Re: UPS anyone?
Peter S Galbraith wrote: I want to put a small UPS on my work system to let it shutdown gracefully when there's a power outage. I'm looking at: Best Power Patriot Best Power Patriot Pro APC Back-UPS BK500M APC Back-UPS BK650M Anyone have any recommendations? Does any of them work `easily' with Debian packages (e.g. using stock cables)? The Debain packages I know of are genpower, bpowerd, apcd and upsd. It'll take a while to look at them all since they mostly conflict with eachother (for obvious reasons). Do you know if bpowerd works with the newer Patriot Pro? I don't have a Patriot Pro, but if the pinout is the same, then it should work. Also, I can't find the cable INT-0027A on their web page. I do find INT-0027 (basic cable for the CheckUPS II Suite for UNIX, AIX, SCO and HP-UX). Is that what I want? My supplier can get it for CDN$52, although he thinks I need INT-0051 (for Windows). Should I get a Patriot or a Patriot Pro, in your experience? I would avoid Patriot at this point. Since I packaged bpowerd, Best has stopped supplying cables for free with their UPS's. If you want a cable, expect to pay $50 to Best. (They used to supply the cables for free with the UPS) Also, bpowerd does not support the INT-0051, which is the only cable Best (now) supports for Windows. I have the beginnings of a cable auto-detect and support for INT-0051, but I haven't finished because the Real World(tm) keeps getting in the way. Also, out of my 3 Patriots that I have owned for the past year, one has died. All in all, I am underwhelmed by Best Power. -Mitch
Re: Binary Newsgroup extractor?
Andrew Holmes wrote: Maybe this is a pointless question, but. Has anyone found a program for linux that will look through binary news groups and download/decode all of the images/attachments according to a filter? I have AUB which works for multi-part messages, but I can't find a way to do single-part messages with binary-attachments. There are several programs like this for NT, (I have used NewsBin quite succesfully), but I can't find one for linux. The only solution I have found is to use slrnpull to get all the messages then use uuencode extract them, but this isn't a very nice solution. TIA, any help is greatly appreciated! Try linleech. The original site appears to be down now, but the Debian packages are available from your local debian mirror. -Mitch
Re: How to find and install Gnome .debs?
In foo.debian-user, you wrote: I have these lines in /etc/apt/sources.list. deb http://http.us.debian.org/debian unstable main contrib non-free deb http://non-us.debian.org/debian-non-US unstable non-US deb http://www.debian.org/~mblevin/gnome-apt ./ deb http://www.debian.org/~bma enlightenment/ deb http://snowcrash.tdyc.com/debian potato rkrusty deb http://www.debian.org/~jim/debian-gtk-gnome/gnome-stage-slink unstable main You can remove the ~mblevin line, as gnome-apt is now in the regular potato distribution. -Mitch -- Nothing recedes like success. -- Walter Winchell
Re: man!
In foo.debian-user, you wrote: When I installed fetchmail and typed man fetchmail it did not work. But when I did a reboot it worked after that. I have now installed pgp and when I type man pgp No manual entry for pgp!! So I did a locate /usr/man/man1/pgp.1.gz. What is wrong ? I don't know what is wrong, because my man can see newly installed manual pages. So, rather than solve your problem, I'll give a crutch. Try running (as root) mandb to update the man caches. This will at least avoid a reboot. I can not reboot my computer every time i install something! I did that in Windows for years. -Mitch
Re: modem speakerphone commands
In foo.debian-user, Erhan Ince wrote: Hi all, I have a 56K US robotics speakerphone modem. I am interested in activating the speakerphone or toggling it off using the modem commands from my visual basic program. I know how to use the MSCOMM port I just don't know how to activate/de-activate the speakerphone. Reason I want to do this is for making the called party listed to a sound file played at the callers computer. Any help will be most appreciated. Are you using the visual-basic package from slink or from potato? I heard the slink version had problems with the MSCOMM port. -Mitch
Re: alien question
In foo.debian-user, you wrote: Hi, when I try to use alien to convert to a rpm file, I got the following: alien: 822-date did not return a valid result. Could someone please help me?? Actually, the full text message of the error is: alien: 822-date did not return a valid result. You probably need to intall the dpkg-dev debian package. So, the obvious question(s) is(are): Did you install the dpkg-dev package? What happens when you type 822-date? What is the result of which 822-date? -Mitch
Re: Debian installation - list of things that make it hard
In foo.debian-user, you wrote: Oh, and a final question that isn't so rhetorical. Is it possible to use apt from behind a corporate firewall that does *not* seem to do IP forwarding or masquerading? I don't know much about firewalls, but I work at a big company where I can pretty much guarantee I have no influence over its configuration. Can apt be told about a proxy in the way that, say, lynx can? set the http_proxy environment variable. See 'man sources.list' for details. -Mitch
Re: PPP troubles.....
In foo.debian-user, you wrote: I started trying to set up a pp connection a few months ago. I can't sit and hack this full time, as I have jobs (in the plural) taking up a *lot* of my time. I've been through ezppp, wvdial, hand hacking, gnome-ppp, and I just haven't gotten anywhere. I have learned a lot about ppp in theory, though. [snip] My first step is to change all files as suggested by my isp. Then what? I am supposing I need to write a script, but am not sure what to put into it, or what order. Be careful about following explicit ISP instructions. Most are written for RedHat, and there may be differences. If you have a URL of the ISP instructions, it may help us to identify where the differences are. -Mitch
Re: PPP troubles.....
Koyote wrote: Be careful about following explicit ISP instructions. Most are written for RedHat, and there may be differences. If you have a URL of the ISP instructions, it may help us to identify where the differences are. Sorry, I thought I had put it in there: http://support.paonline.com/support/OTHER/LINUX.htm For step 8, instead of editing /etc/ppp/chat, edit the file /etc/chatscripts/provider. Also, I would use the following lines instead of what they dictate: TIMEOUT 3 ABORT BUSY ABORT NO CARRIER ABORT VOICE ABORT NO DIALTONE '' \r '' \r '' ATZ OK ATF1 OK ATDT717555 '' \p TIMOUT 50 For step nine, edit the /etc/ppp/peers/provider file, replacing their chat line with: connect /usr/sbin/chat -v -f /etc/chatscripts/provider Also, instead of typing pppd /dev/modem to start the connection, just use the pon command, which is really just a shell script that executes pppd call provider. This is a more flexible setup and allows you to keep several ISPs at once and connect to them with just pppd call myisp, where myisp is just whatever name you want to use to identify the isp (provider in the above example). HTH, -Mitch
Re: Newbie... One more please
In foo.debian-user, you wrote: Any comments about the DEC Alpha vs the UltraSparc? And also discovered that Solaris supports 64 bit processors, whereas Linux doesn't? Linux does support 64-bit processors. Both alphas and ultra-sparcs. -Mitch
Re: Newbie problem
[EMAIL PROTECTED] wrote: In a message dated 5/17/99 5:03:50 PM Central Daylight Time, [EMAIL PROTECTED] writes: 1) What kind of mouse do you have? (PS/2 or serial) I have a Logitech Mouseman 3 button, on serial. 2) What brand of mouse? Logitech 3) What is the setting in XF86Config? You can get this with: [prompt]$ grep -A 8 'Section Pointer' /etc/X11/XF86Config Protocol Mouseman Device /dev/mouse 4) What do you have for /dev/mouse? [prompt]$ ls -l /dev/mouse lrwxrwxrwx 1 root root /dev/mouse - ttyS0 5) Does your mouse work in console mode? Is gpm installed? gpm is installed, but I don't know how to use it in console mode...i'm still a newbie :) Answer those and we can probably track down your problem. -Mitch I hope this helps you help me :) Good. Now check the permissions on your serial port. [prompt]$ ls -l /dev/ttyS0 and the settings for the serial port. [prompt]$ setserial -gv /dev/ttyS0 -Mitch
Re: Can one fake a Debian package?
In foo.debian-user, you wrote: I have QT 1.42 installed from the source tarball. Unfortunately, my KDE installation was in the form of Debian packages, and now I get kde-whatever depends on qt-142 errors from dselect and apt-get. How can I convince my system that QT is installed? Is there a file I can edit? Look at the equivs package. -Mitch
Re: possible bug in sysvinit?
In foo.debian-user, you wrote: The problem is this: I've got a UPS daemon (mgeupsd, not debianized, but I might work on that if I can get this problem resolved.) It works fine when the power fails: it writes FAIL to /etc/powerstatus and sends SIGPWR to init. init then looks at these lines in inittab: If you feel like tweaking the source of mgeupsd, there is a much better (IMHO) way to report power events to init. Just write to the /dev/initctl pipe and you can avoid all status files in /etc and all SIGPWR signals. Much cleaner (and easier to debug...) See bpowerd source for example. Sorry if this isn't directly helpful, -Mitch
Re: Newbie problem
In foo.debian-user, you wrote: I'm trying to run X windows on my computer. I installed all of the packages required for it, but I still have one basic problem: my mouse won't work when I first run XF86Setup. I have gone through every mouse option in XF86Setup, applying them all, and none work. I reinstalled my entire system, and it still won't work. What can I do? I installed all the device drivers related to mice that would work...I have no clue as to where I can go from here. No FAQ or HOWTO has anything on this. I really need help here. Thanks. Give some more information. 1) What kind of mouse do you have? (PS/2 or serial) 2) What brand of mouse? 3) What is the setting in XF86Config? You can get this with: [prompt]$ grep -A 8 'Section Pointer' /etc/X11/XF86Config 4) What do you have for /dev/mouse? [prompt]$ ls -l /dev/mouse 5) Does your mouse work in console mode? Is gpm installed? Answer those and we can probably track down your problem. -Mitch
Re: apt: 0.3.6 and ldconfig: /lib/libNoVersion.so.1 (No such file or directory)
In foo.debian-user, you wrote: ever since I upgraded to apt 0.3.6 I get this error with just about every package I install ldconfig: warning: can't open /lib/libNoVersion.so.1 (No such file or directory), skipping - any thoughts on how to fix this? [prompt]# rm /lib/libNoVersion.so.1 is this a problem with apt or ldconfig? Neither. It is caused by having a dangling symbolic link in your lib directory. ldconfig gives a warning and keeps going, which is the correct behavior. apt simply is calling dpkg, which is calling ldconfig in some postinst scripts, which is the correct behavior. It is (most likely) caused by you having (or used to have) a package on your system that used a library called libNoVersion.so.1.y, which created a symbolic link to specify the latest minor version. Since the symbolic link was created instead of part of the package, it did not get removed when the package was upgraded or removed. This was an error on the part of the package that created the link, which we haven't been able to identify. If anyone has this link (/lib/libNoVersion.so.1) on their system, and it points to an *actual file*, please identify the package that the file belongs to and post it here. -Mitch
Re: Stability
In foo.debian-user, you wrote: I will shortly be taking a two week vacation, and sadly, I have to leave my trusty box behind *schniff* .. It has to stay running, however, because a local user group relies on it for mail and web serving. There is no one else who will have access to the machine, so I need to ensure that it will stay up no matter what. Does anyone have any ideas as to how I should tweak the system to make sure it behaves? I have the obligatory UPS, and on the event of a kernel panic, it'll reboot itself after a ten second delay, but is there anything else? I'd be interested in hearing any ideas you may have. Just make sure that your UPS daemon does the right thing during a powerdown... doesn't send a halt signal until you are sure that the UPS has received a cycle-power signal. I've seen power-strips that allow you to cycle-power from a touch tone phone, but that may be overkill in your case. Consider the following and take heart: [portal]$ uptime 1:50am up 205 days, 13:05, 1 user, load average: 0.08, 0.02, 0.00 -Mitch
Re: sorting this mail
In foo.debian-user, you wrote: hey all I don't suppose the list manager would consider putting somthing like [deb] in the subject line, would he (you)? Sure would help me sort this stuff... You can only sort on the Subject field? What type of program are you using to sort? Others on the list seem to have no trouble sorting the list traffic, and I find the [label] prepended to the subject line to be annoying. -Mitch
Re: libNoVersion.so.1
In foo.debian-user, you wrote: Hi, I keep getting the message below now installing packages using delect. Where does this come from, and how do I get rid of it? ldconfig: warning: can't open /lib/libNoVersion.so.1 (No such file or directory), skipping This is a symlink created somehow in recent potato upgrades. You should be able to safely remove /lib/libNoVersion.so.1, which will get rid of the ldconfig warning. I still haven't seen on this list any verification of which package is creating this link. -Mitch
Re: Broken Select in dselect (typos removed)
In foo.debian-user, you wrote: Hi All. This just started today, after doing a dselect/Update off of www.debian.org stable main, contrib, and non-free. Now when I run dselect and go into the Select option, the program just quits with the message: dselect: failed to create baselist pad: Cannot allocate memory This is on a 32MB AMDK6-233; one partition with 235MB still free. Kernel is 2.0.36. Any clues? Do you have a swap partition? Are you running any other memory-hogging applications at the same time? (like Netscape) -Mitch
Re: Help with Alien
In foo.debian-user, you wrote: Can anyone give me a quick tutorial on how to use dpkg to install rpm's with alien? Thanks... [prompt]# alien -i somepackage.rpm This will convert and install the RPM file. An easier way to get a quick tutorial on alien would by to type 'man alien' at the prompt. -Mitch
Re: Help with apt
In foo.debian-user, you wrote: Not so much dselect, but dpkg yes. I have a dialup connection and sometimes I download a package over more than one dialup session, save it in a directory and install it with dpkg. I have tried apt, but could not get it to install a package from a directory on my hard disk - which is frustrating. It seems to me apt is looking for the same directory structure that the debian mirrors have and I am not going to mirror the whole structure on my hard disk. apt will assume a Debian archive structure unless you end your sources.list line with a directory separater. For example, if your deb files are in /home/jhspies/debian/mydebs/*, you would need to following line. deb file:/home/jhspies/debian mydebs/ HTH, -Mitch
Re: Help with apt
In foo.debian-user, you wrote: Quoting Mitch Blevins([EMAIL PROTECTED]): In foo.debian-user, you wrote: Not so much dselect, but dpkg yes. I have a dialup connection and sometimes I download a package over more than one dialup session, save it in a directory and install it with dpkg. I have tried apt, but could not get it to install a package from a directory on my hard disk - which is frustrating. It seems to me apt is looking for the same directory structure that the debian mirrors have and I am not going to mirror the whole structure on my hard disk. apt will assume a Debian archive structure unless you end your sources.list line with a directory separater. For example, if your deb files are in /home/jhspies/debian/mydebs/*, you would need to following line. deb file:/home/jhspies/debian mydebs/ Mitch Don't you need a Packages file in that dir for that to work? Ooops. You're right. Create a Packages.gz file in the same directory as the debs using the dpkg-scanpackages command. Something like this should work... [prompt]$ cd /home/jhspies/debian/mydebs [prompt]$ dpkg-scanpackages . /dev/null |gzip -c Packages.gz Thanks, -Mitch
Re: Apps For Debian?
In foo.debian-user, you wrote: On Sat, 8 May 1999, Lawrence Wickline wrote: Also need something that will make .gifs (Gimp doesn't seem to do this) Haven't found one yet myself... I wouldn't mind a decent package... I'll probably just wait until Wine gets better and run Paint Shop Pro... Gimp will support GIFs. The capability is just in a separate package for patent reasons. Install the gimp-nonfree package to get GIF and TIFF support. -Mitch
Re: Help with apt
In foo.debian-user, you wrote: Question: So how do I, running from a slink CD, get and install potato's apt? If I add entries for 'unstable' in /etc/apt/sources.list, and then go into dselect, it wants to upgrade my world to potato. Of course, I can always just go download the deb and run dpkg on it, but is there a better way? After changing the references in sources.list, you can... [prompt]# apt-get update apt-get install apt No need to use dselect for this. -Mitch
Re: Thoughts about winmodems
In foo.debian-user, you wrote: Somewhat off-topic, but perhaps applicable If a modem is a device for MOdulating/DEModulation, and a winmodem turns that job over to the software (Windows), is that not grounds for a class-action fraud suit against winmodem manufacturers for marketing them as modems (modulator/demodulators)? Or are winmodems really doing the mod/demod and Windows just does other necessary processing? I bought a big, plastic, insulated box the other day. It was clearly marked as a Cooler. Imagine my surprise when I found out that it needed ice to actually do the cooling. Perhaps we can expand your class-action suit to include these evil manufacturers. -Mitch
Re: Debian 2.1 install help!
In foo.debian-user, you wrote: I just installed Debian 2.1 on a DX2 100 MHz labtop. I am encountering two problems after the installation, and would appreciate your help. The first one, I first used a 3Com ethernet/modem combo card to do ftp install, but since I do not have a permanant IP, I am actually using the modem. However, I was unable to make the switch, or don't know how. In any case, now if I use pon, it will dial the #, and after it connects, the ethernet card will intervene to look for IP address. Is there a way to terminate ethernet card? Secondly, my X server crashes when I startx. I used xf86config, and thought everything should be fine, but just can't seem to get X to work. Is there a way to troubleshoot this problem? The error message is several pages, and basically it says it cannot find a mode. Is it possible to log the error message? Thanks! You can log the error messages with: [prompt]$ startx myfile.txt 21 Then attach myfile.txt to the problem report. Sorry, I don't know much about PCMCIA. -Mitch
Re: Preventing xdm from starting
In foo.debian-user, you wrote: Hi, Since I installed X on my slink dist, xdm starts automaticly when I boot my computer. Normally, this is no problem, but in some circumstances I would like to quit X (ie to rebuild the kernel). My question is: How do I kill xdm, or prevent xdm from starting at boot-time (so I can start X with the startx command)? To stop it until the next reboot, just issue the command: [prompt]# /etc/init.d/xdm stop To remove it completely from your system: [prompt]# dpkg --purge xdm To leave it installed, but keep it from starting on boot: [prompt]# update-rc.d -f xdm remove If you want it available in some runlevels, but not in others, just modify your symlinks with 'update-rc.d'. See man update-rc.d for details. -Mitch
Re: how to keep output of apt-get in a file ?
In foo.debian-user, you wrote: did someone know if apt-get keep it's output somewhere, or the way to do it ? because the long list of thing to do during the install ? I try : apt-get upgrade 12 apt-get_output.txt but I don't see what is wrong. thanks. Try: apt-get upgrade apt-get_output.txt 21 -Mitch
Re: Linux as NFS...
In foo.debian-user, you wrote: Hi All I was just wondering if there was a way to set Linux up as an NFS server without all the hassle of creating an exports file, etc. etc... Just one that I can install go...??? How would you tell the server which directories to serve? The exports file is already there. Just type the name of the directory you want to server in the file (one per line). This is the minimum that you have to do. Next time you reboot (or just reread the config files) it will be running. How much closer to install and go can you get? -Mitch
Re: email from cracklib cron
In foo.debian-user, you wrote: I have a system that I hadn't bothered adding dev/null to the appropriate place in /etc/cron.daily/cracklib to avoid the email (slink distribution). It's on a seldom used system and I just never bothered. I had hoped that the simple fix would be quickly uploaded and I wouldn't need to bother. Well, it finally got to the point that it was driving me crazy so I looked in the Debain mailing list archive and applied the /dev/null. My question is why hasn't this been fixed? Or perhaps it has and somehow my system just isn't up-to-date? Or was it deemed too simple a fix to warrant a new version of cracklib-runtime? Alas, it was a part of Debian for so long that we must now leave it in for backward compatibility. It is now a standard macro for configure to grep the root mailbox for 'cracklib' to determine if the macro is being run on a Debian system. -Mitch -- The above is both unhelpful and untrue --Mitch Blevins
Re: Somebody's scanning my ports or what?
In foo.debian-user, you wrote: Hi all: To continue my new Linux user paranoia, I have just noticed in xconsole that someone's been trying to connect to every port from port 2 thru 1024. It looks like this: Apr 27 20:03:09 main tcplogd: tcpmux connection attempt from [EMAIL PROTECTED] [206.47.37.4] Apr 27 20:03:09 main tcplogd: port 2 connection attempt from [EMAIL PROTECTED] [206.47.37.4] Apr 27 20:03:09 main tcplogd: port 3 connection attempt from [EMAIL PROTECTED] [206.47.37.4] Apr 27 20:03:09 main tcplogd: port 4 connection attempt from [EMAIL PROTECTED] [206.47.37.4] ... ... Apr 27 20:08:13 main tcplogd: port 1024 connection attempt from [EMAIL PROTECTED] [206.47.37.4] This one was the last. Bellglobal is my ISP provider, and I'm connected via ADSL modem. In between these messages sometimes there are the following (I guess that's when existing service was found): Apr 27 20:03:46 main in.telnetd[7141]: connect from cpu.adsl.bellglobal.com Apr 27 20:04:34 main in.ftpd[7145]: connect from cpu.adsl.bellglobal.com Is this within frames of acceptable. I feel like complaining, but don't want to look like an idiot. :) Any comments highly appreciated! This is not acceptable. This is analogous to some stranger on the street coming up and feeling your crotch. I suggest you contact bellglobal and complain. If that does not work, learn proper counter-measures. -Mitch
Re: A few questions
In foo.debian-user, you wrote: [skip 1 and 2... sorry] 3rd. Anyone know of some good text to speech software thats easy to use? I just need to be able to type something in and have it say it to me, and I need to be able to record the sound. I'm looking for something along the lines of the MS speech SDK stuff. Get festival (apt-get install festival)
Re: man-to-text filter
In foo.debian-user, you wrote: Is there a quick and painless way to translate the output from man into plain text? [prompt]$ man somesubject |col -b somesubject.txt -Mitch
Re: dselect has me baffled..
In foo.debian-user, you wrote: ... or I'm just dense today. I am installing slink from the Cheapbytes 4-CD set. The base package install went smooth. I selected the default dialup packages. However, when I tried to install it I got a lot of messages scrolling up the screen that looked like everything was going ok until I got this message on the bottom line: installation script returned error exit status 1 I have searched through the Debian manual that was included on the CD set and I have looked through the archives but haven't found an explanation for this. Could someone let me know what the problem is and point me to a fix? One of the packages barfed when trying to set itself up. To find out which one it is, rerun the Configure option in dselect. It should have a shorter list of scrolling stuff this time, with the culprit clearly identified. Post the results here for more help. -Mitch
Re: deb vs. rpm
In foo.debian-user, you wrote: So if I understand you and others who have replied correctly, the main advantage is the automatic dep-resolustion via ftp. But it seems to me that this has nothing to do with the deb format itself. Instead it is something that results out of Debian making better use of the features of the packageformat. I guess you could write a program like apt-get for rpm too. As I see it after reading the comparison at http://kitenet.net/~joey/pkg-comp.html the rpm format is comparible with the dep format feature-wise. Rpm is even ahead in some (IMHO important) areas like file-dependencies whereas dep only supports package deps. The area in which dep is better in an important area is recommendations/suggestions. So maybe RedHat (and others) only do not make use of the features that rpm offers, while dep-Packagers do. Also with deps you can be sure that they are comptible with your Debian system, something not the case with rpms. Am I missing someting here? That sounds about right. -Mitch
Re: Hit by virus !? Help, please...
In foo.debian-user, you wrote: after my kid was playing games on win95 yesterday evening i was unable to boot into Linux - actually unable to boot into win95 also... Looks like some kind of virus destroyed boot sector with partition table. i can remember approximate partition sizes and order. Is there any way to recover partition table and the system as well ? You got hit with the CIH virus. It was well publicized at least a week prior to the detonation date (yesterday). All reports that I have seen don't hold much hope for recovering without a full reinstall. (hope you backed up your data..) It is an unfortunate fact that if you want to run Windows on a machine nowadays, you must pay tribute to the antivirus gods or suffer their wrath from time to time. -Mitch
Re: Hit by virus !? Help, please...
In foo.debian-user, you wrote: I'm curious about virii and Linux... Am I wrong to assume that Linux is not immune to virii (I don't even know if virii is a word - but it just sounds cool :) ? Obviously the security features of Linux can prevent some virii from affecting certain files on your system... but what about the boot sector? And what if you happen to be su'd or logged in as root when you get (and heaven forbid) execute an infected program? Is there a need for virus scanning software on Linux? My guess is Linux isn't a targe right now because of it's lack of market share - but as more users realize that Linux is better than Windows (imho), I would imagine that virus software will start appearing in our beloved OS as well. Of course Linux is not immune from virii, but it does have many advantages. As you pointed out, the smaller market share makes it less of a target for the virus writers slaving away in the backrooms of antivirus software companies. Virii are written to be small, stealthy, and to spread without much helpful human interaction. This becomes easier when you have a consistant environment to operate in, such as that offered by the millions of binary equivalent versions of Win95 and Win98 that clutter the desktops of the world. With the diversity of the different GNU/Linux distributions that exist, it becomes harder for the virii to hide/spread. Win9x is like a 10-generation, inbred, backwoods, hillbilly family where a common cold can be introduced and wipe out the whole clan. GNU/Linux has a much more robust gene pool. Because of the Unix security model, spreading of virii is harder. Notice how many more viris warnings you see for Win9x than for NT. Data files in GNU/Linux tend to be common ascii text. This would be much harder for a virus to hide in than the corfortable, dark and damp interior of a MSWord file. (data files are a common way for virii to spread) Since GNU/Linux users are not conditioned to blindly run binary-only programs, they are less likely to comply when they get that fateful email with an attached executable and the spiffy subject line of Cool... run me. Fwd to your friends Also, I would hope that if antivirus software does become necessary for Debian users, some smart people would step up and put the virus-cleaners under a Free license, so we can use apt's auto-web-update capabilities to sleep well at night. -Mitch
Re: A lot of questions
In foo.debian-user, you wrote: I have a lot of question: [snip] 2. When I change mode of *.pl (perl), I can excute it in unix, but I cannot excute it in linux, I must use 'perl *.pl' to excute it. I don't know if I need configure something. More than likely, the .pl file has a shebang (#!) that points to a non-existant interpreter. Either change the: #!/usr/local/bin/perl to #!/usr/bin/perl or put a symbolic link for compatibility [prompt]# ln -s /usr/bin/perl /usr/local/bin/perl 3 I don't want to excute xdm at the boot. When I check my /etc/X11/config, it said no-start-xdm. I don't know how to boot to text start. I have installed hamm, and upgrade to slink. After I upgrade to slink, my debian start xdm at boot time. The no-start-xdm is not used anymore. Actually, the whole /etc/X11/config is no longer used and can be deleted. To keep xdm from starting, either remove the package... [prompt]# dpkg --purge xdm or remove the links in /etc/rc#.d to keep it from starting. Hope this helps, -Mitch -- I must get out of these wet clothes and into a dry Martini. -- Alexander Woolcott
Re: slrnpull won't work on Mindspring.com
In foo.debian-user, you wrote: I am trying out a new ISP and have run into a problem I have not seen before. I wonder if anyone else might have run into this? [snip problem] Just wanted to let you know that you are not alone. (I feel your pain) I cannot list newsgroups by wildcards in slrn using Mindspring. I think they've got something funky with their news setup. -Mitch
Re: abiword troubles: improperly installed fonts
In foo.debian-user, you wrote: Running an old AcceleratedX 3.1 Mostly hamm, with slink and potato stuff mixed in, but dpkg is happy. abiword complains about improperly installed fonts. I tried the following as per the /usr/doc/abiword/unix_fonts.txt.gz file: xset fp+ /usr/share/abiword export ABIWORD_FONTPATH=/usr/share/abiword But still no go. Any suggestions? Install the abiword-fonts package. -Mitch
Re: APT setup (was: Re: An idea...)
In foo.debian-user, you wrote: So I looked at the master lists suggested earlier in this thread and tried to restore the non-US settings that got fubared earlier in my setup. I couldn't get it to work. Here's what I've tried: 1) using dselect, I choose apt as my access method. 2) After retyping the sites that work (note to anyone listening: there really should be an add source option in apt- having to re-enter functional sites is dumb, especially since one error means you have to retype everything) I enter http://http.non-us.debian.org/debian as the URL for the new site, stable as the distribution, and non-US as the component. This is the entry that makes sense (since non-US is a directory in stable, and not the other way around) but is not the entry recommended by apt, which says non-US should be the distribution and main, non-free, etc. should be the components (which makes no sense.) Try http://non-us.debian.org/debian as the URL. -Mitch
Re: APT setup (was: Re: An idea...)
In foo.debian-user, you wrote: On Wed, 14 Apr 1999, Mitch Blevins wrote: Try http://non-us.debian.org/debian as the URL. Thanks, Mitch. It works, and I happily have ssh. I'm still curious, though: why is non-US called a distribution in apt, when it is in fact a component? Is this historical or what? I don't really know, Luis. It seems confusing to me, and picking it from the distribution menu in dselect (apt-method) would create a wrong sources.list line, if I grok the directory layout correctly. But I've been wrong before (once or twice). You may want to ask this question on debian-devel, or file a bug against dselect if you think it is really a bug. I never really used dselect to build a sources.list line (until just now), so I never noticed this before. Maybe there is a good reason for it. *shrugs* -Mitch
Re: on bytes-per-inode Q
In foo.debian-user, you wrote: Hello List, May I know at how many bytes-per-inode is the default debian (slink) install at ? Or, in my already working debian system, how do i know how many bytes/inode is it presently set at ? I am about to add a new hd as my /home and i want to make sure I make2fs them in uniform. I would gamble that your bytes-per-inode is 4096, as this is the default. To verify, first find the number of inodes on your system (as root): [prompt]# dumpe2fs /dev/hda1 |egrep '^Inode Count:' Then, find the number of blocks: [prompt]# dumpe2fs /dev/hda1 |egrep '^Block Count:' Then, find the size of each block: [prompt]# dumpe2fs /dev/hda1 |egrep '^Block Size:' Then, do some math: Bytes per Inode = Block Count * Block Size / Inode Count and it should be close to 4096. Allow some error for rounding, overhead, or crap I didn't think of. I'm sure there is an easier way, but DIIKI* -Mitch *Damned If I Know It
Re: APT setup (was: Re: An idea...)
Jens B. Jorgensen wrote: Mitch Blevins wrote: In foo.debian-user, you wrote: On Wed, 14 Apr 1999, Mitch Blevins wrote: Try http://non-us.debian.org/debian as the URL. Thanks, Mitch. It works, and I happily have ssh. I'm still curious, though: why is non-US called a distribution in apt, when it is in fact a component? Is this historical or what? I don't really know, Luis. It seems confusing to me, and picking it from the distribution menu in dselect (apt-method) would create a wrong sources.list line, if I grok the directory layout correctly. But I've been wrong before (once or twice). You may want to ask this question on debian-devel, or file a bug against dselect if you think it is really a bug. I never really used dselect to build a sources.list line (until just now), so I never noticed this before. Maybe there is a good reason for it. *shrugs* Hmmm, I'm not sure where you guys are coming from. I use: http://conan.eecg.toronto.edu/debian-non-US as the site, slink as the distribution, and non-US as the component and it works just dandy. I'm guessing the above which you guys are using works because of symlinks. I typo'd at the top of this message. The URL I meant to print was http://non-us.debian.org/debian-non-US. But that is irrelevant. The question was... Why does the apt method of dselect included non-US as a choice for a distribution? Please try creating an apt line using dselect to see what I am talking about. -Mitch
Re: Netscape immortal?
In foo.debian-user, you wrote: Hi, I have had Netscape Navigator freeze up a couple fo times lately. I have used xkill on the frozen browzer window, but I'm left with a netscape process running that I am not able to kill without rebooting. I have tried to kill it from root also. No go. Still there! Does anyone have a super-kill solution for this, or I will have to reboot again...sigh :) This happens to me often with Netscape 4.51 and 4.08. Although it is immortal when faced with a windowmanager kill signal, is has always died properly when faced with a kill -9 pid -Mitch
Re: What do I do with tarballs?
In foo.debian-user, you wrote: dselect is great, but what do I do when there is no deb package for the soft I want? I know of course how to install somethng from a tarball, my question is just: where do I do it? Where do I put the package etc. and make sure I am not messing up dpkg? Are there any conventions for that? Is there any doc that explains that? dpkg will not put any files underneatch /usr/local/, so you can install in there (this is usually the default, anyway) without worrying about stomping on dpkg, or vice-versa. -Mitch
Re: file owned by what package?
In foo.debian-user, you wrote: netscape navigator 4.08 keeps dieing on me because of an illegal instruction in libBrokenLocale (apparently). i want to find out what package libBrokenLocale belongs to so i can (up/down) grade and see if it fixes netscape. i'm going nuts without the ability to look at the web. libBrokenLocale is in the libc6 package. You can find the package name for a given file using dpkg -S filename anybody have any ideas? I once thought it would be a good idea to have door hinges on the top of the door instead of the sides, much like a giant pet-flap. However, this turned out to be dangerous for the slower moving people who tried to go through the door, like my grandmother (God rest her soul).
Linux Today article
Looks like our own Havoc Pennington has an article featured on Linux Today... http://linuxtoday.com/stories/4604.html
Re: graphic df type of util for Linux?
In foo.debian-user, you wrote: Until I can afford a new hard drive, I find myself typing df often to see how much free space I have on each partition. I think it would be nice to have a little utility that displays graphs of free space per partition, and updates regularly (a graphic output of df, if you will). I'm visualizing bar graphs, here, but I suppose any type of graph would work. Something that could be swallowed in some sort of desktop module would be especially nice (FvwmButtons, Wharf, ...). So maybe I could have three little bar graphs showing the free space for three partitions I specify. Does anyone know if anything of this sort exists? I don't think it is packaged for Debian yet, but you can use http://nui.vlsivie.tuwien.ac.at/gfdisk/
Re: graphic df type of util for Linux?
Paul Lowe wrote: I don't think it is packaged for Debian yet, but you can use http://nui.vlsivie.tuwien.ac.at/gfdisk/ GNOME has a little applet for the panel that does EXACTLY what you want. But...you'll have to install gnome Yes, ignore the crap I said... I meant gdiskfree, located in the gnome-utils package, which can be found the gnome staging area. But you must... (dramatic pause) install gnome. -Mitch
Re: Announce: 3D graphical install for Debian
It works for me... and I only have an AMD/300. Great work, Ben! In foo.debian-user, you wrote: Yup..you are right. It won't boot on my 486dx4/100. Damn..it's got a sweet OpenGL card in it too. -Ian On Thu, 1 Apr 1999, Ben Pfaff wrote: Following up on the success of my VGA 16-color graphical install for Debian, I am pleased to announce the next generation in installation tools for Debian: a 3D install. Right now this installation is still under heavy development, but it is scheduled to replace the text- and 2D graphics-based installs for the upcoming potato release. The rationale behind a 3D install is simple: to make it easier for our users. With 3D glasses and a VR glove, users will be able to select packages by grabbing them out of a bin of available packages and dropping them into a bin of packages to install. There are some minor additional hardware requirements for this install: a Pentium II/333 or equivalent processor, 64 MB RAM, a 3D accelerated video card. Most newer computers will fit these requirements, and others can be retrofitted easily and cheaply to that level. This amount of hardware is only necessary for the install itself, and may be removed later, so these requirements should be acceptable to everyone. For an early demo of this technology, please try the current development boot floppy on my machine at ftp://pfaffben.user.msu.edu/misc/resc1440-3d-19990401.bin Pleased to bring you the future in Debian installs, Ben.
Re: Handheld-Organizer connected
In foo.debian-user, you wrote: I'm looking for a recommendation for an Handheld-Organizer like SHARP Wizard OZ-650 or YO-470. Does someone know about the support in LINUX or can recommend a supported Organizer ? Palm. I gave up my Sharp 2 years ago and haven't regretted it. Debian has good support for synchronizing and programming the Palm Pilot. You can even run Linux in a simulated PalmPilot on Debian if you want (althought it doesn't really do much yet..) -Mitch
Re: Can apt-get work via a proxy?
In foo.debian-user, you wrote: export http_proxy=http://myproxy:port/; apt-get foo.. You sent me this reply a while ago. Up till now I haven't had to use it, but today I needed to, but had this problem with it: Err http://www.au.debian.org stable/main stow 1.3.2-11 407 Proxy authorization required Failed to fetch http://www.au.debian.org/debian/dists/stable/main/binary-i386/a dmin/stow_1.3.2-11.deb 407 Proxy authorization required E: Unable to fetch some archives, maybe try with --fix-missing? The problem is due to the fact that my http proxy sometimes asks for authorization. Under netscape it pops up a dialog box asking for my name and a password. I am presuming this capability hasn't been built into apt. Any chance that it will be soon? just use http_proxy=http://mylogin:[EMAIL PROTECTED]:port/ This is documented in 'man sources.list' -Mitch
Re: Y2K
In foo.debian-user, you wrote: I get a Slackware 2.0.29 Kernel of Linux. I'd like to know if it's Y2K. If not which version is Y2K. Only Debian GNU/Linux is Y2K compliant (any version). All other distros will fail at the end of this year. Please reformat your Slackware system and install Debian as soon as possible. Also, Emacs is NOT Y2K... you should use vi. Hope this helps, -Mitch -- The above is untruthful nonsense based entirely on personal opinion
Re: What DO you lose with Linux ???
frankie wrote: Mitch Blevins wrote: Call me old-fashioned, but I think email should remain a primarily text-based medium. Nothing irks me more that some nut sending an HTML email to the Debian lists. Don't pretend you haven't done it - everyone's done it b4 they realised. I'm sure its some kind of debian-user rite of passage :-) HTML HEAD TITLE Rebuttal /TITLE /HEAD BODY BGCOLOR=#FF BOLDI have INOT/I!!!/BOLD /BODY /HTML
Re: Suggestion for change to debian package format
Mark Phillips wrote: [snip nice suggestion] This has already been discussed and is being addressed in apt. Here's how it will work (if I understand correctly): Apt will keep a boolean flag called 'Auto' for each package installed on your system. 'Auto' is short for 'Automatically uninstall this package when it is not required anymore because of a dependency'. So, when you install package foo, and apt automatically installs libfoo1 and libfoo2 because they are required by foo, then they will be marked with the 'Auto' flag. Later, if you deinstall foo, then apt will also deinstall libfoo1 and libfoo2, provided they are no longer needed by any other packages. This also allows you to easily have groups of packages. For example, you can have a Gnome package that depends on all the various libraries and packages required by gnome. Later, if you decide the gnome sucks, you can just uninstall the Gnome meta-package and all the other unused libs and packages will go away as well. Please note that both the automatic marking of packages as 'Auto' and the automatic removal of 'Auto' packages will be a configurable option in apt, so you won't be stuck with this behavior. Also, you will be able to toggle the 'Auto' flag directly from any of the apt front-ends. Is that what you were looking for? -Mitch
Re: Suggestion for change to debian package format
Mark Phillips wrote: Is that what you were looking for? Almost. This would work for required packages, but what about suggested packages? Perhaps dselect or the apt front end could behave the same (by default) with suggested packages, with the user being able to turn off the auto flag at selection time. I did leave several issues unaddressed. Mostly because I am ignorant about how apt will handle them. Issues are: *) Do you mark Suggests: links with the Auto flag when installed. *) Do you uninstall a package marked Auto only when it doesn't have any dependencies at all (Depends:, Reccommends:, or Suggests:), or do you only consider the strong Depends: link. I do know that in the interactive frontends, that you will be able to toggle the Auto flag as you are installing the packages. So, Yes, the user will be able to turn off (or on) the Auto flag at selection time (no matter what the default is). -Mitch
Re: security issue
In foo.debian-user, you wrote: I have a question regarding security issue with Debian and Linux in general. By now everyone has probably heard about the new Mellissa virus. I know that this doesn't affect Linux because it is related to M$ products only. However, I just wondered if anything of this sort could happen to a Linux system? I know that Linux in general is actually quite secure, but what makes it so? Just some information about Linux security issue would be appreciated. Thank you for any kind of info. This particular virus is spread because the computer is allowed to run scripts just from the act of viewing a document. This could happen to Linux if we ever had a document format that was both popular enough and able to execute unsafe macros. Depending on the configuration and settings, this could happen if we had a MSWord Viewer that ran macros, a Gnome/Baboon format, or even an HTML/Javascript renderer built into the mail client. Actually, this could happen with any buffer overflow exploit found in a very popular mail client. -Mitch This is why email should be limited to text, and everything else is evil (your turn George ;-) )
Re: Suggestion for change to debian package format
In foo.debian-user, you wrote: On Mon, 29 Mar 1999, Mitch Blevins wrote: Apt will keep a boolean flag called 'Auto' for each package installed on your system. 'Auto' is short for 'Automatically uninstall this package when it is not required anymore because of a dependency'. What about a counter for the number of packages that depends on it. If the count is 0, the package can be safely removed. I think this work even for local packages, since you provide your local package increments the counter on packages it depends on. A counter field is totally independant from the Auto flag. Just because a package has no other packages depending on it does not mean that it should be removed. For example, xpilot is an excellent game, and I wouldn't want it removed... but nothing depends on it. The Auto flag is just a boolean to determine if a package should be removed IFF no other packages depend on it. Having /usr/local packages be calculated into the dependency information can be accomplished using the equivs package. An explicit counter is not needed. So, when you install package foo, and apt automatically installs libfoo1 and libfoo2 because they are required by foo, then they will be marked with the 'Auto' flag. Later, if you deinstall foo, then apt will also deinstall libfoo1 and libfoo2, provided they are no longer needed by any other packages. But, what if libfoo1 || libfoo2 is need by some other package? Hmmm... an ambiguous situation if both libfoo1 and libfoo2 have the Auto flag set. I assume that the implementers would either leave both on the system, or deinstall one of them on a first-come first-serve basis. -Mitch
Re: tar oddness
In foo.debian-user, you wrote: Hi all, I was pondering on the problem of splitting output over multiple mountable media, and discovered the following: $ tar czf foo elmfract $ tar czf - elmfract bar $ ls -l foo bar -rw-r--r-- 1 tgakem users 10240 Mar 29 16:59 bar -rw-r--r-- 1 tgakem users7950 Mar 29 16:59 foo The file bar is a valid archive, but contains trailing garbage. Any idea how this comes about? It does not happen if the z flag to tar is not present. I noticed this when I wrote a small program that takes input from stdin (through a pipe), and writes this to different files on different volumes of a mountable medium (say, floppies). When writing to tape (or pipe) gzip will pad to the default block size. The block size of tar is 10K You can see a similar result by getting a file that zips to just over 10K. I have one here called blah. $ tar czf foo blah $ tar czf - blah bar $ ls -l foo bar -rw-rw-r-- 1 mblevin mblevin 20480 Mar 29 10:34 bar -rw-rw-r-- 1 mblevin mblevin 11406 Mar 29 10:34 foo If you want tar to be efficient for small file sizes, you need to change the default block size... but you must also remember to change it during your untar... This is documented in the man pages for gzip (the padding) and tar (the blocksize) -Mitch
Re: Suggestion for change to debian package format
In foo.debian-user, you wrote: This sounds like a great idea.. I have one question tho: Foo depends on libfoo1 and libfoo2 libfoo1 depends on foostuff1 and foostuff2 So, say I remove foo, will it be smart enough to remove foostuff1 and footstuff2 as well as libfoo1 and libfoo2? I don't know. I am not doing the coding. One would assume that since apt can handle recursive dependencies when installing, it should be able to do the same when removing.
Re: What DO you lose with Linux ???
In foo.debian-user, you wrote: Netscape mail (I know, I know) is already standard in most enterprises. There is also Ishmail (commercial) and others. Qualcomm told me that there will be a Unix version of Eudora when you see Satan skiing down the icy peaks of Hell. What is REALLY needed is mail integrated with other applications. Now I'm curious... What type of integration are you thinking of? I agree that most mail clients should be LDAPized soon. I also think that applications (such as browsers) should allow you to easily pop up an xterm with your favorite MUA. Anything else? -Mitch
Re: What DO you lose with Linux ???
George Bonser wrote: What type of integration are you thinking of? Drag and drop from a spreadsheet, word processor, or graphics program ... the embedable opject idea. Having the mail program be able to directly render some standard wp formats, show graphic items within the document, etc. Drag FROM a email to a spreadsheet, database, whatever. Shared document areas and stuff. Sort of like how exchange and outlook are SUPPOSED to work only make it REALLY work. It is a powerful concept, just klunky in its implimentation and its wandering away from existing standards. Easy handling of attachments would be nice (DnD). But I would have a problem with the direct rendering of graphics, WP, spreadsheets, etc... This sounds like an easy way to spread trojans thru buffer overflows, macros, etc. (Witness the Melissa virus) Also, I don't think the OLE/Baboon paradigm lends itself well to email. Other office apps, yes. But email, no. Call me old-fashioned, but I think email should remain a primarily text-based medium. Nothing irks me more that some nut sending an HTML email to the Debian lists. JMHO, -Mitch
Re: ftp mail
In foo.debian-user, you wrote: Does Debian 2.0 (hamm) installation have an ftp and mail program included to use with the ppp? I have the pon and poff controlling my modem. It dials and makes all sorts of wierd noises but then what. I need ftp to avoid floppy problems in getting netscape and X packages. Thanks, Bill There are many ftp and mail programs included. For the command line, you can use the program called ftp for ftp. And you can use the program called (surprisingly) mail for mail. However, dselect has methods build-in for getting packages. Try running dselect and choosing an access method. From there, you can download and install any number of programs for ftp and mail (Netscape, mutt, etc..) -Mitch
Re: How to Email a Binary file from the command line
In foo.debian-user, you wrote: Hi all I am wondering if anyone knows how to send a binary (.tar.gz) file from the command line? I am trying to write a batch (Sorry my old DOS stuff coming back) program that tars and gzipps up some directorys and then emails them to me. All this for the purpose of backing up my config files. I can get the script to tar up the directorys that I want, and then gzip them, but i can't figure out how to email them. You are looking for the mimeit command in the metamail package. example: [prompt]$ cat mytar.tar.gz |mimeit application/x-gtar [EMAIL PROTECTED] This is the subject of the mail -Mitch
Re: How to Email a Binary file from the command line
In foo.debian-user, you wrote: example: [prompt]$ cat mytar.tar.gz |mimeit application/x-gtar/ [EMAIL PROTECTED] This is the subject of the mail Cool, it worked, I think. When it gets to the other end i get several messages all with attachments called Part 1 Part 2. How do I get my original file back again at the other side? (Im in windows using netscape if it makes a difference - if I can just find out what has been done to it maybe I can get it back!) It splits the file up into chunks and sends it one chunk per mail because some mail transports will reject mail over a certain size. I don't know how netscape re-assembles these chunks. You can try making the default chunksize larger so that it is not split. To do this, you can add the -s option to the end of the command, along with the maximum message size. For example: [prompt]$ cat mytar.tar.gz |mimeit application/x-gtar/ [EMAIL PROTECTED] Subject -s 1000 will not split anything unless it is bigger than 10 Megs. mimeit just calls the splitmail command, so see man splitmail for more details. -Mitch
Re: lilo for 96Mb system
In foo.debian-user, you wrote: In earlier e-mail to the list someone advised that I could add the line append=(mem=96M) to /etc/lilo.conf to get the kernel to recognize my newly added memory. When, having done so, I run lilo, I get [snip] append=mem=96M
Re: Staging Areas.
Michael Beattie wrote: On Wed, 24 Mar 1999, Mitch Blevins wrote: In foo.debian-user, you wrote: Can the developers subscribed please tell me/us where the various staging areas are? I keep losing the various posts that have the urls (stupid me). (i.e. gnome, apt, whatever) And if possible, an apt sources.list line too. There is only _one_ (to my knowledge) non-standard location that you need to remember now. That is deb http://www.debian.org/~jules/gnome-stage-2 unstable main This houses all the gnome stuff. The newest apt is now in potato, and gnome-apt is in the staging area above, so this is the only location needed. Thanks.. thats what I am after.. Oh, what about E 0.15? E 0.15 debs are also in the above staging area.
Re: mySQL
In foo.debian-user, you wrote: hi, I am looking for mySQL source. It's not easy for me to download it from internet. in fact, I just can use email to relate with internet. please email a mySQL for linux source to me. thanks. my email: [EMAIL PROTECTED] Hi Chen, The address for mysql source is ftp://ftp.debian.org/debian/dists/stable/non-free/source/devel/mysql_3.21.33b.orig.tar.gz Usually sources are too big for them to be emailed. However, I remember that ORA used to provide an ftpmail service for people who had email access, but no ftp. Here's how it works: Send a message to [EMAIL PROTECTED] with no subject and the word 'help' in the body of the message, and you will get full instructions. Usually, you just put the ftp commands in the body of the email... for example: To: [EMAIL PROTECTED] Subject: reply [EMAIL PROTECTED] connect ftp.debian.org chdir /debian/dists/stable/non-free/source/devel dir binary uuencode get mysql_3.21.33b.orig.tar.gz quit And you will get a uuencoded mail(s) with the desired file. If you don't know exactly where the file is, you can just use the dir commands to search, or most sites have an ls-lR file in the root directory that lists all files contained in the archive. Hope this helps, -Mitch
Re: Coupla quick questions...
In foo.debian-user, you wrote: 1.) I somehow managed to delete my /var/log/news directory and was getting boot errors. I apparently fixed it by recreating the directory as root but that made the /news subdir owned by root:root. Is this correct or should it be owned by root:news? [promt]$ ls -ld /var/log/news drwxr-xr-x 2 news news 1024 Sep 12 1998 /var/log/news
Re: Kernel compile errors
In foo.debian-user, you wrote: Running potato im coming up against the following with kernel 2.2.4 acct.c: In function `sys_acct': acct.c:197: too few arguments to function `filp_close' acct.c:203: too few arguments to function `filp_close' make[2]: *** [acct.o] Error 1 make[2]: Leaving directory `/usr/src/kernel-source-2.2.4/kernel' make[1]: *** [first_rule] Error 2 make[1]: Leaving directory `/usr/src/kernel-source-2.2.4/kernel' make: *** [_dir_kernel] Error 2 Well known error. Either disable BSD-style process accounting, or apply the patch that you should be able to find on the kernel mailing list. (Basically, just add a 'NULL' argument to the 'filp_close' functions at lines 197 and 203) -Mitch
Re: Staging Areas.
In foo.debian-user, you wrote: Can the developers subscribed please tell me/us where the various staging areas are? I keep losing the various posts that have the urls (stupid me). (i.e. gnome, apt, whatever) And if possible, an apt sources.list line too. There is only _one_ (to my knowledge) non-standard location that you need to remember now. That is deb http://www.debian.org/~jules/gnome-stage-2 unstable main This houses all the gnome stuff. The newest apt is now in potato, and gnome-apt is in the staging area above, so this is the only location needed. -Mitch
Re: VB and Active X
In foo.debian-user, you wrote: Someone I know who is learning programming sent me a link to a web page they were working on that had VB and Active X controls. I couldn't see any of it in netscape. Why not? Is it possible to see this stuff on a Linux box? Are LInux users going to be cut off from web based application written in MS languages? cut off is such strong language. I prefer protected. ActiveX and VBScript should not be used for web applications meant for general consumption. Doing so risks not reaching a good percentage of the target audience because: 1) It is not cross-platform. 2) ActiveX has security risks, which prompts many security-conscious users to disable it. Your friend should be using Java/Javascript for client-side interactivity if web-programming is of interest to him. -Mitch
Re: Trying to understand Packages[.gz] files
In foo.debian-user, you wrote: I think this got lost - apologies if this is its second time around. I'm installing Debian gradually; until I get it online I have Windows, in its own partition, set up for Internet use. I use Windows to download Debian packages I'm interested in onto my hard drive, creating a partial mirror of the Debian distribution; I then install those packages by accessing the Windows partition from Debian. What's the significance of Packages (or Packages.gz) files in this case? 1) Does dpkg/dselect/apt assume that all packages listed in a Packages file are available on my hard drive? If so, should I manually create a stripped-down Packages file that describes only the few dozen packages I've downloaded? Yes, and yes. Use dpkg-scanpackages to create the new Packages.gz file.
Re: Where's the CGI bin?
In foo.debian-user, you wrote: I have a normal apache installation; where is the cgi-bin? /usr/lib/cgi-bin
Re: Too many subdirectories
In foo.debian-user, you wrote:
--zYM0uCDKw75PZbzx
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
I made a c program :
#include stdio.h
main()
{
int nbr=3D0;
while(1)
{
printf(%d\n,nbr+=3D1);
mkdir (x);
chdir (x);
}
exit(0);
}
My problem is that there is now too many subdirectories (10,000) and rm said
Memory exhausted when I type=20
rm -rf x
Does anyone know of a C script which can suppress all these directories ?
Thanks.
#include stdio.h
main()
{
while(1) {
if(chdir(x) {
printf(Found the end of these confounded directories.\n);
while(1) {
chdir(..);
if(rmdir(x)) {
printf(Back where we started.\n);
printf(I'll never do _that_ again!\n);
exit(0);
}
}
}
}
exit(0);
}
Re: Two quick questions
In foo.debian-user, you wrote: 1. Is Gnome a window manager for X, or is it a replacement for X? (or it it something else?) Something else. You can use your existing WM with gnome. Gnome just provides libraries for applications and also several utility apps. The gnome-panel is an example of a utility app. Also, the gnome file manager (gmc) will allow desktop shortcuts and the ability to use your desktop as a file repository (like Win95). The gnome libraries also support things like drag-n-drop between applications compiled against them, etc. 2. What is the advantage of using a kernel source package in .deb format as opposed to a generic kernel source in tar.gz format? zless /usr/doc/kernel-package/Rationale.gz -Mitch
Re: Two quick questions
In foo.debian-user, you wrote: Just curious - is there a doc that lists comparisons between GNOME and KDE? Just to to dejanews and do a search under 'flame'. ;-) Seriously, I don't know of any document that does that exactly, but you can read one author's opinions at http://linux.miningco.com/library/weekly/aa03149a.htm?pid=2801cob=home -Mitch -- Any command with less than 48 switches is a Cat in the Hat book. - E. Charters
Re: Enlightenment 0.15 .debs
In foo.debian-user, you wrote: On Sat, 13 Mar 1999, Brian Almeida wrote: Enlightenment 0.15 .debs are out. They will not be uploaded into the distribution, they are placed in the GNOME staging area. This is so that Where's the GNOME staging area? For those of us who like to live dangerously... It's located at http://www.debian.org/~jules/gnome-stage-2/ You can add an entry in your /etc/apt/sources.list like so: deb http://www.debian.org/~jules/gnome-stage-2 unstable main -- Any command with less than 48 switches is a Cat in the Hat book. - E. Charters
Re: I've been cracked! (hamm, 2.0.35)
In foo.debian-user, you wrote: Somebody (through jhb60.jaring.my) wandered into my system, set up a user account for themselves and set up a couple of programs, eggdrop and smurf. I've not been using encrypted passwords, I understand that there are ways to derive the salt that the passwd file uses? The salt is simply the first to characters of the password hash. Assuming you don't use shadow passwords, look in /etc/password at a line: myusername:eNdOfjsu/dsk: The eN is the salt, and it should be different for each password entry. Anyway, this person hid a few files in some interesting places and even replaced my syslogd. Now, when I say hid a few files, there are files that simply don't show up by ls. You can manipulate them but you can't see them unless you ls the entire path. For example, $ ls /usr/lib/fms returns /usr/lib/fms but $ cd /usr/lib;ls fms returns nothing. I have no idea how many files or directories might be hidden this way, nor how I can find out. I've obviously changed passwords and disabled everything foreign that I can find, any suggestions as to what I should be doing about this? A common technique is to replace system commands (ls, ps) with hacked versions that behave in ways that make it difficult to spot/stop the intruder, or that provide a backdoor for the intruder to re-enter. You have been compromised, and you need to reinstall completely. Some data files can be salvaged, but every program must be replaced. Investigate using the tripwire package to detect these things in the future. -Mitch -- Any command with less than 48 switches is a Cat in the Hat book. - E. Charters
Re: still no luck with apt-get
Nathan E Norman wrote: I maintain that particular mirror. [snip successful apt output] Could you have a library problem? The problem appears to be with 206.187.92.15 [prompt]$ nslookup http.us.debian.org Server: ns2.mindspring.com Address: 207.69.188.186 Name:http.us.debian.org Addresses: 208.146.80.105, 206.187.92.15, 209.197.224.62, 207.69.194.216 141.213.4.21 I explicitly set the line in /etc/apt/sources.list to each of the addresses in turn and ran 'apt-get update'. Your mirror, Nathan, was OK. In fact, all of them were working fine except for 206.187.92.15 which gave the offending error messages. Does anybody know who runs this mirror? Can we just drop it from the DNS record? Thanks for your help (and all the others who pointed to mirrors) -Mitch
Re: nfs
You also might try specifying the nfs version in your mount command. In foo.debian-user, you wrote: Howdy: Make sure that the DNS is active on one machine or the other, and that the names are still set up properly... I'm assuming (should I?) that you are running a tcp/ip based net... Also: You may want to try to issue the mount command using the command line option for IP address (I may be thinking about smbmount??) HTH, Brant Under hamm, I could successfully export a file system from a i386 machine running hamm to a SGI running IRIX 6.5 so that the command mount debian:/home /mnt issued on the IRIX 6.5 machine worked giving output something like this. kakapo 16# mount blitzen:/home /mnt mount: blitzen:/home server not responding: Program/version mismatch; low version = 1, high version = 2 NFS version 3 mount failed, trying NFS version 2 mount: blitzen:/home server ok Now when I upgrade the i386 to slink and try the same thing on the SGI I get akapo 19# mount lira:/home /mnt mount: lira:/home server not responding: Program not registered NFS version 3 mount failed, trying NFS version 2 mount: retrying /mnt NFS version 3 mount failed, trying NFS version 2 and it keeps on trying this for ages. Does anyone know a solution to this problem?
Re: Apt 0.3 lost ftp-method ???
In foo.debian-user, you wrote: According to Bob Nielsen [EMAIL PROTECTED]: gnome-apt requires apt ver. 0.3, which does not (yet) support the ftp method. Your options are to use the http method or to downgrade to an older version of apt. The http method works very well and the number of sites which support it is increasing. Thanks for the hint. You are right, substituting ftp by http worked...almost. There seems to be a problem with the non-US part. Here is what I get to see: ~# apt-get update [snip] Err http://ftp.de.debian.org slink/non-US Release 404 Not Found [snip] Why does the http method need the Release file? And why is the one for the non-US part missing? Should I file a bug report? This error is just an eyesore. It will not affect the functionality of apt. You can safely ignore it. I don't know what the timeframe is for all sections and mirrors to have 'Release' files. -Mitch -- X sucks ... The beauty of X is that it sucks independently of the OS. -JC Jaros
Re: still no luck with apt-get
In foo.debian-user, you wrote: I still can't get apt-get update do do anything useful. Anyone have a clue? Here's my /etc/apt/sources.list # Use for a local mirror - remove the ftp1 http lines for the bits # your mirror contains. # deb file:/your/mirror/here/debian stable main contrib non-free # See sources.list(5) for more information, especial # Remember that you can only use http, ftp or file URIs deb http://ftp1.us.debian.org/debian stable main contrib non-free Which apt version are you using? Which distro? (hamm, slink, potato)
Re: mail filtering
In foo.debian-user, you wrote: Hi, I am trying debian after using redhat for awhile and have a question about procmail. With redhat I use fetchmail to download my email and as long as I have a procmailrc set up in my home directory my mail would be filtered automatically. I am using smail with debian since it was the default and actually the setup during install is very slick. The problem is my mail is not being filtered. Do i need A .forward file with debian since I am using smail as opposed to sendmail?? Yes, put a .forward file in your home directory with the folloing line: |exec /usr/bin/procmail -- X sucks ... The beauty of X is that it sucks independently of the OS. -JC Jaros
Re: still no luck with apt-get
Matt Garman wrote: On Thu, Mar 11, 1999 at 04:56:23PM -0500, Mitch Blevins wrote: In foo.debian-user, you wrote: I still can't get apt-get update do do anything useful. Anyone have a clue? Here's my /etc/apt/sources.list # Use for a local mirror - remove the ftp1 http lines for the bits # your mirror contains. # deb file:/your/mirror/here/debian stable main contrib non-free # See sources.list(5) for more information, especial # Remember that you can only use http, ftp or file URIs deb http://ftp1.us.debian.org/debian stable main contrib non-free Which apt version are you using? Which distro? (hamm, slink, potato) I'm using apt version 0.1.9 in the slink distribution. I am trying to update from that mirror without much luck. Try changing the line to be deb http://http.us.debian.org/debian stable main contrib non-free and see if that helps. At least then we can isolate if it is the mirror or your system. -Mitch
Re: still no luck with apt-get
Matt Garman wrote: I am trying to update from that mirror without much luck. Try changing the line to be deb http://http.us.debian.org/debian stable main contrib non-free It appears to be my system :( Same errors: No, it's not your system. I just checked and ftp1.us.debian.org is just an alias for http.us.debian.org. Also, when I try to update from http.us.debian.org I get errors almost identical to yours... and I am running apt_0.3.0 My best guess would be a mirror problem. Any body know some *working* mirrors to test? -Mitch
Re: Custom headers -- outgoing email
Frozen Rose wrote: Yes, Exim will normally write a Sender: header, (although I didn't think it did if one was supplied... hmmm...) Yes, it will rewrite them to match the actual sender. See section 44.11 of /usr/doc/exim/oview.txt [snip] However, if you must change the outgoing sender lines, try adding this line to your rewrite configuration: [EMAIL PROTECTED] [EMAIL PROTECTED] Which will change all occurrences of [EMAIL PROTECTED] to [EMAIL PROTECTED]. You can also use the rewrite rules to rewrite @localhost if you can't get rid of it, although you should be able to. /usr/sbin/exim -brw addr will test the rewriting rules (exim -brw [EMAIL PROTECTED] for example) How do you tie a rewrite rule to a particular route? The documentation wasn't very clear on this. It seems that aaron would want it to only rewrite the sender header going through the smarthost route. Curious, -Mitch
Re: Network Problem
In foo.debian-user, you wrote: I have a small network that looks like: workstation running Linux and Win95 ethernet hub ethernet server runing linux modem isp The server runs ipmasquerade and works fine. The problem is that whenever I login to the server from the ws, no connection is made untill the modem connects to the isp. I first thought the problem had do do with loging in using the server name, but even if I use the ip address as in rlogin 192.168.0.1, same problem. Any suggestions on what causes this behavior. Just a guess. rlogind requests the clients hostname from the source address (see man rlogind). If the client's hostname is not listed in the /etc/hosts file, then it will try to get it from the default nameserver, which I assume is on the other side of your modem at the ISP. The quick-fix is to add your Linux/Win95 client's hostname as an entry in the /etc/hosts of the server, assuming that you don't have any DNS servers running on your local network. -Mitch
Re: what happened to irc.debian.org?
In foo.debian-user, you wrote: I've been trying to connect to irc.debian.org for some time now. It appears to be down. Does anyone have the lowdown on this? I have a feeling the developers are busy getting slink ready. For some reason, irc.debian.org doesn't work thru an IPMASQ connection. Try connecting instead to irc.openprojects.net, #debian. -Mitch
Re: Help. How do I reconfigure my the network?
In foo.debian-user, you wrote: Please I need some help. I have Debian GNU/Linux running on my Sony VAIO PCG-505G which I carry around to different offices. I need to connect to the networks in the different offices so I can download my e-mail and stay in contact with the rest of the world. When I installed Linux I set up the network to work in my main office. It works like a dream. Now I am down in another office, and I can't get the network to work right. I have changed the following files to fit the new network: /etc/hostname /etc/hosts /etc/networks (Did not make any difference at all. Is still looking at the network in the main office.) /etc/resolv.conf Since it did not work I deleted the IP address and default router using route and added a new IP address and default route. After this I was able to ping all the machines in the office, but not the router nor any machines outside the office. Based on this, I am getting the impression that the problem lies in the set up of my default router. Look in /etc/init.d/network Also, check your mailer. It seems to be sending html. -Mitch
Re: Network doesn't work
In foo.debian-user, you wrote: Mitch Blevins wrote: The newer NetGear cards ship with a floppy that contains their own version of tulip.c which is not necessarily the same tulip.c that is included in your kernel tarball. Replace the tulip.c in the kernel tree with the one included on the floppy and recompile, reboot, etc. Maybe that will work. -Mitch Are these decent NICs? I'm planning a home network, and I saw NetGear cards for about $30 US each. I read that Linux was one of the supported OSes, and read on their website that they had their own tulip drivers. I've heard a lot of recommendation for tulip-based NICs. I've never had problems with them. They're fast and dirt cheap. I have also heard many other people sing praises to NetGear. -Mitch
Re: Strange 'find' result
In foo.debian-user, you wrote: Can somebody explain this to me? $ find /cdrom -iname wx* $ find /cdrom -iname wxx* /cdrom/debian/hamm/hamm/binary-i386/libs/wxxt1_1.66d-2.deb Why does the first 'find' query give no results? Are you quoting the argument to avoid shell expansion? $ find /cdrom -iname 'wx*'
Re: Network doesn't work
In foo.debian-user, you wrote: AAAH!!! Well I made it back from Frye's and bought a Netgear Fast Ethernet NIC, a Netgear Fast Ethernet Hub, and I'm still using the Linksys pcmcia fast ethernet NIC for my laptop. I recompiled my desktop kernel to recognize the Tulip module for the new NIC. It was detected, etc. I ifconfiged (I think correctly because no errors). Now when I telnet from my laptop to my desktop, I get 100% packet loss. When I switch the configuration back to the way it was it ranges from between 88% and 97% packet loss. Do I go buy new cables now? Is there a way to fix this. Thanks everyone. Happy Valentine's Day. The newer NetGear cards ship with a floppy that contains their own version of tulip.c which is not necessarily the same tulip.c that is included in your kernel tarball. Replace the tulip.c in the kernel tree with the one included on the floppy and recompile, reboot, etc. Maybe that will work. -Mitch
Re: Slink, xbase-common and xdm...
Hogland, Thomas E. wrote: Interesting - if I can ever figure out how to get into the system I'll do that :-) There's no way to shut down and reboot clean - xdm won't do anything till you log in g... Can't you just Ctl-Alt-F2 to another console? -Mitch
Re: How to find files by text/subdirectories
In foo.debian-user, you wrote:
Lance Hoffmeyer [EMAIL PROTECTED] writes:
What is the easiest way to locates files (say HTML) by text in their
documents?
find ~/somedir -iname *\.htm* -exec grep -i some text {} \; -print
Also look at the 'rgrep' package.
rgrep some text /somedir
-Mitch
Re: svgalib
In foo.debian-user, you wrote:
I have now decided to use svgalib but already have run into trouble !
/* programme svgatest.c */
#include stdio.h
#include vga.h
main()
{
/* code */
}
#: gcc svgatest.c -O2 -lvga
Programme won't compile because
a) Can't find vga.h
b) Can't find vga library
Yet dselect claims svgalib is installed. Do I need to install svgalib-bin ?
I have vga.h in the usr/src tree - is this where it should be ?
Where is the vga library that needs to be linked ?
Thanks for your help.
Install svgalibg1-dev
Re: Wooooooo Hooooooooooo!!!!!!!
In debian-user, you wrote: [snip poorly formatted message] Please do not post messages in html format. And also please do not post lines over 76 columns wide. It makes it difficult for most folks on this list to read/respond to your messages. Hope your next problem/solution goes as well for you as the last. :) -Mitch
Re: Where is emacs and vi command line editing set?
In foo.debian-user, you wrote: A long long time ago my root account mysteriously switched to using vi command-line editing instead of the default emacs, and I've been too lazy to get around to fixing it until now I guess. Where is this behavior configured? Christopher Mine is set in ~/.inputrc with a line like... set editing-mode vi
Re: suspicious connections
In foo.debian-user, you wrote: I have a stand-alone machine, with dialup ppp connection (using diald). I think someone was trying to hack me today, and I'd like advice on how to find out whether they succeded, and what to do about it. I'd also appreciate suggestions on the easiest way to prevent, or at least monitor, such activity in the future. You can get the iplogger package, which will log every tcp connection made to your machine. The lsof package is also useful for finding out if a service is running on a port on your own machine. I believe lsof is kernel version dependent, so you may have to expirement some example: [bash]$ lsof -i :22 COMMAND PID USER FD TYPE DEVICE SIZE/OFF INODE NAME sshd32211 root6u inet 0x0149ac0c 0t0 TCP *:ssh (LISTEN) Also, if you are paranoid, I would suggest getting the tripwire package. This will monitor your system for changed system files. -Mitch
