From: Digestifier To: Linux-Admin@senator-bedfellow.mit.edu Reply-To: Linux-Admin@senator-bedfellow.mit.edu Date: Fri, 29 Oct 93 08:13:18 EDT Subject: Linux-Admin Digest #129 Linux-Admin Digest #129, Volume #1 Fri, 29 Oct 93 08:13:18 EDT Contents: Re: PLIP Setup Problems (Alan Cox) Re: Emacs 19.19 installs (Michael Lipka) Re: PLIP Setup Problems (Tim Towers) Re: Linux with 2 Ethernet Cards - Routing? (Brad Isley) Subject: mayday! (m'aidez!) (Claude Morin) Summary: Questions in non-help groups (Byron A Jeff) finger not reporting last login time! (Roth Mark Daniel) Re: Trick to backup Linux to a Sun tape drive (tip) (Doug McLaren) Re: syslogd hangs inetd (Ronan Mullally) Re: NTP / XNTP ? (Kohjin Yamada) Trick to backup Linux to a Sun tape drive (tip) (jbm@speedy.login.qc.ca) Re: Linux creating files I can't delete? (Rene COUGNENC) Re: Help with routing under Linux (Alexander P. Komlik) ---------------------------------------------------------------------------- From: iiitac@swan.pyr (Alan Cox) Subject: Re: PLIP Setup Problems Date: Thu, 28 Oct 1993 15:50:42 GMT In article <2alc2m$due@risc1.rz.fh-heilbronn.de> uhl@sun1sun1.rz.fh-heilbronn.de (Thomas Uhl) writes: >Is there anybody who ha setup a connection via PLIP sucessfully? Please >send me your setup files (/etc/hosts, rc.net, ...) via e-mail. The pl13 and earlier plip driver is a dud. To run plip get the newer pl13 alpha test for pl14 (and remember its an alpha test so you may need to do some debugging!) or join the net channel and install net2e-beta1 which is also serious debugging needed stuff. > >What transfer speed can I expect from PLIP? Perhaps it makes more sense to >buy a D-LINK adapter? I get 25Kbyte a second over PLIP between a 386DX40 and a 386SX33. With a pair of faster machines the speed should go up. Alan iiitac@pyr.swan.ac.uk ------------------------------ From: lipka@lip.hanse.de (Michael Lipka) Subject: Re: Emacs 19.19 installs Date: Thu, 28 Oct 1993 10:59:41 GMT Reply-To: lipka@lip.hanse.de In article <1993Oct24.025803.14122@creeper.atl.ga.us> wjg@creeper.atl.ga.us (Joey Gibson) writes: ||| Intel 386 (i386-*-isc, i386-*-esix, ... Sorry folks... It wouldn't help to cancel previous my article? :-)) ------------------------------ From: tim@lorien.demon.co.uk (Tim Towers) Subject: Re: PLIP Setup Problems Date: Thu, 28 Oct 1993 18:19:28 +0000 Thomas Uhl posted a query regarding setting up PLIP Firstly - PLIP does not work with Net-2d so probably forget it I would like to hazard a guess that there are quite a few people out here on the net who would like to know the answer to questions like this so reply by mail would be inappropriate. If you want the Net-2e code and (supposedly) working PLIP wait for linux 0.99.14 or join the net mailing list. Tim -- +-----------------------------------------------------------------------------+ | Tim Towers | Don't believe everything you read | tim@lorien.demon.co.uk | | +44 952 811506 | or everything you write.. | PC Linux machine YEAH! | +-data/voice/fax---------PGP-key-available-----------------IRC-Grimwiz--------+ ------------------------------ From: brad@slammer.atl.ga.us (Brad Isley) Subject: Re: Linux with 2 Ethernet Cards - Routing? Date: Thu, 28 Oct 1993 19:27:26 GMT In article , Justin Botelle wrote: > >Justin >(I don't necessarily need them to route... or even want them to route for that >matter, but I do want people from our network to be able to do nice things I need routing. Does anyone have first-hand experience getting routing to work with 2 SMC Elite cards? (They seem popular!) Thanks much in advance. -- brad@slammer.atl.ga.us (Brad Isley) +1 404 925-9663(H) 493-2484(W) ------------------------------ From: klode@empress.com (Claude Morin) Subject: Subject: mayday! (m'aidez!) Date: 28 Oct 1993 18:34:21 GMT Well, I guess I'm another fool who shouldn't really be trying this, but please read on...I swore I wasn't going to do this, in fact I've been wrestling with things for about a week an a half, but here goes: # stty PANIC I'm providing technical assistance for my wife's presentation about interactive Internet services. I've set up a SLIP connection from my home Linux box to an Internet service provider (ISP) in order to do a live demonstration of IRC, telnet, and xmosaic. We can do everything but xmosaic by dialing in with kermit, but we *really* want to show xmosaic. 'term' is not useful because the X11 drawing commands have to come across the connection (too slow). The presentation is in 18 hours (!!!!!!!) and things aren't working properly. I've administered ethernet networks for many years, so I figured setting this up would be easy...NOT! So much for my technical skills... I'm using the official 0.99.13 + Matt Dillon's big set of patches + the endian fix + Matt Dillon's dev_delque() patch + Brian Edmonds' routing patch. Furthermore, Brian was kind enough to send me a copy of his linux/net source tree; it didn't make any difference. Please note that his connection runs at full speed. I have net-0.26.tar.z from tsx-11. [Since I'm so close to the deadline, I'm including my setup scripts after the description, just in case someone wants to see them...] I have one major problem: the connection seems unusually slow. Here's some information: My Linux box is called "tolly"; the ISP end is called "pp.uunet.ca". When 'ping'ing pp.uunet.ca, I'm seeing delays of ~500ms. Furthermore, when connecting from tolly to empress.com (connected directly to pp.uunet.ca), I'm seeing turnaround times of > 1000ms for interactive use (hit key, wait one+ sec, see char on screen). Output seems to "stall" for large amounts of data (> 1K or so). Transmitting a packet from my end (pressing a key) seems to restart output. When output *does* occur, it seems very fast. I'll get a block quickly (full-bore transmission, it seems), see a long pause, then another block quickly. I'm NOT dropping packets, nor am I seeing transmission errors. Lastly, if I try "telnet tolly chargen" or "telnet localhost chargen", the connection gets closed after 102 lines, unless I have another network connection going (ie busy), in which case I could get 200, 300, or even 500 lines before it closes. Hmm, how about some silly questions: Is this performance normal? I doubt it. Is my machine misconfigured? Likely :-) Some not-so-silly questions: Does this have anything to do with CSLIP? I doubt it, since stuff is getting through, but... Is there a recommended set of modem settings for 14.4K, v.42/v.42bis, Hayes-compatible modems? Miscellaneous questions, not connected to my problems: I've been using pp.uunet.ca as my nameserver. Acceptable? Thanks for any help anyone can give me. I'll be pulling an all-nighter trying to get this thing going tonight. Claude P.S. Please excuse spelling and grammar mistakes; I haven't the time to check for them :-) ===== Claude Morin that's "klode", not "clod"...french :-) Empress Software Inc. Markham, Ontario, Canada ===== from /etc/rc.local if [ -x /etc/rc.net ]; then /bin/hostname tolly /etc/rc.net MULTI else /bin/hostname tolly fi ===== /conf/net/hosts 127.0.0.1 localhost 255.255.255.255 broadcast 142.77.254.115 tolly.uunet.ca tolly 142.77.254.1 pp.uunet.ca pp ===== /conf/net/NETCONF # I tried writing my own rc.net, without any change in behaviour # Our hostname (not necessarily our IP address!) and domain name. host tolly domain uunet.ca # Our gateway to the world. #gw xs4all.hacktic.nl # Loopback - always enabled. iface lo localhost 255.0.0.0 127.255.255.255 up # Routes. route localhost route 127.255.255.255 # Start up any servers needed. rpc portmap server inetd server syslogd ===== /conf/net/pp.uunet.ca.dip main: echo on # First of all, set up our name for this connection. get $local tolly.uunet.ca # Next, set up the other side's name and address. get $remote pp.uunet.ca # Set the desired serial port and speed. port cua0 speed 19200 # Reset the modem and terminal line. # This seems to cause trouble for some people! # reset #init ATZX4S50=0TS0=0 # Prepare for dialing. #echo on send ATZ1\r wait OK 5 if $errlvl != 0 goto error send ATM0Q0V1E1X1\r wait OK 2 if $errlvl != 0 goto error dial dt if $errlvl != 0 goto error wait CONNECT 60 if $errlvl != 0 goto error # We are connected. Login to the system. login: sleep 3 send \r\n\r\n wait ice: 10 if $errlvl != 0 goto error send ip\n wait ogin: 10 if $errlvl != 0 goto error send \n wait ord: 5 if $errlvl != 0 goto error send \n loggedin: # Set up the SLIP operating parameters. get $mtu 1500 # Set Destination net/address as type 'default' (vice an address). # This is used by the 'route' command to set the kernel routing table. # Some machines seem to require this be done for SLIP to work properly. # default # Say hello and fire up! done: print CONNECTED to $remote with address $rmtip mode CSLIP goto exit error: print SLIP to $remote failed. exit: ===== /conf/net/resolv.conf nameserver 142.77.254.1 ===== /conf/net/host.conf order hosts,bind multi on ===== /conf/net/inetd.conf # # inetd.conf This file describes the services that will be available # through the INETD TCP/IP super server. To re-configure # the running INETD process, edit this file, then send the # INETD process a SIGHUP signal. # # Version: @(#)/etc/inetd.conf 3.10 05/27/93 # # Authors: Original taken from BSD UNIX 4.3/TAHOE. # Fred N. van Kempen, # # # # # Echo, discard, daytime, and chargen are used primarily for testing. # echo stream tcp nowait root internal echo dgram udp wait root internal discard stream tcp nowait root internal discard dgram udp wait root internal daytime stream tcp nowait root internal daytime dgram udp wait root internal chargen stream tcp nowait root internal chargen dgram udp wait root internal # # These are standard services. # ftp stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.ftpd -l telnet stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.telnetd smtp stream tcp nowait root /usr/sbin/tcpd /usr/lib/umail/umail -bS nntp stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.nntpd # # Shell, login, exec and talk are BSD protocols. # shell stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rshd login stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rlogind exec stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rexecd talk dgram udp wait root /usr/sbin/tcpd /usr/sbin/in.ntalkd ntalk dgram udp wait root /usr/sbin/tcpd /usr/sbin/in.ntalkd # # Pop et al # pop-2 stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.popd pop-3 stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.popd # # The Internet UUCP service. # uucp stream tcp nowait uucp /usr/sbin/tcpd /usr/lib/uucp/uucico -l # # Tftp service is provided primarily for booting. Most sites # run this only on machines acting as "boot servers." # tftp dgram udp wait root /usr/sbin/tcpd /usr/sbin/in.tftpd -s /tftpboot bootps dgram udp wait root /usr/sbin/tcpd /usr/sbin/in.bootpd # # Finger, systat and netstat give out user information which may be # valuable to potential "system crackers." Many sites choose to disable # some or all of these services to improve security. # finger stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.fingerd systat stream tcp nowait guest /usr/sbin/tcpd /bin/ps -auwwx netstat stream tcp nowait guest /usr/sbin/tcpd /bin/netstat #-f inet # # Time service is used for clock syncronization. # time stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.timed time dgram udp wait root /usr/sbin/tcpd /usr/sbin/in.timed # # End of inetd.conf. ===== /conf/net/networks loopback 127.0.0.0 uunet 142.77.254.0 test-net 192.0.2.0 ===== END OF TRANSMISSION ------------------------------ Crossposted-To: comp.os.linux.development,comp.os.linux.misc From: byron@cc.gatech.edu (Byron A Jeff) Subject: Summary: Questions in non-help groups Date: Thu, 28 Oct 1993 22:39:06 GMT My effort has petered out after a couple of weeks. I received a few basic classes of responses: 1) "I think my question is appropriate for the group." 2) "There is too much traffic in c.o.l.help." 3) "I post where the gurus are." 4) "I wanted more visibility/faster response." 5) "I didn't know." In addition from my observations many of the groups are write only becuase the same questions come up over and over and over again. If posters just scanned the last week of messages, much of the time they would find an answer and would not have to post. Also many questions/answers can be found in the FAQ's and HOWTO's. People don't read them much of the time. It seems to me that this a bad situation and it's not going to get any better because the querants don't do any basic research before posting. After running this survey for a couple of weeks my conclusions are the same: 1) Don't answer inappropriate questions. 2) Direct people to the correct answers (HOWTO, FAQ, whatever). 3) Redirect questions to comp.os.linux.help. Comments welcome. Survey responses upon request. Followup to c.o.l.misc only. Later, BAJ --- Another random extraction from the mental bit stream of... Byron A. Jeff - PhD student operating in parallel! Georgia Tech, Atlanta GA 30332 Internet: byron@cc.gatech.edu ------------------------------ From: roth@ux4.cso.uiuc.edu (Roth Mark Daniel) Subject: finger not reporting last login time! Date: 29 Oct 93 05:10:05 GMT I am having a weird problem where finger doesn't report the last login time of a user you finger on my system. Instead, it says "Never logged in." (If the user is logged in when you finger him, you do get the appropriate message about when he logged in and his idle time, if any.) I suspect that this is a problem with fingerd, rather than finger, because when I finger a user at another system from my system it works fine. When other systems finger at mine, it has the problem. Does finger/fingerd read this information from wtmp or lastlog? What should the permissions be on any relevant files? Here are my system specs: I'm running Linux 0.99.13 on a 486DX/33 with the libc 4.4.4 and finger 5.2.3. Most of my networking stuff comes from the SLS v1.03 distribution of the Linux 0.99.12 kernel (I've upgraded just about everything else). Any help would be appreciated! -- Personal religious issues include: | Interface: CLI, eventually VUI, no GUIs! OS: UNIX (Linux 0.99PL13 on my PC) | (except X-Windows, of course.) Still recovering from MS-DOS... | Micro: PCs! No computer w/o a text-mode Language: C++ Shell: tcsh | can possibly be taken seriously! ------------------------------ From: dougmc@utpapa.ph.utexas.edu (Doug McLaren) Subject: Re: Trick to backup Linux to a Sun tape drive (tip) Date: 29 Oct 93 05:36:01 GMT Useful 'hack', indeed. But Gnu tar will do it all for you :) tar -cvf user@machine:/dev/tape ... if user's account on machine has the appropriate .rhosts file to allow root to rlogin/rsh as them ... % cat ~user/.rhost machine_to_backup root % We do this at work for backups ... works great ... -- dougmc@utpapa.ph.utexas.edu "Anarchy means having to put up with things that really piss you off." ------------------------------ From: ronan@scamall9.iol.ie (Ronan Mullally) Subject: Re: syslogd hangs inetd Date: 28 Oct 1993 23:10:57 -0000 SuperGirl (root@calvin.iaf.nl) writes: > >recvfrom unix: Bad file number > >Currently, syslogd is off (even though inetd still routes through > >tcpd)... although, I don't like the idea of leaving syslogd off. > Hmm, this is no solution, you simple won't see them anymore :-( > But i admit, i'm having them too, using an 13e kernel... I've been getting the same error fairly frequently as well, with a variety of kernels - an extract from one of my syslog files: Oct 26 20:55:24 scamall9 syslogd: recvfrom unix: Bad file number Oct 26 20:55:55 scamall9 last message repeated 55845 times Oct 26 20:56:56 scamall9 last message repeated 111926 times Oct 26 20:57:57 scamall9 last message repeated 111472 times Oct 26 20:58:58 scamall9 last message repeated 113394 times Oct 26 20:59:20 scamall9 last message repeated 40665 times Oct 26 20:59:20 scamall9 syslogd: exiting on signal 15 I'm not entirely sure what causes it, put once it starts trying to log messages 100,000 times a second, the load average rockets, and _anything_ that tries to use syslog will hang (including a root login in order to kill the syslogd process...). -- Ronan Mullally 320 Sutton Park, Sutton Dublin 13, Ireland. ------------------------------ From: kohjin@marina.prug.or.jp (Kohjin Yamada) Subject: Re: NTP / XNTP ? Date: 28 Oct 93 16:31:59 GMT In article <2abl3mE35c@uni-erlangen.de> kardel@faui46b.informatik.uni-erlangen.de (Frank Kardel) writes: >fee@cxf111.rh.psu.edu (Chuck Fee) writes: > >>XNTP 3.3 (which apparently has some linux support built in) is dying >> with undefined symbols (_adjtimex in xntpd/nyp_loopfilter.0 >> and _adjtime in lib/systime.o) These symbols seem to be defined >> in include fiels in /usr/include/sys and /usr/include/protocols. > |Ok, the story is this. It works because Torsten Duwe and I did the |port from a previuos port made by Philipp Gladstone (thanks!). There |are just a few minor things left over. First of all use xntp3.3a.tar.Z |as base. Additionally you need patch level 13(something | g i believe) or |wait until pl13 comes out officially. Then the only problem left (sorry for that) |is the the distribution contains a file include/sys/timex.h which must be |removed (or re-named) for Linux. This naming conflict stems from the way |adjtime() is supported already (with pl13) in Linux and is not yet supported |by all other OS around out there (remember xntp has been ported to quite a few |Unix versions). I hope that this issue can be cleared up in the future. Things |tend to get more complicated the more parties are involved in it (Base line code |production, porting to >>10 platforms, vendors, OS bugs, ...). | Xntp version 3.3a (beta) compiles with ALPHA-pl13l.tar.gz fairly fine. Official patchlevel 13 doesn't. It doesn't contain timex.h in /usr/include/linux. The deamon xntpd doesn't run, though. It doesn't complain at all but there is no xntpd running in the process. I have some trouble in tickadj too. It fails in "nlist fails: can't find/read /vmunix or /unix". There is /zImage but no /vmunix nor /unix, do I miss something? If you could e-mail me or post the file Config.local and associated files, it would help many. If there is any suitable ftp site for Linux xntp exsists, it will be fine too. [deleted the here on] --kohjin -- *----/----* Kohjin Yamada, JR1EDE [kohjin@marina.prug.or.jp] Q-----T------H 504-55 Shimo-Yamaguchi, Hayama, Miura, Kanagawa, Japan *------/|-------* Phone:+81-468-75-6750 Fax/Modem/Voice:+81-468-76-1176 ------------------------------ From: jbm@speedy.login.qc.ca Subject: Trick to backup Linux to a Sun tape drive (tip) Date: Thu, 28 Oct 1993 23:36:45 GMT I found a nifty trick to backup the Linux file systems (both local and NFS mounted) to a Sun workstation with an Exabyte tape drive. The problem: Linux root user can't spawn remote shells on the Sun workstation, because he doesn't know the root password on the Sun. Hence, no simple way to write data (as root) to the Exabyte tape drive on the Sun (/dev/rst1). Root user on Linux can't use rmt tape protocol either. The solution: Ordinary user acount under Linux with same UID and login name as Sun account can spawn remote processes, no problem. The problem is that the tar backup should be done by root, in order to read all files under Linux, but that the data needs to be written to /dev/rst1 on the Sun by the unprivileged user. The Trick: Create a pipe with "mknod /tape p", and then start a user process to read from the pipe and write to the tape drive with (from a user shell with rsh privs on the Sun) "rsh Sun_machine dd of=/dev/rst1 to create a file named "-v." My problem is that I can't erase it, since > rm doesn't seem to recognize it, and neither does null... Any suggestions? > -rw-r--r-- 1 root root 372 Oct 26 23:47 -v plux:/home/rene man rm RM(1L) RM(1L) NAME rm - remove files (...) GNU rm, like every program that uses the getopt function to parse its arguments, lets you use the -- option to indicate that all following arguments are non-options. To remove a file called `-f' in the current directory, you could type either rm -- -f or rm ./-f -- linux linux linux linux -[ cougnenc@renux.frmug.fr.net ]- linux linux linux ------------------------------ From: apkom@lx12.ukrcom.kherson.ua (Alexander P. Komlik) Subject: Re: Help with routing under Linux Date: Wed, 27 Oct 1993 14:06:31 GMT Reply-To: apkom@lx12.ukrcom.kherson.ua Alex Liu (aliu@aludra.usc.edu) wrote: : Following Linus' suggestions I went and checked my /etc/networks file : as well as upgraded my libc to the current one (v4.4.4). So now : my flags look correct. (They are N routes) : it should be, I think) yet as soon as I start "routed", networking : to other machines stops working. But you forgot about /etc/gateways file! I did not find any doc's or man's in my SLS1.03 discribed the 'routed' demon and /etc/gateways. But I know that this demon is controlled by this file. ------------------------------ ** 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 ******************************