From: Digestifier To: Linux-Admin@senator-bedfellow.mit.edu Reply-To: Linux-Admin@senator-bedfellow.mit.edu Date: Thu, 21 Oct 93 16:15:03 EDT Subject: Linux-Admin Digest #119 Linux-Admin Digest #119, Volume #1 Thu, 21 Oct 93 16:15:03 EDT Contents: Re: Sending mail to MX forwarded mail addresses with Smail (Robert W. Brewer) Re: UNIX sysadmin FAQ- proposal and volunt (Dave Platt) NEC SCSI CDR-84 CD-ROM no (Francois Vigneron) Re: Auto-reply robot for Linux? (Hans de Hartog) Problems after installing SLS to SLACKWARE (Wolfgang Thronicke) Single user? (Tod Olson) newuser shell script (Mike Sperry) A number of network problems (J Dominick H Layfield) Send-mail HELP (Kim C. Callis) Pb. with libc.so.4 file (or link) (Richard Migneron) Re: lilo, vmlinuz zImage (Piercarlo Grandi) Re: a lost+found magic! (Jim Dodd) Re: Linux/X/Motif (DAVID L. JOHNSON) Help with routing under Linux (Alex Liu) Security for dialin (P. David Gardner) ---------------------------------------------------------------------------- From: rbrewer@rwb114.rh.psu.edu (Robert W. Brewer) Crossposted-To: comp.os.linux.help,comp.os.linux.misc Subject: Re: Sending mail to MX forwarded mail addresses with Smail Date: 20 Oct 1993 02:27:50 GMT Reply-To: rbrewer@psu.edu Daniel M. Coleman (dcoleman@mbs.telesys.utexas.edu) wrote: >Such is one case. Is there anyway to get linux Smail to send to such >addresses without having to manually specify the fowarded-to host? In /usr/lib/smail/routers, comment out the "match-inet-addrs:" section, and uncomment the section further down which starts "match_mx_hosts:". match-inet-addrs tells the mailer to look up the IP address of the host the mail is destined for, and then connect directly to its smtp daemon. This completely fails for hosts with an MX record. The man page states that match_mx_records is a fully BIND compatible mail router that does all the proper lookups of MX records. I can now send mail to UUCP hosts and other such beasties that I couldn't touch before. All this is documented in the man page, but it takes a whilel to wade through it. -Rob -- Robert W. Brewer Ask me how Jesus saves! Finger for PGP public key... ------------------------------ Crossposted-To: comp.unix.admin,comp.sys.sun.admin,comp.sys.sgi.admin,comp.admin.policy From: dplatt@ntg.com (Dave Platt) Subject: Re: UNIX sysadmin FAQ- proposal and volunt Date: Wed, 20 Oct 1993 02:50:07 GMT >rh>> - Mounting /var/spool/mail via NFS >rh>> - don't >rh> >rh>But why not? Because of file locking? Sun recommend it... Or at least >rh>tell you how with SunOS.... > >That's something I would like to know too. We do it here for years now without >any problems. We have Sun and HP clients and Sun Servers and as I said: NO >PROBLEMS. So what's this all about? NFS-mounting of /var/spool/mail seems to work very well, if two conditions are met: [1] All of the NFS-client machines send _all_ "local" mail to the NFS server, for delivery by the server's sendmail/smail daemon. [This avoids the need to give the client machines trusted-root access to the spool directory, and also avoids sendmail/sendmail collisions during delivery.] Current versions of SunOS sendmail (and Solaris too, I assume) have this feature built in... regardless of what the local-delivery ruleset may say, SunOS 4.1.x sendmail does SMTP delivery to the NFS server if it realizes that it is an NFS-client of /var/spool/mail. [2] Some form of network file locking is used to keep the mail user-agent (mail, mush, etc.) from tripping over the mail transfer agent (sendmail/smail), so that mail is not accidentally lost if a user-agent tries to update a mailbox while sendmail is trying to extend it (or vice versa). SunOS seems to do both of these things just fine (these days... things were not so rosy in earlier versions of SunOS). I suspect that Linux would require manual configuration of sendmail/smail to meet the requirements of rule [1], and I don't know whether it supports locking as required in [2]. -- Dave Platt dplatt@ntg.com ..or.. ...netcomsv!ntg!dplatt USNAIL: The 3DO Company, NTG division 2470 Embarcardero Way Palo Alto CA 94303 ------------------------------ Subject: NEC SCSI CDR-84 CD-ROM no From: francois.vigneron@cld9.com (Francois Vigneron) Date: Tue, 19 Oct 93 14:59:00 -0600 @SUBJECT:NEC SCSI CDR-84 CD-ROM not recognized AD>My Linux system doesn't recognize my NEC CDR-84 SCSI CD-ROM drive. At boot, AD>I see my Adaptec 1542 accessing the drive (LED flashing), also when AD>Linux tests for SCSI devices, the drive is accessed but not recognized. AD>I tried to remove/add terminators to every SCSI device - no avail (all AD>other devices work perfectly). Anybody else who had this problem? AD>Or somebody, who uses the drive successfully? I use a NEC CDR-84 CDROM Driver on my Anonymous UUCP site. I have had enormous problems with it. The answer is to use this drive as the last device on the SCSI chain and to disable "sync mode" + "Parity" mode on the Adaptec controler. On my system, I have 2 HA1540b: - 1st SCSI controler: IBM 1Gb SCSI disk, GigaTape (DAT SCSI Unit) - 2nd controler: NEC CDROM, Hitachi CDROM. I tried to have all 4 devices on the same controler: it was a real PITA. I haven't got an e-mail address, but here is my FAX number: +33 1 42 47 06 11 (France, Paris). --- * SLMR 2.1 * My Internet adress is : 192.69.69.69. Girls are welcome. * Cam-Mail * Modula.bbs@Top50.cld9.com ------------------------------ From: dehartog@ccult1.comcons.nl (Hans de Hartog) Subject: Re: Auto-reply robot for Linux? Date: Wed, 20 Oct 1993 09:11:31 GMT dgardner@netcom.com (P. David Gardner) writes: >I need an auto-reply robot for Linux; does one exist? This is where >someone outside would send an email message to info@domain.whatever, and >then when received, domain.whatever would return a canned message or file. I have the following tiny mail-server running on Linux. It's very insecure but it's only local so I don't bother. The trick is as follows: create an account to do the auto-reply (in my case: mailsrvr). In the home-dir of that account make the file .forward that contains the line: "| /usr/users/mailsrvr/ms" This way, any mail is piped thru the shell-script ms which looks like: #!/bin/sh cd `dirname $0` trap "rm -f $$.00 $$.01" 0 csplit -s -f $$. - '/^$/' fgrep -is mail-server-request $$.00 && for fl in `cat $$.01` do if [ -f "$fl" ]; then ( uuencode `basename "$fl"` < "$fl" ) 2>&1 elif [ -d "$fl" ]; then ( ls -alF "$fl" ) 2>&1; else cat ms.hlp; break; fi done | mail `sed -n 's/^From: *//;s/.*<\(.*\)>.*/\1/p' $$.00` exit 0 The help-file (/usr/users/mailsrvr/ms.hlp) looks like: Dear Requester, Your request should only contain pathnames, separated by whitespace. No pathname may refer to a special file, named pipe, socket, fifo ... Directories will be listed (ls -alF), files will be uuencoded and the results concatenated. To separate responses, send separate requests. The rest of your request was ignored. What you want is not the mail-server functionality, but someway or another, you have to figure out the return address from the header you get; so, what you need is something like this line: mail `sed -n 's/^From: *//;s/.*<\(.*\)>.*/\1/p' [the-header-you-get]` followed by the file you want to send as a response. Good luck! HdH. -- _____________________________________________________________________________ Hans de Hartog, dehartog@comcons.nl, Voice: +31 348033100, Fax: +31 348033181 Committed Consultancy BV, Korenmolenlaan 1b, 3447 GG Woerden, The Netherlands ------------------------------ From: wolf@hador (Wolfgang Thronicke) Subject: Problems after installing SLS to SLACKWARE Date: 20 Oct 1993 10:24:59 GMT Hi, I recently installed Slackware 1.0.4 (A+X series) and added the SLS 1.03 TeX series. After this most X binaries seem to be broken or part of the dynamic libs... the error message I get is: Corrupted fixup table after starting an X program. Has anyone encountered similar problems? Wolfgang wolf@cadlab.de ------------------------------ From: tao@stax.lib.uchicago.edu (Tod Olson) Subject: Single user? Date: Wed, 20 Oct 1993 20:27:05 GMT I'm running SLS 1.03. How do I boot single user without first going multi-user? Please respond by mail. Thanks. -- Tod A. Olson "The lawyers like it, but you'll ta-olson@uchicago.edu never get it past advertising." ------------------------------ From: mps@colmiks.com (Mike Sperry) Subject: newuser shell script Date: Thu, 21 Oct 1993 00:32:59 GMT Hi, Does anyone know where I can get a script (or program) that queries new callers for name, etc, that works under Linux (SLS 1.03)? It would be nice if it left the data in a text file that I can simply read. Thanks! Mike (please mail to uunet!colmiks!mps or just post) -- Mike (if you mail, mail to uunet!colmiks!mps) ------------------------------ Crossposted-To: comp.os.linux,comp.os.linux.help From: ct93010@oxford.ac.uk (J Dominick H Layfield) Subject: A number of network problems Date: Thu, 21 Oct 93 10:14:42 BST PLEASE REPLY TO u90rjm@uk.ac.ox.prg =================================== I've been having a few difficulties setting up loopback links for tcp/ip on my Linux (0.99 pl8) box. Wonder if anyone can help ... I've set up the kernel to include all the network stuff. I've then set the machine up so that it runs /etc/rc.net on startup - successfully starting portmap, inetd, named etc. No errors are reported to /usr/log/*. I've also set all the config files for named up using the /etc/hostcvt.build script. So far so good ... I can then rlogin into my own machine IF I AM ROOT. But I get 'unknown host' messages if I am any other user. If I run named with debugging switched on I can see the message missed 'nodename' as '' when I attempt to reference my machine as a non-root user. Apparently root doesn't seem to use named! If anyone is interested I can send the named config files to them by mail, and a debug log!! The second network problem is that a loopback mount: mount :/dev/hda7 /mnt fails with an 'RPC timed out' message, resulting in mountd continuously retrying ... any ideas? I'm using the networking stuff that came with SLS 1.01 which I believe is pre "net-2" (whatever that means). I really don't want to be upgrading just so I can do loopback stuff. Its not essential to me, but it _should_ work, and talk isn't that great without it (and I _do_ have async terminals, so talk is a bit useful!). Regards. ------------------------------ From: callis@noc.usfca.edu (Kim C. Callis) Crossposted-To: comp.mail.sendmail,comp.mail.uucp,comp.os.linux Subject: Send-mail HELP Date: 21 Oct 1993 10:21:16 GMT Here is the scenario. I just applied for and received a domain name under the .us domain. So I turn to my friendly neighborhood System Admin (you know, the guy they pay alot of money to figure out things) and I ask him to set me up for a UUCP mail feed. The very informed response i got was to figure out myself how to do it and let him know. I figured the first place to start would be sendmail. I figured that there would have to be some rule written which would take mail which is sent to jdoe@callis_group.sf.ca.us (which of course is being routed to USF) and send it to the uucp spool area for my Linux box to call in and pick up. So how would I do that, and what files would have to be updated? The sysadmin has already placed me in the DNS (I thought that was unnecessary), and i thought that a MX record needed to be created. Secondly, you UUCP types, I need to know how to set up a machine to receive UUCP mail request, and forward them accordingly. I have read Managing UUCP and Usenet from O'Reilly and Associates, but it covers calling out under UUCP, not dialing in. Any help would be appeciated. -- ******************************************************************* * Kim C. Callis * * Univ. of San Francisco, San Francisco, CA * * EMAIL: callis@usfca.edu * ******************************************************************* LINUX: Right now! The choice of a new generation.... ------------------------------ From: adopt@CAM.ORG (Richard Migneron) Subject: Pb. with libc.so.4 file (or link) Date: 21 Oct 1993 08:31:29 -0400 Hello, I just finished installing the SLS release out of tsx-11.mit.edu (the whole thing, a,b c, d, s, t and X), everything went perfectly, I did it in about 30 min., swapping disks the fastest I could (I got to get to work), this gives a new meaning to the words 'Disc Jockey' :-) Anyhow, when I try to boot from the HD, it goes L then I then it freezes. When I try from the floppy boot disk, it goes on pretty good until almost to the end (it is very close to where the login prompt would be) and finally says the following : init: can't load library '/lib.so.4' No such library Now, I suspect that either the lib is not there or the link was badly or not constructed. I tried to mount the partition from the A1.5 disk to create the link myself (or correct it) but it didn't like any of the 'mount -t ' commands with either e2fs linuxext or other type of mount partition type. Did I get a bad disk or file, I hear you ask. I think not since all the sizes on the net and on my disks compare and no errors occured during the install. Anybody out there can help me out with this. Love what all of you are doing for Linux, when will there be a Linux for IBM RS/6000 (AIX sucks eggs). - Richard ------------------------------ From: pcg@aber.ac.uk (Piercarlo Grandi) Subject: Re: lilo, vmlinuz zImage Reply-To: pcg@aber.ac.uk (Piercarlo Grandi) Date: Thu, 21 Oct 1993 15:24:25 GMT >>> On Tue, 19 Oct 1993 21:47:26 GMT, rfrankel@us.oracle.com (Rick >>> Frankel) said: Rick> I was wondering the following: 1. How do you make linux boot off Rick> of a partition w/ lilo. There is rather well done manual for this... Rick> 2. What is the relationship between vmlinuz and zImage. Are both Rick> actually needed (i.e., if lilo boots w/ zImage, why is vmlinuz Rick> needed?) Well, you can make lilo boot with vmlinuz, why not? BTW, one thing which is rather crucial but is not clearly singled out in the manual is that in order to boot Linux LILO must have three files for its own use: a boot map {this it builds when you run it} a boot loader {this is usually boot.b} a Linux kernel {this is usually a copy of zImage} and that since LILO builds pointers to their blocks, if any is altered LILO must be rerun to recompute those pointers. I have found that the most flexible option for booting is actually botting from a floppy; this sounds backwards, but it's actually rather convenient. It is also easy: * fdformat /dev/fd0H1440 * mkfs -t ext2 /dev/fd0H1440 * mount -t ext2 /dev/fd0H1440 /mnt creates and mounts an empty partition on a floppy. The I copy /etc/lilo/boot.b to /mnt/lilo.boot, zImage to /mnt/vmlinuz, and I specify in /etc/lilo/config the appropriate lines (I call the map /mnt/lilo.map); so in the end one has in the root of the floppy the following files: "lilo.map lilo.boot vmlinuz". Other kernel images can be added; it's useful to have a generic kernel with almost every device configured in, and I call it "vmlinuz.any". That's all. One can not only put several alternative kernels ona floppy, have different boot floppies, and so on. It is also useful to have another floppy with a miniroot filesystem on it, for disaster recovery; to use it one simply boots the usual floppy, but with a different root directory (the floppy), and perhaps with a suitable RAM disk size. Having a boot floppy has at least two advantages, apart from flexibility in choosing what/how to boot: if you have another OS, you need not change its booting arrangements at all, and installing Linux can be totally transparent; there is no problem with limitations of the disk BIOS, e.g. wrt more than 1204 cylinders, as the boot is entirely from the floppy, and once the Linux kernel is running, there is no problem. ------------------------------ From: jimd@netcom.com (Jim Dodd) Subject: Re: a lost+found magic! Date: Thu, 21 Oct 1993 16:55:08 GMT Rene Angel Sepulveda (rene@inf.utfsm.cl) wrote: : Hello! : I have a lost+found directory into a partition (/user) : but I only have one disk! : : My disk is a IDE 540 Mb. I did 3 partitions on it, they are: : : /dev/hda2 / ext2 defaults : /dev/hda3 /user ext2 defaults : /dev/hda1 none swap sw : Why I have a lost+found directory inside of one disk? : I have others linux machines and they have not lost+found : directory. : $ cd /user : $ ls : $ user1/ lost+found/ user2/ user3/ : user4/ user5/ user6/ : How can I fix it?? I'm not sure why you have a lost+found on only one partition. But it is not a problem. In fact, you should create one for each partition. when you run e2fsck, if it finds files, ie inodes, that are allocated but have no entry in any directory, e2fsck can reconnect it if you desire. It is done by creating a file in the lost+found directory with a name that correstondes to the inode number. This gives you a chance to "save" the file. I would recommend you create lost+found directories on _ALL_ of your partitions. this can be done by the following script: mkdir /lost+found i=100 while [ i -gt 0 ] do touch tmp$i i=`expr $i + 1 ` done rm tmp* You do need to do the touch part to insure that there are enough slots already in the directory, so e2fsck doesn't have to try and allocate additional blocks for these directory entries when the free block list might be messed up. Hope this helps. -- Jim Dodd email: jimd@netcom.com ------------------------------ From: dlj0@ns1.cc.lehigh.edu (DAVID L. JOHNSON) Subject: Re: Linux/X/Motif Date: 21 Oct 93 16:03:03 GMT In article , danw@hebron.connected.com (Dan Wilder) writes: >etxias@mega.ericsson.se (Inge A. Suhr) writes: > >>Is there any Motif stuff available for Linux, if so, what and where ??? >>Pointer to any descriptive summary of requirements is appreciated.... > >>Ciao // IASuhr > >Mail sales@metrolink.com, they have a port of 1.2 for Linux. Needs >Xfree86 and maybe XS3, see the X faq for details, a suitable >graphics card, tweaking, etc. It does not require XS3. I'm using it quite happily with SLS-1.02/XFree-1.2 on a graphics card which shall aremain nameless to avoid flames about not dealing with a certain company (I bought it innocently!). Rumor was that the Metolink install >had a few warts, maybe they've made it slicker since the post. I had absolutely no problems, beyond getting it customized the way I wanted it. It ran fine right out of the box, and in my pinion is a great product. I got it early on, as well. I don't know of any problems with installation. -- David L. Johnson ID: dlj0@lehigh.edu Department of Mathematics Lehigh University, Bethlehem, PA 18015 Telephone: 215-758-3759 (office) 215-282-3708 (home) #include ------------------------------ From: aliu@aludra.usc.edu (Alex Liu) Subject: Help with routing under Linux Date: 21 Oct 1993 12:03:07 -0700 (Hi, I posted trough the Mailing lists and got an overwhelming response. ZERO messages, so here it goes again, over here.) Hi, (I checked the NET2_HOWTO and I see no mention of ROUTED). Basically I am trying to use a Linux machine as a router (gateway?) I have 2 networks, using IP address: 198.187.207.xxx and 150.8.4.xxx. And I want to route them trough a Linux box. THis is a 386DX25 with 4MB of RAM. (Should only be running as a router, so no X or any heavy duty things) I installed to SMC Elite 16 cards. eth0: 0x300,IRQ10, ADDR 0xD0000 eth1: 0x280,IRQ11, ADDR 0xD4000 These are 16 bit cards with 16K of shared mem. Now, I configured computer using IFCONFIG and ROUTE and both ports work ok (independantly) I can ping to the network in eth0 and telnet and everything. The same is true if I connecto to hosts in the eth1 networks. When I enter route to lists the kernel routes I guet: Destination net/address Gateway address Flags RefCnt Use Iface 150.8.4.0 * UH 0 86 eth0 198.187.207.0 * UH 0 6 eth1 localhost * UH 0 101 lo Now, the fun part. When I try using "routed". The routes change to: Destination net/address Gateway address Flags RefCnt Use Iface 150.8.4.0 * UH 0 86 eth0 198.187.207.0 * UH 0 6 eth1 150.8.4.0 utx.unitrx.com UGN 0 0 eth0 198.187.207.0 utx.unitrx.com UGN 0 0 eth1 localhost * UH 0 101 lo (utx.unitrx.com is the name of the localhost) When routed is running I can't neither send nor recieve. Or at least when I do a "ping", i get nothing. That is if a enter "ping someotherhost" I get nothing in return but looking at the lights on the network card (It has the RX, TX LEDs, I see those lights blinking. These lights are usually off if we are not active) However, doing a ping localhost works ok. After a while (say more than 1 minute) the routes seem to reset to the following: Destination net/address Gateway address Flags RefCnt Use Iface localhost * UH 0 101 lo I have try running routed with the -t and logfile option and this is what I get. ====================================================================== Tracing enabled Tue Oct 19 15:05:59 1993 Tracing packets started Tue Oct 19 15:05:59 1993 Tue Oct 19 15:05:59: ADD dst 127.0.0.0, router 127.0.0.1, metric 1, flags UP state ==>(cont line) PASSIVE|INTERFACE|CHANGED|EXTERNAL timer 0 ADD dst 150.8.0.0, router 150.8.4.6, metric 1, flags UP state ==> PASSIVE|INTERFACE|CHANGED|INTERNAL|SUBNET timer 0 ADD dst 150.8.4.0, router 150.8.4.6, metric 1, flags UP state ==> INTERFACE|CHANGED|SUBNET timer 0 ADD dst 198.187.207.0, router 198.187.207.129, metric 1, flags UP ==> state INTERFACE|CHANGED timer 0 sendto: Invalid argument REQUEST to 150.8.4.255.0 Tue Oct 19 15:05:59: sendto: Invalid argument REQUEST to 198.187.207.255.0 Tue Oct 19 15:05:59: RESPONSE to 150.8.4.255.0 Tue Oct 19 15:06:29: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:06:29: sendto: Invalid argument RESPONSE to 150.8.4.255.0 Tue Oct 19 15:06:59: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:06:59: sendto: Invalid argument RESPONSE to 150.8.4.255.0 Tue Oct 19 15:07:29: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:07:29: sendto: Invalid argument RESPONSE to 150.8.4.255.0 Tue Oct 19 15:07:59: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:07:59: sendto: Invalid argument RESPONSE to 150.8.4.255.0 Tue Oct 19 15:08:29: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:08:29: sendto: Invalid argument /* * After a while (i.e., after the system removes the routes) */ Tue Oct 19 15:08:59: CHANGE metric dst 150.8.4.0, router 150.8.4.6, from 1 to 16 CHANGE metric dst 198.187.207.0, router 198.187.207.129, from 1 to 16 RESPONSE to 150.8.4.255.0 Tue Oct 19 15:08:59: sendto: Invalid argument RESPONSE to 198.187.207.255.0 Tue Oct 19 15:08:59: sendto: Invalid argument ====================================================================== Anybody has an idea what is going on? This is my rc.inet1 file if it is helpful. #! /bin/sh HOSTNAME=utx.unitrx.com # # Add loopback device /etc/ifconfig lo localhost /etc/route add 127.0.0.1 # Configure sales network side /etc/ifconfig eth0 150.8.4.6 netmask 255.255.255.0 broadcast 150.8.4.255 /etc/route add 150.8.4.0 # Configure R+D network side /etc/ifconfig eth1 198.187.207.129 netmask 255.255.255.0 broadcast 198.187.207.255 /etc/route -v add 198.187.207.0 # End of rc.inet1 ====================================================================== Please, somebody help me. I would really appreciate it if you do I am completely stump on this. Also, while we are at it. How do you do subnetting under Linux? -- _____________________________________________________________________________ Alejandro Liu |EMail: aliu@usc.edu |All mispellings are intentional 424N Electric Ave. #C |Voice: 818-293-8696 |Anything mentioned here is not Alhambra,CA 91801 |Data: 818-293-XXXX |necessarily true. ------------------------------ From: dgardner@netcom.com (P. David Gardner) Subject: Security for dialin Date: Thu, 21 Oct 1993 17:43:54 GMT I guess I'm just not understanding my books right, but I still need to clamp down security for a dial-up Linux system I'll be bringing online shortly. I've not been able to lock down the /etc directory files, and users can copy them freely to their directories. They can't write back to /etc because I do have that bit set off properly, but that's as far as I can lock down the directory because when I kill read access, Pine and Elm, etc. can't see the /etc/passwd file to tell who's calling. Am I being too paranoid? Do I even need to worry about users copying the passwd file? I know even if they know enough to nab the shadow file, they won't get a decipherable password to get in, but I'd rather not let casual users even get a list of users on the system (for reasons not obvious, but it's what I need to do). I've run Cops 1.04 and taken care of most of the other problems it reported (except it still reports userid anonymous as being too long). Any help, assistance, etc. appreciated from those who have brought up a dialin Linux/SLS 1.03 system. Dave -- -- Dave Gardner dgardner@netcom.com ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Admin-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.admin) via: Internet: Linux-Admin@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Admin Digest ******************************