH9 to test it on. I think the main part was LD_LIBRARY_PATH=., where . presumably contains the latest compiled CVS of libc. for those who did not yet get the clue, please look at the following page: http://www.talkaboutprogramming.com/group/comp.programming.threads/messages/39627.html look at this question, and the first immediate answer (http://www.talkaboutprogramming.com/group/comp.programming.threads/messages/39634.html) Anyway, it was bleeding-edge developer discussion, not something I'd try with libc on any system unless it was a system I didn't care to reformat and re-install afterwords. which is precisely why LD_ASSUME_KERNEL=2.4.1 will _avoid_ using the NPTL threading library (= bleeding edge, very bloody), and cause the dynamic-loader to use the linux-threads library (=older, safer, better tested, good for the user). do yourself a favour, and re-read that page. you _will_ see the light ;) -- guy refusal to thoroughly read information, is futile For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On Tue, 16 Dec 2003, Oleg Goldshmidt wrote: guy keren [EMAIL PROTECTED] writes: that's not what he was writing. he said that the benchmark should be perfrmed on one of those systems. he didn't say these systems contained the NPTL code from CVS head. get yourself one of those systems, and test. Sorry, misattributed. Linus did test Fedora and responded to Ulrich ;-) what linus did test, was that on the system he was using, _avoiding_ the use of NPTL (via the LD_ASSUME_KERNEL=2.4.1 hack) solved the problem. linus was merely interested in making sure this was NOT a linux kernel problem - beacause the linux kernel is his concern, while the NPTL code is somebody else's concern. after he isolated the problem and showed it was not a kernel issue, his work was done. So in the 7.2 vs 9 case, can the client grab the kernel RPM from Fedora Core 1, rpm -ivh it reboot, and compare again? it's not a kernel issue - it's a libraries issue. do read that post again, to see the light ;) I did correct myself in a followup, didn't I? ;-) indeed, but it was important to make sure people don't try doing what your first post suggested, which is why i sent two replies - one for each letter. and please DON'T suggest that people take binaries frm one distro and install on another - some people on this list will notnotice your question mark, thinkthis is the way to go, and find their systems unable to boot any longer. people tend to read what they want see :0 The rpm -ivh kernel*.i686.rpm procedure on redhat usually works for me. ofcourse, when you take the kernel for the same distro for which it is intended, then it should be ok. but taking a kernel that was meant for a _different_ distro, is a bad idea. There is little danger that the system won't boot any longer - the old kernel stays there. unless the new kernel is so incompatible, that it could break the disk's contents, in which case it could cause unfixable problems (that is, unfixable, unless you make a re-install). Glibc is trickier, as I mentioned in a followup to myself. Don't try that at work... but do try the LD_ASSUME_KERNEL=2.4.1 method. it causes the system to switch to an _older_ version of the threading library, which does work. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Various performance problems
On Tue, 16 Dec 2003, Alon Weinstein wrote: Muli Ben-Yehuda wrote: On Tue, Dec 16, 2003 at 09:21:27AM +0200, Alon Weinstein wrote: cat /etc/ld.so.conf, assuming it hasn't been changed since the last time someone ran ldconfig. Looking at /etc/ld.so.conf: /usr/kerberos/lib /usr/X11R6/lib /usr/lib/qt-3.1/lib /opt/opengroupware.org/Libraries/ix86/linux-gnu/gnu-fd-nil /opt/skyrix/system/Libraries/ix86/linux-gnu/gnu-fd-nil /usr/lib/sane /usr/lib/mysql No tls or i686... greping for either 'tls' or 'i686' in etc/sysconfig or etc/rc.d (I'm just reaching here, not really sure where or if I should be looking) got me nothing. and in case you missed the previous posts for some reason: you can not know. they made some stupid(?!!?) hacks in glibc, makint it imposible for you to know, as a user, which version of the library is actually in use, other then by writing test programs, stracing programs, etc. so do try running your program with: LD_ASSUME_KERNEL=2.4.1 program parameters... and see what you come up with. How can Ichange that default? (perhaps using an ugly hack -- ln -s /lib/i686 /lib/tls ? ;) ) not sure if this will work - it _could_ be that the dynamic loader expects to find certain symbols (=functions, variables) in the tls libraries, which it does not expect to find in the non-tls libraries, causing your system to break fully and completely (until a rescue operation is performed). you will need to set the 'LD_ASSUME_KERNEL' variable very early during system boot - that will make _most_ processes launched with the non-tls libraries. however, the 'init' process, is launched before any of the system's initialization scripts are launched - and the 'init' process is the process that launches the terminal handler processes (getty, mgetty or similar), and those launch your login shells, etc. and if i were you, i wouldn't have messed up with the launching of the 'init' process, unless i had: 1. a very good idea how to restore the original state, in case of problems. 2. a few hours to spare. 3. the will to write a small C program, that will be launched _in place of_ the regular init program, set the LD_ASSUME_KERNEL environment variable, export it, and then launch the original init program. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problems with multiplay IP's machine
On Tue, 16 Dec 2003, Elad Almadoi wrote: Hey! I have an IBM xSeries 335 runing redhat 7.3 Till few a days ago all was runing as it cant get better Looks like from it self, a problem started: The machine have many IP's, the main one is configured as eth0 and the others as eth0:0 eth0:1 eth0:2, etc... All the eth0:* won't reply to ping, the main one does reply here is an idea: 1. stop using the old way of defining additional IP addresses (i.e. the eth0:* devices). instead, start using the 'ip' command to add addresses (i.e. 'ip addr add AAA.BBB.CCC.DDD/24 dev eth0', change '24' to the proper number of bits that compose your network's netmask). 2. problems don't just start to happen out of the blue. something is causing them. now, drop this saint peter attitude, and try to genuinely think - what _did_ change in the machine's configuration? was it rebooted on the day where the problem started happening? perhaps someone changed a config file a month ago, without checking that it works, and only after a recent reboot or 'service network restart' or a similar operation, the new config file came into effect? 3. All the IRC servers runing (even on main IP) after few hours disconnecting the sockets, with the reason Connection reset by peer Nothing for the pings when I try to lookup on tcpdump, like they are not getting to the machine BUT process can listen to that IP's and it's posiblle to connect them from the internet No firewall installed\runing Anyone got any idea? Thanks! did you try to reboot the machine? did you try to go over the config files and checking for their correctness? even if you think nothing changed at all - check. saying but i didn't change anything, is usually a good way to avoid solving problems. you change soemthing. or someone else with root access changed something. or some process that you are running accepted a change from other devices on the network (i'm not talking about break-ins, rather about a miss-configuration), etc. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Fwd: Linux 2.6.0
On Thu, 18 Dec 2003, Nadav Har'El wrote: On Thu, Dec 18, 2003, Muli Ben-Yehuda wrote about Fwd: Linux 2.6.0: Haven't seen this mentioned here yet, so, without further ado... Does anybody know of a good summary of the differences between the latest Linux 2.4 versions, and 2.6? hold your horses... remember the dragons, mike (and kernel 2.4.0). according to the bug-fix list, the kernel is not realy stable yet. fixing races that can cause oops of various sorts? give it a few versions before you really try to use it. it took 2.4 more then half a year, from the 'stable' kernel, until a kernel that was actually stable (perhaps it was closer to a year? my memory is getting too short for this). based on past experience, it'll take a few more month before 2.6 will get to a stage where you'll dare put it on a production system that needs to work under some kind of stress.. this is not a prediction - just a catious look at the past. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: KDE: Mutex error...
On Thu, 25 Dec 2003, Amichai Rotman wrote: I am running KDE 3.1.3 on an MDK 9.2. Every time, after closinga KDE app, I get the following error: Mutex destroy failure: Device or resource busy I see it when I use the console to run a app (i.e. KAppfinder) Any ideas, pointers? please search for the following (exact) text on google: Mutex destroy failure: Device or resource busy KDE 3.1.3 read the _entire_ thread you'll find in the first result - you'll see the light. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
and what's with pine+imap? (was: Re: Suggentions for server side spam control)
i tried checking for the possibility to have spam filtering with the following configuration: remote mail server, accessed using 'pine', via an imap server. - thus, i cannot install a spam-filter on the remote server. - the local procmail is never activated, and thus seems to be un-useable here. - i can't use fetchmail - this is imap, not pop3. - couldn't find a way, in pine's configuration, on how to set up a filter that passes the message via an external program. - searching for a solution using google, as well as reading spamassassin's documentation, just shows solutions that assume you can set spamassassin to run via procmail. this does not seem to work for my setup. is there any solution, _WITHOUT_ replacing the mail client, and without reverting to pop3? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: and what's with pine+imap? (was: Re: Suggentions for server side spam control)
On Wed, 31 Dec 2003, Alon Altman wrote: On Wed, 31 Dec 2003, guy keren wrote: i tried checking for the possibility to have spam filtering with the following configuration: remote mail server, accessed using 'pine', via an imap server. - thus, i cannot install a spam-filter on the remote server. - the local procmail is never activated, and thus seems to be un-useable here. - i can't use fetchmail - this is imap, not pop3. - couldn't find a way, in pine's configuration, on how to set up a filter that passes the message via an external program. - searching for a solution using google, as well as reading spamassassin's documentation, just shows solutions that assume you can set spamassassin to run via procmail. this does not seem to work for my setup. is there any solution, _WITHOUT_ replacing the mail client, and without reverting to pop3? IIRC, fetchmail supports IMAP, so use fetchmail+procmail and then either use the downloaded mail locally, or use IMAP to upload the mail back to the server. this setup defeats the purpose of using imap in the first place - to be able to see all messages _without_ downloading the messages themselves. i'm beginning to think i'm asking for the imposible - to filter the letter, i need to first download it. however, i should be able to filter out by the message headers that _are_ downloaded by imap, thus eliminating a large part of the spam, and only then downloading the rest of it for further inspection... oh, well. no spam solution for me... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [JOB OFFER] Adwise Seeking - PHP Professional
On Sun, 18 Jan 2004 [EMAIL PROTECTED] wrote: I think your complains regarding Microsoft Word format are ridiculous. I don't what you to send me documents in Open Office or in PDF formats as it will make another unnecessary loop. ok, lets put this in a different perspective. you are looking for Linux people (or you wouldn't have been looking for them on this list). quite a few of the people that use linux, do not have microsoft office, and are thus unable to send you a resume in the format you want to get. thus, you cut out some people. it's even feasible that the more a person is familiar with linux, the less chance that they have microsoft office. so you're cutting out the people that you most want to send you their resumes. here is a simple idea: how about accepting resumes written in html? they are easy enough to create for linux users (and user of other platforms as well). and they are easy enough for you to process (microsoft word, office 2000 or higher, can import html quite easily - especially the simple html that is likely to be found in a resume). so, what do you chose - to keep people antagonized towards your company, or to be a little more flexible? -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Motif vs. Lesstif
On Sun, 15 Feb 2004, Eli Marmor wrote: I believe that this thread was started because of a small misunderstanding: There are TWO BiDi/Hebrew Support implementations: IBM's, and mine (AKA El-Mar). IBM sub-contracted parts of the project to Israeli and Egyptian companies (Applicom in Israel). I've had no connection with that project (Applicom was an important customer of me, and they even purchased my implementation - after finishing the project for IBM, and were a Source Licensee - but all of this doesn't have any connection to their project for IBM. IBM was a customer of me too, but also with no connection to their BiDi/Hebrew/Arabic project). eli, the only reason why such a miss-understanding could have resulted, is because you decided to start talking about this project, and made a few statements that imply of your involvement. i don't realy see how reading about your clients and their clients and their sub-contractors, has got anything to do with clearing things up. next time, you can avoid such mis-understandings by not creating them in the first place. and no, please don't give me your phone number to call you because you have no time to read e-mail ;) you are entitled to a reply - and this time i will try to stop myself from responding to it. hopefully the hint is clear, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Zope as a wiki system
On Sun, 22 Feb 2004, Shachar Shemesh wrote: We are looking for a good Wiki type solution for hamakor's web site. Being the reknown paranoid I am, I'm looking for a well known solution, prefereably one that I can just apt-get install. For reasons I would rather not go into at the moment, if it has to use a database system, I prefer it to be Postgresql to MySQL. I was wondering about Zope. I know it is more of an infrastructure than a system, but it's supposed to have a wiki system (as well as 2^37 other things), and it is, by all means, a well known solution. I was more wondering about whether it is not going to turn out to be a total overkill, and an administration nightmare? I have not administrated it before. zope was used as iglu's engine in the past. it was dropped, because it was hard to install and manage (i.e. there was an update pending for a long, long, long time, before eventually shlomi (i think) set up a replacement, and zope was dropped. it's not 'open' enough to extend, and not too many people around know how to program for it (at least that was the case on iglu). i state all of this not because i found it hard to manage (i never eevn tried), but rather because of the fact that it was never realy managed by anyone except chen. so you do the math... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Need help - arguments for Why not VB+ASP+Access
On Sun, 22 Feb 2004 [EMAIL PROTECTED] wrote: My gliding club is going to write their own software, after years of being dependent on an ancient Magic software that no-one could update and did a fraction of the requirements. A club member who knows MS stuff valunteered to write it in VB+ASP+Access. this is the single most important argument - that there is someone that said they will do it. there's no good way to beat that in a gliding club. the next best thing you can do, after vulanteering to do it yourself (or finding someone else to vulanteer doing it on linux), is try to talk to this guy personally, convince him that he wants to learn about linux (it will give _him_ an edge over people who only knows windows, and open new opportunities for him), and then tell him that this _is_ a good opportunity to start learning (i.e. he is over the part of finding an idea for a project, which serves a real purpose). ofcourse, you'll need to talk to him privately before the meeting with the other club members - and then have him on your side during that meeting. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
bidirectional file transfer software?
i am looking for a file-transfer program (for linux), which can run one upload and one download simultaneously, on a _single_ TCP connection (sort of the TCP equivalent of the BModem protocol used on BBS-es years ago). i assume this will need to be a client+server application (since standard file transfer protocols such as ftp are not 'full-duplex' in this sense). - it needs to run on the command-line (i.e. NOT an X/svgalib based application). - it needs to show the temporary transfer rate as transfer progresses, seperately for each direction (upload/download). is there such a thing in existence? searching for 'file transfer' + 'bidirectional' on freshmeat/google gives nothing. using 'full duplex' gives a lot of info about full-duplex ethernet NICS and switches. by the way, you might ask what's wrong with using two seperate TCP connections - one for upload and one for download?. think of a slow link, where a mass download in one direction slows down the ACKs for the upload, and vice versa, giving an overall poor performance. (and i don't want/can't get into playing with linux's priority queues for this). thanks, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: bidirectional file transfer software?
On Thu, 4 Mar 2004, Eran Tromer wrote: Questionable motivation aside, you can use ssh -L or -R to multiplex several virtual TCP connections on a single physical TCP connection, and run some file transfer protocol on each. i thought about this, but: 1. ssh does encryption which adds overhead - is there a way to disable encryption in ssh?? 2. which file transfer program would i b able to run on top of an ssh tunnel? i don't think i'll be able to tunnel an ftp connection, for instance... -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: bidirectional file transfer software?
On Thu, 4 Mar 2004, Ely Levy wrote: I don't know any client that does what you said but maybe udp one would solve your problem as well? not likely, since an udp file-transfer program would have to implement its own acks, which will bring me back to the same problem (i.e. acks will be delayed by the queuing of data). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: bidirectional file transfer software?
On Thu, 4 Mar 2004, Shachar Shemesh wrote: That is not the case here. TCP/IP assumes that packets do not get corrupted along the way(1). It therefor assumes that, if a packet is lost, it's beause of congestion. As such, the rate at which ACKs arrive is tightly bound into the TCP/IP rate limiting algorithm. This means that your assumption is wrong for the case at hand, as a saturated connection is going to be losing lots and lots of packets. Guy's assertation that using a single TCP connection for both directions seems, to me, like a correct one, as the ACKs for one direction piggyback on the data for the other. this is indeed exactly what i'm trying to achieve here. i'll re-iterate: - i'm talking about slow links - so i want a minimal protocol overhead - not in terms of processing time, but rather in terms of added data overhead (how much extra _data_ does ssh's encryption send along with the original data it delivers? same for scp?). - i don't want to use compression at all here, since i want to do this for measurement of raw transfer rate. this, in order to be able to test raw transfer speed over a link, when it is loaded - so i need something that manages to use the full bandwidth available - hence the need to avoid artificial delaying of ACKs in one direction, by data which is sent over the other direction. and hence the need for the application to show me the temporary transfer rates in real time. That aside, there is a TCP option called Delayed ACK, but it is aimed at something completely else. If Guy manages to find a client that does what he's looking for, it will be a good idea to turn it on, however. delayed ACKs don't _realy_ interest me, since an empty ACK packet is about 40 bytes long. assuming an MTU of around 1500 bytes, then each 1500 bytes of data (actually 1460 bytes of data) will cause 40 bytes in the other direction to be sent - which is within the expected error range for measurements (it's about 2.7% of the 'usefull' bandwidth). the same goes for selective ACKs. thus, the only reason that the ACKs worry me, is beacuse they can only piggyback the same TCP connection for which they are ACKing. if there are other active TCP connections, which fill the router's queue, then cause the transfer to be slowed down. from what i see so far, there's no ready-made solution for this - i'll see if i manage to get something done with scp over ssh... thanks, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kernel compilation
On Tue, 9 Mar 2004, David Sapir wrote: Thanks for your answer. I found the config files you mentioned. I will try to just enable encryption compilation. I would also like to state that it is the first time I'm trying to compile the kernel, so forgive me if my questions seem stupid or obvious. before asking questions, i suggest you look at something on the net that explains the compilation process. i just recently gave a lecture about this in haifux - check the slides at http://www.haifux.org/lectures/88-sil/ as well as the earlier lecture at http://www.haifux.org/lectures/86-sil/kernel-modules-drivers/ this _might_ help you get started. hope this helps (and sorry for the shameless plug). -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Runtime changing of timezones
On Tue, 16 Mar 2004, Nachum Kanovsky wrote: If I use tzselect to change my time zone, is there any way to get applications that are already running to notice the timezone change, and update their clocks appropriately? perhaps by sending them some signal? As an example application, I would like to get the gnome clock to be updated without having to restart gnome. I am also running a custom application, and I would like them to all display the correct time. what _are_ you trying to achieve? are you trying to adjust the machine for daylight saving time changes? or you're running a test machine and testing some software that need to be tested in different timezones? in the former case, you are not supposed to touch the time-zone at all, but rather make sure that the TZ file you have is updated with the dates of daylight saving time's changes, and then the switch will be made automatically - man 'zic' for more information, and look at ftp://ftp.cs.huji.ac.il/pub/tz/israel (look at the 'README' file) for easy instructions for the madness called israeli summer-time ;) if you're talking about the later, i'm afraid tzselect cannot help - since (i pressume) you're using it to set the TZ environment variable - and this change does not effect existing processes. you _might_ be able to apply such a change to them by attaching a debugger to them and using it to invoke the relevant library function - but i'm not sure how various applications will react to such a change. if you change the time zone of the entire machine - perhaps this will effect currently-running applications. -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problems configuring NFS-server
On Fri, 2 Apr 2004 [EMAIL PROTECTED] wrote:
Hallo friends and list members.
I have some problems configuring NFS-server.
Trying to mount the server I get this :
$ mount_nfs 10.0.0.8:/usr/BSD/usr
mount_nfs: bad MNT RPC: RPC: Timed out
this calls for running a sniffer on the open-bsd machine, listening for
all traffic going between its IP and the server's IP, then running the
'mount_nfs' command again. you will probably see that eventualy the BSD
machine sends packets that do not get any replies. based on the port
number to which the BSD machine tries to connect, you'll know who the
culprit process is.
you must learn to use a sniffer and strace whenever applicable (strace is
not applicable here, since the IP packets are sent by obsd's kernel, not
by the user-space mount_nfs command).
as for the suggestion that was raised regarding checking connectivity to
port 2049 - you can do this using netcat ('nc' is the name of the command
- somehting like 'nc -u host port').
--
guy
For world domination - press 1,
or dial 0, and please hold, for the creator. -- nob o. dy
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
