Re: Want to creat a CD with indexed HTML content
Thanks for the pointer Richard Pat On Thursday 27 Feb 2003 12:54 pm, Richard Hector wrote: On Thu, Feb 27, 2003 at 11:32:16AM +, Pat Colbeck wrote: Hi I am working on a little project to produce some reference materail on a CD. Basically it will be a canned web site. The idea being that you could stick it in your CD drive and browse all the content easily. What I would like to do is have an option to search teh CD without the use having to index it themselves. Does anyone know of a search engine that you can install on Windows that will let you point it at a pre made index file and that is free. Preferably one that is compatable with indexing produced by HTdig as the live copy of the web site is running on Woody with HTdig and Apache. Just a thought - rather than a Windows app, could this be done with Java or Javascript (will a browser load a Java applet from a local disk?) or similar, and therefore remain internal to the browser, and (mostly) cross platform (and on-topic :-). http://www.google.com/search?q=client-side+search+engine came up with some interesting hits that could be useful. Richard -- Pat Colbeck Cisco CCIE :2305 E-Mail :[EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Arbitrarily setting From: header in Mutt
Hi If you are using different mail addresses when replying to specific people or companies you can use send hooks to automatically set a different my_hdr dependant on who you are mailing. If like me you use a different E-mail address for work and for mailing lists like this one and have your mail sorted into different folders you can use folder hooks to set a my_hdr dependant on which mail folder you are in when you compose the mail or reply to mail. It would look like this:- folder-hook +Spool/debian-user my_hdr From: [EMAIL PROTECTED] This says if I am reading mail in folder /Mail/Spool/debian-users then change the From header to [EMAIL PROTECTED] You can also change all the other variables etc with send and folder hooks in the same way. Pat -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Re: Stop dselect installing packages
Thanks Colin Thats exactly the thing I was looking for. Pat On Wed, Feb 06, 2002 at 10:43:18AM +, Colin Watson wrote: You could move the cursor down to the line saying Available packages (not currently installed) and press '_'. That will select to purge all packages which aren't yet installed. -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Stop dselect installing packages
Hi I did a minimal install on a server here that is just going to be an FTP and DNS server in a test lab. I did the initial install to reboot and then quite the install so far so good. Unfortunately when I go into dselect to install proftp etc and the choose install dselect also wants to install a whole raft of other stuff like latex and emacs stuff I real don't need on this server. I guess these packages are part of some kind of base package set but how do I tell dselect that I dont want them. I just want to be able to choose individual packages I actually need. Thanks Pat -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Re: Stop dselect installing packages
Thanks Colin Is their a way to say to dselect just forget about all the packages you currently have marked for install from the command line so next time you run it it only has a list of allready installed packages and ones you can install. Its just that their are thousands of packages in there and its really easy (not to say time consumingi) to miss some when scrolling through the whole lot looking for marked packages. I am also thinking about those times when I accidentally select a package in dselect and it selects a whole load of dependancies and I press the wring key to return to the top level menu. It would be nice at this point to just quit and then have a way of wipping out the selections. Thanks Pat On Tue, Feb 05, 2002 at 10:30:07AM -0600, Colin Watson wrote: On Tue, Feb 05, 2002 at 03:47:03PM +, Patrick Colbeck wrote: I did a minimal install on a server here that is just going to be an FTP and DNS server in a test lab. I did the initial install to reboot and then quite the install so far so good. Unfortunately when I go into dselect to install proftp etc and the choose install dselect also wants to install a whole raft of other stuff like latex and emacs stuff I real don't need on this server. Those are the packages marked as standard. Remove the selections once in dselect (use the '_' key) and it won't ask you again. -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Installing Woody from CD
Hi Whats the current status of the Woody CDs. I have downloaded all 8 dated the last week in January, if I actually blow CDs of these will I actually be able to do a clean install of Woody. I just wanted to cjeck before I wasted 8 CDRs. Thanks Pat -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Re: Installing Woody from CD
Hi Dimitri Do you mean you installed Potato then upgraded to Woody from the Woody CD's or from the Internet. I have Potato 2.2r3 CDs so that would be an OK way for me to go (actually I have got Potato installed on my laptop just to re familarize myself with Debian as we have all been using SuSE for the past 18 months or so). Pat On Mon, Feb 04, 2002 at 10:29:38AM -0600, Dimitri Maziuk wrote: * Patrick Colbeck ([EMAIL PROTECTED]) spake thusly: Hi Whats the current status of the Woody CDs. I have downloaded all 8 dated the last week in January, if I actually blow CDs of these will I actually be able to do a clean install of Woody. I just wanted to cjeck before I wasted 8 CDRs. Netinst image I d/l'ed last week looks for Release file in the wrong place (couple of dir levels up from where the file actually is), so you can't install the base system. Without that you can't install lilo make your box bootable. (Shirley including extra couple of Mb of base.tgz in the 65Mb image is not that hard, eh?) Also, 3c59x module is included in only one image: udma-ext3 (I wonder who came up with that bright idea... every Dull here has a Vortex NIC). But that doesn't really matter because of the above. I ended up burning a potato CD and doing dist-upgrade to woody. On the positive side, that worked without too many problems. Dima -- Surely there is a polite way to say FOAD.-- Shmuel Metz Fornicate Off And Decease. -- Rik Steenwinkel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- --- Pat Colbeck E-mail: [EMAIL PROTECTED] Tel:I'm not telling ---
Re: IPX over the Internet
As previous posters have indicated you definately cannot pass IPX over the Internet without some kind of tunnel. Internet routers only understand IP so when they see an IPX frame they would just drop it. What you have to do is encapsulate IPX in IP packets and the the IP packets can traverse the Internet ant the IPX can be de-encapsulated at the other end. Pat
Re: COM21 is killing me with ARP
It really doesnt matter that there is a whole class B address space as you should only get arped when someone om the same class B needs to know your mac address. Once the arping device has your mac address it should cache it so it doesn't have to arp for it again for a long time. All the other people on the calls B shouldn't be trying to find your MAC address as they theoretically should only be talking to your service providers DSLAM. It sounds like somebody has screwed up at the service provider configuring their routers they have probably:- i) Configured a really small arp cache timeout value so the service provider router is permanatly having to re arp for the mac addresses of all the DSL modems or ii) Configured a static route via a broadcast interface (eg etherent) on the cental router. This is a really bad thing as instead of just arping for the next hop address the router will arp every time it needs to send a packet to any address on the network the route is for to try and determine the gateway to that address. This is a really good way to crucify network performance , static routes pointing at interfaces rather than next hop addresses should only be used on point t point networks (leased line etc). Pat -- Cisco CCIE #2305
Re: PDAs
Dont write off EPOC devices so quickly. PSION are doing quite well at the moment with the Revo and they are very popular here in the UK. I use a 5MX and its very good with a slot for flash expansion cards and a keyboard you can actually use. It syncs really well with Outlook etc (good for work where MS is the corporate standard and the MIS guys havent even heard of postscript as a file format). There are currently no utilities for syncing with Linux but you can mount its file system from LInux (over infra red even) and copy stuff backwards and forwards. Wether its better than a PALM type device depends on what you want. If you just mainly want a contacts database and a diary then go for a PALM they are small and sync nicely with Linux PIMs however if you want to actually type on the thing and maybe use it as a dumb terminal (I often setup routers using my PSION) or keep PDFs etc for instant reference on it (you can stuff 320Mb flash cards in a PSION - but not the REVO it has no expansion slots) choose a PSION. Pat
Re: CPAN updating Perl version
Hmm this raises another similar issue. Xemacs has its own packaging system that lets you download the latest elisp packages does anyone know the implications of this with apt ? Pat
Re: COM21 is killing me with ARP
Hmm point taken. I hadn't actually thought about the router actually running out of arp cache space. Bryan == Bryan Andersen [EMAIL PROTECTED] writes: Bryan Actually it does matter. When Joe user turns off their box Bryan it nolonger can answer requests for it's ethernet adderess. Bryan This means a bunch of requests for it's arp address. So Bryan when someone scans the network you get bombarded by arp Bryan requests, and the caches naturally gets trashed durring Bryan this. If you know a provider does this you can realy hose Bryan up their network by bombarding them with random addresses Bryan in their network space. To keep from having this trash a Bryan network the router really needs to have enough cache Bryan entries to store all hosts on the network. Many routers Bryan just can't handle that for a class B network. They really Bryan should break their network up into class Cs. Pat
Komba and user permisions
Hi I have just compiled Komba on 2.2r3 with KDE 2.2 from kde.debian.net and it works bar when I mount an NT share neither root or a user can see any files on the share. ls returns this bash-2.03# whoami root bash-2.03# ls NTserver ls: NTserver: Permission denied bash-2.03# df seems to work though bash-2.03# df Filesystem 1k-blocks Used Available Use% Mounted on /dev/hda2 5992876 1469256 4219192 26% / /dev/hdc1 29544228470908 27572564 2% /home //UK_YK1_SRV_99/Users 24585728 15334400 9251328 62% /home/pcolbeck/NTserver bash-2.03# I followed the directions in the Komba handbook and created a new group for smbmounting and added myself and root to it then changed the permisions on smbmount* and smbumount* to -rwsr-x---1 root sambamou 569 May 8 20:20 /usr/bin/smbmount -rwsr-x---1 root sambamou10492 May 8 20:21 /usr/bin/smbmount-2.0.x -rwsr-x---1 root sambamou 252636 May 8 20:21 /usr/bin/smbmount-2.2.x Any pointer gratefully appreciated. Pat
Slow server due to reverse lookup
Hi I have a tftp/ftp server running 2.2 in a lab environment. This thing is basically for people to ftp files to then tftp them down to routers etc for testing the code. Unfortunately both FTP and TFTP are extremely slow to connect (like over 20 seconds) and often the clients timeout. Once connected its fine, also if the clients host address is in the hosts file then connection is very quick. I am assuming that the server is doing a reverse lookup on all incoming tcp conections. Since the test lab has no DNS and the machines can have an ip address in a range that covers several thousand addresses (its a claa B network) I really don't want to type all the ip addresses into the hosts file with dummy names. Is there anyway to turn of the reverse lookup and make life easier as this is a secure network not connected to the Internet ? Thanks Pat
Re: Slow server due to reverse lookup
I don't want to setup DNS as the possible range of IP addresses is huge. I would have to add entries for all the RFC 1918 address space, that would be hundreds of thousands of records to add to a DNS server. This is a lab where network engineers replicate routing issues or test new routers/switches etc so networks come and go all the time and they can be anywhere within private address space. Thanks Pat Ethan == Ethan Benson [EMAIL PROTECTED] writes: Ethan On Mon, Jun 11, 2001 at 11:38:37AM +0100, Patrick Colbeck Ethan wrote: I am assuming that the server is doing a reverse lookup on all incoming tcp conections. Since the test lab has no DNS and the machines can have an ip address in a range that covers several thousand addresses (its a claa B network) I really don't want to type all the ip addresses into the hosts file with dummy names. Is there anyway to turn of the reverse lookup and make life easier as this is a secure network not connected to the Internet ? Ethan why don't you just setup dns? it would make your (and Ethan everyone elses) life alot easier. Ethan -- Ethan Benson http://www.alaska.net/~erbenson/
Re: Slow server due to reverse lookup
Thanks Bud DNS in nsswitch was causing most of the problems but proftpd was still really slow even after that. For reference in case anyone else runs into this proftpd does reverse DNS and Ident against incoming addresses but you can switch it off in proftpd.conf thus: UseReverseDNS off IdentLookupsoff Now the thing connects instantaneously for all allowed methods :) Mind you I wouldn't recommend doing this on a server connected to the internet or one that had any data that you really cared about on it. this one is just a staging and temporary storage area for router/switch firmware so I dont really care if someone manages to break it. Thanks everyone for the suggestions. Pat Bud == Bud Rogers [EMAIL PROTECTED] writes: Bud On Monday 11 June 2001 05:38, Patrick Colbeck wrote: I am assuming that the server is doing a reverse lookup on all incoming tcp conections. Since the test lab has no DNS and the machines can have an ip address in a range that covers several thousand addresses (its a claa B network) I really don't want to type all the ip addresses into the hosts file with dummy names. Is there anyway to turn of the reverse lookup and make life easier as this is a secure network not connected to the Internet ? Bud You probably want to take dns out of /etc/host.conf and Bud /etc/nsswitch.conf. Both have good man pages that will Bud explain what you need to do. You might also want to set up Bud NIS. Bud -- Bud Rogers [EMAIL PROTECTED] http://www.sirinet.net/~budr Bud All things in moderation. And not too much moderation Bud either. Bud -- To UNSUBSCRIBE, email to Bud [EMAIL PROTECTED] with a subject of Bud unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: laptop instillation via a PCMICA cd rom
Hi I just had a similar problem with a Libretto. I solved it by creating a small DOS partition and using packet drivers and packet driver type ftp software to ftp the basic files needed for an install (see the Debian install documents on www.debian.org). Basically you need the kernel and a couple of driver disks and the base system archive. Once I ran the basic install off the DOS partition I switched the rest of the install to FTP and installed from rest of the packages from a Debian mirror site but you could equally well FTP from the CDs mounted on a local machine. Also with a PCMCIA CDROM you could probably mount that and install from it once the base system was installed from the DOS partition. Good luck Pat
Slow Ximian Gnome 1.4
Hi I 'upgraded' my Gnome on 2.2.r3 to Ximian 1.4 and it now runs like a dog. I have a P400 with 128Mb RAM and a Matrox G200, running top with just a gnome terminal open shows 50% CPU idle with XSVGA and Nautilus being the biggest CPU hogs (even though all Nautilus windows are closed). On the same system with KDE 2.1.2 and a Konsole open running top and Konqueror open I get 98% CPU idle ! Is everyone else seeing something similar or do I have some kind of problem on my box ? Pat
Network starts before PCMCIA
Hi I just installed 2.2r3 on a Libretto and it works fine except that on boot it starts the networking before it starts PCMCIA so the ethernet card is not initialised. I can do a /etc/init.d/network resart and it all bursts into life. Is there a recommended way of fixing this issue ? Thanks Pat
Re: Install from single floppy
Carl == Carl Fink [EMAIL PROTECTED] writes: Would have taken a lot of floppies . Luckily I have found a packet driver for my Xircom card and am currently ftping all the required files to the Libretto using CUFTP. If anyone else is going to try this be aware that FreeDos doesnt work with Xircom packet drivers, they won't find a free IRQ but if you boot from a system disk formatted with Win98 they work fine apart from the fact that they then kill off your floppy drive but at this point you really dont need it. Now to have some fun ... Thanks Pat Carl On Wed, May 16, 2001 at 01:57:36PM +0100, Patrick Colbeck Carl wrote: I am going with the DOS partition option. I have installed FreeDos in a little 50 mb partition and that works OK. Now I have to get the reuired Debian files on there (loadlin kernel etc) and it looks like the best way to do this is via NFS over PLIP by booting the Libretto off a tomsrbt rescue disk and NFS mounting my other laptop over PLIP ! So I am off to buy a parralel null cable on the way home ! Carl Why? If FreeDOS recognizes your floppy drive, you can copy Carl the files onto the DOS partition using floppies (or a single Carl floppy if you're really short). -- Carl Fink [EMAIL PROTECTED] Carl -- To UNSUBSCRIBE, email to Carl [EMAIL PROTECTED] with a subject of Carl unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Install from single floppy
Hi I have just aquired a Toshiba Libretto 70CT onto which I want to intsall 2.2.r3 and wipe out the nasty corrupted Win98 that is currently on the hard drive. Unfortunately this cute little thing has a strange PCMCIA floppy and the install barfs loading the root disk (basically it cannot see the floppy drive). There is all sorts of info on installing RedHat on one of these using zip drives etc (which I dont have) but I was wondering is there a way to fo a network install using a PCMCIA card or perhaps PLIP that would use only a single 1.44Mb floppy to kick it off ? As an alternate method are there any instructions for creating custom install disks so I could build a kernel that suppoprted this weird floppy device ? Thanks Pat
Re: Install from single floppy
Hmm after losts of digging and squinting at web pages written in Japanese (which I dont understand at all) it seems to come down to this. 1. The Debian boot disk kernels wont read the floppy after the first disk. 2. Hence you must install from somewhere else. 3. Options are Zipdisk or DOS partition. I am going with the DOS partition option. I have installed FreeDos in a little 50 mb partition and that works OK. Now I have to get the reuired Debian files on there (loadlin kernel etc) and it looks like the best way to do this is via NFS over PLIP by booting the Libretto off a tomsrbt rescue disk and NFS mounting my other laptop over PLIP ! So I am off to buy a parralel null cable on the way home ! I will report back hopefully tommorow on my progress. Pat
Re: domain name: internet vs. intra-net
The recommended way to cope with DNS on a LAN using private addressing that is connected to the internet somehow eg via NAT is to use a thing called splt primary DNS. Basically what you do is this. 1. Have a DNS server on the publically addressed side that lists the stuff that the internet should be able to see (ie your mail server web server etc) but nothing that has a private address. This is setup as a primary DNS server. 2. Have another DNS server on the private LAN that lists only the stuff on the private LAN. This is setup as a slave to the first server. Thus anyone on the internet gets a resolution of your domain via 1) and those on the private LAN get resolution from 2) which if it does not have a name cached forwards the request to 1). Usually on a big LAN these two DNS servers are separate machines but it is possible (though I have never done it) to run them as two logical servers on the same physical server so long as the server is I guess dual homed into both the private and public LAN. Try looking here for example configs etc :- http://www.dns.net/dnsrd/ Have fun Pat
Re: module sound with potato 2.2 r2
This raises a point that has been puzzling me for a while. I sthere a list anywhere of all the groups in debian, what they are for and when users should be added to each group ? I tend to do it by cat group to see what groups there are then guessing but maybe it is documented somewhere ? Pat On Wednesday 11 April 2001 14:34, Robert Voigt wrote: Many thanks I have the sound as root but I have had put the permissions of the devices /dev/dsp and /dev/hdd to 666. I don't know if it is suitable but I have the sound ! Nice to hear that it works. But changing permissions is usually not nice. The right way to go is to add all users that should have access to the soundcard to group audio (or whatever it's called, can't remember exactly). Just type as root addgroup user group
Re: telnetd slow to respond
Yup I think the telnetd tries to do a reverse lookup on the incoming telnet session. Adding your ip address and machine name to the hosts file of the machine you are telneting to will speed it up a lot. Same thing applies to proftpd so maybe its a function of the inetd super server ? Pat On Thursday 05 April 2001 13:40, Jason P. Holland wrote: This is normally associated with DNS problems. check and make sure you have /etc/resolv.conf setup correctly. Also, if you need a static entry or two, add them to /etc/hosts. Jason I'm having difficulty telnetting into my system. It appears that I get a connection right away, but it takes a long time to receive the login prompt and often I don't receive a prompt at all. Any advice would be appreciated. Bill
gv with neXtaw anyone?
I seem to remember that in the docs that come with either gv or neXtaw it says that it is not possible to link gv with neXtaw and gives the reason why. Sorry I can't remeber exactly why though. Pat Philip Lehman writes: I'm trying to build gv using neXtaw, but it segfaults. Linking to neXtaw works fine with Xaw based apps for me, but I'm not sure about replacing Xaw3d with neXtaw. Is this at all possible or am I getting somthing wrong here? -- Philip Lehman [EMAIL PROTECTED]
3Com 509B Ethernet card working in Windows but not in Linux
I have used these cards with RedHat and Debian with both 2.0 and 2.2 kernels and they work fine. One thing to look out for is that they are ISA cards with some plug and play functions. The plug and play functionality is useless. Get the 3c5x9cfg program from the dos intsallation disks that come with the card. Boot your PC and go into BIOS setup. Next go int the Plug and Play settings and reserve an innterupt for non plug and play cards. Boot to DOS and then run the 3c5x9cfg program. There should be an option to manualy setup the card, turn of plug and play and then set it to the interupt that you reserved in the bios setup. It should work fine then so long as you have the correct module to load or compile the drivers into the kernel. Pat
Not quite there with printing
Hi I have narrowed my printing problem down to the fact that on boot the /dev/printer device is not being created. If I remove generate a simple printcap using apsfilter or magicfilter for a local printer it works fine. However if I try and chain apsfilter to a Novell remote printer it doesnt create the /dev/printer when I reboot. Strangely it works OK until I do reboot. Does anyone know the correct way to link a Novell printer to a Linux print filter, I followed the ipx howto etc and this is my printcap using apsfilter :- ascii|lp1|ljet4-a4-ascii-mono|ljet4 ascii mono:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-a4-ascii-mono:\ :lf=/var/spool/lpd/ljet4-a4-ascii-mono/log:\ :af=/var/spool/lpd/ljet4-a4-ascii-mono/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-ascii-mono:\ :mx#0:\ :sh: # lp|lp2|ljet4-a4-auto-mono|ljet4 auto mono:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-a4-auto-mono:\ :lf=/var/spool/lpd/ljet4-a4-auto-mono/log:\ :af=/var/spool/lpd/ljet4-a4-auto-mono/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-auto-mono:\ :mx#0:\ :sh: # raw|lp3|ljet4-a4-raw|ljet4 auto raw:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-raw:\ :lf=/var/spool/lpd/ljet4-raw/log:\ :af=/var/spool/lpd/ljet4-raw/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-raw:\ :mx#0:\ :sh: # remote|remotep|LJ4 remote on ethernet:\ :lp=/dev/null:\ :sd=/var/spool/lpd/remote:\ :if=/usr/local/bin/nprint-script: :mx#0:\ :sh: # Thanks Pat
Re: Not quite there with printing
Patrick Kirk writes: Why not check out the ultimate printing database for Linux...http://www.picante.com/~gtaylor/pht/printer_list.cgi?predef=CJ Nice resource but it still doesnt tell my how to link in the nprint batch file to lpd and apsfilter. Pat
Printing pronlem with Potato
Hi I am trying to print to a Netware prionter using Potato and apsfilter. I had this running fine using the Debain supplied kernel but I have now compiled my own 2.2.10 kernel and I can't print anymore. lpc reports the following when I try and start the printer:- lpc start all ascii: printing enabled lpc: connect: Connection refused couldn't start daemon lp: printing enabled lpc: connect: Connection refused couldn't start daemon raw: printing enabled lpc: connect: Connection refused couldn't start daemon remote: printing enabled lpc: connect: Connection refused couldn't start daemon lpc Printing anything also gives connection refused. My printcap looks like this :- ascii|lp1|ljet4-a4-ascii-mono|ljet4 ascii mono:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-a4-ascii-mono:\ :lf=/var/spool/lpd/ljet4-a4-ascii-mono/log:\ :af=/var/spool/lpd/ljet4-a4-ascii-mono/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-ascii-mono:\ :mx#0:\ :sh: # lp|lp2|ljet4-a4-auto-mono|ljet4 auto mono:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-a4-auto-mono:\ :lf=/var/spool/lpd/ljet4-a4-auto-mono/log:\ :af=/var/spool/lpd/ljet4-a4-auto-mono/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-auto-mono:\ :mx#0:\ :sh: # raw|lp3|ljet4-a4-raw|ljet4 auto raw:\ :lp=/dev/lp0:\ :sd=/var/spool/lpd/ljet4-raw:\ :lf=/var/spool/lpd/ljet4-raw/log:\ :af=/var/spool/lpd/ljet4-raw/acct:\ :if=/var/lib/apsfilter/filter/aps-ljet4-a4-raw:\ :mx#0:\ :sh: remote|remotep|LJ4 remote on ethernet:\ :lp=/dev/null:sh:\ :sd=/var/spool/lpd/remote:\ :if=/usr/local/bin/nprint-script: Finally lsmod reports the following:- dasterdly:~# lsmod Module Size Used by 3c509 5812 1 parport_pc 5620 1 (autoclean) lp 4476 1 parport 6820 1 [parport_pc lp] ipx12188 3 vfat 11196 0 (unused) fat25120 0 [vfat] ncpfs 26000 0 (unused) dasterdly:~# I can't see any boot errors in any log files or any other lpd errors for that matter. This is driving me nut. What did I do wrong with my kernel. Pat PS I can see all the Novell servers with slist.
Re: network problem?
All VLAN stuff on switches makes use of oversize frames (or giants) as well. Basically they TAG the VLAN identifier and such like to a standard ethernet frame thus if the frame is already 1500 bytes long it becomes technically an illegal size. So if you have some switches with VLANs you may see oversize frames on the LAN but they should only be interswitch communication. Pat
GTK VIM debs
Hi Does anyone know the address of the site that has VIM debs compiled against GTK ? Ta Pat
XF86Setup problem
Same thing with my Potato system. Pat
tacacs and radius
Chad A. Adlawan writes: hello everyone ! i friend was asking me whether i can help him convert his slackware system to debian (since imthe only debian user in our group and their slak admin jumped ship) from tacacs to radius ... whatever those are. can someone please help point to me where i should look for more info regarding this ? also in question is for an ISP billing system, any suggestions? TIA, Chad Radius and Tacacs are protocolols for providing authentication and authorisation over a network. Usually they are used to authentcate a dialin users name and password when he/she connects dials in to an access server or router and then authorise the services on that access server (ie can the user have ip over ppp or ip and ipx or perhaps dialback as well). You can also use them to creat a virtual dial in profile for the user which is downloaded to whichever access box they dial in to (saves ISPs and such like configuring all the users on each box). They give central control and admin to this process, basically you configure the deamon on a central (usually Linux or Unix) box and do minimal config on the dialin routers, just point them at the daemon. Radius was developed by Livingston (now Lucent I think) and is pretty much in the public domain and almost all routers etc support it. Tacacs or these days Tacacs+ is Cisco Systems proprietry but since they have such huge market share you still see a lot of it. Cisco provide source code that will compile on Linux and also sell a supported Unix and Windows NT version. There are many implementations of Radius. Which is the best Radius or Tacacs+ depends on your envioronment, mixed vendor equipment then Radius is best but if you are an all Cisco site then Tacacs+ has more options for their equipment. Hope that clears it up for you. Pat
3C905B 10/100 card falls back to 10MB
Hi My job is networks and I can give you a major piece of advice, autonegotiation sucks ! You can run into all sorts of problems especially if both end of a link try and autonegotiate. I would ask your network manager to hard set the port on the switch you are connecting to to 100Mb full duplex and that should fix the problem (I am running Debian Potato on a laptop with an autonegotiating card set to auto and it works fine like this also RedHat 6.0 with a 3c509 set to auto and its fine). If you can't persuade the network manager to make the change the hard set your card tp 100Mb using the utils that come with the card. Pat Bill Von Elm writes: Hello All, I recently had to move a system from a 10MB ethernet segment to a 100MB segment. I edited /etc/init.d/network to re-assign the IP parameters and that worked just fine, but I can't seem to get the card (a 3Com 3c905B 10/100) to run at 100MB. During system boot-up the 100MB LED on the card is lit, but as soon as the filesystems are mounted the 100MB LED goes out and the 10MB LED comes on. Everything else works OK except the speed is wrong. I'm running the Debian 2.1 distribution and the 2.0.36 kernel. I'm kind of at a loss as to what to look at here- ifconfig does not address the speed of the interface and I can't find anything that looks like it might (I thought this speed negotiation stuff was done in hardware). If anyone can point me in the right direction I'd sure appreciate it!
Re: Emacs without X in Debian?
Xemacs certainly doesnt need X. It used to be called Lucid Emacs (no reference to X). The version in the Debian debs is compliled with X support so if you launch it in X it will run as an X app but if you start it in a real console it will run fine or in an Xterminal do Xemacs -nw and it will stay as a console app. Infact that is how I use it to read my mail on my work PC from hoem over dialin as the X version of Xemacs takes too long to load up over a modem line. Pat
RE: I am not impressed with Debian so far.
Hey, my hard drive did the sudden thrashing thing last night too. Its never done it before (well it has in NT but not in Linux). All I was doing was reading mail remotely over a dialup line using xemacs in a kterm in KDE 1.1.1 (from snowcrash). It stopped after a while (about 4 minutes) and has been fine since. This never happened before in RedHat or with Hamm. Is this a KDE thing perhaps ?. I am running on an AST M series Laptop which has 48Mb ram and a 2GB Linux partition with about 1300MB free and a 92MB swap file. Pat
XFS XFree86 3.3.3
Hi Well I figured this out eventually. Just in case anyone else is having problems with this here is how it works. xfs will start automatically. You should remove all references to the actual fontapths in XF86Config and replace them with this FontPath unix/:7100 The actual fontpaths are set ine /etc/X11/xfs/config If you want trutype support as well then you need also to referen xfstt in XF86Config like this FontPath unix/:7100,unix/:7101 Then copy your truetype fonts to /usr/share/fonts/truetype. Perhaps this should be in an FAQ somewhere as xfs and xftt are started automatically but its very unclear how to use them. Also an XF86Config generated by XF86Setup does not reference xfs. The info is their but spread through several readmes and man pages. Maybe the documentation will be better in Potatoe, but my system is just Slink with only XFree86 upgraded. Pat
Weird PPP problem
Hi I have just installed Slink on my laptop with KDE 1.1.1 debs on top of that. I am having a weird problem with KPPP. If I run as root it works fine but if I run a a user it fires up the modem OK connects to the access router and starts the ppp deamon. The weird thing is that then it NAKs the access routers request to ue CHAP authentication and then the link drops so I get a pppd died unexpctdly error. Why should the user affect whether or not I can use CHAP ? Yours puzzled Pat
XFS XFree86 3.3.3
Hi I have upgraded my Slink XFree86 to 3.3.3 using the debs. I have two questions about this. 1. Since this uses XFS should the fontpaths be defined in the config file for XFS instead of XF86Config and if so what should the top of XF86Config look like (ie the bit where it calls XFS) ? 2. Does this version of XFS support truetype fonts like the one in RedHat 6.0 or do I still need to use XFSTT and again if so can someone send me the relevent bit of XF86Config that calls XFSTT. Thanks Pat PS If anyone is using KDE 1.1.1 on an AST M series laptop do the upgrade to XFree86 3.3.3 as it get rid of all the horrible marching ants round the top of the windows - thank god as it was driving me mad.
Potatoe - usable ?
Hi After trying to get slink working with gnome and enlightenment and finding lots of annoyances such as gnome-apt causing the ftp method to fail and not being able to turn of enlightenments pager I was thinking of upgrading to potatoe. So what the consensus of opinion, is potatoe usable at the moment ? I dont mind some problems but is it reasonably stable ? Pat
Slink - just want to upgrade XFree86 and Gnome
Hi Well after several months in RedHat land (just to make it easy to keep up with Gnome) I want to move back to Debian. I have some nice new Debian 2.1 CDs but unfortunately 2.1 doesnt have XFree86 3.3.3 and I need it on both my PCs (Matrox G200 and a laptop with a GL7555). The question is how do I set up dselect or apt just to upgrade XFree86 and Gnome (and obviously any other dependencies). I don't want to upgrade the whole installation to unstable just the two bits mentioned. I would also like to retain the abilty to do an apt-get update apt-get install to fixup the system with the latest patches rather than for it to go get everything in unstable (which is what it did when I just added the unstable dir to apt sources). Thanks Pat
Going back to RedHat (but only for a while)
Reply-To: Hi Well after using Debian for the last three months I am going back to Redhat, queue flame war ... Well actually this is only going to temporary as I want to track the latest Gnome stuff and this is much easier on RedHat unless you have time to complile all the assoiciated libraries etc every other day. I just wanted to say thanks to all who helped me move to Debian in the first place and offer some thoughts after running both systems. I have run Redhat exclusively for the three yesr (4.0 - 5.1) then Debian for the last three months (Hamm upgraded to Slink). I just installed RedHat 5.2 this week. I use it every day for my work so I guess I get a good feel for how a distribution feels. Well RedHat is slick, very slick. It installs easily in 15 minutes and autodetects everything including a plug and play sound card. It also let me configure a printer to the company Novell network through its GUI config tool. Graphics card support is very good including latest Matrox products (G200). However once installed it leaves a feeling of lack of depth. The number of applications available in the distribution is limited compared to Debian. Sure you can get them from the contrib directory at Redhats ftp site but they aren't tested by anyone you just have to trust who ever decided to post them. I would rather compile my own stuff than do that. Also the documentation for RedHat is no where near as good as Debian. If you buy official RedHat then you get a very nice thick installation manual but the documentation in /usr/doc doesen't seem as good as Debians. Linuxconf wil let you config most of the basic system stuff (user accounts host files, etc) and it seems to work pretty well now (unlike in 5.1). However you wont learn how stuff actually works by using it. In conclusion RedHat is a good solid very proffesional distribution taht is easy to install. Its perfect for the market they seem to be aiming at these days (ie corporate server). However you will learn more with Debian and it has many more WELL TESTED and integrated packages. So to all those Newbies struggling along with Debian keep it up its worth it. Don't go to RedHat you may get it up and running the basics faster but after that its not as good. Debian looks like it will soon catch in this field though what with APT and so on. I will be junking RedHat as soon as possible and coming back to Debian as soon as possible. Probably once Gnome 1.0 is out and Potatoe moves to freeze. Again thanks to all those who have helped me out during the last few months and see you all again soon. Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Colour 15 bpp colour depth in X
Reply-To: Hi I am running Slink on my AST 5200M laptop. It has a Cirrus 7555 graphics card built into it. I took a XF86Config from the Cirrus and Xwindows web page and it works fine except even though I have added -bpp 16 to startx it is actually running at 15bpp. The culprit seems to be a weight 555 statement in the XF86Config. However if I remove this line all the colours look washed out and the display seems very dim. It runs NT at millions of colours so I guess it should do X at 16 bpp. Does anyone know where I should start looking to fix this problem ? Thanks Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Re: Break Key
On Wed Feb 03, 1999 at 01:12:24PM +, Daniel Marquez-Klaka wrote: Hello, How can i send an break through an xterm ?? Example: when i telnet to an cisco, the ctrl-D key don't work. Daniel Press ctrl ] This will take you bake to the telnet prompt, the type send brk press enter That will do it (works for a 2503 via a Dec Terminal server anyway) Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Re: LILO hassles
On Wed Feb 03, 1999 at 07:34:08AM -0600, Andrew Ivanov wrote: Hello, I just changed from suse-linux to debian. My installation is simmilar to the yours, only that my first hd has 1233 cylinders. If i start linux i get an : LILO UNCOMPRESSING KERNELOK LOADING LINUX and than it hangs. I had a similar problem. Try compiling your own kernel and do a make zImage instead of bzImage (Debian default). Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Printing problems
Hi I haven't made any changes to my printing setup but all of a sudden I have started getting this error when trying to print no connect permissions I am using Debian 2.0 upgraded to Slink and I do an update by APT every week. My last update was yesterday perhaps this may have casued the problem ? Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Re: Printing problems
On Tue Feb 02, 1999 at 03:39:23PM +0300, Heikki Vatiainen wrote: Patrick Colbeck wrote: I haven't made any changes to my printing setup but all of a sudden I have started getting this error when trying to print no connect permissions I am using Debian 2.0 upgraded to Slink and I do an update by APT every week. My last update was yesterday perhaps this may have casued the problem ? You probably have lprng_3.5.2-1.2 installed which has a bug with permissions. lprng_3.5.2-1.3.deb fixes the bug and it should available at the mirrors. If that does not help, check the mailing list archives and see what Chuck Stickelman [EMAIL PROTECTED] wrote under subject LPRng - Solved!! // Heikki Thanks very much the new lprng fixed the problem straight away. Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Re: mailsorting with mutt
On Sat Jan 30, 1999 at 08:19:13AM +1100, [EMAIL PROTECTED] wrote: G'day to all, could someone help this slow newbie to sort my mailing lists i'm using mutt and sifting through my mail manually is now quite a chore i would like to sort them ie: deb-user'luv'etc into there own mailbox when fetching mail perhaps a copy of a .m uttrc might be helpfull for the commands?? anyhow any help would be greatfully appreciated i uuse smail and fetchmail is the command i use to retrieve mail Thanks for your time:-) Craig McVean.. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null I use the same combination but use procmail to sort the mail into different mailboxes. Under my user area i have the following diectory structure .mutt - contains muttrc and aliases files .procmail - contains procmail recepies Mail\spool Mail\current OK first you need a .procmailrc something like this:- #Set on when debugging VERBOSE=off #Replace `mail' with your mail directory (Pine uses mail, Elm uses Mail) MAILDIR=$HOME/Mail #Directory for storing procmail log and rc files PMDIR=$HOME/.procmail LOGFILE=$PMDIR/log INCLUDERC=$PMDIR/rc.maillists INCLUDERC=$PMDIR/rc.else And a .forward file thus:- |IFS=' ' exec /usr/bin/procmail -f- || exit 75 #yourusername Next lets look at the recipies in the .procmail directory, first the rc.maillists:- :0: * ^(Resent-)?Sender: (owner-)?mutt-users(-request)[EMAIL PROTECTED] * $ !^(To|Cc|From):.*$pat.colbeck spool/Mutt :0: * ^(Resent-)?Sender: (owner-)?debian-user(-request)[EMAIL PROTECTED] * $ !^(To|Cc|From):.*$pat.colbeck spool/Debian :0: * ^(Resent-)?Sender: (owner-)?majordom-wmaker-owner(-request)[EMAIL PROTECTED] * $ !^(To|Cc|From):.*$pat.colbeck spool/WindowMaker :0: * ^(Resent-)?Sender: (owner-)?oldtools(-request)[EMAIL PROTECTED] * $ !^(To|Cc|From):.*$pat.colbeck spool/Old-Tools :0: * ^(Resent-)?Sender: (owner-)?gnome-list(-request)[EMAIL PROTECTED] * $ !^(To|Cc|From):.*$pat.colbeck spool/Gnome :0: * ^To: .*playground.sun.com spool/LandRover :0: * ^To: .*playground.sun.com spool/LandRover This delivers all mail from the maillists to ~/Mail/spool/maillistname My posix expresion programing is not the best and the occasional mail missed thes filters and gets dropped in my inbox so check them yourself. Next we have the recipie rc.else which should catch anything not matched by rc.maillists :0: * ^From: .*cisco.com spool/Cisco :0: * ^From: .* spool/INBOX OK now we have all the mail in different spool directories under ~/Mail/spool I like to have it like that incoming to a spool directory and the automove it to a current directory as I read it. This lets me use V under Xemacs as well if I feel that way inclined (not often these days). Here is my .muttrc which is just a link the real muttrc under ~/.mutt # # This rc file just sources the real one in .mutt # source ~/.mutt/muttrc Now for the relevent parts of my muttrc ### # # Pat's Muttrc # Nothing goes in if I dont know what it does. # ### reset all # reset all variables to system defaults. set move=yes# Move mail-spool to $mbox? # # Paths: # (unset implies the compile time defaults) # set folder=~/Mail # where are the mailboxes (+ and = are shortcuts) # # Mailinglists: # lists rro uk-lro oldtools gnome-list wmaker debian-user # # mailboxes filename [ filename ... ] # (check the files for new mails. Space cycles between them) # mailboxes +spool/INBOX +spool/Cisco +spool/WindowMaker \ +spool/Gnome +spool/Mutt +spool/LandRover +spool/Old-Tools \ +spool/Debian mbox-hook +spool/INBOX +current/INBOX mbox-hook +spool/WindowMaker +current/WindowMaker mbox-hook +spool/Cisco +current/Cisco mbox-hook +spool/LandRover +current/LandRover mbox-hook +spool/Mutt +current/Mutt mbox-hook +spool/Mutt +current/Mutt mbox-hook +spool/Gnome +current/Gnome mbox-hook +spool/Old-Tools +current/Old-Tools mbox-hook +spool/Debian +current/Debian # # Aliases: # (called in ~/.muttrc to load them only once) # set alias_file=~/.mutt/aliases # add the Aliases here source ~/.mutt/aliases # and include all existing aliases Well I hope this clears things up a bit for you. Pat -- Patrick Colbeck email: [EMAIL PROTECTED] Senior Analysttel: you dont seriously expect me to Azlan Ltd give that out on the internet do you ?
Database Newbie
Hi I want to make a small multi user database but am not a programmer so I don't know C. I have found mysql and the xmysql administration program and it looks like I could get my head arround creating the actual database itself. However the user interface is a differnet matter, does anyone know if there is an easy to use X client for this database or perhaps I should use something else entirely. Any suggestions will be gratefully recieved as this is a whole new area for me. Pat
Re: Would like to drop Pine for Mutt
And if you start mutt with mutt -y it will startup showing a list of mailboxes along with how much mail is in them. Pat On Sun Jan 24, 1999 at 02:47:39PM -0700, JD wrote: Quoting Peter Berlau ([EMAIL PROTECTED]): On Sun, Jan 24, 1999 at 11:19:45PM +0200, Daniel Mashao wrote: Hi Daniel, I would like to drop pine to move to mutt but my main problem is that mutt does not seem to have a folder listing like pine. I like the folder listing as it allows me to move through all my folders and see if they have new messages. You can add the folders in Your ~/.muttrc like mailboxes ! = jmb != thomas !=rainer ! =debian-de ! =debian-us !=debian mailboxes ! =isdn ! =~/Mail/slink ! =tex ! =Mail/mbox ! =sahrens I don't know if the '!' is always needed, but it works Actually, the only folders that you need to define in this manner, are the ones that you want Mutt to watch for new mail, such as various folders that you may have procmail dump mail into. Beyond that, anytime you are changing folders, or saving to a folder, just hitting '?' will give you a list of all available folders. And lastly, no, the ! is not needed except the one time. That merely tells mutt to monitor your spool for new mail. Beyond that, the '=' tells mutt to expand from your defined Mail directory. the above statement could be condensed down to something like: mailboxes ! =jmb =thomas =rainer =debian-de =debian-us =debian =isdn You get the picture. This also assumes that you want mutt to watch ALL of these folders for new mail. You would also want to: set folder=~/Mail so that mutt know where to start expanding from... HTH, jdk -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Netscape Toolbar Icon
On Fri Jan 22, 1999 at 06:54:05PM +1100, Hamish Moffatt wrote: On Fri, Jan 22, 1999 at 10:27:58AM +0700, UF Marpaung wrote: I've installed Netscape and Navigator45 in my Debian system. I use WindowMaker and Enlightenment. But why icon images in Navigator's toolbar look bad ? only 2 color (black and white) Any suggestion ? What color depth are you using in X? Probably 8 bit (default). Netscape is a colour hog. It also tends to do this with some X servers in 24 bit colour. Try 16 or 32 bit colour instead. Pat
Gnome
Hi people I want to install Gnome on Slink but don't want to scatter new versions of the libraries (eg gtk) arround anywhere. Does anyone know if you can compile it like KDE where it looks for all libraries relative to a partciular install path (eg /usr/local/gnome) so the apps that come with Slink can continue using the old libraries. Also this would be usefull for cleanup if when the next release is out (rm -Rf /usr/local/gnome). Thanks Pat
Re: Help with Smail/Fetchmail/Mutt!!!
On Thu Jan 21, 1999 at 01:25:18PM +, J.L.Gomez Dans wrote: Hi! After some tinkering about, my university has duly informed me that no mail will be delivered from outside our campus by smtp, that I need to fetch it from a POP server. As this goes, I'd like to configure everything so that my e-mail goes out with nice headers, and appears to be sent through the POP server. I know loads of people have that configuration, and I was just wondering if someone could point me in the right direction! Cheers! Jose OK here is how I do it. First Smail, change /etc/smail/config so the relevent part looks like this visible_name=esc.azlan.co.uk more_hostnames=localhost -domains hostnames=dasterdly.esc.azlan.co.uk The first line gives the domain name of your univecity and the fourth is your machine name. This will set all outgoing smtp mail so it appears to be from the mail server defined by the mx record in the DNS of your domain. Now config fetchmail using the .fetchmailrc in your home directory thus # Configuration created Mon Aug 3 13:48:10 1998 by fetchmailconf # set postmaster pcolbeck poll mail.esc.azlan.co.uk with proto POP3 user username-at-pop3 there with password password is pcolbeck here Finally you can tweek mutts config (.muttrc) thus my_hdr From: [EMAIL PROTECTED] (Patrick Colbeck) This will get mutt to make the from address anything you want. Hope this helps Pat
Re: Help needed with laptop
I had a similar problem with an AST M series laptop. Turned out it was the way Debian compile the kernel for the distribution. Try compliling your own kernel and make sure you make a zImage kernel rather than a bzImage kernel and hopefully you should be OK. Pat On Thu Jan 14, 1999 at 06:43:45PM +, tracheotomy bob wrote: Hallo all, I relly hope you can help me with this REALLY annoying problem. I recently bought a Sony VAIO 747 (impulse) and I want to put Debian on it (I have 2.0 (hamm)). I can put Red Hat 5.1 on it easily. But Debian is a bitch. Sometimes it starts and sometimes it doesn't. It loads the root.bin alright but after loading most of the kernel, the laptop reboots. Sometimes it doesn't. If it does start then I can install the CD OK, but on the reboot I get a 1FA prompt. If I press 1 for /dev/hda then I get a prompt of 1234F. It doesn't what I do the machine will not start with Debian on it. Red Hat works perfectly so it's not the hardware. If the CD boots then I'm assuming that the kernel works OK with the hardware, but the fact that it aborts during the 'loading kernel' stage is suspect. I'm inclined to think that it may be the way lilo is operation. Red Hat seems to install lilo into /dev/hda and debian seems to install to /dev/hda1. Maybe I'm wrong here I could use that cleared up. Anyway, any help with this will be much appreciated. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
dosemu and freedos
Hi Looking through past messages of this list it seems that the freedos shipped with Debian (Slink) doesn't support lredir. Is their any point shipping it then as this seems to prevent any access to your hard drives. Also the setup-dosemu program appears to be missing, does anyone know if there is a deb with it in or do I need to get it from the sources. Finally is it possible to add a program to the hdimage.first using mcopy, whenever I try it I get an error thus:- dasterdly:/home/pcolbeck/tsmain/USERS/PCOLBECK# mcopy ./NET33X.EXE g: Internal error, dirty end too big 8800 6e00 1a00 28160 Streamcache allocation problem:G 3 dasterdly:/home/pcolbeck/tsmain/USERS/PCOLBECK# Thanks Pat
Re: removing ^M from a text file
I seem to remember that crypt++ with emacs/xemacs was supposed to do this but I just tried it and it didn't. I may be doing it wrong though, does anyone have any comments ? Pat On Tue Jan 12, 1999 at 03:13:41PM +, M.C. Vernon wrote: Dear all, I had the misfortune to be stuck with MSWrite over the vac, and although I told it to save as plain text, it didn't. I can delete most of the cruft, but there is ^M at the end of each and every line. I've tried catdoc, word2x, and sed, but none of them will remove them. Can you suggest anything, please? please cc me as I'm not on this list ATM. Thanks, Matthew -- Elen sila lumenn' omentielvo Steward of the Cambridge Tolkien Society Selwyn College Computer Support http://www.geocities.com/Area51/Chamber/8841/ http://www.cam.ac.uk/CambUniv/Societies/tolkien/ http://pick.sel.cam.ac.uk/ -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: colored comman prompt
It certainly is. Check out http://bash.current.nu/ Themed bash prompt no less, with all different colours and escape chars etc. Pat On Thu Jan 07, 1999 at 09:57:48PM +0800, [EMAIL PROTECTED] wrote: is it possible to make the command prompt colored? like the escape sequence that you do in DOS? [off topic] how to insert an escape character in vi*? k e c h i e -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Papersize and SO5
Hi I am trying to print to a HPLasterjet 4 Plus from StarOffice 5.0. I have my papersize set to A4 under Debian and also in StarOffice but whenever I try to print the printer gives an error saying it wants letter size paper loading ! I am using the generic postscript driver in SO. Has anyone seen anything similar to this. NB WordPerfect 8 using its internal printer driver works fine. Pat
startx
Hi Some files have gone off my PC (I was upgrtading to slink). The most obviuse is startx does anyone know which package its in. I have tried reinstalling xbase but that didnt help. Pat
Made a mistake in dselect
Reply-To: Hi I made a mistake in dselect and unselected xbase, I aborted out without removing any packages. Is there a way of resyncing dselect so taht everyting I actaully have installed is selected so next time I run it it doesnt remove X ? Thanks and seasons greetings. Pat
What does this mean with dselect
Reply-To: Hi Can anyone tell me what messages like this mean when I use dselect with the apt access method. Checking system integrity...ok The following packages have been kept back gobjc smail g++ egcc doc-linux-html wxhelp doc-linux-text Thanks Pat
Re: Mutt colours
Oops sorry, I think my original post wasn't very clear. I know how to change mutts colours I have been using mutt on and off for two years. Its just that I really like that particular colour combination and wondered where it came from. Just the colour section of muttrc would be great. Pat On Wed Dec 16, 1998 at 09:00:56PM +0500, [EMAIL PROTECTED] wrote: Subject: Mutt colours Date: Wed, Dec 16, 1998 at 04:44:29PM + In reply to:Patrick Colbeck Quoting Patrick Colbeck([EMAIL PROTECTED]): Reply-To: Hi This isn't really important but here goes anyway. I have over the last week or so being introducing myself to Debian and playing with Hamm and Slink. At one point my Mutt mailer was running with a nice colour setup (not one I made rather it was installed by one of the Mutt debs I used) that used quite a lot of green rather than the usual red and blue setup. I have reinstalled my machine with Hamm and a few bits of Slink and now it has gone back to the old red and blue config. Does anyone know where the other colour scheme came from as it was much easier on the eyes. For the life of me I can't figure out which mutt dep it was in. Pat Pat The colors (sorry colours) are setup in your .muttrc file. The Manual covers it very well. Here is an example from my .muttrc color attachment green black # .. color treeredblack # index color header brightyellow black ^Cc: color header brightyellow black ^Date:# pager header color headercolor4 color6 ^Subject: color indicator white blue # index color normal white black # pager body color quoted brightyellow black # pager body color quoted1 brightcyan black color quoted2 brightgreen black color signature redblack # pager body color status white blue # index status bar default: black white color index red default ~z10k HTH -- The problem with program verifiers is that they tend to cheat at toy problems in order to get results. ___ Wayne T. Topa [EMAIL PROTECTED] -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Smail not delivering mail to local domain
Reply-To: Hi Recently I posted about smail being able to deliver mail via smtp to remote domains but not to users on my own domain unless they existed on the box smail was running on. Basically I wanted mail for these users to be delivered to the company mail server. Several people have offered solutions using smartuser or smartrelay but for once I have come up with a simple solution myself so incase anyone else has the same problem here it is. In /etc/smail/config as setup by debian there is a line like this hostnames=hostname.domain:domain eg hostname=workstation1.somecompany.com:somecompany.com change it to hostname=hostname.domain.com eg hostname=workstation1.somecompany.com Now it should work properly and hand off all mail for users not on workstation1 to whatever dns turns up as an mx record dor the domain. If anyone has any comments about any negative effects of this solution please mail me. Also I really think that this problem should be in a Debian FAQ (not necesarly with my solution - I am no mail expert) as it is a really common config. The solution using sendmail is in the sendmail FAQ and RedHats installation manual but its been a pain to sort with Smail as there seems to be very little documentation for Smail (possibly I just haven't found it yet) Pat
Mutt colours
Reply-To: Hi This isn't really important but here goes anyway. I have over the last week or so being introducing myself to Debian and playing with Hamm and Slink. At one point my Mutt mailer was running with a nice colour setup (not one I made rather it was installed by one of the Mutt debs I used) that used quite a lot of green rather than the usual red and blue setup. I have reinstalled my machine with Hamm and a few bits of Slink and now it has gone back to the old red and blue config. Does anyone know where the other colour scheme came from as it was much easier on the eyes. For the life of me I can't figure out which mutt dep it was in. Pat
Re: Debian too difficult, Red Hat?
OK well I guess I am qualified to put in my two hapeth here as I have just moved to Debian after using RedHat since 4.0. RedHat is VERY easy to install in some ways. Its hardware detection is very very good and it takes about 15 minutes to do the whole thing. It does however have some major disadvantages 1. There menu system for X doesnt change depending on what packages you have installed so say if you have no Xemacs the menu option will still be there (this may have changed in 5.2 though as I have'nt tried this yet. 2. Upgrading to latest packages is not automatic you have to download them from the errata ftp site one by one after readig which ones you need. 3. A lot is left unconfigured, RPMs are quite good but they doen't go into config mode after they install the way debs do, so say you install sendmail it won't then ask you some questions to config it for your PC you will just have to hack away by hand. 4. LinuxConf which they now use for almost all system config is a complete pile of *. Sorry to those who like it and the author as its a great idea but it just annoys the hell out of me. Try using it to setup IPX connectivity sometime and see what I mean. If it every gets sorted out though it will be nice. 5. The printing subsystem is a complete mystery to anyone who isn't a wizard at reading very long bash scripts. I thing it uses nenescript but god knows exatly how its bolted together. It works fine if you have a local printer thats in the supported list but for anything else its a nightmare. 6. RedHat change a lot about the way a package installs eg where it puts its files and also config scripts, well so does Debian to some extent but Debain tend to have Readmes to tell you what they have changed RedHat doent you have to guess or examin the patches in the SRPMS. Eventually I started to feel constrained and annoyed by RedHat, its a nice shinny slick system in some ways but Debian seems to have more depth. Also changing over at a later date will give you another learning curve (which I am going through right now :) ) as things are different enough between the two to throw you. Don't get me wrong RedHat is not a bad distribution and they have done a lot to raise the profile of Linux,I liked it a lot better than SuSe or Slackware but I think Debian is better and gives more control. Also whilst RedHat may initially seem to be easier in the medium term and long term it really isn't. Pat - in opinionated mode. On Wed Dec 16, 1998 at 08:46:01AM -0800, Clyde Wilson wrote: I agree with you Kent. Debian is much too difficult to start out with. Redhat removes a lot of options to give you a working system without much configuration on your part. Later, when you are shooting for guru-ship you can go to Debian and really get into it. Both systems are a tremendous amount of fun!!! On Tue, 15 Dec 1998, KTB wrote: Hi, thanks to all the people who have offered advice with configuring X-windows. I have not been successful and am brain dead at this point. I also tried hooking up to the internet with the same result. I chose the debian release because I wanted to learn more about computers (I have only used a pc off and on for the past year) and I like the philosophy behind Debian. I am wondering if Debian is just too difficult for me at this point. I am wondering if maybe I should try Red Hat, I have heard it is easier to install, and then come back to Debian. Does this sound like a logical progression to anyone? I don't have experience with either one so I just don't know the best course to take.
Smail and POP3
Reply-To: Hi Well I have got my X working now (finally found the old SUSE Matrox server on the Webi its gone from the SUSE hompage). Now I have a smaill Smail problem. I am using Smail to send mail and fetchmail to pull incoming mail from the company POP3 server. I have set my visable domain to be esc.azlan.co.uk useing the smail-config program and mail going to other domains is OK. However if I try and send any mail to anyone who is in my own domain (esc.azlan.co.uk) Smail bounces it back at me claiming that user does not exist. What I think is happening is that it is looking for the user on this PC and not trying to forward it to mail.esc.azlan.co.uk. When I used to use sendmail there was a DM option you could set so as to masquarade as the mail server for a domain. Basically anything going to another domain was sent direct from your machine with a header rewrite (replacing [EMAIL PROTECTED] with [EMAIL PROTECTED]) and any mail sent just to the local domain was sent to the main domain mail server. Is there a way to get this behaviur with Smail ? Pat PS Thanks for all the help I have had from everyone so far in migrating from RedHat to Debian, even after using RedHat for a couple of years its more of a culture shock than you would think changing to a new distribution.
XFree86 3.3.3
Reply-To: Hi To change my main PC at work from RedHat 5.1 ro Debian 2.0 I need to be able to run XFree86 3.3.3 as I have a Matrox G200 graphics card (currently I am running the SUSE special xserver for this card). Are there and debs available for 3.3.3 unstable or whatever or do I just go get the binary release from ftp.xfree86.org ? Pat -- - Patrick Colbeck Senior Technical Analyst _ Azlan Ltd Tel:+44 (0)1904 691997 Lion House Fax:+44 (0)1904 692112 4 Pioneer Business Park Clifton MoorEmail:[EMAIL PROTECTED] York England YO3 4TN _
OFF TOPIC slashdot
Reply-To: Hi Sorry about the offtopic post, please reply to me privately. Does anyone know if there is a problem with slashdot.org at present. I can't resolve its ip address using the root domain servers. I can however resolve it using some of the web hosted DNS checkers on peoples web sites. Also once I get its ip address this way I can't ping it or browse it etc. Traceroute shows the trail dying somewhere in the USA, (I am in the United Kingdom). I would e-mail Rob Maldera directly but since I cant resolve his address I cant mail him ... Pat -- - Patrick Colbeck Senior Technical Analyst _ Azlan Ltd Tel:+44 (0)1904 691997 Lion House Fax:+44 (0)1904 692112 4 Pioneer Business Park Clifton MoorEmail:[EMAIL PROTECTED] York England YO3 4TN _
Unsure of how Debian is updated
Reply-To: Hi I am converting from RedHat to Debian and in the main am pleased with Debian. One thing I can't quite work out though is how Debian handles updated packages. For example XFree86 3.3.3 is out now, will this eventually be integrated into the 2.0 release or are the packages in this release frozen except for bug fixes ? One of RedHats nice features is that there is the upgrade drectory on their ftp server and you can just see all the packages that have been upgraded since the release of that version of RedHat is there an equivalent for Debian ? Pat
kde and jpeglib6a
Reply-To: Hi I am trying to install kde on debian 2.0. When configuring kdelibs it returns an error Tou need jpeglib6a. Please install the kdesupport package kdesupport is just a collection of standard libraries not anything specific to kde so I would rather use .debs to install these. I have installed libjpeg-progs_6a-11.deb libjpegg6a_6a-11.deb libjpegg-dev_6a-11.deb I presume these are the libraries its looking for but it still doesn't pick them up. Any ideas ? Pat -- - Patrick Colbeck Senior Technical Analyst _ Azlan Ltd Tel:+44 (0)1904 691997 Lion House Fax:+44 (0)1904 692112 4 Pioneer Business Park Clifton MoorEmail:[EMAIL PROTECTED] York England YO3 4TN _
Boot problems
Reply-To: Hi A few days ago I reported a problem booting from the stock 2.0.34 kernel distributed with Debian. My laptop just rebooted after displaying loading Linux .. I fixed it by compiling my own kernel fronm the 3.0.36 sources and though well there you go a 2.0.34 bug. Not so I am afraid. Today I tried using the kernel-package to produce a debianised custom 2.0.36 kernel and guess what this one wont boot either ! If I do a straight make zImage. make modules etc it works fine using the same config file.It looks like a problem with the way the debian kernel packager works. Has anyone else seen this ? Pat
dselect problem on install
Reply-To: Hi Has anyone seen this problem with Debian 2.0 (from Cheapbytes) before ? It does the initial install fine and the reboots and asks you what type of install you want eg Scientific workstation etc and then runs deselect. When it gets to installing the packages all goes fine until it starts to ask about creating the XF86Confing file. At this point if I answer YES or No it no longer responds to any keyboard input. I can switch to another vty and kill -s 9 the current post install process and then when I switch back to the original vty it has moved on to the next package and proceded OK until it needs some user input but the keyboard is still locked out. Any ideas ? Pat
Re: Lilo and MBR
Right on target Noah ! I downloaded 2.0.36 for kernel.org this morning and recompiled, what do you know it boots perfectly now. Well now to learn the intricies of dselect, an easy to understand intuative package... not. Pat On Tue Nov 24, 1998 at 11:50:44AM -0500, Noah L. Meyerhans wrote: -BEGIN PGP SIGNED MESSAGE- On Tue, 24 Nov 1998, Patrick Colbeck wrote: Following up to my prevous mail is the lilo or mbr shipped with Debian 2 faulty ? When trying to boot from hda3 (hda1 is win98, hda2 is swap) it just says loading Linux the reboots. It never gets to vmlinuz. This is like what happens when it cant find vmlinuz when you forget to run lilo. it boots fine off the CD with rescue root=/dev/hda3 and RedHat 5.1 works fine. I didn't see your original mail, so I might not have all the facts. However, your problem actually looks a lot like a kernel problem that shows up on some motherboards based on the Intel TX chipset. If that's what you have, then you might need to try a more recent kernel, like 2.0.36. You should check the kernel mailing list archives for more details. I remember this problem being discussed around the beginning of this year. noah
Install problem
Reply-To: Hi I am having problems installing Debian 2.0 from a Chapbytes CD. It installs OK but on rebooting I get a message that says 3FA and if I press return I get the lilo prompt. If I press return again I get the Loading Linux .. then the thing just reboots. Its an AST Ascentai M serias laptop (200MMX) and it was previosly running RedHat 5.1 with no problems. I have Win98 installed on hda1 with the Linux swap on hda2 and the root partition on hda3. The install was done totally from CD as you cant have a CD drive and floppy drive in the laptop at the same time. If I then use the dustribution CD as a rescue disk and tell it that hda3 contains the root partition it boots fine and I can rerun lilo but then again rebooting from hard drive gives the same error. Yours stumped Pat
Lilo and MBR
Reply-To: Hi Following up to my prevous mail is the lilo or mbr shipped with Debian 2 faulty ? When trying to boot from hda3 (hda1 is win98, hda2 is swap) it just says loading Linux the reboots. It never gets to vmlinuz. This is like what happens when it cant find vmlinuz when you forget to run lilo. it boots fine off the CD with rescue root=/dev/hda3 and RedHat 5.1 works fine. This is my first Debian install so maybe I am just missing soething that is different from the way RedHat does it. Pat