Re: export won't work
On Sun, 01 Sep 2002 19:04:07 +1000 Russell <[EMAIL PROTECTED]> wrote: > Why won't export work? When i run this > bash script on debian 3: > > #!/bin/sh > ABC=def > export ABC > exit 0 > > when i type 'export', then ABC doesn't > appear in the list of environment variables. That's unsurprising. Environment variables are passed to _children_, not parents. So when you run that script, it creates an environment variable ABC, then exits. It's got no children at all :) I think what you want, is to source the file. 'source /path/to/script'. That'll load it in the current environment/process. Of course, with that 'exit 0' there, it'll also exit you from your shell ;) Might want to get rid of it. msg00205/pgp0.pgp Description: PGP signature
Re: netmasks
On Sat, 21 Sep 2002 14:33:06 + Martin Clarke <[EMAIL PROTECTED]> wrote: > a question i was asked recently on a topic that I thought > i understood untill i was asked > > Given the hypothetical subnet 136.206.16.128 and netmask > 255.255.255.128, state the valid range of IP addresses that could be > assigned to hosts on the subnet?(also another was 64.122.34.83 & > netmask 255.255.255.240 :( ) > > i was used to looking at plain netmasks ie 255.255.255.0(ff00) etc > > which I understand fine but my grasp of anything else is less than > good... > > any help is appreciated.. 'apt-get install ipcalc; ipcalc 136.206.16.128/255.255.255.128; ipcalc 64.122.34.83/255.255.255.240' :) msg03085/pgp0.pgp Description: PGP signature
Re: Lost Mozilla
On 22 Sep 2002 14:41:36 +0100 Brian Webb <[EMAIL PROTECTED]> wrote: > Help,Debian Newbie.Have installed Debian 3.0 OK. Problem, Mozilla was > working OK. Now I cannot launch it, I get message 'No such file etc'. > Tried apt-get to install &/or remove it, but can neither install or > remove it.I'm now stuck! > Thanks in advance. The package you want (the one that contains the 'mozilla' command) is mozilla-browser - are you sure you have it installed? If so, what command are you running (exactly), and what's the (exact) error? msg03086/pgp0.pgp Description: PGP signature
Re: keeping checksums of every file installed
To quote Robert Waldner <[EMAIL PROTECTED]>, # # Hi! # # I´ll soon have to give out the root-pw of one of my boxes temporarily, # but I´m a little paranoid... # # Does anybody know of some program/script that could make checksums of # each&every file installed and keeping this list somewhere safe so that # I can compare it when I take the password away from my colleague to # make sure there´s no trojan or something installed? # # (The joke is that there´ll be a security audit and they want to have # the root-pws for that...) I think something along the lines of; find ./ -type f | xargs md5sum > outputfile.txt Would work. Later, do the same, and then 'diff -u outputfile.txt newoutputfile.txt', to see the differences. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: upgrade to a lib in testing
To quote "Robin Rowe" <[EMAIL PROTECTED]>, # Hi. I don't seem to have it right. I get an error on 'apt-get install # libsdl1.1-dev': # # Reading Package Lists... # Building Dependency Tree... # E: Couldn't find package libsdl1.1-dev Don't read the other reply ... Use deb-src, trust me. Here's what you do; 1) Temporary add a deb-src entry pointing to testing. You've done this already. 2) As root, do 'apt-get -b source libsdl1.1'. That'll build libsdl1.1 .debs. 3) As root, do 'dpkg -i <.debs>', where <.debs> are the names of the .deb files generated by 'apt-get -b source'.. If you just add a 'deb' line pointing to testing, and then 'apt-get install something' from it, you're going to run into big problems, trust me. And you'll get little sympathy :( David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: IMPS/2 problems after kernel 2.4.1 update
Okay, first of all, please use paragraphs. They exist for a reason :) Since it's too much trouble for me to break things up to present a nice clear, response, you'll have to deal with my babbling. First of all, if your mouse isn't initialized as an IntelliMouse, it pretends it's a regular PS/2 mouse. Second, gpm gets to your mouse before X, so you have to be positive that your mouse type is set to "imps2" in /etc/gpm.conf . Third, it sounds like both GPM and X are trying to use /dev/psaux. Check to see what /dev/mouse is linked to. Since you're using GPM and X, specify explicitly in XF86Config[-4] that the mouse device is /dev/gpmdata, not /dev/mouse . Now, try rebooting, after making sure all those config options are set properly. If it still doesn't work, set X to use /dev/psaux, type ImPS/2, 'rm /etc/rc2.d/*gpm' to stop 'gpm' from being started on boot, and then re-boot. Hopefully that will work. If it doesn't, you have problems that arn't caused by X or gpm. The reboots are necessary, and they have to be hard-boots. This is because of the mouse initialization thing. Hope that helps :) To quote Dave Bresson <[EMAIL PROTECTED]>, # # # Hi there, # # # I'm having a problem with my Microsoft Intellimouse (an intellimouse with # two buttons and a scroll wheel that uses the ps/2 interface) since going # from a 2.2.x kernel to 2.4.1. The way i have setup (and how everything # worked perfectly before this upgrade) was that gpm was set to read from # /dev/psaux using the imps2 protocol. It then repeats this to /dev/gpmdata # which i then have /dev/mouse linked to. My XF86Config-4 is setup so that # it knows the mouse is at /dev/mouse, and uses the imPS/2 protocol. I also # have a line for the Zaxis stuff to get the scroll wheel to act like # buttons 4 and 5 in X so that the wheel works. Anyway, all this is well # and good because it was this exact setup that worked before the new # kernel. Anyway, since upgrading the kernel, gpm no longer works on # /dev/psaux with the imps2 protocol (hence, X doesn't work either...). The # mouse just goes all crazy moving all over the place, and pushing random # buttons and what not. So i set gpm and X to use the regular ps/2 # protocol, and everything was fine, however i get no third button or # scroll wheel anymore, which sucks. Just in case this was just a gpm # problem, i tried having X read directly from /dev/psaux using the IMPS/2 # protocol, but unfortunately got the same bad result. Does anyone have any # idea what might be the problem here? It had to have been something i # configured wrong in the kernel...however i don't know what, since it just # made the IMPS/2 stuff not work, while the regular PS/2 still works...very # strange. Any ideas, or tips? # # # thanks, # # # # dave # # # # # -- # To UNSUBSCRIBE, email to [EMAIL PROTECTED] # with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] # David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: IMPS/2 problems after kernel 2.4.1 update
To quote Dave Bresson <[EMAIL PROTECTED]>, # On Fri, 16 Feb 2001, David B. Harris wrote: # Yeah, sorry about that...my babbling doesn't translate well to paragraphs. # Ah well... :) # > First of all, if your mouse isn't initialized as an IntelliMouse, it # > pretends it's a regular PS/2 mouse. # # Right. What do you mean by 'initialized'? You mean initialized as in # what gpm.conf and XF86Config-4 have it set as? Or something else? Well, it depends on what gets to the mouse first. If X gets to the mouse first, then it'll be initialized as whatever XF86Config-4 has it set to. If gpm gets to it first, it'll be initialized to whatever gpm.conf thinks it is. Anyways, I'm sure this is all very confusing, and it took me nearly a month to figure out what was going on :) Basically, you *have* to make sure that whatever gets to the mouse first initializes it to the highest degree possible. The means using the protocol that most closely matches what the mouse actually uses. # > Third, it sounds like both GPM and X are trying to use /dev/psaux. Check # > to see what /dev/mouse is linked to. Since you're using GPM and X, # > specify explicitly in XF86Config[-4] that the mouse device is # > /dev/gpmdata, not /dev/mouse . # > # # Well, since X wants to use /dev/mouse , i made it a sym link to # /dev/gpmdata (since that is what gpm is set to repeat to) I guess i # probably shoulda stated that it was just a link in my previous email. I had thought this ... but since 'gpm' and X were conflicting, I wasn't sure. Are you using devfs with devfsd? Double-check the /dev/mouse is a symlink to /dev/gpmdata. Using devfsd, changes arn't always kept over a reboot. # > Now, try rebooting, after making sure all those config options are set # > properly. If it still doesn't work, set X to use /dev/psaux, type # > ImPS/2, 'rm /etc/rc2.d/*gpm' to stop 'gpm' from being started on boot, # > and then re-boot. Hopefully that will work. If it doesn't, you have # > problems that arn't caused by X or gpm. # # Okay, i had tried this before, however it was without the reboot, so i # tried it again. As it turned out, this fixed it. Pointing X straight to # /dev/psaux, and not starting gpm at all did it. Which leads me to believe # that gpm doesn't work with the new kernel 2.4.1 i set up, since the old # configuration worked just fine with 2.2.18...very strange. Any ideas on # why this might be? Yup, I figured a reboot would fix it. :) It's that bloody initialization thing. I don't think it's a problem with gpm and 2.4.x. I'm running 2.4.1-ac15, with devfs, but without devfsd and it works fine. I think that if gpm.conf is set to use the imps2 protocol, it should work. Just to humour me, try explicitly stating in /etc/X11/XF86Config-4 that X should use /dev/gpmdata. Make sure 'gpm' starts at boot time, and re-boot. If that doesn't work, feel free to contact me for more help :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Antwort: Rebooting is foolish ....
To quote William T Wilson <[EMAIL PROTECTED]>, # On Fri, 16 Feb 2001 [EMAIL PROTECTED] wrote: # You don't need to reboot to change the hostname, either. The command is # 'hostname'. # # You need to reboot to change the partition table of a disk with mounted # filesystems, and you need to reboot to recompile the kernel, and you need # to reboot for hardware upgrades. That's about it, really... You also need to re-boot for some hardware re-initialization. See the recent thread on XFree 4.0.2 and an IntelliMouse-compatible mouse. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: ssh and devices
To quote Rick Rezinas <[EMAIL PROTECTED]>, # Hi, # # I'm using testing # I was having an issue (think I saw a post regarding this recently), # where a regular user couldn't use ssh due to ' no controlling tty' # # I found that the issue was an inability for nonroot to write to /dev/tty # and changed o+w for /dev/tty # # Now all seems happy, but I was curious if this is actually a good fix or # creates a security issue...and why this would've been changed. Also, what # packages are in charge of which devices...is there a debian standard for this # or do package maintainers just create and modify devices as they see fit? I think the "correct" fix would be to change the group of all the tty* devices to the "tty" group. Then, when you log in, it gets temporarily changed to your own group ID. So, permissions should be rw-rw, with "tty" as group, "root" as owner. I think. :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Antwort: Rebooting is foolish ....
To quote Carel Fellinger <[EMAIL PROTECTED]>, # On Fri, Feb 16, 2001 at 04:06:03PM -0500, David B . Harris wrote: # ... # > You also need to re-boot for some hardware re-initialization. See the # > recent thread on XFree 4.0.2 and an IntelliMouse-compatible mouse. # # I doubt it. Any trouble I've had with PS2 mouses not being properly # initialized could be solved by "/etc/init.d.gpm restart" preceded by # the occasionally physical reconnecting of the mouse. But YMMV:) Don't doubt it. I didn't just make it up to prove that a reboot *could* be required to re-initialize hardware :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Modules, modprobe, and 2.4.1 kernel
To quote "Dean A. Roman" <[EMAIL PROTECTED]>, # Hello all, # #I recently configured, compiled, and installed the 2.4.1 kernel. So # far things are going ok with it. # # However, modprobe has stopped working correctly, and can't find any # modules. Chances are, you need a newer version of modutils. Here's a quick way to get the appropriate version, made for your particular platform(Debian Potato, I'm guessing). 1) Add a deb-src line in /etc/apt/sources.list pointing to "testing", as opposed to "potato" or "stable". Remove this line when you're done. 2) As root, 'apt-get -b source modutils'. 3) As root, 'dpkg -i '. There you go :) If you read /path/to/kernel/sources/Documentation/Changes , you'll see the versions of all the programs you need to run a 2.4.x kernel. # Thanks for any help you can give! No probs. In return for this help, please refrain from posting your question multiple times to the mailing list. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Where are the iso images for woody?
To quote <[EMAIL PROTECTED]>, # I guess the subject is clear enougth, does woody come with XFree4.0 and # kde2? There are currently no Woody ISOs. Whether Woody will have XFree86 4.0.x or KDE2 will be determined when Woody is frozen, or when the part of Woody that those packages are in is frozen. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Kernel compile comments
# Comments or suggestions welcome but it looks like I have wasted my time and # should have stayed with RH With an attitude like that, you should have stayed with Red Hat. Or better yet, Windows. When you having nothing installed other than Windows, Office, and Internet Explorer, it's fairly stable. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: open ports remaining
To quote Glenn Becker <[EMAIL PROTECTED]>, # # All, # # I have been trying to secure my Debian box, which enjoys a DSL # connection. I've been going through /etc/inetd.conf, commenting out # services, and K'ing others in the /etc/rc2.d/, until what I have left is # the following (output from nmap): # # Interesting ports on localhost (127.0.0.1): # PortState Protocol Service # 22 opentcpssh # 25 opentcpsmtp # 53 opentcpdomain # 111 opentcpsunrpc # # So, the questions: I have read some about sunrpc in the list archives but # have not found how to close the port; don't know what domain is, and am # confused smtp is still there since I've commented out the line in # inetd.conf ... I have set my hosts.deny to # # ALL EXCEPT sshd: ALL # # ... I guess then port 22 is not a concern? Personally, I've never put a whole lot of faith into just turning services off. Now, I could be wrong. What I usually do is use 'ipchains' to set up a firewall. Not that it helps you here, but it's something to think about :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Simple Questions
To quote "Leonard Leblanc" <[EMAIL PROTECTED]>, # Hey All, # # I have just a quick (and probably stupid) question. I was under the # impression that you could just place scripts in the cron.daily, # cron.monthly, and cron.weekly directories and that those scripts would be # run. Now either there is an error in my script or I'm not doing something # right. Can someone give me a procedure for adding a job to the cron # schedule...? You are correct there... A couple of things to look for: Is the script executable? 'chmod +x
Re: ide cdrom -> scsi emu problems
To quote Jerrud <[EMAIL PROTECTED]>, # [EMAIL PROTECTED]:~$ mount /cdrom # mount: wrong fs type, bad option, bad superblock on /dev/cdrom, #or too many mounted file systems #(aren't you trying to mount an extended partition, #instead of some logical partition inside?) # [EMAIL PROTECTED]:~$ Don't know about that ... # [EMAIL PROTECTED]:~$ mount /dev/hdc # mount: can't find /dev/hdc in /etc/fstab or /etc/mtab # [EMAIL PROTECTED]:~$ mount /dev/hdd # mount: can't find /dev/hdd in /etc/fstab or /etc/mtab # [EMAIL PROTECTED]:~$ Looks like you don't have entries for /dev/hd[c,d] in your /etc/fstab . If that's the case, you have to specify the mountpoint, ie: 'mount /dev/hdc /cdrom' or 'mount /dev/hdd /cdrom'. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: grep and memory problems with kernel 2.4.1
To quote Thomas Braun <[EMAIL PROTECTED]>, # i do cd / # # grep -r hallo * # # und then cames a memory enhausted and the network is down. Well, since you're specifying -r, it's going recursively through subdirectories ... I don't know for sure, but maybe it's running into some problems with some of the devices in /dev? :) Remember, most of them are just like files, you can 'grep' them all you want, even if it's not always a good idea ;) I doubt this would be specific to 2.4.1, though. Have you tried it in 2.2.x? Do you get the same errors(or similar ones) there? David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: grep and memory problems with kernel 2.4.1
To quote Joris Lambrecht <[EMAIL PROTECTED]>, # the /dev directory indeed just lists a list of names wich are linked to # device driver files through the inode table # # so in fact you're grep-in the output of the /dev, if this contains some # control chars it might hang your grep command, you should* be able to kill # this from another console # Umm... More than that; since "-r" is passed to grey, and those devices are, for the most part, regular files, they themseves are grepped. For instance, 'grep -r hello /*', will eventually lead to grepping /dev/hda. You'll be grepping your entire bloody harddrive. :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Kernel compile comments
To quote Ross Boylan <[EMAIL PROTECTED]>, # Hey, give the guy a break. Of course he's aggravated, although it may # be that only a little RTFM would clear things up. I wasn't really the aggrivation that bothered me... I've dealt with many a frustrated newbies. What does bother me is guilt-trips. Intended or not, that fellow tried to make people bad, by saying he is considering switching back to another distribution. If you stop and think about it rationally(even I didn't do that at first), it's not a wholly useless idea. No distribution is right for everyone. Debian is a community, after all. I contribute by helping everyone I can(mostly in #debian). So, if this fellow wishes to call upon that community for help, I don't think it's unreasonable to ask he not try to emotionally manipulate people, whether that be threatening to use another distribution, or whining about how tough things are. Sorry, just the way I feel :( David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Debian or Linux 7???
To quote Steve Rudd <[EMAIL PROTECTED]>, # Hi! # # I am frustrated with the linux 2.2 kernel. I have had two hacks in 3 months # and I am going broke rebuilding my server. # # I went out and bought Redhat 7, and got hacked 6 weeks later. # # I have been placed in contact with a guy who wants me to use Debian. But if # it based upon the same kernel as redhat, how is it going to be more secure? # I checked and found that # # from (http://www.securityfocus.com/) # Security risks for years: 1997-2000 respectively: # Debian 3, 2, 32, 45, 12 # RedHat 6, 10, 49, 85, 20 # # So Debian is about twice as good as redhat, but that is not real reassuring. # # I am considering joining the debian family, but am a bit concerned about # security. # # Just how much more secure is Debian than redhat? A little lesson in security: A computer is as secure as the admin is thoughtful and thourough and knowledgable. Red Hat can be just as secure as the most secure Debian box, and vise versa. The Linux kernel itself is only a relatively small part of a fully-functioning server. It is, of course, incredibly important :) Anyways, the kernel itself is actually pretty darned secure; it's the programs underneath it you need to worry about. You need to secure your web server software, your file sharing software, everything. *that's* where you're getting hacked - not in the kerel. Anyways, if you're running a business, I think you should either spend some money and hire someone to come in and help you secure your server, or you/your admin should start reading Unix administration guides. A competent admit can secure whatever distribution of Linux is put in front of him. An incompetent admit won't be able to secure anything. Debian, in a default installation, is more secure than Red Hat, but that isn't saying much. The default installation of Debian wouldn't work too well as a full-blown modern webserver, for instance. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Debian or Redhat 7???
To quote Steve Rudd <[EMAIL PROTECTED]>, # Well first, I repent of calling Linux 7: Redhat 7. Yes I am new. I have # been maintaining my own box from a su level for about 3 months. That is why # I was calling in an expert to install Debian tomorrow. It has become quite # obvious to me that I am way over my head in trying to get my server secure. That may be the case... Let me tell you something from my own experience. I've done tech support for going on twelve years, off and on. I'm not talking about customer tech support ... I'd call myself an admin, but I'm more of a troubleshooter. Of course, I have friends and family I always lend a hand to. Anyways, you seem fairly secure with yourself, so you probaly don't need to hear this, but I can't be sure, so I'll say it anyways. I'd say, without a doubt, 90% of the people I've ever met can get a $60,000 CDN a year job working with computers, with just a bit of tutelage. Usually I only need to work with a complete computer-illiterate person before they have the basic skills necessary to troubleshoot a problem. After that, the only thing you can get is experience. Anyways, I strongly encourage you to continue asking questions on the list here, in the #debian IRC channel on irc.openprojects.net, and to your friends. After a few months, if you're willing to put in the effort, you'll probably have enough general knowledge to start reading (and understanding!) various HOWTOs and program documentation packages. At that point, the only thing holding you back would be the amount of time you could spend reading and learning. Anyways, it's always best to administer your own box. Failing that, it's best to have a full-time admin, who's responsible for your box. Anyways, have fun :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: 2.4.1 install successful - what's up with modconf?
To quote Aaron Brashears <[EMAIL PROTECTED]>, # Thank you to all who let me know some more information about kernel # 2.4.1. It installed flawlessly, runs smoothly so far, and if anything # seems lightning quick. I'm looking forward to reformatting some of the # partitions as reiser and play around with it for a while. # # However, I'm really confused by one last issue. When I run modconf, it # doesn't think I have any modules available. However, the rest of the # module utilities seem to work ok. For example: I imagine it's just that 'modconf' hasn't been updated to be 2.4.1-compat yet... No biggie, I'm sure it'll happen(and I'll do it myself if I have to, to make sure I'm not blowing hot air ;). David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Running Deb on old PCs
To quote dickson chow <[EMAIL PROTECTED]>, # 1. How can i install debian at its bare minimum? (HD space is only # 340mb. sorry if this is a dumb question. i'm new to linux and the linux # structure is totally alien to me. i'm trying very hard to learn it # though :) What I would do is extract base2_2.tgz (the base Debian system) to the target Debian root partition. I'd set up LILO, make it bootable and whatnot. There, now you're got the most minimum Debian system available(I think, anyways :)... It's pretty small, the tarball is only 15M. Do get walked through the configuration, 'dpkg-reconfigure base-config'. That should run pretty much everything you'd get during a normal installation routine. After that point, 'apt-get install' very very few packages. Only the ones you absolutely need. I'm not sure if it's possible to 'apt-get remove' anything that was in base2_2.tgz; I think all those packages are marked "essential" and hence can't(or shouldn't, anyways ;) be removed. # 2. is it possible to get mgetty to fax out a document recieved from # another computer on a lan (ie. from a windowz computer)? I believe there are some Win32 clients around ... or I might be thinking of Hylafax. I'm not sure ... anyways, check http://www.freshmeat.net , it's a great place. You could also check http://www.appwatch.com , but I don't think the latter has any more software listed than the former. # 3. is it possible to get mgetty to email me the voice messages and # faxes? or at least notify me by email that i have new voice messages and # faxes? i'm guessing i need to write a complex script for that? Yes. But I'm not sure if the script would be terribly complex. 'apt-get install mgetty-docs' on a computer that has a wee bit more hard drive space, then poke around. I don't think it's all that difficult, from what I've read in the docs. # 4. maybe check faxes and voices messages using a web browser? that way i # can check my messages when i'm out of time or something hehe. That'd be more complex; you'd probably have to write a web front-end ... and you'd also have to install some sort of webserver :( At 340M, you're already going to be limited to how many voice messages you can store with only the base Debian config; adding big stuff like a webserver might make matters worse. # 5. would it be easier to compile my own kernel and get all this to work? Compiling your own kernel and getting the goals you've set forth accomplished have rather little to do with one another :) Anyways, you *are* going to have to work a fair bit to get this to work, there's no "mgetty-voice-fax-web-frontends-with-emailing-scripts-and-magic-wand" package :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Running Deb on old PCs
To quote Gavin Hamill <[EMAIL PROTECTED]>, # Just my two cents' worth here... I did a semi-manual potato install # recently (i/e/ selecting individual packages from the menu rather that # just the groups) and have ended up with a full mail and # web/SSL/PHP/mod_perl system with quite a few extra odds and ends in under # 200Mb... 340 should leave plenty of room even for a kernel recompile :) Thanks for the reply :) I'm not sure how much the voice/fax data will take up though. Worst case scenario is that it's uncompressed. At that point, you can look at around 1.5-7M per page per fax(depending on quality), and I think it's about 10M per mintue of uncompressed audio, although I'm not sure of that. Suddenly, that extra 140M doesn't feel quite so roomy ;) To the original poster: make sure the faxes and voice messages are compressed :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: after 'su -', 'Can't open display'
To quote James Sinnamon <[EMAIL PROTECTED]>,
# My apologies for a question that should have been answered over and
over
# again on this
No problem at all :) But check the archives next time. Since you have a
specific string to seach for("Can't open display"), a search would
provide good results.
# When I start X Windows, using the KDE window manager, I change to root
X Window System. X11R6. X11. X Windows System. *Not* X Windows. For
trademark reasons :) Commonly just reffered to as "X". Also, KDE isn't a
window manager, it's a desktop environment. It has an application
launcher/panel, a file manager, desktop controls, and indeed it does
have a built-in window manager :)
# (with su - )
# for administrative tasks. However I seem unable to run X window
# applications.
Try just using 'su', instead of 'su -'... It might fix your problem.
# Whatever X application I try to run, I inevitably get a message
similar
# to "... unable to open display".
#
# Previously on other distributions of Linux, I have used, as root :
#
# export DISPLAY=localhost:0.0
#
# and, prior to that, as the normal user that started the X window
# session:
#
# xhost localhost
#
# This somehow doesn;t work on Debian Linux (unstable distribution).
If just using 'su' doesn't work, and when you're in 'su', $DISPLAY is
set (if 'echo $DISPLAY' returns anything, it's likely set correctly),
you might need to run 'xhost +localhost' as the user who started the X
session. You'll have to do this before you try to run any X apps as
root, and it'll have to be done each time you log in and start KDE.
David Barclay Harris, Clan Barclay
Aut agere, aut mori. (Either action, or death.)
Re: rc.local
To quote Vittorio De Martino <[EMAIL PROTECTED]>, # Having experience of Linux RedHat, is there anyone out there able to tell me # where is the equivalent of the rc.local file in Debian and where can I find # it? There's isn't any rc.local per se, but executables in /etc/rc.boot/ will be run on boot-up, after everything else. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: mouse works under gpm, but X screws it up
To quote Peter Jay Salzman <[EMAIL PROTECTED]>,
# append="-R imps2" <-- this line was me fooling around
Get rid of that, and instead add a this line:
repeat_type=raw
#Option "Protocol""imps/2"
I don't know if it's case-sensitive, but IMPS/2 works for me... It might
make a difference.
#Option "Device" "/dev/psaux"
Yeah, make it use /dev/gpmdata - gpm and X are fighiting over the mouse.
# when i try to change gpm to use /dev/gpmdata, the mouse stops working
under
# gpm. i've been at this for awhile. can some kind soul tell me how
to get
# the mouse working under both console and X? it's a bummer having to
kill
# gpm everytime i want to start X up.
Maybe I don't quite understand what you're saying, but here's how things
work, ideally:
'gpm' uses the appropriate protocol, ie: "imps2".
'gpm' repeats absolutely everything("raw" repeat_type) to /dev/gpmdata
X reads /dev/gpmdata, instead of the real mouse device.
Everything is happy. :)
David Barclay Harris, Clan Barclay
Aut agere, aut mori. (Either action, or death.)
Re: LILO and big HD's
To quote Raffaele Sandrini <[EMAIL PROTECTED]>, # Hi # # I have win and lin insalled on my computer # unfortunally windows must be on the first partition wich is 20 GB's big. the # next 20 GB's are for debian. Is there any chance to get LILO working under # this circumstances (perhaps a new version or something like that)? Yup, the more recent version of Debian have large IDE disk support. You need to add the line: lba32 to the top of your lilo.conf to enable the code for it. Woody(testing) and Sid(unstable)'s LILO packages are fairly recent, but Potato(stable)'s are pretty old. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: LILO and big HD's
To quote Peter Jay Salzman <[EMAIL PROTECTED]>, # don't have this problem, just for my own knowledge... # # doesn't this require a quasi-new bios? since lilo can only bootstrap using # whatever services the bios allows. True enough, it does require a BIOS with EDD support. The LILO sources have a bit on it, I think. Also, I think almost any Pentium-compatible BIOS has it, so you're looking at five year old computers :) If you've got a five year old computer, the $200 spent on a hard drive with which LILO would need EDD support could be better spent elsewhere :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: LILO and big HD's
To quote Pann McCuaig <[EMAIL PROTECTED]>, # On Sun, Feb 25, 2001 at 17:22, Raffaele Sandrini wrote: # # > I have win and lin insalled on my computer # > unfortunally windows must be on the first partition wich is 20 GB's big. the # > next 20 GB's are for debian. Is there any chance to get LILO working under # > this circumstances (perhaps a new version or something like that)? # # Reaches everywhere on my 30G disk partitioned like so: # # That is, all 4 partitions are bootable via lilo. Hmm... Try putting a kernel on the last partition, and then get LILO to boot from it. If it works, would you mind sharing your lilo.conf? David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: LONG FILE NAMES
To quote "D. Hoyem" <[EMAIL PROTECTED]>, # attempted to put them in the /mount directory, one # would go there but the other would not... both were # labeled vfat:win95 and I assume that you can not have # 2 of the same in /mount. The problem is that when I You can only mount one partition per directory. If you want to mount both under /mount, you'll need to create two directories under /mount, and mount them each individually there. # open those drives to copy something from them to # Debian they use the short file name ie comm~1.deb # instead of communicator-base-275_4.75-2.deb, it is # this way if I use gmc or open the directory and use # ls. How do I correct this? When I look at a folder # on the cdrom they are the long file nameHelp. Be sure you're mounting them with the "vfat" filesystem type. Check /etc/fstab if that's where you've put in their information. It's probably set to "fat16" or "msdos" right now. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Considering Debian
To quote "Rainer Mager" <[EMAIL PROTECTED]>, # So, to make a long story short, how is the Debian package manager in this # regards? Do the same problems exist? Well, understand that there are two parts to the Debian packaging system. 'apt', and 'dpkg'. 'dpkg' is roughly equivalent to function, if not in features. 'apt' is a suite of utilities, and is generally what a user would use. Debian has what are called repositories; there are local mirrors in pretty much every country(I think ;). You can 'apt-get install ', and it will resolve dependancies for you. It'll install everything needs, within limits. Sometimes two packages conflict, and sometimes you'll need to manually over-ride it(so that the other package gets removed). Anyways, it's pretty slick, compared to 'rpm' :) # How easy is it to mix package installed # pieces to pieces installed by hand? It's bad. Very bad. :) To make a long story short, just don't. If you *must* install something manually, pass "--prefix=/opt --sysconfdir=/opt/etc" or something to ./configure, to keep your own compiled apps seperate from the packaging system. A *lot* of problems that Debian users have is because they're mixing Sid or Woody(unstable Debian branches) packages in a Potato(stable Debian, currently) system. This causes numerous problems. A lot of problems are also caused by people side-stepping the package management system. Basically, if 'apt-get' and 'dpkg' don't want to install a package, you shouldn't. Things will break, and break badly. Thank being said, there are some 6000 packages or so for Debian's Sid branch. That's an awful lot of packages :) Most are pretty up-to-date, too, except for a few notable exceptions like Mozilla(which is currently stalled at M18). It's not too often that you'll have to do a manual installation of an app. If the occasion arises, it's not too hard to make your own .deb out of a source tarball, but I'll leave that exercise up to you :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Considering Debian
To quote "Rainer Mager" <[EMAIL PROTECTED]>, # David (and all), # # Thanks for the reply. The part about mixing hand built stuff with pacages # in concerning as I do this quite often. The number of available packages is # encouraging but, nonetheless, I know occasions will arise. I've had to # build/install by hand X, glibc, postgres, the kernel, gcc, freeamp, and # others because of needing bleeding edge versions that fix bugs or because, # in debugging the current version, I needed a non-stripped binary. # Although the automatic installation abilities of apt sound nice, I find # that I usually want to actually download the, in my case, RPM so that I can # use it on multiple machines. That is, I question the benefit of this for me. # # Anyway, I wall continue to explore this and, once again, thanks for the # info. Well, unless you need CVS stuff, there's the unstable variant of Debian, Sid. Heck, it even has some CVS stuff :) For a while, KDE2 was from CVS... Anyways, there are plenty of bleeding-edge packages in Sid. Woody is supposed to be more "tame" than Sid, and Potato is really meant for servers, where things *can't* change out from underneath you. It sounds like you have the requisite skills to run a Sid system, so I'd suggest it. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Eek! X won't go away!
To quote MaD dUCK <[EMAIL PROTECTED]>, # also sprach kmself@ix.netcom.com (on Fri, 02 Mar 2001 08:42:15PM -0800): # > $ startx & exit # > # > ...which is what I use (do what I mean, not what I say). I can assure # > you there are no console sessions on this box. # # this will kill my X immediately and log off. the only way i got it to # work is by startx, ctrl-alt-f1, suspending, and then exiting. # obviously though, X remained suspended... # # not even startx & sleep 10 && exit works. Try 'nohup startx > /dev/null 2>&1 & exit', and replace /dev/null with a logfile of your choice. If you omit it, "nohup.out" will be used instead. Of course, you can keep it as /dev/null , but that might make debugging more difficult ;) I'd also like to point out that [G,X,K,W]DM is built for this sort of thing ... David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: New to Linux
To quote "The Grand Oral Disseminator" <[EMAIL PROTECTED]>, # Hi... # # I'm new to Linux, let's start with saying that! But I am thnking about installing Linux on my PC (Intel PIII 667MHz)... Can you give me some information about which distribution (A free distribution, since it is just my goal to have a mere 'meeting' with Linux, to get to know it - if you wish) would be best for me (I know about the Debian Distribution), where I can downlaod it, how I should install it, what other items I need, etc.. I think it depends on how much you know about computers, how quickly you learn new things, and how much you're willing to read. The last point is very important. People who know next to nothing about computers, and arn't too interested in learning should stick with Red Hat or Mandrake; they take care of a lot of setup chores which a lot of people just don't get. Most hardware in Debian has to be configured manually, at least partially. If you don't learn new things too quickly, I'd suggest Red Hat. It does a pretty good job of adding an eyecandy layer to everything, but underneath it it's a Linux system, no matter what. And no matter what people say, it's still designed primarily as a server OS. Using Red Hat, you'll likely be able to get up and running quickly, using a nice GUI, and then you can slowly get yourself accustomed to the hairier aspects of Linux. If you're unwilling to read much, you might as well not use Linux at all; you'll be stuck in GNOME or KDE, and you'll be missing out on the best things Linux has to offer. If you're a quick study, you don't mind reading, and you're familiar with the hardware within your computer, I strongly suggest Debian. It takes a little while to learn where everything is, and how you accomplish tasks, but once learned, Debian is extremely easy to use. Things are laid out, generally, with a fair bit of forethought. There are lots of little, very handy, Debian-specific utilities that are invaluable. Maintaining a Debian system is generally easier too, if you're using the stable variant. With Debian, "stable" means more than "it won't crash", it also means "we won't change things out from underneath you without due notice". On a Potato(the stable variant of Debian) system, you can generally safely upgrade the system without paying attention. I've also found that Debian's packages are generally of much higher quality. Since it's a community of volouteers, most people just package the apps they use on a daily basis, the apps they're most fond of. Since they only have one or two, and since they'll be using it themselvse, I find that they take to time to really make things right. Anyways, just my thoughts :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Strange /dev/pts/ problem
To quote Francois Gelis <[EMAIL PROTECTED]>, # I am running libc6 version 2.2.1-1, with a kernel 2.4.0 in which I have # configured the use of the devfs architecture. The problem sounds like it might be because you're not running 'devfsd'. 'apt-get install devfsd', and reboot. If you're not using Woody or Sid, read http://dharris.freeshell.org/linux/index.html (the first blurb) to see how to (relatively) safely getting a Woody/Sid package on your Potato box. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: sylpheed in potato
To quote "Marcelo Chiapparini" <[EMAIL PROTECTED]>, # Hi everyone! # # It is possible to install sylpheed in potato without updating any library? # In the Debian site sylpheed belongs # to unstable, but the shylpheed web page says that it runs under potato... Well, what you can do is build the Sid package for your Potato system. Check out http://dharris.freeshell.org/linux/index.html . The first blurb there explains how to do it. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Output cron in mailbox
To quote "mark's-debian" <[EMAIL PROTECTED]>, # Hello, # # I setup a cronjob to fetch these mailinglists messeages from my mailserver. Everytime the job is run I get an email with the output from this job. How do I stop this. Because I'm afraid I'm loosing valuable diskspace Add "> /dev/null 2>&1" to the end of the command; that'll route both standard output and standard error to /dev/null. That should catch everything. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: SSH with IPmasq
To quote ray p <[EMAIL PROTECTED]>, # I have a firewall (floppyfw) based that I am trying to do SSH through to # my Debian box on the other side and it will not work. I have uncommented # the lines that are supposed to allow forwarding to the box behind it in # the scripts and still everytime I try to connect I get connection # refused. I know that SSH is up and running on my Debian box and can # connect to it from other machines on my home network. But I can not get # to it from the other side of the firewall. Any ideas or suggestions please? I think this is a problem with your router, not Debian. That said, could you supply some more information? Specifically, any and all firewall rules that you have set up on the router(to get a list, run 'ipchains -L -n'). If you could attach whatever scripts the floppyfw uses to set up the firewall, that would be ideal. Also, you might consider going with a full-featured router-on-a-floppy distribution. the LRP(Linux Router Project) is good. For something a bit more friendly, check out the Coyote Linux distribution, which is also based on the LRP. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: more with wget..
To quote Roberto Diaz <[EMAIL PROTECTED]>, # bash-2.03$ wget -c --timeout=90 --wait=50 # ftp://sunsite.auc.dk/pub/os/linux/debian-cdimage/2.2_rev2/i386/binary-i386-3.iso # --18:18:25-- # ftp://sunsite.auc.dk:21/pub/os/linux/debian-cdimage/2.2_rev2/i386/binary-i386-3.iso #=> `binary-i386-3.iso' # Connecting to sunsite.auc.dk:21... connected! # Logging in as anonymous ... Logged in! # ==> TYPE I ... done. ==> CWD pub/os/linux/debian-cdimage/2.2_rev2/i386 # ... done. # ==> PORT ... done.==> REST 316073944 ... done. # ==> RETR binary-i386-3.iso ... done. # Length: 263,397,416 [-52,676,528 to go] (unauthoritative) # # [ skipping 308650K ] # 308650K -> ,, ,. .. .. .. [120%] # 308700K -> .. .. .. .. .. [120%] # 308750K -> .. .. .. # # What is happening? I've seen this a number of time, when resuming downloads. I imagine it's a wget bug. In any case, it never caused me any problems :) The files were always uncorrupted. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: GDM background image
To quote [EMAIL PROTECTED] (Robert Martinovic), # Speaking of Gdm, has anyone else found that although in # /etc/gdm/gdm.conf i put my .gtkrc in for Gdm to load with my gtk # theme, it still loads with the default theme (which is really ugly)? Make sure the gtkrc file is readable by GDM. I think it runs as root, but I could be mistaken. If there's something wrong reading the file, GDM uses the default theme. Right now I've got it set to Marble3D, and it works fine. David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: Can't reach ipchains from 2.4
To quote Jonathan Markevich <[EMAIL PROTECTED]>, # I'm trying to compile in ipchains emulation support in kernel 2.4, but make # xconfig has it greyed out and I have no idea what I need to enable to get to # it. Can someone point me in the right direction? You need to disable all 'iptables' support in order for IPChains compatibility to be available. Or, you have to compile everything iptables/ipchains-related as modules. Only one type can be in use at a time :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
Re: quick howto-command questions?
To quote "john smith" <[EMAIL PROTECTED]>, # 1. How can I find out the total number of files (also hidden) in the current # directory? # # 2. How can I find out the total number of executable files (also hidden) in # the current directory? # # 3. how to find the total number of files of a given an extension? # (ex."*.tar.gz") # # 4.how to list files alphabetically that end in "*.c"? You can use 'find' for all of those. 'man find' :) David Barclay Harris, Clan Barclay Aut agere, aut mori. (Either action, or death.)
System.map irritation
Hey there :)
I'm new to the list(but I've actually subscribed ;), and I'm having a
bit of an irritant considering an error I get often:
{module_list} {module_list_R__ver_module_list}
Warning: /usr/src/linux/System.map does not match kernel data.
The error appears when I run things like 'ps', 'top' and friends. Now,
I've never had this error when compiling a custom kernel before, and it
really irritates me :( Before it's suggested, copying System.map to
/boot/ just doubles the error; one for
/usr/src/linux/System.map, and one for /boot/System.map(or
System.map-2.2.18d, which is my kernel version).
This has never shown up before, and all the searching I've done says to
copy the file to /boot, which of course doesn't do anything :)
Anyways, thanks in advance for any help. I'm an intermediate user(I've
been around computers since before I could write[even block letters],
even though I've only been using Linux for about four-five years) and
don't mind reading, so feel free to get technical :)
Dave
Re: Removing Helix packages
To quote Hall Stevenson <[EMAIL PROTECTED]>, # Well, for some stupid reason, I decided to give the Gnome packages from Helix a try. After re-starting X-Windows, I noticed that my Eterms don't have scrollbars and scroll-arrows. My first assumption is that it's related to Helix's version of the various imlib packages. Now though, I can't remove them, at least without removing a bunch of other packages that depend on them. # # I've tried different "--ignore-depends=???" options, "--force", etc with no luck. I'm sure there is a way to force the removal of *just* the items I specify, isn't there ?? If so, could someone explain ?? Yup - use dpkg. 'dpkg --force-depends '. Dave
Re: init scripts
To quote kmself@ix.netcom.com, # Netiquette cop. # # I tend to consider a CC: request appropriate. It'd the "please reply # off list" that I'll comment on. Strangers mailing me *directly* # off-list with questions or followups get their knees bashed in . You'll have to forgive me, then ;) Some lists I'm on don't have a reply-to-all stuff, so I just "Reply", whereas some lists *do* have reply-to-all stuff. It's hard sometimes early in the morning trying to remember which button to push ;) Dave
Re: X error
To quote Bob Nielsen <[EMAIL PROTECTED]>, # I had a problem with woody several days ago where X would die # immediately after starting, before the window manager took over. In # trying to fix it, I really messed up my X installation, so I decided to # downgrade to potato and did a fresh installation of 2.2r2. I did this # and everything was working fine (including X). # # Since I had started with a somewhat minimal installation I just added a # lot of new packages and now X dies immediately after starting with no # messages, which is the same behaviour I had before with woody. # # System is a K6/2-350 with 128 MB, Cirrus 5430-based video card with # 2048k videoram. # # Any ideas as to where I should be looking? Lemme think. /etc/X11/Xwrapper.config /etc/X11/XF86Config[-4] and, if you're using GDM, /etc/X11/gdm/* (take a good look into the scripts in subdirectories there). Just a few thngs.
Re: Problems with kernel recompiling
To quote Vadim Kutsyy <[EMAIL PROTECTED]>, # I want to recompile the kernel. I config and compiling went without any # problem (I am trying 2.2.18pre21 static on woody). Hwever, when I am # restrarting computer, Xserver starts, but gdm doesn't, so all what I am # getting is gray screen whithout anything at all. # # Any recomendations? I think there might be something wrong with GDM's scripts - but I could be wrong. There are a number of things that could cause those symptoms :( But the first thing I'd do is look in /etc/X11/gdm , and the subdirectories therein. They contain scripts that GDM runs. If something is wrong with one of them, who knows? :) Dave
Re: newbieDoc: Debian runlevels intro
To quote will trillich <[EMAIL PROTECTED]>, # please run your eyeballs over them -- and any of the other # documents we've got there -- and feel free to post some feedback # to the newbieDoc project at # # http://www.eGroups.com/messages/newbieDoc Well, I must admit I don't like this phrase: 'That "fits-and-starts" syndrome just doesn't happen with Linux.' I think it should be ammended to 'That "fits-and-starts" syndrome just doesn't happen with a well-configured Linux computer.' It might make it have less of an impact, but it's also true :) Dave
Re: copying a file system across the network
To quote Lindsay Allen <[EMAIL PROTECTED]>, # # Past messages have detailed how to copy a file system from one drive to # another, but now I need to do it across the network. Of course I need to # preserve soft and hard links and device files. # # I do have a tape device here, but at 525Mb it's nowhere near big enough # for the job. I have ssh but not rsh. # # Is there a straight-forward way of doing this? Thanks for any pointers. I'd suggest one of two methods: 1) 'umount ; tar c / > partition.tar' -or- 'umount ; (tar c / | gzip -9) > partition.tar.gz' if you've got more CPU power than network bandwidth. 2) and I'm not sure about this, but 'scp' would be good here. It will do on-the-fly compression. Just read the docs to double-check that it handles links and device files properly. It's part of 'ssh'. Dave
Re: System.map irritation
To quote Ethan Benson <[EMAIL PROTECTED]>, # well i also used gcc-2.95.2 as that is also what is in potato. # # i should note one other detail, i build the kernel on straight potato # (with potato modutils) and install the package on a different machine # that has upgraded modutils. but i don't see how that could make any # difference as modutils are not used during building AFAIK. There are a number of things the kernel uses modutils and friends for - either at build time, or run-time. /usr/src/linux/Documentation/Changes tells us what version we want. I've had a few difficult-to-track bugs that were the result of bad versions of some of those utilities. With everyone's permission, I'm going to bring this to the kernel guys' attention. I've ruled out any "easy" fixes on this end; I think it's a kernel problem. A fresh build of 2.2.17 does not cause the error, while 2.2.18 does. I've also heard reports that test12 had the error, where test11 and prelease didn't. So, are we agreed that we should bother the kernel guys? Dave
Re: Burning with PCMCIA ATAPI/IDE
To quote "Hall Stevenson" <[EMAIL PROTECTED]>, # > Now to burn with cdrecord, you need a SCSI device. Is there an # > easy way to make the beast emulated as an SCSI device ? # > # > I tried append="hde=ide-scsi" in lilo.conf, but this don't work. # # One thing to try is to run "modconf" and add "SCSI Emulation" (and # possibly another option or two... sorry, don't remember right now). # # I could *not* get it to work with the default debian kernel. I tried # adding some modules with modconf, added a similar line to # /etc/lilo.conf, and so on. It seems to me I was missing the "sr_mod" # module and couldn't load it via modconf. I ended up compiling a custom # kernel and it works fine now. I've got IDE-SCSI emulation working, and this is what I had to do: Added: append="ignore=hdb ide-scsi=hdb" to the appropriate section in /etc/lilo.conf. Of course, 'hdb' is probably different on your machine ;) After I had booted, I also needed to 'modprobe ide-scsi'. I think. I don't have a burner, but I borrow a friend's for backup every month. So I never added it to /etc/modules or anything. Dave
Re: potato instalation (despair)
To quote "Antonio A. Lobato" <[EMAIL PROTECTED]>, # # I already tryied many ways to install the potato packages(because I # already installed it Basic System from hard disk), but it`s how if my cdrom # reader don`t existed. # Also it`s impossible I acess the internet from Linux because my modem is # the US Robotics Winmodem, and is not compatible with Linux. It remains the # follow alternative: # # download the packages that I want from the internet to my Windows partition # and install it from the hard disk. I don`t want install much packages. Only # the basic: X11, gcc, gv, latex, xterm. # # -What do you think of all it ? # # -How do I install this packages from a local hard disk ? Well, my bet would be that you'd end up downloading dozens(if not more) packages, just to have those programs you mentioned. However, if you go that route, use 'dpkg' to install them. For instance, "dpkg -i /mnt/Downloads/Debian-deb/*", where /mnt is you mounted Windows partition, and /Downloads/Debian-deb is where you downloaded the packages to. Dave
Re: Where is libblacs-pvm.so ???
To quote "Christopher W. Aiken" <[EMAIL PROTECTED]>, # I just installed Debian 2.2_r2 and chose some # packages from the "simple" setup. I got a lot # of the same errors reported during the package # configuration step. "ldconfig" also reports # the same error. # # The error indicates that /usr/lib/libblacs-pvm.so # is not available. Sure enough libblacs-pvm.so is # a symb link to libblacs-pvm.so.1 and libblacs-pvm.so.1 # is not available. # # Does anyone know where libblacs-pvm.so is? What it is # used for? Hmm... For what it's worth, I have a fully functioning Sid(unstable) Debian setup running, and I don't have this file. I don't think it's something to do with compiling - I compile lots of apps and I got all my development packages a while ago - and libblacs-pvm.so wasn't one of them... Dave
Re: Straight to woody after fresh potato install?
To quote Jon Pennington <[EMAIL PROTECTED]>, # csj wrote: # > # > The first of two questions: Is it OK to specify woody as the apt source after # > a fresh potato installation (I just grabbed, as someone suggested, the 2.88mb # > floppy install image)? I want to have woody with the least number of # > downloaded packages. # # Sure. I'd recommend changing your sources.list, apt-get update, then # install debconf, then apt-get dist-upgrade. # # > Second question: where is the best place to get the latest gnome packages? # > I'm planning to do some gtk compiling. Which gnome and what (virtual?) # > package do I specify to get a complete development environment. This query # > has been prompted by my inability to compile the lastest xcdroast (0.98?) # > under potato. # # I just use the packages at Helixcode. I don't even use the Debian GNOME # packages if I can help it. I would just like to point out that it is really quite difficult to get rid of Helix GNOME in favour of the regular Debian packages if at some point in the future you decide Helix is too buggy. I installed Helix without difficulty; and it ran just fine. However, I have a rather spartan desktop, without even the GNOME panel. I long ago got rid of GNOME's session manager(sort of slow and bloated, that one), so really all I'm running is Sawfish and some 'gkrellm' monitors. So, I decided to get rid of Helix and switch to the Debian packages. I did it, but it wasn't fun at all. Dave
Re: mouse configuration
To quote Mithras <[EMAIL PROTECTED]>, # Excuse me if this question is really basic. # # After installing Debian for the first time (but not Linux), I # discovered the mouse wasn't working. # # It didn't move at all at first, and I found last night that /dev/mouse # was just symlinked to some weird device, but not ttyS0 like it should # (for my 3-button Logitech mouse). When I started X, however, the # mouse moved only in occasionally sudden leaps to the left & downward, # while also sending imaginary mouse clicks (pop-up menus appeared). # # Could I have chosen the wrong mouse device to link to? I've heard # that GNOME can be sluggish, but this doesn't appear to be a system # grinding along (other symptoms, like hard drive activity, or slow # screen draws, are not evident). (My cpu is an AMD K6 233, with 80M.) # I cleaned the mouse wheels, so it couldn't be that either. # # Note: I did notice an X error on the console about accessing the mouse # after I quit last night, but alas I don't have that message written down. # # Any suggestions would be appreciated. I think the problem is 'gpm'. I know it is possible to have 'gpm' and X share the mouse, but I've never found a way(and to all those who are reading this; /dev/gpmdata isn't sufficient). You might want to consider stopping 'gpm', by doing a '/etc/init.d/gpm stop', and see if you X mouse problems go away. If they do, you might try to disable 'gpm' from being started at boot time by using the 'update-rc.d' script. # Besides this mouse problem, I'd just like to say how *pleased* I've # been with Debian. The default X environment (what I've been able to # see of it) and Xdm are gorgeous, and when I discovered that xdm & my # mouse problem conspired to keep me from a console or xterm prompt, I # really appreciated finding the shell prompt available from the boot # CD. My past experience has been with old Red Hat & Slackware # distributions, which simply didn't look as *pretty* as Debian. # (Trivial perhaps, but it is somehow satisfying when your favorite # operating system doesn't look like a sow's ear.) Agreed :) I've rebooted maybe two dozen times in the past few days just so I could see the Debian theme of the Linux Progress Patch :) You're right, veeery satisfying : Dave
Re: Many install questions
I'll try to help where I can :) To quote "David B. Small" <[EMAIL PROTECTED]>, # 1) Have other folks had trouble installing from potato CD's, or is it # likely mine are bad. I can't say for sure - I only used one CD, but it went flawlessly(after I figured out how to get through the install process ;). # 3) Should there have been a file manager/explorer included, when I # installed? (There wasn't...if none come standard, are there # recommendations? Eazel/Nautilus? Konqueror? Others? Come to think of it, I don't think there is. However, I do have some suggestions. a) 'mc': Console app, very popular. b) 'gmc': GNOME version of 'mc', also fairly popular. Ageing technology, will be replaced by Nautilus as soon as is feasible. c) 'konqueror': KDE file manager, does HTML fairly well(I use it as my daily browser). Fairly stable, but not terribly feature-full as a file manager. Think Windows Explorer. d) 'dfm': Old, but still useful. Light on resources, uses X. I use it for desktop icons(when I need them). e) 'nautilus': Unstable and slow, but very very pretty. Will use Mozilla for HTML, with appropriate decrease in stability and decrease in speed. f) 'filerunner': 'mc'-like, uses X and Tcl/Tk(or just Tk, dunno for sure). I used this for a fair while. Powerful and easy to use. g) 'emelfm': Like 'filerunner', but more up-to-date, and uses GTK+ as its widget set. When don't use command-line tools, this is what I use. # 4) Should there have been a GUI package manager? (or is apt-get/dpkg # the only way to go?) More options: a) 'dselect': Old and difficult to use for most. b) 'console-apt': Also known as 'capt', relatively easy-to-use. I use this for the most part. Console based. c) 'gnome-apt': GNOME frontend to 'apt'. Not entirely bug-free, but useful nevertheless. d) 'stormpkg': GTK+/GNOME-based package manager. I find it very useful. Similar interface to console-apt. When I'm going to be working on package management for more than three minutes, this is what I use. # 6) How can I get my Logitech USB mouse to work with this system? # (It's less important, since I still have a psaux mouse from MS) I don't know for sure, but kernel 2.2.18 *does* have support for USB. Never had to use it myself, though. I imagine the hardest part would be to get X to use the right device/protocol. Hope I helped, Dave
Re: mouse configuration
This thread about mouse configuration got me to try to get 'gpm' and X co-operating again, and I was successful this time. Let me explain my setup. I have a Logitech mouse of some sort, a Mouse Man Plus, I think. It's got two buttons, plus a wheel that also acts as a third button(plus up-and-down, which are in reality two more buttons, for a total of five). In my /etc/X11/XF86Config-4 file, I had the mouse protocol set to MouseManPlusPS/2, which worked well. I had full use of my wheel, etc., etc.. In my /etc/gpm.conf file, I had the mouse type set to "ps2", as there existed no setting specifically for a Logitech Mouse Man Plus. Normally, when gpm started before X, and X was configured to use /dev/gpmdata, I had the use of the regular three buttons on my mouse, but no wheel(which was really irritating). If I made X start first, using /dev/psaux, and then killed X, started 'gpm', changed X to use /dev/gpmdata, everything worked all right. Obviously something odd, eh? :) Well, I changed /etc/X11/XF86Config-4 to use /dev/gpmdata, and using the IMPS/2 protocol(Intellimouse PS/2). I also set the mouse type in /etc/gpm.conf to 'imps2', also designed for the Intellimouse PS/2. Now everything works great. I find it odd that using IMPS/2 everywhere makes a difference; since /dev/gpmdata should be an exact duplication of the mouse's data, but maybe there was some initialization thing going on that my mouse didn't like. Anyways, I'm sure this is mouse-dependant, but take it for what it's worth :) Dave
Re: XDM startup screen
To quote Koen Colpaert <[EMAIL PROTECTED]>, # After some experimenting with Suse, Slackware and Mandrake I turned to # Debian. After installing and configuring X I was presented with a grey # xdm-display as a loginscreen. I was wondering if there aren't any better # background images available and where can I find them. Check out http://x.themes.org . It has lots of miscellaneous themes; for instance, it has XDM themes :) Dave
Re: back to stable
To quote Knud Sørensen <[EMAIL PROTECTED]>, # apt-get got over 40 packets to install including X windows 4.0 Not surprising. gnome-db is even now developmental software; it would require lots of new GNOME stuff : # 1) So, How do I return to stable ?? You remove each and every one of those 40 packages that were installed, and you pray :) You did keep a list of those packages, didn't you? :) # 2) Is there any database front-end which run on stable and support # oracle ? I dunno about that :( Dave
Re: Packages list.
To quote <[EMAIL PROTECTED]>, # # Is there a quick way to rebuild the Packages file under the binary-i386 # tree? In other words, I would like to download the files I want for # Debian to put on one CD, like KDE 2, netscape 4.76? In other words, is # there an easy way to make a custom CD for Debian, or do you have to go # through the origional packages file and pick out the entries. Check out http://cdimage.debian.org . It's sort of an odd layout, like it asks you question about what you intend to do. So tell it you want to "burn many CDs to sell/give away", and then choose the option that is not "burn official CDs", which will lead you to: http://cdimage.debian.org/ch32.html . That'll get you a script which I think might fit the bill. Dave
Re: back to stable
To quote Knud Sørensen <[EMAIL PROTECTED]>, # That what i have tried to but some of the packets # pre or post remove scripts fails! # # So, is the any way to remove a packet without running them. ? Not that I know of, off the top of my head. 'man dpkg' to check for sure. Dave
Re: formatting new partition
To quote john gennard <[EMAIL PROTECTED]>, # It seems formatting can be done with 'mkfs' - most of the manpage # is (hopefully!) clear, but I have seen no guidance on what block # sizes would be appropriate or what the advantages/disadvantages # of different ones are. 512 and 8192 may well be the extremes of # choice - but what should I select and why? Well, most of that is done automatically. If you have Windows on the machine, you should use 'mkfs.vfat', so it can read it too. If you don't have Windows, stick with ext2, and run 'mkfs.ext2 /dev/'. It should Just Work(tm). :) Dave
Re: FYI: using logitech cordless keyboard extra keys in X4
# I left the mute key out, as I haven't found a way yet to actually
toggle
# it on and off (I know you can mute it with amixer set Master mute and
# unmute with amixer set Master unmute, but you'd need an extra script
to
# have it toggling between those values).
Here's a quick script to make mute-toggling work:
--- Start script ~/bin/toggle-mute.bash ---
#!/bin/sh
if [ -f /tmp/muted.${USER} ]; then
aumix && rm -f /tmp/muted.${USER}
else
aumix && touch /tmp/muted.${USER}
fi
--- End script ~/bin/toggle-mute.bash ---
Re: System security question
To quote Martin Bishop <[EMAIL PROTECTED]>, # Netstat shows the following services on my home machine: # Is it safe to leave it as it is? I've closed of all services in # inetd.conf and I'm using Firestarter as my firewall. I'm a little # paranoid that my system is listening for connections. But I don't # know enough to tell if these are internal services or if these # services are listening to outside connections. # # What do you guys advise? Well, Linux is a very network-oriented operating system, so you'll always have things listening for connections. Those particular ones look fairly normal. HOWEVER, you should do a couple of things: 1) Read up on hosts.allow and hosts.deny , and then edit them in /etc . Make sure you make them secure. 2) Use a real firewall(not that Firestarter isn't nice, I use it myself to display firewall logs), such as "rc.firewall"(recently renamed to "rcf"), which you can find on Freshmeat. Read all the instructions very carefully, and set it up. If you have a nearly-inpenetrable firewall, you don't really need to worry so much about what's listening for connections on your machine. Of course, always worry, but you won't need to fret over it so much. Dave
Re: Debian and Pentium 4
To quote Philipp Schulte <[EMAIL PROTECTED]>, # The german magazine "c't" (http://www.heise.de/ct) explitely says that # there is no problem with Debian and the Pentium 4 on the Dell # Dimension 8100. # RedHat and Suse caused some problems. That's probably because Debian uses(at last check) 2.2.18pre kernels. Dave
Re: AGH! Hard disc faults!
To quote Matthew Sackman <[EMAIL PROTECTED]>, # The hard disc is a 10Gb 7200rpm 2Mb cache Seagate that is less than 6 months # old. I'm fairly furious that it has developed an error this quickly. It's the nature of the beast, unfortunatly :( However, often times the manufacturers are really quite good about it. I had an IBM DJNA(7200RPM, 14Gig, 2M cache) which died about a month after warranty(not that the warranty even applied to me - I bought it from an OEM), and they happily replaced it. Dave
Re: How to help out.
To quote <[EMAIL PROTECTED]>, # I am willing to donate some of my spare time to Debian if they will have # me. What needs work? Lots of things ;) I'm fairly new, but I've offered to write man pages for some packages that don't have them(packages that I use, of course). I imagine if you have a few favorite packages that either arn't packaged, or have no package maintainer, I'm sure nobody would mind if you took them over :) Dave
Re: marks in syslog
To quote "Marco C ." <[EMAIL PROTECTED]>, # Hi buddies! # Why sometimes in my logs magically appears the line # date,time host -- MARK -- # What does this MARK mean? # Thanks in advance, marco For the life of me, I can't remember what they mean. But I'm pretty sure it's a command flag - 'man syslogd' shoud tell you, I think. Dave
Re: Need help with XFree86
To quote Russ Cook <[EMAIL PROTECTED]>, # Howdy! My wife bought me a new computer for Christmas, an E-machine # with the i810 chipset. I was able to get X11 working under TurboLinux # (the CD came with the LAN card I bought), but I'm having difficulty # with Debian. Attached are the XF86Config file and the XFree86 log file. # Any help would be most appreciated. I can't help you, but I'd like to thank you for taking the time to make logs and attach them :) May not seem like much trouble, but it's done far too seldomly. Dave
Re: KDE 2.2
To quote [EMAIL PROTECTED], # # I'm running 2.2r2. I downloaded the KDE deb's from sourceforge. When I # try to install I get a failed instalation from libqt2.2 saying it needs a # newer version.Well the libqt that I got was in the same directory from # sourceforge. So I went to get the newest qt from troll's ftp site. Just # my luck, no deb's. I down loaded the tarball, but even though I set the # environment in the .profile under root it still won't install because it # can't find the path. Does any one have, or can send me a link to the # newest qt in deb form, or at least tell me what I'm doing wrong. # # Please email me directly, My email server is down, so I'm using this one # temporarily. Check out http://kde.tdyc.com , they have Potato KDE2 debs, and I'd bet they've got QT debs to go along with it. Dave
Re: runlevels/boot stuff
To quote "M.B.Midden" <[EMAIL PROTECTED]>, # Im looking for some info about runlevels. Is there a howto for it? What is # the best runlevel for a server en when should u switch ( how can you switch) # ( i tried man runlevel ofcourse ) etc I'm sort of bored, so here's a little tutorial ;) Each runlevel has an associated directory in /etc . /etc/rc0.d/* : Used to halt the system. /etc/rc1.d/* : Used to put the system into single-user mode(after another runlevel has already been reached). Similar to "Safe Mode" from MS Windows. /etc/rc2.d/* : Regular multi-user runlevel. /etc/rc3.d/* : Regular multi-user runlevel. /etc/rc4.d/* : Regular multi-user runlevel. /etc/rc5.d/* : Regular multi-user runlevel. /etc/rc6.d/* : Used to reboot the system. /etc/rcS.d/* : General not used manually, it's used to put the system into single-user mode right after the kernel has been started. Usually entered by typing "linux single" at the LILO prompt. Runlevels 0, 1, 6, and S generally arn't modified by the user, which leaves runleves 2, 3, 4, and 5. Within all of these directories are a number of symbolic links to scripts within /etc/init.d . When the symbolic link's name start with "S", that means the appropriate script is executed with the argument "start". So, for instance, if there is a symbolink link /etc/rc2.d/S20gpm -> /etc/init.d/gpm , then '/etc/init.d/gpm start' is run. You'll notice there's a number between the "S" and the script name. That's the priority by which the script is run. The scripts are executed serially, lower numbers starting first. When a symbolic link's name starts with "K", the appropriate script is run with the "stop" argument. The same rules about priorities are used. When you enter a runlevel, all the scripts are called. Runlevel 6, for instance, calls(at the very end) /etc/init.d/reboot, so the computer reboots. When you type 'reboot' at the command line, you're basically entering runlevel 6. By default, your Debian system enters runlevel 2 on boot. Also by default on your Debian system, runlevels 2, 3, 4, and 5 are exactly the same. There's no difference between them until you make them different. Now, to answer your question :) By default, the user-defined runlevels are all the same, so there's no "best server runlevel". However, you can manually change the symbolic links(either using strict 'ln' and 'rm' commands, or(prefferably) using 'update-rc.d'), so that runlevel 3, for instance, starts all your regular Internet/Intranet services(apache, inetd, pop3d, and friends), and you can make runlevel 2 start your regular desktop(no apache, no identd, nothing you don't use) with X and your font servers and everything else you regularily use. Okay, I've wasted enough of your time ;) For some good, in-depth(but still understandable) information, check out the man pages for 'init' and 'update-rc.d'. You can also read the man page for /bin/init's configuration file, 'inittab', but I find t somewhat more difficult to understand. Oh, and if you do decide to change which services go into which runlevels, always use 'update-rc.d'. :) Dave
Re: runlevels/boot stuff
To quote Ethan Benson <[EMAIL PROTECTED]>, # On Fri, Jan 05, 2001 at 07:37:25PM -0500, David B. Harris wrote: # # > /etc/rcS.d/* : General not used manually, it's used to put the system # > into single-user mode right after the kernel has been started. Usually # > entered by typing "linux single" at the LILO prompt. # # wrong, rcS.d is run once at boot, regardless of what runlevel your # entering. single AND multiuser mode. (unless you enter single user # mode with the -b switch at boot, this stops ALL initscripts from # running which is more in line with BSD single user mode.) My mistake :)
Re: lilo.conf
To quote Russell Coker <[EMAIL PROTECTED]>, # I am working on the Debian package of lilo and am writing code for # auto-generating lilo.conf files. # # Below is an example of the type of lilo.conf that can be generated. The # debconf asks whether you want boot or boot-menu as the boot loader, it asks # what VGA mode you want, what parameters to append, and what DOS partitions to # setup with the "other=" settings. The rest is inferred from the running # system. # # If your lilo.conf has things that aren't in this which you think should be in # debconf then please do an anonymous ftp to ivanova.coker.com.au and upload it # to the incoming directory. I'll try and make the next version do what you # require. Doesn't look to shabby :) I'm assuming that the paths to drives and partitions are gleaned from current system settings, right? :) Anyways, I have 'lba32' in my main lilo.conf section. For more recent LILOs(anything after 21.6, I think - though I could be off), it'll allow for booting kernel images past the 8.4G(1024th cylinder) mark. Thanks for your work, Dave
Re: lilo.conf
To quote Russell Coker <[EMAIL PROTECTED]>, # On Saturday 06 January 2001 14:55, you wrote: # > I request that you provide support for me telling lilo that I don't # > want it touching my lilo.conf at all. Also, you should consider how # > to deal with changes that the user makes by hand to lilo.conf. If you # > don't regenerate existing lilo.conf files then you should not have # > much of a problem. If you do, then it gets fun. # # One thing that I am considering is to have a file /etc/lilo.conf.user file # (or maybe some other name). This file will be included into the debconf # generated settings just before the "image" parts. Then you could put in any # other settings you want. # # I'll put in an option to not change lilo.conf in the next version. I'd also like to suggest that a file be made which will override any of debconf's settings. That way, if I can't have lba32(for instance, if it does bad things), I can make sure it never gets there. Dave
Re: runlevels/boot stuff
To quote [EMAIL PROTECTED] (Miquel van Smoorenburg), # You could have just pasted /usr/share/doc/sysvinit/README.runlevels.gz # here .. Does it matter? I was bored, and I think it was helpful. Anyone care to disagree? Dave
Re: GNOME Mail Client
To quote Raghavendra Bhat <[EMAIL PROTECTED]>, # Brad Burns posts: # # > Can anyone recommend a good GNOME Mail Client ? # > # # U mean a mail client based on the GTK libs ? # Plump for Sylpheed. It is good. I use Sylpheed too, and I like it. Dave
Re: making a boot disk
To quote D-Man <[EMAIL PROTECTED]>, # # I now have Debian bootable from loadlin (which is how my system is set # up). # # Now I want to make a backup boot disk on a floppy. I tried running # mkboot, but it gave some errors and the disk wasn't bootable # afterwards. # # # Warning: device 0x1604 exceeds the 1024 cylinder limit # Fatal: geo_cmp_addr: Cylinder number is too big (13647 > 1032) # # # Debian is installed on the last 2GB of a 10GB drive so I understand # that I can't boot directly off the debian partition. (Actually, with # my BIOS I can't boot off the second IDE bus anyways) A floppy isn't # too big, and I want to use the floppy to boot. A boot disk is one way to go. You might also try adding 'lba32' to the top of your lilo.conf. Assuming a recent enough LILO, you shouldn't get that error. I use lilo-21.6.1, installed to its own directory(/temp/lilo), and I've but 'lilo' on hold as far as apt and dpkg are concerned. Dave
Re: trouble getting fetchmail to start SOLVED
To quote Ross Boylan <[EMAIL PROTECTED]>, # By the way, is there any reliable way to capture all the messages # which go to the console on system startup? dmesg and the various files # in /var/log do not get everything that shows on the screen--for # example, fetchmail's complaint when it couldn't find a system. Here's what I do. All various Sxx
Re: Ad blocking with Squid
# - I don't want to run both Junkbuster and Squid because I would see # this as wasteful. I guess that's somewhat true. I have both of them running. JunkBuster forwards its requests to squid(on port 3128), so that only the stuff junkbuster lets through gets downloaded. Some of my family members don't like Junkbuster, so they just use port 3128, bypassing it. Ah well, that's just the way I've got it set up. I've got a fairly beefy router(with 48M of RAM), but I also play all my MP3s on that machine(decoded, raw audio sent to my desktop - I've got more bandwidth on my 100Mbit network than I've got CPU power in my desktop), and do a bunch of other things without any slowdown. Dave
Re: gpm and wheelmouse
To quote Rick <[EMAIL PROTECTED]>, # Option "Protocol""Intellimouse" Just to be sure, the protocol for an Intellimouse-compatible mouse on the PS/2 port is "IMPS/2". I dunno if your mouse is on a serial, USB, or PS/2 port, but I figured it was worth mentioning. Dave
Re: gpm and wheelmouse
To quote "Rick" <[EMAIL PROTECTED]>, # my post was slightly premature (or maybe just got my thoughts in order, as # I investigated further into gpm.conf and found that where I had Hey, no worries :) Personally, it took me more than a month to get gpm+X working properly, although it wasn't wholly my fault(mouse protocol irregularities). So no problem ;) Dave
Re: [2.4.0] migration to devfs
To quote Ethan Benson <[EMAIL PROTECTED]>, # > 2.) boot. fsck will fail. do manual fsck, remount / rw, edit /etc/fstab: # > /dev/ide/host0/bus0/target0/lun0/part1 /boot ext2 defaults 0 2 # > /dev/ide/host0/bus0/target0/lun0/part2 none swap sw 0 0 # > /dev/ide/host0/bus0/target0/lun0/part5 / ext2 defaults 0 1 # > /dev/ide/host0/bus0/target0/lun0/part6 /local ext2 defaults 0 2 # > /dev/ide/host0/bus0/tagret1/lun0/cd /cdrom iso9660 ro,user,noauto # # all i can say is if this hideous thing is ever forced down our throats # i will switch to another OS. Don't let the door hit you on the way out :) Dave
Re: [2.4.0] migration to devfs
To quote Ethan Benson <[EMAIL PROTECTED]>, # i fail to see how typing: # # /dev/ide/host0/bus0/target0/lun0/part1 # # instead of /dev/hda1 or /dev/wd0a whenever i need to do anything # related to raw devices is a performance improvment. nor is writing # huge kludgy initscripts or bloated daemons just so i can do: He didn't say that it was a performance improvement. He was asking. No need act like it was a personal attack. # anyway thats just my rant on the subject, if you like devfs use it, # but leave it an OPTION so i can leave it off. (and not an `option' # like proc has become where you have the option to turn off and have a # useless broken system) You might want to consider changing your attitude a bit. These people are putting in how much time for free? If devfs becomes one of those "must-have for a useable system" options, then put your effort where your mouth is and work on the code. Show some gratitude. Say, every now and then, "Thank you Linus, you've given me a Free Unix-close which I use." That's not so hard, is it? Dave
Re: W sprawie ISO
To quote "Micha³ Rymaszewski" <[EMAIL PROTECTED]>, # Czesc # Chcialem sie dowiedziec skad mozna sciagnac najnowsza dystrybucje Debiana w wersji stable w formacie ISO # # Z gory serdeczne dzieki za pomoc http://cdimage.debian.org That'll help you out :) Dave
Re: [2.4.0] migration to devfs
To quote Ethan Benson <[EMAIL PROTECTED]>, # On Sat, Jan 06, 2001 at 07:46:53PM -0500, David B. Harris wrote: # > You might want to consider changing your attitude a bit. These people # # oh i want choice, my attitude is clearly flawed afterall if i did not # want choice i would still be letting MS or Apple tell me what to do. Yes, your attitude *is* clearly flawed. Whether or not you get a choice is irrelevant. These people put their blood and sweat into something which you directly benefit from(which no recompensation for you, either, I bet). And you have the *gall* to demand ANYTHING? That's the flaw. Not that you want choice - but that you demand it from people who have done nothing but work for you, for free. Dave
Re: [2.4.0] migration to devfs
To quote Joey Hess <[EMAIL PROTECTED]>, # So if you have a problem with something, talk to the authors. Spewing # bile across the lists of an unrelated project is just going to be # conterproductive for you in the long run. I apologize for my part in this argument. It really upsets me to see this, and I should have taken it off-list. Sorry for the inconvenience(sp?), Dave
Re: Cannot Open /dev/dsp device
To quote Aaron Maxwell <[EMAIL PROTECTED]>, # Hi. xanim complains that it cannot open the /dev/dsp # device. Funny thing is, I'm in the audio group and the # device's permissions seem legit: # Can't Open /dev/dsp device # ^C # yomama[12]% If some other program is using /dev/dsp, you'll also get that error. Check to see if 'esd' is running, it'll snatch up /dev/dsp. I know there are ways to make it release it every so often, but I can't remember how. Dave
Re: Cannot Open /dev/dsp device
To quote Aaron Maxwell <[EMAIL PROTECTED]>, # > If some other program is using /dev/dsp, you'll also get that error. # > # > Check to see if 'esd' is running, it'll snatch up /dev/dsp. I know there # > are ways to make it release it every so often, but I can't remember how. # # esd is not running, I'm certain. I don't have any apps running that # might grab dsp that I'm aware off. # # btw, I'm running sawfish with kde, mixed woody and potato. Well, I know you've probably already checked, but in the interests of thoroughness ;): is "arts" running? It's the KDE sound server. I apologize if you've already checked - I can't help it :) Dave
Re: lilo.conf
To quote Russell Coker <[EMAIL PROTECTED]>, # http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark Hey, I didn't know you wrote bonnie++ :) Great utility, thanks :) Dave
Re: [2.4.0] migration to devfs
To quote Stefan Nobis <[EMAIL PROTECTED]>, # No, but Andreas stated clearly that he don't want to use devfsd. And the above # are the internal names of devfs and the device drivers. The other names like # /dev/discs/disc0 and the like are the user friendly naming scheme which is # brought to you with devfsd. So if you don't use devfsd you don't get the new, # shorter names but only the very long internal names (which are deprecated to # use). Incorrect; on my system, the shortened names exist without devfsd. Observe: [EMAIL PROTECTED] /home/david]# ps aux | grep devfs | grep -v grep [EMAIL PROTECTED] /home/david]# mount -t devfs none /dev [EMAIL PROTECTED] /home/david]# ps aux | grep devfs | grep -v grep [EMAIL PROTECTED] /home/david]# ls /dev/discs/ total 0 lr-xr-xr-x1 root root 30 Dec 31 1969 disc0 -> ../ide/host0/bus0/target0/lun0 lr-xr-xr-x1 root root 30 Dec 31 1969 disc1 -> ../ide/host0/bus1/target1/lun0 [EMAIL PROTECTED] /home/david]# ps aux | grep devfs | grep -v grep [EMAIL PROTECTED] /home/david]# Tada :) They do exist. It's still longer than '/dev/hda1', but '/dev/discs/disc0/part1' isn't all that bad. Dave
Re: [2.4.0] migration to devfs
To quote [EMAIL PROTECTED], # Can you give us a rundown on how to get this to work? I followed the # instructions in the README but the permissions and owner/group bits never # stayed the way I wanted them. (eg: root.audio for all of /dev/sound, # root.video for all of /dev/v4l, etc). I'm using the devfsd from unstable # now. # # Also, I use the NVIDIA binary XFree86 4.0.1 driver. This driver needs # five entries in /dev to work. I can't get devfs to keep those entries # there across reboots. I made a script to do make those entries every time # the system is booted but I'm sure there is a prorper way of doing it. Since I don't know much about your system, all I can do is tell you how I got it to work. I'm running unstable(Sid), so I just did an 'apt-get install devfsd'. If you're not running Sid, I imagine you can pick up the source package, and build it(using dpkg-buildpackage). You could also temporarily change the 'deb-src' line in your /etc/apt/sources.list to point to "unstable", and do an 'apt-get -b source devfsd'. That installed a script(s?) in /etc/init.d, which start devfsd at boot-time. Of course, you have to have the kernel automatically mount devfs in /dev, which is available as an option in the kernel pre-compilation configuration. If you didn't select it there, add "devfs=mount" as an argument to the kernel, either through /etc/lilo.conf, or at the boot-time LILO prompt. Then, it Just Worked(tm). All ownerships and permissions are as I'd expect, including /dev/sound and friends. If this doesn't work for you, would you mind giving us a bit more information? Some things of note: What devfs-related options you configured in your kernel, how devfs is being mounted to /dev, how devfsd was installed, and anything else that is related. :) Dave
Re: When was /dev/gpmdata put into use??
To quote Carel Fellinger <[EMAIL PROTECTED]>, # On Sun, Jan 07, 2001 at 11:53:36AM -0600, John Foster wrote: # > I am trying to isolate a problem with my ps2 mouse. The pointer function # > does not work in console or xterm. It will cut and paste OK. I have # > checked all of the various configurations and read all of the related # > docs. I have determined that the problem did not exist prior to # > reinstalling my systems from scratch using the current stable debian # > 2.2r2. I have not noticed the /dev/gpmdata prior to this install. I have # > alwasy used /dev/psaux for both XF86Setup and for gpmconfig. The docs # > say to use /dev/gpmdata for XF86Setup and /dev/mouse or /dev/psaux for # > gpmconfig. I am certain that the hardware is OK. I want to try an older # # As explained recently in an other thread you can't share a ps/2 mouse # between processes (gpm and X), but if you feel lucky you might try # removing the line "repeat_type=" from /etc/gpm.con. The better option # is to use "/dev/mouse" in XF86Setup and have "repeat_type=raw" in # /etc/gpm.conf and to top this off make sure that /dev/mouse is linked # to /dev/psaux. (look to my prior posts this week for more) I think he means that /dev/mouse should be linked to /dev/gpmdata . Dave
Re: Package problems
To quote Thibaut Cousin <[EMAIL PROTECTED]>, # 1) Konqueror depends on lesstif. Is it normal ? I don't see where lesstif is # used... In order to support Netscape plugins, KDE has to use Motif/Lesstif. # 2) Bramhs can only be used by root because of a /dev/rtc: permission denied. # Can the permissions be changed, or at least a warning be given during # installation ? You can always change the permissions on a device :) I don't know why Bramhs uses the real-time clock, but oh well. # 3) kFilereplace compiles correctly (for debian-source package) but I get an # error during install : Dunno bout that one :) Dave
Re: didn't do it quite right...
To quote Rino Mardo <[EMAIL PROTECTED]>, # On Sun, Jan 07, 2001 at 07:03:26PM +0100 or thereabouts, tom wrote: # > Howdy Guys, # > # > I just ugraded from 2.0.36 to 2.2.18 (hey, I'm slow...) on a # > potato system. I did not do it the "debian way", however, and # > I still appear to have 2.0.36 installed, and running top give # > me ye olde "warning: /boot/System.map has an incorrect kernel # > version" # > # > I looked at lilo.conf, made one change. Here are all the steps I # > took: # > # > 1. unpack source to /usr/src/linux/* # > 2. cd /usr/src/linux/ # > 3. make menuconfig # > 4. make dep # > 5. make clean # > 6. make bzImage # > 7. make modules # > 8. make modules_install # > 9. mv System.map /boot/System.map # > 10. cd /usr/src/linux/arch/i386/boot/ # > 11. mv bzImage /boot/ # > 12. cd /boot # > 13. cp vmlinuz vmlinuz.safe # > 14. cp bzImage vmlinuz # > 15. lilo # > # # what i always do is rename /lib/modules/2.2.x to /lib/modules/2.2.x-n # whenever i have modules And that works? From what I understand, 'modprobe' and the kernel rely on /lib/modules/ to be the same as the kernel version(ie: 'uname -r'). Dave
Re: Debian. Gonna give it a whirl :)
To quote Sebastiaan <[EMAIL PROTECTED]>, # > Pfff, I really don't know how many Kbs of packages you need to # > upgrade. The Xfree86 packages changed since 2.2r0, IIRC, so it could # > be quite a few Kbs.. # # Too much. It takes already an hour to download the kernel source (ideal # line), and that is only 15MB. You will be better off buying the cd's. # # Depending on your demands, you can buy only two new CD's and download some # other stuff you might need. This is all a matter of patience, and in some cases, cost. In North America, local regional calls are free - they don't cost anything. Likewise, unlimited internet access can usually be bought for about $20US a month. I've been using Linux pretty much exclusively for three years, I guess, and I've only bought two boxed distributions. Even then, I had downloaded the ISOs before I decided to pay the distribution makers. Sure, it took ~18 hours for a 640 meg CD image, but hey, I can just leave my computer to do its thing. Dave
Re: Debian. Gonna give it a whirl :)
To quote John Hasler <[EMAIL PROTECTED]>, # David B. Harris writes: # > In North America, local regional calls are free - they don't cost # > anything. # # For me that "local region" is about ten miles in diameter and includes # exactly one (unreliable) ISP. # # > Sure, it took ~18 hours for a 640 meg CD image, but hey, I can just leave # > my computer to do its thing. # # Not everyone can afford a separate phone line for the computer. I wasn't saying that :) But, most of the posts I saw were along the lines of "don't even try it". Just wanted to say that, yes, it is possible. Assuming you have a) the means, and b) the patience. Dave
Re: nis/nfs/automounter
To quote Daniel Hearn <[EMAIL PROTECTED]>, # Hey, having a problem with my debian box here, was wondering if anyone # has had this problem before. I have made my box an nfs server, and I # have my home directory located on it. I want to be able to automount my # home directory when I log into another machine on the network. I am # using NIS for authentication. I can authenticate successfully, and I # can even mount my home directory from another box. The only problem is # that unless I mount it on the Local machine, it is read-only. So # basically, I just want to be able to write to my home directory from # another computer. Any clues? Let me know if you need more info. Yeah, check out 'man exports'. The file you want is /etc/exports(which you probably already know). You probably want something like: /home other.computer(rw,no_root_squash) Dave
Re: whoops -subject meant to be software RAID
To quote John Griffiths <[EMAIL PROTECTED]>, # OK i've built a server to run a Linux/Apache/Mysql/PHP setup on. # # It works pretty well and we're happy with it. But i've statred hearing about Software RAID. Our budget won't run to SCSI or a RAID controller. There are relatively cheap IDE-based RAID solutions available, if you want to consider that. # But we could stick in a second HD on the other IDE controller. Yup. # Would this be worthwhile? How hard is it to set up a software raid? what do I do? does it matter that i want the whole OS as well as the data storage mirrored? Fairly easy. Read the appropriate HOWTOs are http://www.linuxdoc.org Dave
Re: whoops -subject meant to be software RAID
To quote John Griffiths <[EMAIL PROTECTED]>, # ># Would this be worthwhile? How hard is it to set up a software raid? # >what do I do? does it matter that i want the whole OS as well as the # >data storage mirrored? # > # >Fairly easy. Read the appropriate HOWTOs are http://www.linuxdoc.org # > # >Dave # > # OK.. thanks for pointing me there. # # so on a debian potata box would i need to recompile the kernel? or could i # # #apt-get install kernel-patch-2.2.10-raid # #apt-get install raidtools2 # # and then use the raidtools? # # www.debian.org package lists makes no use of RAID1 which is the level i'd want to go for... As a rule, I don't use prepackaged kernels. However, looks like those apt-get lines would do the trick. Unless you're mounting your root filesystem as RAID, you can use RAID modules. So long as /lib/modules isn't a RAID array, RAID support doesn't need to be compiled into the kernel. So, you might only need to add modular support for RAID in your current kernel sources, and then add the appropriate modules to /etc/modules , before your RAID partition is mounted. I am, unfortunatly, unfamiliar with what Debian does concerning RAID. As always, documentation is your best resource. Have fun :) Dave
