How are these packets getting created?
Hi, I have a router / switch that sits between my Debian machine at the cable modem. It does NAT. I also have iptables running on my Debian system, and I noticed that the following packets were being dropped by the Debian firewall : Debugging sambaIN= OUT=eth0 SRC=192.168.1.103 DST=65.75.178.249 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=33063 DPT=12345 WINDOW=5808 RES=0x00 SYN URGP=0 Debugging sambaIN= OUT=eth0 SRC=192.168.1.103 DST=65.75.178.249 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=33067 DPT=12345 WINDOW=5808 RES=0x00 SYN URGP=0 Debugging sambaIN= OUT=eth0 SRC=192.168.1.103 DST=65.75.178.249 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=33067 DPT=12345 WINDOW=5808 RES=0x00 SYN URGP=0 Debugging sambaIN= OUT=eth0 SRC=192.168.1.103 DST=65.75.178.249 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=TCP SPT=33067 DPT=12345 WINDOW=5808 RES=0x00 SYN URGP=0 Debugging sambaIN= OUT=eth0 SRC=192.168.1.103 DST=69.61.33.146 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=54529 DF PROTO=TCP SPT=33123 DPT= WINDOW=5808 RES=0x00 SYN URGP=0 Given that I wasnt doing anything to connect to these machines, the question then is : are these packets really coming from my Debian machine (ip = 192.168.1.103), or are these forged packets? And if they are forged, how are they getting across the NAT router? I do have other machines on my internal network that I dont trust at all. How would I go about finding where these packets are coming from? MAC addresses? Thanks, Jor-el -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Trouble with CSS (Mozilla)
Hi, I am trying to duplicate the effects shown at http://www.meyerweb.com/eric/css/edge/complexspiral/demo.html . In Mozilla, I see all the effects perfectly, however, when I try to duplicate the effects with my test code, I fail miserably. Here is what I have so far : trial.html : ?xml version=1.0? !DOCTYPE html PUBLIC -//w3C//DTD XHTML 1.0 Strict//EN http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd; html xmlns=http://www.w3.org/1999/xhtml; lang=en xml:lang=en head titleTrial page/title link rel=stylesheet href=try.css type=text/css / style type=text/css !-- body { position: relative; margin: 0; padding: 0; } div#content { color: #EED; postion: absolute; top: 26px; left: 150px; } -- /style /head body pContent not in div tag/p div id=content pHey there!/p /div /body /html and try.css : body {background: black url(shell-bg.jpg) 0 0 no-repeat fixed;} div#content {background: #468 url(shell-blue.jpg) 0 0 no-repeat fixed;} The result that I see is that the Content not in div tag appears to be visible and the Hey there! is barely visible. I dont see any background image appear (and yes, the shell*.jpg images are in the same directory and readable). Any ideas? Thanks, Jor-el -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
dependancy not found (apt-get)
Hi, I am trying to use my apt preferences setup to install straw from unstable onto my stable system. Here is what I get : trillian:/etc/apt# apt-get install -t unstable straw python-gnome2 python2.2-gnome2 libnautilus2-2 Reading Package Lists... Done Building Dependency Tree... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: Sorry, but the following packages have unmet dependencies: libnautilus2-2: Depends: libgnome-desktop-0 (= 2.0.3) but it is not installable E: Sorry, broken packages trillian:/etc/apt# apt-get install -t unstable straw python-gnome2 python2.2-gnome2 libnautilus2-2 libgnome-desktop-0 Reading Package Lists... Done Building Dependency Tree... Done Package libgnome-desktop-0 has no available version, but exists in the database. This typically means that the package was mentioned in a dependency and never uploaded, has been obsoleted or is not available with the contents of sources.list E: Package libgnome-desktop-0 has no installation candidate (1) What does this mean and (2) Is there hope for me? Thanks, Jor-el -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: kill with regex?
On Sun, 19 Jan 2003, Colin Watson wrote: On Sun, Jan 19, 2003 at 05:32:19PM +0100, Philipp Gruemmer wrote: Hugh Saunders wrote: ps x gives a list of xine's which i would like to kill [...snip...] 2) how do i get kill or bash to realise its an expression? Is there any reason not to use 'killall xine'? That's how I do it ususally... 'pkill xine' is better if you have a new enough version of procps, since 'killall' does ... let's say surprising things on System V-ish variants of Unix. Hi, A more general solution would be : produce list of pids to kill | xargs kill Regards, Jor-el -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: kill with regex?
On Sun, 19 Jan 2003, Antonio Rodriguez wrote: How would you produce this list? Jor-el wrote: On Sun, 19 Jan 2003, Colin Watson wrote: On Sun, Jan 19, 2003 at 05:32:19PM +0100, Philipp Gruemmer wrote: Hugh Saunders wrote: ps x gives a list of xine's which i would like to kill [...snip...] 2) how do i get kill or bash to realise its an expression? Is there any reason not to use 'killall xine'? That's how I do it ususally... 'pkill xine' is better if you have a new enough version of procps, since 'killall' does ... let's say surprising things on System V-ish variants of Unix. Hi, A more general solution would be : produce list of pids to kill | xargs kill Antonio, ps x | grep xine | awk '{print $1}' ...would work for the original poster. So would killall, but killall would work only when the process names are all the same. If, for instance, you wanted to kill all processes belonging to a particular user, one would do : ps auxw | awk '{if($1 == username) print $2}' | xargs kill (where username stands for the username that you are targetting) Regards, Jor-el -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Static route config
Hi, How do I configure a static route into the startup scripts / files so that the route is set at boot time? I dont think /etc/network/interfaces will work for this because none of the interfaces on the machine belong to the destination net. Here is the config I am working with : eth0 - 192.168.0.0/24 -- Uplink to the router eth1 - 192.168.1.0/24 tr0 - 192.168.2.0/24 I need to route packets destined for 192.168.10.0/8 onto eth1. I can define the route on the command line - just havent figured out where to define it so that the system comes up on a reboot with this route also defined. Thanks, Jor-el
Re: Problem with postfix relaying
On Mon, 18 Feb 2002, martin f krafft wrote: also sprach Jeff [EMAIL PROTECTED] [2002.02.18.0103 +0100]: I have brackets around this in my main.cf. I do not know if this is your problem, but this line is all I needed to do: relayhost = [my.isp.mail.server.com] Brackets are not necessary, so that's not the problem. Are you sure your hostname and origin settings are correct? Are on a network that the server will listen too (your IP address)? brackets will make it use the A record or IP address straight, rather than asking postfix to figure out the MX record first. do host my.isp.mail.server.com and obtain the IP from the result. then, do these two: ping ipaddress telnet ipaddress 25 report what you see. also, what's the entries in /var/log/mail.log, and the output of the `mailq` command? Hi, Thanks to all who replied. I've discovered (thanks to Martin's suggestions) that the reason why my mail server is talking to texlog.texas.rr.com instead of the smtp server that I specified (smtp-server.austin.rr.com) is because the DNS server for my host (run by the ISP) states that texlog is the mail server for my host. Witness the following translated tcpdump output (see attached script to see how I got this output) that I get when I send a mail : 08:17:39.870965 24.28.XX.XX.33467 24.93.35.62.53: 10894+ MX? smtp-server.austin.rr.com. (43) (DF) EG1@@mG]#53.V*smtp-serveraustinrrcom 08:17:39.982657 24.93.35.62.53 24.28.XX.XX.33467: 10894* 1/4/5 MX texlog.texas.rr.com. 10 (224) (DF) [EMAIL PROTECTED]G5f*smtp-serveraustinrrcom [EMAIL PROTECTED]@9H]#f]# x]#]#!T]#? 08:17:39.983107 24.28.XX.XX.33467 24.93.35.62.53: 10895+ A? texlog.texas.rr.com. (37) (DF) EA1@@mG]#5-/*texlogtexasrrcom 08:17:40.056209 24.93.35.62.53 24.28.XX.XX.33467: 10895 1/4/4 A 24.93.35.223 (196) (DF) [EMAIL PROTECTED]G5~*texlogtexasrrcomH]#9^Mns1austin9ns29ns2E9ns1~]# A]#Z]#!l]#? 08:17:40.056649 24.28.XX.XX.54994 24.93.35.223.25: S 1776713462:1776713462(0) win 5840 mss 1460,sackOK,timestamp 469512701 0,no p,wscale 0 (DF) EO@@PDG]#i~ 1 08:17:40.153124 24.93.35.223.25 24.28.XX.XX.54994: S 2682070823:2682070823(0) ack 1776713463 win 24616 nop,nop,timestamp 357429 443 469512701,nop,wscale 0,nop,nop,sackOK,mss 1460 (DF) E@@=]#G''i~`(_ M1 08:17:40.153224 24.28.XX.XX.54994 24.93.35.223.25: . ack 1 win 5840 nop,nop,timestamp 469512711 357429443 (DF) E4O@@PKG]#i~'( 2M and as you can see, this is why postfix is talking to the texlog server (24.93.35.223). However, when I use dig, here is what I get : trillian:~# dig -t MX my_isp_assigned_hostname.austin.rr.com ; DiG 9.2.0 -t MX my_isp_assigned_hostname.austin.rr.com ;; global options: printcmd ;; Got answer: ;; -HEADER- opcode: QUERY, status: NOERROR, id: 33340 ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0 ;; QUESTION SECTION: ;my_isp_assigned_hostname.austin.rr.com. IN MX ;; AUTHORITY SECTION: austin.rr.com. 3600IN SOA ns1.austin.rr.com. rradm.austin.rr.com. 2002021801 3600 600 86400 3600 ;; Query time: 108 msec ;; SERVER: 24.93.35.62#53(24.93.35.62) ;; WHEN: Thu Feb 21 08:25:38 2002 ;; MSG SIZE rcvd: 87 ...which I dont understand since the above output doesnt show texlog.texas.rr.com as the mail server for my host. So now I have two mysteries : (1) Why is postfix ignoring my relayhost parameter and querying DNS and (2) Why am I not getting similar answers to what postfix gets when querying my ISP DNS server? For reference, I enclose again all non-commented lines in /etc/postfix/main.cf : command_directory = /usr/sbin daemon_directory = /usr/lib/postfix mynetworks = 127.0.0.0/8 smtpd_banner = $myhostname ESMTP $mail_name (Debian/GNU) debugger_command = PATH=/usr/bin:/usr/X11R6/bin xxgdb $daemon_directory/$process_name $process_id sleep 5 program_directory = /usr/lib/postfix myhostname = trillian.megadodo.umb myorigin = /etc/mailname mydestination = trillian.megadodo.umb, localhost.localdomain relayhost = smtp-server.austin.rr.com mailbox_command = procmail -a $EXTENSION mailbox_size_limit = 0 canonical_maps = hash:/etc/postfix/sender.map Changes to my config since I first reported this problem : (1) I changed the map file to point to canonical_maps. It was originally pointing to sender_canonical_maps. (2) I edited /etc/mailname to put the FQDN of my hostname (trillian.megadodo.umb). Previously 'hostname' would report 'trillian'. It now reports the FQDN. Thanks, Jor-el sniffer.pl Description: tcpdump output xlater scarfed from the 'net
Problem with postfix relaying
Hi, In my current setup, my MUA talks directly to the smtp server of my ISP to send mail. I wanted to change that to where it would send mail to an smtp server running on my machine (postfix) which would then relay it to my ISP's smtp server. Unfortunately, for some reason, postfix seems to be talking to another smtp server. Here are the relevant entries in main.cf : program_directory = /usr/lib/postfix myhostname = trillian.megadodo.umb alias_maps = hash:/etc/aliases alias_database = hash:/etc/aliases myorigin = /etc/mailname mydestination = trillian.megadodo.umb, localhost.localdomain relayhost = smtp-server.austin.rr.com mailbox_command = procmail -a $EXTENSION mailbox_size_limit = 0 #recipient_delimiter = + sender_canonical_maps = hash:/etc/postfix/sender.map According to the above config, I would expect postfix to talk to smtp-server.austin.rr.com . However, the tcpdump traces I have been running show that it has been talking to texlog.texas.rr.com - which promptly bounces my mail saying - Relaying denied. What am I doing wrong? Thanks, Jor-el
Re: bringing out variable
On Wed, 9 Jan 2002, Sebastiaan wrote: #!/bin/sh # DISPLAY=:1 export DISPLAY When I execute this script, the value is set correctly in the script environment, but after execution 'echo $DISPLAY' still results the old value. I have also tried 'set DISPLAY', without succes. Any hints? Sebastiaan, Where are you doing the 'echo' from? From within the script, or from the command line after the script has been run? If the latter, then what you see is to be expected. When any program or shell script is run, a new shell is first spawned and the program / shell gets executed within that shell. Any 'export' that you do from the script will therefore only be visible to processes that are descendants of the shell running that script. The shell from which it was invoked will not have its env. changed at all. In order to do what you want, you need to use the 'source' builtin function of bash (man bash and search for 'source filename'). The alternative syntax for 'source' is to use '.' in place of the word 'source' . Regards, Jor-el
Re: xhost errors under root and from remote hosts
On Tue, 8 Jan 2002, François Chenais wrote: I run xhost Francois : xhost access control enabled, only authorized clients can connect Note what it says - ^^^ and now read 'man xauth'. xhost will only allow you to enable / disable access control. If access control is enabled (as is your case), you will need to use xauth to authorize clients. Regards, Jor-el
Re: trying to run VMwareworkstation 3.0.0-1455..says I have wrong kernel headers
On Fri, 4 Jan 2002, Walter Tautz wrote: during the run of vmware-config.pl it keeps asking for the location of the kernel headers, so I tell it /usr/include but it complains: (I am running testing (woody)): - Begining of config log-- What is the location of the directory of C header files that match your running kernel? [/usr/src/linux/include] /usr/include The directory of kernel headers (version 2.4.13) does not match your running kernel (version 2.2.19). Even if the module were to compile successfully, it would not load into the running kernel. Walter, That output you quote above is somewhat surprising to me. When I installed it, instead of pointing to /usr/src/linux/include , the default was pointing to something like /lib/modules/2.4.15-pre1/build/include/ for me. I was a bit surprised by that, but it appears that 'make modules_install puts the headers there. I didnt have any other modules, but I did a 'make modules_install' and things worked out fine for me. Regards, Jor-el
Did something break with gnome 1.4?
Hi, I did an apt-get upgrade a couple of days ago, and I noticed that I was getting updates to GNOME which would take me to v1.4 . However, I didnt restart my window manager till yesterday - which was when I discovered that enlightenment got totally hosed. After trying to fix it in various ways, I gave up and tried uninstalling and reinstalling GNOME and enligtenment. GNOME by itself gives me a functioning window manager. But adding enlightenment causes the window manager to come up without any active mouse bindings - thus rendering it useless. Has anyone come across this problem? Thanks, Jor-el
Re: Can't access aliased ip address
Chad, On Wed, 2 Jan 2002, Chad Morgan wrote: This doesn't appear to be the problem. When I ping ip address C, from machine A (my computer at a remote location) tcpdump icmp -i eth0 yields only a request. However, once I ping ip address B from Machine A it will show both the request and the reply. This is puzzling. Your first sentence breaks my hypothesis and the second sentence strengthens it. Perhaps if you could provide your routing table more legibly, it would be of more help (use the -n option of route). The hostnames in your prev. post were truncated and it was hard to guess how the routing was setup. Also try the following : from B / C, do 1. traceroute A 2. traceroute A -s C On A, monitor the traffic using a filter for src = A or dst = A and post the results. Also, I just looked at the syslog, which I probably should've done earlier and found the following: Jan 2 15:23:46 hostname kernel: Packet log: input DENY eth0 PROTO=1 MACHINEA:8 IPADDRC:0 L=92 S=0x00 I=0 F=0x4000 T=43 (#9) I dont know what this log entry means. Its possible that you have a firewall problem but your symptoms are more indicative of a routing problem. Perhaps the output of 'ipchains -L -v -n' would help (Note: I run iptables and I'm guessing that its options are similar to ipchains. The -n will produce numeric, rather than symbolic output). Regards, Jor-el
Re: Can't access aliased ip address
On Wed, 2 Jan 2002, Chad Morgan wrote: eth0 Link encap:Ethernet HWaddr 00:A0:CC:23:A3:AC inet addr:216.86.213.93 Bcast:216.86.213.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:241367 errors:0 dropped:0 overruns:0 frame:0 TX packets:260291 errors:0 dropped:0 overruns:0 carrier:0 collisions:41 txqueuelen:100 Interrupt:10 Base address:0x6000 eth0:1Link encap:Ethernet HWaddr 00:A0:CC:23:A3:AC inet addr:216.86.213.94 Bcast:216.86.213.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Interrupt:10 Base address:0x6000 snip Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface adsl-gte-la-216 * 255.255.255.255 UH0 00 eth0 adsl-gte-la-216 * 255.255.255.255 UH0 00 eth0 192.168.0.0 * 255.255.255.0 U 0 00 eth1 216.86.213.0* 255.255.255.0 U 0 00 eth0 default adsl-gte-la-216 0.0.0.0 UG1 00 eth0 Note: the the first two entires are the hosts associated with eth:0 and eth0:1 Now, I'm a remote location but when logged into this machine I can ping all ip addresses. From my machine I get no responce when I try and ping the address associated with eth0:1 however when I use tcpdump host 216.86.213.94 I get the following result while pinging that address from my machine so it looks like it is getting something but not answering. 11:31:07.232889 ca-brea2a-102.stmnca.adelphia.net adsl-gte-la-216-86-213-94.mminternet.com: icmp: echo request (DF) Chad, It sounds like you've fallen prey to the perils of assymmetric routing. For simplicity, I'll refer to your (my machine) external server to be machine A, eth0 to be ip-address B, and eth0:1 to be ip-address C. Here is how things are happening : Machine A sends echo request to C. C replies. The routing table on your multihomed server says that the packet needs to travel out via eth0. The outgoing echo reply therefore will contain the source address of B. A is not listening for replies from B. It is listening for replies from C. Hence your problem. To see if I am right, filter tcpdump on the icmp protocol, rather than the host of C. If you see echo replies from B coming in, and echo requests for C going out, what I said is correct. Regards, Jor-el
Where are the man pages for the C library?
Hi, I can find the man pages for things like strcpy, printf, and things like that on my system. I installed the glibc6-doc but that only had info documentation in it. Where are these man pages to be found? Thanks, Jor-el
Re: Securing bind..
Russell, On Sun, 30 Dec 2001, Russell Coker wrote: Also don't allow recursion from outside machines. Why does this help? Another possibility is to have the port for outgoing connections be something other than 53 (54 seems unused) and use iptables or ipchains to block data from the outside world coming to port 53. Security through obscurity? Quite frankly, I find this strategy annoying. Nothing annoys me more than finding out that I have to open up yet another hole in my firewall so that I can access another idiot who has set up his webserver at port 1080, or 8080 or whatever his fancy pleases. If your service isnt secure at the IANA designated port, why would it be secure at the new one? Of course, in the case of DNS servers, you could be OK, since you do want to lessen the number of folks who use your services (right?). But in general, I consider this to be poor advice. Regards, Jor-el
Re: multiple copies of getty running Newbie #61
On Sun, 30 Dec 2001, Ian Balchin wrote: Hi, I see in top that there are always copies of getty running - 4 copies that I can see all owned by root. Is this normal? Ian, 'man getty' produces (among other things) : DESCRIPTION getty opens a tty port, prompts for a login name and invokes the /bin/login command. It is normally invoked by init(8). Does that answer your question? Look at your /etc/inittab and check to see how many times getty is set to spawn. You should see that many instances on your system. Regards, Jor-el Note : It doesnt hurt to RTFM.
Re: Securing bind..
Russell, On Sun, 30 Dec 2001, Russell Coker wrote: Lots of good stuff snipped Please read my messages carefully before flaming me. Ack! My apologies. Poor reading and poor wording. DNS cache machine sents out requests from source port 54 (not obscure - every administrator of every DNS server on the net can easily discover this). Not sure I follow what you are saying here. Are you saying that it is pretty easy for a DNS admin to figure out what port you are running the DNS server on (if so how?) or are you saying that port 54 is a well agreed upon port for this purpose. I doubt very much that it is the latter, since http://www.iana.org/assignments/port-numbers states that port 54 is assigned to XNS (whatever that is). Regards, Jor-el
Where did my memory go?
Hi, Here is the output of 'top' on my system : 23:52:59 up 23 days, 13:21, 1 user, load average: 0.00, 0.00, 0.00 22 processes: 21 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.6% user, 0.4% system, 0.0% nice, 99.1% idle Mem:255804K total, 229276K used,26528K free,10584K buffers Swap: 586144K total, 8384K used, 577760K free, 161836K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 23757 jorel 15 0 968 964 776 R 0.9 0.3 0:00 top 1 root 6 0 544 508 476 S 0.0 0.1 0:30 init 2 root 9 0 00 0 SW0.0 0.0 0:05 keventd 3 root 19 19 00 0 SWN 0.0 0.0 0:03 ksoftirqd_CPU0 4 root 19 19 00 0 SWN 0.0 0.0 0:03 ksoftirqd_CPU1 5 root 9 0 00 0 SW0.0 0.0 1:14 kswapd 6 root 9 0 00 0 SW0.0 0.0 0:00 bdflush 7 root 9 0 00 0 SW0.0 0.0 0:21 kupdated 107 daemon 9 0 136 5656 S 0.0 0.0 0:00 portmap 271 root 9 0 356 324 288 S 0.0 0.1 0:08 syslogd 273 root 9 0 272 148 148 S 0.0 0.0 0:00 klogd 361 root 8 0 176 108 108 S 0.0 0.0 0:00 inetd 370 root 9 0 260 176 128 S 0.0 0.0 0:03 cron 373 daemon 9 0 132 6464 S 0.0 0.0 0:00 atd 391 root 9 0 172 108 108 S 0.0 0.0 0:00 getty 392 root 9 0 172 108 108 S 0.0 0.0 0:00 getty 393 root 9 0 172 108 108 S 0.0 0.0 0:00 getty 394 root 9 0 172 108 108 S 0.0 0.0 0:00 getty 11737 root 9 0 184 148 148 S 0.0 0.0 0:00 dhcpcd 9485 root 9 0 388 324 324 S 0.0 0.1 0:00 getty 12216 root 9 0 556 476 456 S 0.0 0.1 0:12 gpm 23692 jorel 12 0 1392 1392 1064 S 0.0 0.5 0:00 bash Can someone explain to me why (1) I am swapping even though the RSS column adds up to a small fraction of 256MB (2) Why the summary stats say I'm using 229MB even though there isnt anything running that accounts for that memory usage? The kernel is 2.4.16 - if it matters. Thanks, Jor-el
Remote X connections (was Re: Weird log message)
On Fri, 28 Dec 2001, Karsten M. Self wrote: Be default, X11 doesn't allow TCP/IP (remote) connections in more recent versions, though this can be overridden. Karsten, How can this be overridden? Where can one find info about this restriction and how it is enforced? Thanks, Jor-el
Re: Where did my memory go?
On Sat, 29 Dec 2001, Noah Meyerhans wrote: On Sat, Dec 29, 2001 at 01:20:31AM -0500, P Prince wrote: Linux does preemptive swapping. AFAIK, you can't turn it off, and if you could, you probably shouldn't. It will swap out dormant parts of memory, just in case you start up something big, like X, or Netscape, or Emacs. (he he) Under, say, 128MB of RAM, this makes a lot of sense, but when you go to half and gig and beyond, it becomes less useful. However, under such situations there is no penalty for swapping. The swapped pages are cached in RAM. The idea is that the kernel is getting a head start just in case it actually *needs* to swap later. If it does, all it has to do is re-allocate memory, rather than swap pages out to disk and then re-allocate memory. If it had to do that, then a situation in which some program suddenly tried to allocate memory would cause the system to grind to a halt as pages are swapped out. Noah, In that case it sounds to me like there is a genuine problem on my system. Even if the swap is cached in RAM, this doesnt explain the stats reported by top : the 8M of swap + the total of the RSS column doesnt come close to the 229M that is reported as being used. Are there any memory leak issues being reported with the 2.4.16 kernel? Thanks, Jor-el
Re: ping myself
On Fri, 28 Dec 2001, David Gardi wrote: It depends. If you ping localhost or 127.0.0.1 then no. If you have set up your NIC with an IP other than the loopback ie 192.168.0.1, and ifconfig ethx shows it together with a MAC address, and you ping that, then yes. David, Wrong! _All_ locally defined ip-addresses are internally routed to lo. Pinging any ip address that a host is setup to receive packets for, doesnt prove a thing about the functioning of the NIC. As an experiment, ping any non-lo ip-address on your system. If ping reports n packets Tx'ed and n packets Rx'ed, your lo interfaces RX and TX counts will have increased by 2n. Regards, Jor-el
Re: missing posts to list
On Fri, 28 Dec 2001, k l u r t wrote: there seems to be some posts to this list that are missing. when i tried to back track and look up some past posts on http://lists.debian.org and usenet, some posts are just not there. Klurt, How do you know they are missing? Perhaps you mean the following phenomenon : If one goes to http://lists.debian.org/debian-user/2001/debian-user-200112/thrd9.html on that page, one can see for example : Re: Where do you RTFM ? Jesse Goerz Possible follow-up(s) Re: Where do you RTFM ? Wendell Cochran If you are referring to the 'Possible follow-up(s)' as missing posts, then relax. That is due to people responding off the list and those emails being quoted in later posts on the list. I've also noticed that the Outlook Express newsreader has the annoying bug of not displaying certain read posts after a few days have expired. I only mention this since you mention the usenet. However, your MUA is pine, so you are most probably not on an MS platform and referring to that problem. Regards, Jor-el
Re: missing posts to list
On Fri, 28 Dec 2001, Colin Watson wrote: What posts are you thinking of? Your own? Be careful that the mail-to-news gateway is one-way, so anything posted to this list by news won't make it back by mail. Colin, I just remembered this thread from debian-devel which may be relevant re: missing posts : http://lists.debian.org/debian-devel/2001/debian-devel-200112/msg01941.html Regards, Jor-el
Re: problems with eth0 set-up
On Wed, 26 Dec 2001, Rachel Andrew wrote: ifconfig: eth0Link encap: Ethernet HWaddr 00:A0:CC:5E:69:C1 inet addr: 10.0.0.16 Bcast: 10.0.0.255 Mask 255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU: 1500 Metric:1 RX Packets:0 errors:0 dropped:0 overruns:0 frame:0 TX Packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 Interrupt:11 Base Address:0xdc00 lo Link encap: Local loopback inet addr:127.0.0.1 Mask 255.0.0.0 UP LOOPBACK RUNNING MTU:3924 Metric:1 RX Packets:4 errors:0 dropped:0 overruns:0 frame:0 TX Packets:4 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 route: Kernel IP Routing Table Destination Gateway Genmask Flags Metric RefUse Iface 10.0.0.00.0.0.0 255.255.255.0 U 0 0 0 eth0 0.0.0.0 10.0.0.10 0.0.0.0 UG 0 0 0 eth0 Rachel, It looks like packets are being sent out from your machine, but no reply is being received. Two possibilities are (1) you have a firewall and it is blocking the inward traffic. (2) Your router configuration is such that traffic to 10.0.0.16 is not being sent out over the right interface. In the case of (2), the routing table from your router would be of help in debugging. Regards, Jor-el
Re: Mounting floppy. Newbie #61
On Thu, 27 Dec 2001, Ian Balchin wrote: I note that when you copy a file to /floppy it seems to be incredibly quick. Is it in fact written to the floppy at the same time as it appears on /floppy ? Ian, All I can say is Ouch!. Unlike the DOS world, Unix in general, and Linux in particular, do not guarantee that data will be written to disk immediately after the in-memory copy of the disk blocks have been updated. Unixes, in general, cache such info, and typically write out to media (floppy disks, hard drives, etc) only when its efficient to do so. Thus, when you make drastic changes to the contents of a floppy disk, the in-memory copy of those disk blocks gets updated, and this happens quite fast. However, unless this copy is written out to the floppy itself, the changes will not be permanent. There is a standard way that Unixes support to force the write to media. This is called the 'sync' command. If you had done 'sync' after changing the contents of your floppy, you would have seen that your system was writing to the floppy - which does take a significant amount of time. Given the symptoms you describe, I would guess that you didnt do a 'sync' before you popped the floppy disk, and thus your backups were never written to floppy before you popped out the drive. Regards, Jor-el
Re: scientific graphing program with histograms
On Thu, 27 Dec 2001, David Wright wrote: Can anyone suggest a scientific graphing program that can do histograms. * i don't know if guppi can, but there is no debian package David, It can. It also is being packaged (but is not currently ready). I compiled the darned thing myself and can give you pointers if you are so inclined. I dont know if plotutils will work for you though - there is a Debian package, but it doesnt have a man page (only info) and info is broken on my system, so I cant give you any details. Regards, Jor-el
Re: What's a debian kid look like?
On 23 Dec 2001, Brian Nelson wrote: MH [EMAIL PROTECTED] writes: So, what are you doing here? He shouldn't be here, so make him go away with procmail: :0 * ^TO_debian * (^X-Mailer:.*Microsoft|^Content-Type:.*html) /dev/null I never saw his noise until you quoted it... Brian, Now thats plain mean. There are lots of polite folks deserving of answers, who post to this list from Microsoft platforms. Dont throw out the baby with the bathwater. Regards, Jor-el
Re: VMWare
On Tue, 18 Dec 2001, Marcelo Chiapparini wrote: Hi all, I would like to know if anyone in this list is using VMWare Workstation in order to run flavours of Windows with Debian GNU Linux as the host machine. Marcelo, Not quite what you asked for, but I am running VMWare on Debian testing / unstable and am running Redhat 7.2 in one of the VM sessions. Why? Some of the non-free software (DB2 etc) on linux comes with rpm installs and alien didnt quite do the trick. I've got a dual Pentium-II 400MHz system with 256 Mb of RAM and with 128 MB allocated to the VM session, with KDE and DB2 loaded, the VM started to swap. I didnt observe any response times problems though. Which means all my problems could be fixed with more memory. One niggling limitation though : the VMWare version that I am using (3.0.0 build 1455) has a max of 204MB that can be allocated to a VM session. If you need more than that for a VM session, you are SOL. Regards, Jor-el
Re: Apache doesn't start
On Mon, 17 Dec 2001, Dragón wrote: Here the attachement with the strace output. Dragon, There is something truly wierd on your system. On looking at the strace output closely, I see that apache is getting read errors on files like /etc/passwd and /etc/group . Why is this happening? There are a number of other errors that are occuring on your system, but these seem to be the start of it all. Perhaps if you find out what is going wrong with these files, the mystery may be solved. Regards, Jor-el
Re: URGENT - Bind broken by potato- woody
On Tue, 18 Dec 2001, john wrote: Chain POSTROUTING (policy ACCEPT) target prot opt source destination SNAT all -- 192.168.0.0/24 0.0.0.0/0 to:202.92.79.119 SNAT all -- 192.168.2.0/24 0.0.0.0/0 to:202.92.79.119 Forward is Accept This box: options { directory /var/cache/bind; // query-source address * port 53; forwarders { 192.168.0.142; // the secondary 202.92.79.116; // what I used to resolve off }; }; If I run nslookup and set server 202.92.79.116 John, This wouldnt have anything to do with the fact that your firewall rules are for 202.92.79.119 whereas the DNS server is at 202.92.79.116, would it? Regards, Jor-el
Re: Apache doesn't start
On Sun, 16 Dec 2001, Dragón wrote: Has anybody got a clue why it doesn't work? Program received signal SIGSEGV, Segmentation fault. 0x8066430 in ap_get_local_host () (gdb) bt #0 0x8066430 in ap_get_local_host () #1 0x8063aec in ap_fini_vhost_config () #2 0x8055c37 in ap_read_config () #3 0x805d9c8 in main () #4 0x400dd64f in __libc_start_main () from /lib/libc.so.6 (gdb) Dragon, Is your localhost interface working correctly? What is the output of 'ping localhost' and 'host 127.0.0.1' ? Regards, Jor-el
Re: Apache doesn't start
On Sun, 16 Dec 2001, Dragón wrote: --- Jor-el [EMAIL PROTECTED] wrote: On Sun, 16 Dec 2001, Dragón wrote: Is your localhost interface working correctly? What is the output of 'ping localhost' and 'host 127.0.0.1' ? Regards, Jor-el Dragon, Well, it is quite obvious from what you sent that your incorrectly setup localhost interface is the problem. The question that remains now is where exactly is its failure? Do an 'ifconfig -a' and send the output. Also, if you are using ipchains, send the output of 'ipchains -L -v' and if using iptables, send the output of 'iptables -L -v' . Regards, Jor-el
Re: Apache doesn't start
On Sun, 16 Dec 2001, Dragón wrote: Thank you very much Jor-el it looks like the kernel is the problem, isn't it? Yours. --- Jor-el [EMAIL PROTECTED] wrote: On Sun, 16 Dec 2001, Dragón wrote: --- Jor-el [EMAIL PROTECTED] wrote: On Sun, 16 Dec 2001, Dragón wrote: Is your localhost interface working correctly? What is the output of 'ping localhost' and 'host 127.0.0.1' ? Regards, Jor-el Dragon, Well, it is quite obvious from what you sent that your incorrectly setup localhost interface is the problem. The question that remains now is where exactly is its failure? Do an 'ifconfig -a' and send the output. Also, if you are using ipchains, send the output of 'ipchains -L -v' and if using iptables, send the output of 'iptables -L -v' . Regards, Jor-el Dragon, No, the problem is with your localhost interface which isnt up. Here is what my 'lo' interface looks like in the 'ifconfig' output : loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:9241 errors:0 dropped:0 overruns:0 frame:0 TX packets:9241 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 Note the status which says that it is 'UP' and 'RUNNING' - something which is not present in your output. I would suggest that you simply run 'ifup lo' and look at whether the interface has come up using 'ifconfig' . If it has, then retry with Apache again. Regards, Jor-el
Re: Apache doesn't start
On Sun, 16 Dec 2001, Dragón wrote: - No, the problem is with your localhost interface which isnt up. Here is what my 'lo' interface looks like in the 'ifconfig' output : loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:9241 errors:0 dropped:0 overruns:0 frame:0 TX packets:9241 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 Note the status which says that it is 'UP' and 'RUNNING' - something which is not present in your output. I would suggest that you simply run 'ifup lo' and look at whether the interface has come up using 'ifconfig' . If it has, then retry with Apache again. # ifup lo Ignoring unknown interface lo=lo Dragon, I'm afraid you've reached the limits of my expertise. What you have quoted above is an error that needs to be fixed, but I have no clue on how to fix it. Perhaps it has something to do with the dummy0 device you also are using. I hope someone with better expertise than me can solve it for you. Regards, Jor-el
Re: OT:programmin questions
On Sun, 16 Dec 2001, Eric G. Miller wrote: On Sun, 16 Dec 2001 18:54:59 -0500, David Bellows [EMAIL PROTECTED] wrote: [snip] not a programmer as you'll quickly notice. My problem is that I have to enter a key followed by ENTER. This is obviously cumbersome when Probably the input is being buffered. But you need to be able to read each charachter immediately. I believe there's a way to turn off input buffering on STDIN via bash, but at the moment I don't recall the magic to do so... I think you might want to setup the read function to only read one character at a time (by default, I think it is line buffered). Look up how to do those things, and I think you'll be on your way... Hi, I think, this is what you were thinking off : http://groups.google.com/groups?hl=enthreadm=6aff58%24mun%241%40defiant.cistron.nlrnum=1prev=/groups%3Fq%3Dgroup:comp.os.linux.development.system%2Bauthor:miquels%2540cistron.nl%26hl%3Den%26as_drrb%3Db%26as_mind%3D25%26as_minm%3D1%26as_miny%3D1998%26as_maxd%3D26%26as_maxm%3D1%26as_maxy%3D1998%26rnum%3D1%26selm%3D6aff58%2524mun%25241%2540defiant.cistron.nl Regards, Jor-el
Re: Apache doesn't start
On Sun, 16 Dec 2001, Dragón wrote: --- Christoph Simon [EMAIL PROTECTED] wrote: On Sun, 16 Dec 2001 14:25:21 -0600 (CST) Yes that helped! Now it is UP and RUNNIG But Apache is still giving Segmentation Fault. Dragon, What does the strace / Apache debug show about where it is failing? Jor-el
Re: ftp oddity in a masqueraded machine
On Sun, 9 Dec 2001, Américo Rocha wrote: Self-explanatory: Will not open connection to 192.168.0.8 (only to 212.54.148.10) Anonymous user logged in. No data connection: No such file or directory Also from that debian box (192.168.0.8) Is it saying it won't download anything to it, but only to the machine acting as NAT ? Americo, (Sorry, I dont know how to type out an 'e' with the accent on it) Dont assume anything is self-explanatory. It may be self-explanatory to you because you are intimately familiar with your set up. In this case, for instance, I am assuming that you have two machines - one is a masquerader (with an external ip address of 212.54.148.10) and a masqueradee with an ip address of 192.168.0.8 . Or these may be two interfaces on the same machine - but who will know unless you give us those details? Also, needed is where you are trying to ftp from and which is the ftp server and whether you have a firewall which has rules for ftp. Regards, Jor-el
Re: crc error after uncompressing the kernel
On Sat, 8 Dec 2001, Guy Durand wrote: what does it mean when you get a crc error after the lilo prompt and the kernel start uncompressing. It happened with my laptop and I am able to boot the original kernel bot not the new kernels I had compiled after. Badiane, Are you sure your kernel compile succeeded? If for instance, you were doing a 'make zImage', the build process complains at the very end that the kernel is too big or something like that. In that case, I've seen the zImage file actually being produced, but its worthless - you will get failures like the one you just described. If that is the case, do a make bzImage. Regards, Jor-el
Re: IPTABLES and DHCP
On 6 Dec 2001, Randolph S. Kahle wrote: My question is this -- How do I synchronize the IP address for eth0 that dhcpcd secures with the rule in iptables? It seems to me that I need to run some script or take some action each time dhcpcd secures a new IP address. Randy, Here is what I did : I installed ipmasq and disabled it (remove the symlinks if you are not using filerc - if you are, its much easier). The ipmasq package has a number of utils like default-if, and ipofif that can help you in your task. default-if for instance returns the net if that has the default route. If you are using DHCP, then the default route will be set for that interface. ipofif can be applied to the output of default-if to tell you the ip address that has been assigned to you. Regards, Jor-el
Re: prolonged number crunching
On Thu, 6 Dec 2001, Alec wrote: Hi I have a program that produces output to STDOUT and will probably take several days to run. I already figured that it's better to run it using at utility and collect the results by email. This way the program will not be bound to any specific terminal and, say crashing X, will not interrupt it. I'm wondering if there's also a way to insure against shutting down of the machine, i.e. enable saving of the session (RAM) at shutdown and restarting from where it stopped when the computer is brought up? Alec, Mainframe programmers are very familiar with this requirement : long running jobs that span are not very unusual in the mainframe environment. The solution goes by the generic name of checkpoint-restart. A quick search on the web for that revealed the following link : http://www.science.uva.nl/research/scs/Software/ckpt/ I havent tried the software, so this isnt a recommendation. But it might be just what you need. Regards, Jor-el
Kernel lockups - how do I troubleshoot?
Hi, I am running testing / unstable on a dual CPU Tyan Tiger 100 motherboard with 2.4.15pre1 . My machine has locked up on me twice in the past week leaving no clues in the syslog or anywhere else. At the point where it locked up, the screen was in power saving mode and couldnt be brought back live by using the keyboard or mouse. I had to use the reset to get it to reboot. I'd greatly appreciate it if anyone could either tell me how to or point me to any info which will atleast allow me to gather info on the crash, so that I can post it to linux-kernel. Thanks, Jor-el
Re: ARGH!! Now It's WORSE :-(
On Wed, 5 Dec 2001, Stan Brown wrote: I'm runing potato + Progeny + 2,4 kernel on 2 machines one at home, and one at work. Somehow I have installed something on the machine at work that spits out an erro meesgae about every 5 minutes about an I/O error on the floppy (which I NEVER use). I've posted about this twice in the last week or so, but never goy any useful answers. So this morning I wnet through dselect trying to compare packages between teh machines, I dleted a few packages on the machine at work in an effort to get rid of this (since it make using the console impossible). Now it's worse! Now in addiation to that error I'm geting cdrom: open failed Stan, Sounds like something is expecting a floppy in the drive. If you have a spare floppy, put one in the drive and see if a read or write takes place. If it does, then run lsof during the floppy access - this should allow you to track down which process is trying to access the floppy drive. Hope that helps, Jor-el
Re: Kernel lockups - how do I troubleshoot?
On Thu, 6 Dec 2001, Paolo Falcone wrote: [EMAIL PROTECTED] wrote: I'd greatly appreciate it if anyone could either tell me how to or point me to any info which will atleast allow me to gather info on the crash, so that I can post it to linux-kernel. if it's specifically a kernel crash (as in Oops) you can use'the ksymoops utility for the kernel developers to help them track the source of your problem. Unfortunately, I cant tell if it is an oops or not because the display remains stuck in power saver mode. But thanks for the info. I doubt if it's really the kernel at fault there, though... have you tried checking the hardware? If everything still doesn't work despite the long hours of troubleshooting, then you might say that you've narrowed the problem to the kernel. Not sure how I would go about checking the hardware. The only diag routines that I have come across are for NIC's. And those too only seem to run under DOS (sigh!). Any suggestions? Thanks, Jor-el
info is broken for me
Hi, I recently switched to using testing / unstable. Now when I try to use info, I get : [EMAIL PROTECTED]:~$ info info info: Cannot find node `Top'. I tried uninstalling info and reinstalling it, but this didnt seem to fix the problem. Any clues as to what the problem could be and how to resolve it? Thanks, Jor-el
Re: Problem posting with trn
Colin, Thanks for the reply (again). I discovered that the NNTPSERVER variable was NOT being honoured (what a relief to be able to use the British spelling of 'honour' again!). In fact, even though pretty much everything you said matched the docs, the programs were refusing to cooperate. On this theory that this was a bug, I did dpkg --purge of all news and trn related stuff I installed, and then installed inewsinn first before installing trn. Lo and behold, things started to work right away. Even then there were a number of bugs that I noted - the /etc/news/innd.conf file was being ignored - the values for the ORGANIZATION and NNTPSERVER were being read from the /etc/news/server and /etc/news/organization files - thus contradicting the documentation in the inews man page. Now, however, I've come across some rather annoying Pnews (isnt that still the news injector, or is it 'inews' when inewsinn is installed?). Here is what I got when I was replying to a newsgroup posting and I used the 'F' (quote original when replying) command : Check spelling, Send, Abort, Edit, or List? s Article not posted -- more included text than new text (Article not posted.) Garrh! I dont need a program to act like a grandmother. Is there anything I can do to turn off this behaviour (ah-ha! British spelling again!). Thanks, Jor-el On Wed, 15 Aug 2001, Colin Watson wrote: On Tue, Aug 14, 2001 at 11:10:12AM -0500, Jor-el wrote: If you don't have a local news server, there's no reason for you to have /var/lib/news/active at all, so no. Is it possible that you haven't set the NNTPSERVER environment variable correctly (it should be the hostname of your news server)? The inews from inewsinn should honour that environment variable and not try to look in a local active file if it's set. Cheers, -- Colin Watson [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Problem posting with trn
Hi, I am experimenting with trn, and am posting to the test newsgroup. The following env. variables are set : AUTHORCOPY points to my pine sent-mail folder, and REPLYTO is set to my email address. Here is a screenshot of my post attempt : == 0 unread articles in test -- read now? [ynq] Getting overview file. End of newsgroup test. What next? [qnp] Subject: Testing testing testing Distribution: (leaving cbreak mode; cwd=/home/devel/pgmr) This program may post news to many machines. Are you absolutely sure that you want to do this? [ny] y Prepared file to include [none]: Your article's newsgroup: test[no description available] Check spelling, Send, Abort, Edit, or List? s Article appended to /home/devel/pgmr/nsmail/sent-mail (re-entering cbreak mode) What next? [qnp] inews: No valid newsgroups in test inews: article in /home/devel/pgmr/dead.article dead.article contains : Newsgroups: test Path: not-for-mail From: Jor-el [EMAIL PROTECTED] Subject: Testing testing testing Date: 14 Aug 2001 07:01:55 GMT Organization: Megadodo Corporation, Ursa-Minor Beta Lines: 1 Reply-To: [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] Many many machines Why am I getting the error No valid newsgroups in test? Any help would be appreciated. Thanks, Jor-el
Re: Problem posting with trn
Hi, Well, I installed inewsinn, and here are the results : Subject: Test 9 from outer space Distribution: (leaving cbreak mode; cwd=/home/devel/pgmr) This program may post news to many machines. Are you absolutely sure that you want to do this? [ny] y Prepared file to include [none]: Your article's newsgroup: test[no description available] Check spelling, Send, Abort, Edit, or List? s No such newsgroups as test. (Article not posted.) Check spelling, Send, Abort, Edit, or List? Yes, there is a /var/news/lib/active file, and it doesnt contain the newsgroup test in it. Am I expected to manually update this file each time I subscribe to a new newsgroup? After installing inewsinn, I also edited /etc/news/inn.conf but Pnews doesnt seem to be reading that file. Is there no way to do this other than to install a local news server - something which I definitely do not want to do. Regards, Jor-el On Tue, 14 Aug 2001, Colin Watson wrote: On Tue, Aug 14, 2001 at 02:08:12AM -0500, Jor-el wrote: I am experimenting with trn, and am posting to the test newsgroup. The following env. variables are set : AUTHORCOPY points to my pine sent-mail folder, and REPLYTO is set to my email address. Here is a screenshot of my post attempt : [...] What next? [qnp] inews: No valid newsgroups in test inews: article in /home/devel/pgmr/dead.article dead.article contains : Newsgroups: test Path: not-for-mail From: Jor-el [EMAIL PROTECTED] Subject: Testing testing testing Date: 14 Aug 2001 07:01:55 GMT Organization: Megadodo Corporation, Ursa-Minor Beta Lines: 1 Reply-To: [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] Many many machines Why am I getting the error No valid newsgroups in test? Any help would be appreciated. Looking at the error message, I take it you're using the version of inews in the 'inews' package. That looks in /var/lib/news/active to find the list of possible newsgroups. If you have a local news server, make sure 'test' is in its active file. If you don't, you'll need to install inewsinn or inn2-inews instead. -- Colin Watson [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Netscape wrapper
Hi, I have to use a (closed-source) proprietary solution call Aventail connect to VPN to work. On Linux, I have to prefix every command I run with the command avconnect - for example : avconnect telnet some host . This has the effect of socksifying (or whatever the right term is for VPN), the socket calls. The documentation for this product says that if I use Netscape in this manner, not to use the wrapper script, or it can cause Netscape to crash. Sure enough, Netscape does crash with a bus error every now and then. I looked at the wrapper script in question trying to figure out what the heck it does, and I must confess defeat. Not only could I not figure out the purpose of the script and what it was doing, I also have no idea as to what exactly I am to invoke, if trying to execute Netscape without the wrapper. Maybe there is a way to modify the wrapper itself to do what I need to do, but I hope someone can give me a clue in that matter. Thanks, Jor-el
Re: exacutables on nfs mounts
Andrew, This sounds like the noexec flag is in effect. This would prevent executing programs on an NFS mount. Do mount and see if the noexec flag shows up as one of the options listed. Regards, Jor-el On Thu, 19 Jul 2001, Andrew Dixon wrote: Hi All, I've got a perl script here that won't run in my home director (which is nfs mounted) but will run if I put it in a file on the local disk. The /etc/exports file on the server looks like this: /home (rw,no_root_squash) and the /etc/fstab on the client looks like this: northbridge:/home /home nfs rw,user 00 Any ideas? Thanks, Andy -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: installing kernel-debs
Joerg, By a coincidence, I had opened a bug yesterday on this very topic. Here is the link which also contains the maintainer response. Regards, Jor-el On Fri, 20 Jul 2001, Joerg Johannes wrote: Hello list I have, in the last few weeks, compiled a lot of kernels (with make-kpkg) Well, I have installed these kernel-image-blah.deb's, but without removing the old ones. dpkg -l shows ii kernel-image-2. # -- dpk -l does not show more of the package version ii kernel-image-2. ii kernel-image-2. ii kernel-image-2. ... How can I safely remove the old ones (whose files were overwritten by the newer ones) ? I could make a 2.4.5 kernel and remove all the 2.4.3.deb's, but will this work? won't dpkg complain about files that are not there but should be there? joerg -- Did you know that if you play a Windows 2000 cd backwards, you will hear the voice of Satan? That's nothing! If you play it forward, it'll install Windows 2000. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where is the kernel source?
Nathan, Thanks for the reply, but the package you mentioned doesnt contain the source either. Yes, I know I could get 2.2.18, but the point is that I want to get the source of the kernel that I installed from the Debian CDs. It seems to me that if I cant find the source to that anywhere, that would be a GPL violation (the binary distributor has to provide the source code). Regards, Jor-el On Thu, 19 Jul 2001, Nathan E Norman wrote: On Thu, Jul 19, 2001 at 04:40:24PM -0500, Jor-el wrote: Hi, The kernel image that my system uses appears to be from the package kernel-image-2.2.18pre21-ide . One would expect that the source code to this .deb would be found in the corresponding source package - i.e in kernel-source-2.2.18pre21-ide . But it isnt there. www.kernel.org no longer has the pre patches stored there, so I dont know where to go get 2.2.18pre21 source code. Any ideas? 2.2.18pre21 was a pre-release candidate for 2.2.18. When 2.2.18 was officially released, it was preferred to any pre versions. You should get 2.2.18. I'm not sure about the ide patch ... I now run 2.4.x :) Cheers, -- Nathan Norman - Staff Engineer | A good plan today is better Micromuse Ltd. | than a perfect plan tomorrow. mailto:[EMAIL PROTECTED] | -- Patton
Re: installing kernel-debs
Joerg, Of course, it would be good if I had included the link too, wouldnt it? http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=105964repeatmerged=yes Regards, Jor-el On Fri, 20 Jul 2001, Jor-el wrote: Joerg, By a coincidence, I had opened a bug yesterday on this very topic. Here is the link which also contains the maintainer response. Regards, Jor-el On Fri, 20 Jul 2001, Joerg Johannes wrote: Hello list I have, in the last few weeks, compiled a lot of kernels (with make-kpkg) Well, I have installed these kernel-image-blah.deb's, but without removing the old ones. dpkg -l shows ii kernel-image-2. # -- dpk -l does not show more of the package version ii kernel-image-2. ii kernel-image-2. ii kernel-image-2. ... How can I safely remove the old ones (whose files were overwritten by the newer ones) ? I could make a 2.4.5 kernel and remove all the 2.4.3.deb's, but will this work? won't dpkg complain about files that are not there but should be there? joerg -- Did you know that if you play a Windows 2000 cd backwards, you will hear the voice of Satan? That's nothing! If you play it forward, it'll install Windows 2000. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Where is the kernel source?
Hi, The kernel image that my system uses appears to be from the package kernel-image-2.2.18pre21-ide . One would expect that the source code to this .deb would be found in the corresponding source package - i.e in kernel-source-2.2.18pre21-ide . But it isnt there. www.kernel.org no longer has the pre patches stored there, so I dont know where to go get 2.2.18pre21 source code. Any ideas? Thanks, Jor-el
pine and trn interaction
Hi, Does anyone know how to set trn up so that when a post is made to a newsgroup, a copy of the mail is stored in the sent-mail pine folder? Thanks, Jor-el mophobia, n.: Fear of being verbally abused by a Mississippian.
Re: can someone please explain to me error in execution path?
On Mon, 3 Jan 2000, Joseph de los Santos wrote: I installed enlightenment but when I try to run a program I get the error There was an error running the program: eeyes THis program could not be executed. this is most probably because this program is not in the path for your shell which is /bin/bash. I suggest you read the manual page for that shell and read up how to change or add to your execution path. I tried adding the default hidden directory of /.enlightenment to the path of my /bin/bash but it would not let me...I even made permission to enlightenment directory read/write to all. maybe I'm doing it all wrong? can sompebody please lend a helping hand. Thank you. Joseph, You shouldnt have to add the .enlightenment directory to your path. From the error message, it just seems that you havent installed the eeyes package from the 'graphics' section. Try installing it and let us know. Regards, Jor-el Carmel, New York, has an ordinance forbidding men to wear coats and trousers that don't match.
Re: Mirror transformation of the Postscript files
On Sun, 26 Dec 1999, Alexey Petrov wrote: Hi All, I have need in mirror transformation of the Postscript files for printing on the transparence. I should be glad for any idea. Alexey, This may sound like a dumb question - but cant you just flip the transparency onto its other side? Regards, Jor-el You have all eternity to be cautious in when you're dead. -- Lois Platford
Re: Detecting IBM 13.5GB IDE HDD---anyone?
On Sat, 25 Dec 1999, aphro wrote: are you running the latest BIOS ? 7.9GB is quite common on older boards as a limitation (32GB is another barrier) make sure you got the latest bios and the drive should be in LBA mode i believe. I just installed corel 1.0 on an IBM 18GB ide and it went fine (AOpen AP5T-3 i430TX mainboard) nate On Sat, 25 Dec 1999, Alan E. Davis wrote: adavis Installing potato. My system is as follows: adavisMainboard: ASUS P5A adavisCPU:K6-II 400 adavis adavis What is sfdisk? Is it on the install disks---apparently not? I have attempted to follow the Large Disk HOWTO and jumpered the drive as a 15 Head drive, as suggested. Otherwise it was detected at 500MB. adavis adavis I only see 7.9 GB. Several different BIOS parameters don't improve the situation. adavis adavis Has anyone on this list succeeded at this? adavis adavis Alan Davis adavis [EMAIL PROTECTED] adavis adavis adavis -- adavis Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null adavis Nate, This was a question originated by Wichert Akkerman on linux-kernel around the beginning of this month. If you look at the archives of the linux-kernel around at time, you should find a lot of comments regarding this problem. Here is one of them : http://boudicca.tux.org/hypermail/linux-kernel/1999week50/0753.html Regards, Jor-el You have all eternity to be cautious in when you're dead. -- Lois Platford
Installing an older compiler
Hi, I had to install some packages from unstable on my Slink system, and the dependencies forced an upgrade of the compiler to gcc 2.95.2). However, the 2.0 kernels wont compile with this. Is there a way to also install the standard Slink compiler on my system so that I can use it to compile the kernel? TIA, Jor-el You have all eternity to be cautious in when you're dead. -- Lois Platford
Re: sgmltools-2 and texinfo
On Mon, 13 Dec 1999, Steve George wrote: Hi, This should do the trick: jade -t tex -d /usr/lib/sgml/stylesheets/nwalsh-modular/print/docbook.dsl myfile.sgml myfile.tex where myfile.sgml is the DocBook SGML file you want to convert and myfile.tex is the output filename you want to use. Steve On Sun, Dec 12, 1999 at 08:52:34PM -0600, Jor-el wrote: Hi, Does anyone know if there is a conversion path for converting a docbook sgml document to texinfo format? Steve, Thanks. The path you specified was not quite correct on my semi-slink and semi-potato system, but the idea worked. Regards, Jor-el If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol
Re: DocBook Tools?
On Mon, 13 Dec 1999, Eric Gillespie, Jr. wrote: Where can I get the tools db2html, db2ps, db2rtf, etc.? They are used in generating documentation for Gnome, but I can't find them. I've grepped the Contents file for slink and potato, but don't see them. Eric, These sound remarkably like the programs packaged in the unstable sgmltools-2 package. Except, instead of the 'db' prefix, they have an 'sgml' prefix. Perhaps, you would prefer an unstable Debian package to an rpm? Regards, Jor-el If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol
sgmltools-2 and texinfo
Hi, Does anyone know if there is a conversion path for converting a docbook sgml document to texinfo format? TIA, Jor-el If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol
Re: Mailing list headers [Was Re: ssh vs telnet - which is faster?]
On Fri, 10 Dec 1999, Martin Schulze wrote: Kenneth Stephen wrote: On Thu, 9 Dec 1999, Steve Lamb wrote: Friday, December 10, 1999, 4:04:47 AM, RAVIKANT wrote: Exactly. See also http://garcon.unicom.com/FAQ/reply-to-harmful.html Furthermore, I think that the original problem of getting two copies of the same mail can be solved easily by using procmail. They're still transmitted twice. If you want to get the receipient to reply only to the list or to you personally, use the Mail-Followup-To: header. Set it to the list so modern mail programs will use that address if you hit group reply. Joey, This is an excellent idea. Thanks! However, I couldnt help but notice a reply from Steve Lamb to your link on debian-devel (dont know why he did that since the Reply-To is set to this list). Here is the link that he mentions : http://www.unicom.com/BBS/bbs_forum.cgi?forum=replytoread=38-00.msgsession =385142a809215317use_last_read=onlast_read=0 I must say that this represents the usual attitude I've seen from close-minded admins (who thankfully are few and far between) : if the behaviour that I dislike is allowed by the RFC's, the hell with the RFC's! Specifically, I'd like to address point 4 of his reply to the link that you quote. Once this is demolished, I think that all of his other responses become laughable : I quote : This is the most valid of the points. However, it is also quickly dismissed. Because three events must happen for this to become a valid arguement. 1: The person must be sending from an account that s/eh cannot recieve mail from. While this may have been true in the past, in most cases people are reading/sending mail from the account they want the mail go to. In the 7 years I've been using email and the 2 years of working at an ISP I have only seen the reply-to used twice. Both times it was because a person was switching to a new account and wanted an effective way to get people in the habit of sending to the new address. In such cases the mailing lists are usually the first to move to the new address. Aside from a 1-2 day overlap, there really is no danger of missing any mail. 2: The person reading the message wants to reply. The vast majority of people who read mailing lists lurk. They don't reply. In their case it doesn't matter what the reply-to is set to. It is only when the person replies that it matters. 3: The person replying wants to make it a private reply. Most replies go back to the list. So there has to be a reason to send a private reply. End quote. In the following the you refers to Steve Lamb, not Joey. Subpoint 1 : Just because you havent had experience with such behaviour doesnt give you a mandate to wipe out such RFC allowed behaviour. As I've said in another email : if you dont like it, write up another RFC and get it approved. Subpoint 2 : And if the reader does want to reply, then what? Is it your holy decree that such subscribers should not be replied to? Subpoint 3 : What??!! This doesnt make sense at all. Surely you mean the person sending the email wants the replier to make it a private reply. In that case, it is really not upto the list-admin to prevent this behaviour. Regards, Jor-el -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Ain't that something what happened today. One of us got traded to Kansas City. -- Casey Stengel, informing outfielder Bob Cerv he'd been traded.
Re: How to apply stylesheets to docbook?
On Sun, 5 Dec 1999, Aaron Van Couwenberghe wrote: Hi all - I have been trying to figure out how to make something useful out of docbook documents. I'd like to render to pdf, latex, groff, and html, but available documentation is archaic and conflicting. Where do I start? I'd previously been trying with jade, but I caught wind that sgml-tools gained support for docbook 3.1. This doesn't appear to be true... Executing sgml2latex on a minimal document: ..~/docs$ sgml2latex DB-article.sgml Processing file DB-article.sgml /usr/bin/sgmlsasp: can't open /usr/lib/sgml-tools/dist/article/latex2e/mapping': No such file or directory Executing jade on the same minimal document: ..~/docs$ jade -t tex DB-article.sgml -d /usr/lib/sgml/stylesheet/dsssl/docbook/nwalsh/print/docbook.dsl -o DB-article.tex [about 500 lines of errors follow] Even norm walsh's book doesn't offer any help here. Ideas anyone? ;) Aaron, I cant tell you how to get things working with the sgml-tools package. I can tell you that I've recently got things to work with the sgmltools2 package from potato. Note that the apt-get install of sgmltools2 fails unless you remove the existing sgml-tools package (I've filed a bug report). With sgmltools2, the formats suppported are html, ps, dvi, rtf, txt, ld2db, and jadetex. The man page claims that passing of options to jade is also possible, but I havent tried this. The Docbook 3.1 DTD is definitely supported as I am writing using that DTD. You might also try http://www.debian.org/~bortz/SGML-HOWTO/ Regards, Jor-el Ain't that something what happened today. One of us got traded to Kansas City. -- Casey Stengel, informing outfielder Bob Cerv he'd been traded.
Bold using docbook and other problems
Hi, I am currently using the sgmltools2 package from Debian, and the corresponding latest and greatest docbook packages from Debian Potato. However, I cant seem to get a few basic things to work the way I want it to. My reading of the docbook documentation suggests that the way to get the bold effect when converted to html is to use the emphasis tag with a role of bold or strong. This I do by emphasis role=bold /emphasis but the result that I get in html is still in italics - which is no different than what I get with plain emphasis tags. What is the correct way to do this. The actual problem that I am trying to solve is to have the following text Example n--- formatted in bold (in html) some blurb --- ordinary emphasis some code listed --- am using the programlisting tags some more blurbs --- ordinary emphasis Here is what I tried : exampletitle/titleprogramlisting/programlisting/example parasome blurb programlistingsome code listed/programlisting parasome more blurbs What I get in html is Example n -- in bold some blurb -- ordinary text some code listed-- formatted exactly as typed some more blurbs-- ordinary text Here are some problems as I see it : 1) The programlisting tags are supposed to preserve all whitespace according to the documentation. However, it seems to skip tabs. As a result, I've had to convert all tab characters to an equivalent number of spaces in the sgml text. 2) The docs indicate for the example tags that the title tags are mandatory. However, they do not state that the programlisting tags are mandatory, whereas the sgmltools converters treat this as a mandatory child for example. 3) Note that there is an extra blank line between the Example n title and the blurb that follows. How can I get rid of it? Thanks in advance, Jor-el Ain't that something what happened today. One of us got traded to Kansas City. -- Casey Stengel, informing outfielder Bob Cerv he'd been traded.
No. of lines / page in ps output
Hi, When I use the sgml to postscript converter, the number of lines that appear in one postscript page is quite low and not appropriate for an A4 size page. Where is the pagesize specified in sgmltools? Is it in a place which can be controlled by the user? Thanks in advance, Jor-el Ain't that something what happened today. One of us got traded to Kansas City. -- Casey Stengel, informing outfielder Bob Cerv he'd been traded.
sgml and DocBook
Hi, I wrote an entire document using the debiandoc DTD before I realized that I picked on the weaker and less prevailing DTD. I would like to convert my entire document to docbook. Unfortunately, the docbook package that I have (originally the Slink package, but I upgraded to the potato sgmltools2 package) doesnt seem to be set up perfectly. When I try to convert the following test document : !doctype DocBook system book ? /book into html by doing '$sgmltools -b html try.sgml' (where try.sgml is the above file), I get : /usr/bin/jade:OSFD0:1:24:W: cannot generate system identifier for document type DOCBOOK /usr/bin/jade:OSFD0:1:24:E: reference to entity DOCBOOK for which no system identifier could be generated /usr/bin/jade:OSFD0:1:0: entity was defined here /usr/bin/jade:OSFD0:1:24:E: DTD did not contain element declaration for document type name /usr/bin/jade:OSFD0:2:5:E: element BOOK undefined /usr/bin/jade:OSFD0:2:5:E: element DOCBOOK undefined /usr/bin/jade:OSFD0:4:8:E: end tag for DOCBOOK omitted, but its declaration does not permit this /usr/bin/jade:OSFD0:2:0: start tag was here Can anyone tell me what is wrong? I have the vague idea that I need to set up my catalog correctly, but have no idea how. TIA, Jor-el Did you know ... That no-one ever reads these things?
Re: Why did 'man dpkg' quit working on me?
On Sat, 27 Nov 1999, John Pearson wrote: On Fri, Nov 26, 1999 at 07:58:35AM -0600, Jor-el wrote Let me answer your questions in reverse: You don't *really* have a standard Slink install. My guess is that somewhere along the line, you must have installed a new dpkg. The slink dpkg has its man pages under /usr/man, which is the convention adopted for slink; this has chaged for Potato to /usr/share/main, and I'm guessing that you either installed a 'potato' version of dpkg, or rebuilt a potato release from source. I could be wrong, it may also be that upgrading to the potato version of man-db moves your man pages (but then, they all would have moved). The way you fixed it is fine if you anticipate running a 'mixed' installation (some slink, some potato) and won't hurt in any event. Alternatives would be to downgrade to the Slink version of dpkg (but some potato packages, e.g. enlightenment, may require a later version), or to build a 'slinkified' version of the dpkg package, modified to reflect slink conventions: i.e., use /usr/info, /usr/doc and /usr/man in place of /usr/share/info, /usr/share/doc and /usr/share/man (not necessarily a complete list, but they are the changes that spring to mind). John P. -- [EMAIL PROTECTED] [EMAIL PROTECTED] Oh - I - you know - my job is to fear everything. - Bill Gates in Denmark -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Hi, Thanks for the replies - John and Lindsay. I think your theories are on target. I recently began exploring the world of apt-get, and when I did an 'apt-get dist-upgrade' on stable, I think one of the packages that got upgraded was dpkg. This was probably a version that has the new changes. I wish though, that there is a way I could tell what date and time a current package was installed, so that I could be sure that this is indeed the cause. Regards, Jor-el Interestingly enough, since subroutine declarations can come anywhere, you wouldn't have to put BEGIN {} at the beginning, nor END {} at the end. Interesting, no? I wonder if Henry would like it. :-) --lwall
Why did 'man dpkg' quit working on me?
Hi, I am working on a box which has a standard Slink install. I used to be able to pull up the man page of dpkg by doing 'man dpkg'. Yesterday, all of a sudden, this command quit working on me, and I cant figure out why. The man page was still around - located in /usr/share/man/man8, but man wasnt picking it up unless I used the '-M' flag to tell it where it was. I tried rebuilding the man database by doing a 'mandb -c', but this didnt solve it either. In the end, I had to end up editing /etc/manpath.config and added /usr/share/man to the MANDATORY_MANPATH and then rebuilding mandb, before my problem could get solved. I have two questions : (1) did I solve this the right way? and (2) Why did this fail in the first place? Any ideas? Regards, Jor-el I drink to make other people interesting. -- George Jean Nathan
Re: quicktime4 player for linux?
On Sat, 6 Nov 1999, T.V.Gnanasekaran wrote: any player that can play qt4 and the qt4 streaming media? -gnana 2.2.13|potato Hi, Does xanim not do this? Regards, Jor-el Automobile, n.: A four-wheeled vehicle that runs up hills and down pedestrians.
Re: Page length using a2ps
On Fri, 27 Aug 1999, Richard E. Hawkins wrote: I have a text document that is formatted with page length = 68 rather than the 65 which is the default for a2ps. However, whenever I try to format the text using a2ps into a 68 lines per page document (using the --lines-per-page option), it doesnt seem to make a difference : I still get a postscript document with 65 pages. What am I doing wrong? Try changing the order of the options you use; iirc, that's one of the ones that fussy about its placement. You might also try chars-per-line instead. a2ps can be just a bit cranky at times. rick -- Richard, I received many answers to my question (and thanks to all who contributed), but yours (order of options) turned out to be the right one. Thanks a lot, Jor-el He who is intoxicated with wine will be sober again in the course of the night, but he who is intoxicated by the cupbearer will not recover his senses until the day of judgement. -- Saadi
Re: Files: Debian-to-dos.
On Thu, 26 Aug 1999 [EMAIL PROTECTED] wrote: Is there a way to save a linux-file to a dos-diskette in a dos-readable way? (Say I wanted to show you guys a config-file or something when I'm mailing from an NT-box, not wanting to type the whole thing, or I wanted to take a file from f. x. StarOffice with me to work...) Please don't flame a pitiful 'nix newbie. I'm really trying to learn, and I already love the efficiency of linux compared to my horrible WinNT box at work (breaks down several times a day...) Regards Vitux Vitux, A one-line solution with no special programs required would be : sed 's/$/control-v followed by control-m/' filename output_file This would convert from Unix format to DOS format. The reverse operation would be acomplished by : sed 's/control-v followed by control-m$//' filename output_file (The '$' is literally present in both cases) Regards, Jor-el He who is intoxicated with wine will be sober again in the course of the night, but he who is intoxicated by the cupbearer will not recover his senses until the day of judgement. -- Saadi
Page length using a2ps
Hi, I have a text document that is formatted with page length = 68 rather than the 65 which is the default for a2ps. However, whenever I try to format the text using a2ps into a 68 lines per page document (using the --lines-per-page option), it doesnt seem to make a difference : I still get a postscript document with 65 pages. What am I doing wrong? Thanks, Jor-el He who is intoxicated with wine will be sober again in the course of the night, but he who is intoxicated by the cupbearer will not recover his senses until the day of judgement. -- Saadi
Re: Collecting mail for other users with fetchmail?
On Wed, 25 Aug 1999, Phillip Deackes wrote: Jor-el [EMAIL PROTECTED] wrote: Phillip, Its seems to me that you havent read the manpage. Specifically, the sections about the '-m', the '-u', and the user authentication. It seems to me you didn't read my original query. I did read the manpage, -m appears to refer to the mail delivery agent. I use exim as an MDA. -u relates to the login name with the IP. As I explained, my IP's POP3 server requires that I log in with scgf, my *hostname*. The same hostname is used by everyone on my system, hence mail collected with this login could belong to anyone. As it happens, I found the answer in the Linux Gazette - they have an excellent article on Setting up Exim for a home network (http://www.linuxgazette.com/issue42/stumpel.html I added this to my ~/.forward file: if $header_to: contains sah1 then deliver sah1 This causes exim to deliver mail addressed to [EMAIL PROTECTED] to the correct inbox for sah1, ie. /var/spool/mail/sah1 I appreciate your reply, Jorel, but I am somewhat annoyed at your implication that I would not bother to read the most obvious manpage. It appears that many Linux users are just too quick to point the finger at others who they believe have not read the documntation. Phillip, Please accept my apologies for the unwarranted aspersions. You are right : I didnt completely understand your question. Jor-el He who is intoxicated with wine will be sober again in the course of the night, but he who is intoxicated by the cupbearer will not recover his senses until the day of judgement. -- Saadi
Re: Collecting mail for other users with fetchmail?
Phillip, Its seems to me that you havent read the manpage. Specifically, the sections about the '-m', the '-u', and the user authentication. Regards, Jor-el On Wed, 25 Aug 1999, Phillip Deackes wrote: I use currently use fetchmail to collect mail from my IP's POP server. There are three users on this system. My IP allows email addresses in the form [EMAIL PROTECTED] uk, therefore allowing each user on my system to have their own email address. To log on and collect mail, the 'hostname' is given then my password. *All* mail for that hostname is then downloaded by fetchmail - the problem is that any mail for other users appears in *my* mailbox (/var/spool/mail/gsmh). How can I ensure that mail for another user on my system gets into the correct mailbox? Many thanks. -- Phillip Deackes Debian Linux (Potato) -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null He who is intoxicated with wine will be sober again in the course of the night, but he who is intoxicated by the cupbearer will not recover his senses until the day of judgement. -- Saadi
Re: compiling error: xauth not found
Horacio, 'xauth' is indeed present in Debian - you just havent installed it. It is part of the package xbase-clients on Slink. Regards, Jor-el Be careful of reading health books, you might die of a misprint. -- Mark Twain On Sat, 21 Aug 1999, J Horacio MG wrote: I'm trying to compile ssh1 and found the following error while running ./configure. I su'ed root and copied the tarball to /usr/local: $ su ... /usr/local/ssh-1.2.27# ./configure ... checking for xauth... no configure: error: configuring with X but xauth not found - aborting Since there's no xauth, but .Xauthority in Debian, how do I tell the configure script about this? BTW, I know there is a ssh1 deb package, and even a ssh2 deb package, but I'd like to do it by hand. TIA -- Horacio [EMAIL PROTECTED] Valencia - ESPAÑA -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: compiling error: xauth not found
Horacio, Maybe something went wrong on your system during or after the install. But I do suggest you do your homework more thoroughly. Here is what I get : debian$pwd /mnt/cdrom/dists/slink/main/binary-i386/x11 debian$dpkg -c xbase-clients_3.3.2.3a-11.deb | less drwxr-xr-x root/root 0 1999-02-23 02:54 ./ drwxr-xr-x root/root 0 1999-02-23 02:54 usr/ drwxr-xr-x root/root 0 1999-02-23 02:54 usr/doc/ drwxr-xr-x root/root 0 1999-02-23 02:54 usr/doc/xbase-clients/ -rw-r--r-- root/root 9972 1999-02-23 02:54 usr/doc/xbase-clients/copyright -rw-r--r-- root/root 290 1999-02-23 02:54 usr/doc/xbase-clients/README.Debi an -rw-r--r-- root/root 23295 1999-02-23 02:54 usr/doc/xbase-clients/changelog.D ebian.gz drwxr-xr-x root/root 0 1999-02-23 02:54 usr/X11R6/ drwxr-xr-x root/root 0 1999-02-23 02:54 usr/X11R6/bin/ -rwxr-xr-x root/root 27499 1999-02-23 02:54 usr/X11R6/bin/Xmark -rwxr-xr-x root/root 4616 1999-02-23 02:54 usr/X11R6/bin/appres -rwxr-xr-x root/root 6020 1999-02-23 02:54 usr/X11R6/bin/atobm -rwxr-xr-x root/root 33876 1999-02-23 02:54 usr/X11R6/bin/bdftopcf -rwxr-xr-x root/root 10708 1999-02-23 02:54 usr/X11R6/bin/beforelight -rwxr-xr-x root/root 70376 1999-02-23 02:54 usr/X11R6/bin/bitmap -rwxr-xr-x root/root 4496 1999-02-23 02:54 usr/X11R6/bin/bmtoa -rwxr-xr-x root/root 10696 1999-02-23 02:54 usr/X11R6/bin/dga -rwxr-xr-x root/root 40240 1999-02-23 02:54 usr/X11R6/bin/editres -rwxr-xr-x root/root 24108 1999-02-23 02:54 usr/X11R6/bin/fsinfo -rwxr-xr-x root/root 30180 1999-02-23 02:54 usr/X11R6/bin/fslsfonts -rwxr-xr-x root/root 28752 1999-02-23 02:54 usr/X11R6/bin/fstobdf -rwxr-xr-x root/root 18664 1999-02-23 02:54 usr/X11R6/bin/iceauth -rwxr-xr-x root/root 81748 1999-02-23 02:54 usr/X11R6/bin/mkfontdir -rwxr-xr-x root/root 12596 1999-02-23 02:54 usr/X11R6/bin/oclock -rwxr-xr-x root/root 7308 1999-02-23 02:54 usr/X11R6/bin/resize -rwxr-xr-x root/root 5984 1999-02-23 02:54 usr/X11R6/bin/sessreg -rwxr-xr-x root/root 77044 1999-02-23 02:54 usr/X11R6/bin/setxkbmap -rwxr-xr-x root/root 3588 1999-02-23 02:54 usr/X11R6/bin/showrgb -rwxr-xr-x root/root 1813 1999-02-23 02:54 usr/X11R6/bin/startx -rwxr-xr-x root/root 84972 1999-02-23 02:54 usr/X11R6/bin/x11perf -rwxr-xr-x root/root 2712 1999-02-23 02:54 usr/X11R6/bin/x11perfcomp -rwxr-xr-x root/root 26432 1999-02-23 02:54 usr/X11R6/bin/xauth Note the last line. Regards, Jor-el Be careful of reading health books, you might die of a misprint. -- Mark Twain On Sat, 21 Aug 1999, J Horacio MG wrote: $ dpkg -l | less ii xbase-clients 3.3.2.3a-11miscellaneous X clients may be xauth is somewhere else? Jor-el dijo: 'xauth' is indeed present in Debian - you just havent installed it. It is part of the package xbase-clients on Slink. On Sat, 21 Aug 1999, J Horacio MG wrote: I'm trying to compile ssh1 and found the following error while running ./configure. I su'ed root and copied the tarball to /usr/local: $ su ... /usr/local/ssh-1.2.27# ./configure ... checking for xauth... no configure: error: configuring with X but xauth not found - aborting Since there's no xauth, but .Xauthority in Debian, how do I tell the configure script about this? BTW, I know there is a ssh1 deb package, and even a ssh2 deb package, but I'd like to do it by hand. Regards, -- Horacio [EMAIL PROTECTED] Valencia - ESPAÑA
How does one activate sound on Slink
Hi, I wasnt too concerned about audio when I installed Slink. Now, however, I feel the sudden desire to make my machine talk. The only question is how. None of the audio modules required have been installed to disk, and 'modconf' doesnt have any audio drivers listed. The install guide doesnt make any mention of sound. How the heck do I get started? TIA, Jor-el There's some entertainment value in watching people juggle nitroglycerin. -- Larry Wall in [EMAIL PROTECTED]
Re: Exim and Mail Forwarding
-BEGIN PGP SIGNED MESSAGE- Graham, You may recieve some better advice from someone else, but it seems to me that you need to run DNS for your local network. The MX records for the local domain would tell the respective mailservers which machine they need to deliver external mail to. In the case of your machines on your local network, the MX records would be setup so that non-local users on the local network would be delivered to the Exim server on the appropriate machine on your local network. All internet mail (if originating from machines other than your gateway machine) would have to be delivered first to your gateway machine, which would relay it to wherever. Regards, Jor-el PS. There are some additional complications that you will discover if you are doing address rewriting. My research indicates that Exim is really a poor fit for dialup machines. I will be switching to Postfix as soon as the Debian package in Potato matures. There's some entertainment value in watching people juggle nitroglycerin. -- Larry Wall in [EMAIL PROTECTED] On Thu, 5 Aug 1999 [EMAIL PROTECTED] wrote: Hi, I have just finished setting up my internet gateway server using the Setting Up Mail for a Home Network Using Exim tutorial in the July(43) issue of Linux Gazette. However I have another linux machine that will need to send mail to its local users, the users on the internet gateway and the internet. Do i set it up so that the internet gateway server is the smarthost for this machine and then use the aliases/.forward file to forward mail from the internet gateway server to this second server or is there a better way. The problam I have currently is that the two machine cannot send email to each other but everything else works, telnet, ftp etc. Any ideas?? Regards Graham -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBN6l1APrE9j2ZpWNBAQGCQQP+KTusazlSu9KT0Ak0jozdh86CAzVYRNP8 Jau6fzj5i2t+pxjS8OynApIslWe6VFG/tPJDemQoMqzClsQR52GPA+s8k0dEd+fO VNfI+w24B0UZASZw8t0ZFiidzsRF57jWpv8FPsYTQ0HV1kklUj04E4D9y6e5BM7a vWXg6uKKV2M= =UJIi -END PGP SIGNATURE-
Re: 2.2.x kernel and PCI eth0 problem
-BEGIN PGP SIGNED MESSAGE- Rupert, I assume that you are running Slink. Have to installed the upgrade versions of the packages that are required for your to be running the 2.2.x kernels? I belive the net-tools package is one of the packages that need to be ugraded, so I am not at all surprised to hear that networking is failing for you. There is a URL available somewhere on the Debian website which tells you about the packages that need to be upgraded. Unfortunately, I do not have a URL handy - so you'll have to search. Regards, Jor-el Everywhere I go I'm asked if I think the university stifles writers. My opinion is that they don't stifle enough of them. There's many a bestseller that could have been prevented by a good teacher. -- Flannery O'Connor On Mon, 2 Aug 1999, Rupert Brooks wrote: Hi, I have configed and compiled both the 2.2.10 and 2.2.5 kernels, and my ne2000 clone PCI ethernet card does not work with either. With my 2.0.34 kernel, the card worked fine. The module ne2k-pci loaded without a hitch and dhcpcd configured the ip information flawlessly. ( I use an ADSL connection which requires DHCP to be used to set up the IP address.) I have tried both kernels with NE2000 support compiled in. In this case, the card appears to be recognized on startup, but DHCPCD does not create new files when it runs. DHCPCD does not report any errors either. ifconfig without arguments shows the loopback device. Ifconfig listing all devices lists the eth0 device and it looks ok, but none of the IP information is configured. Note that I have also used dhcpcd eth0 to try and force dhcp to use the ethernet card. If I configure the kernel to use loadable module support for the ethernet card, then the module ne2k-pci won't load. When using insmod to try to load it, it will complain aboutn several symbols not being resolved. From their names they look like the ethernet card hooks in the kernel. After that lengthy description - I have a couple questions Does anyone out there have a ne2000 PCI ethernet card working under a 2.2.x kernel? If so, did you have to do anything wierd to get it to work? I am not sure what is not working, because there have been so few error messages. Can anyone recommend some diagnostic tests? TIA Rupert Brooks -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBN6boavrE9j2ZpWNBAQHr7wQAhRVXTfZcyb+oRJegJ/K7BOOT+mS06H8+ Ptmz35+BkgSJddrYScmO/8bxzZHNoinhVDd2KwPfYEnJQFvIf2thP40rqXj4wJA8 w5/AIQJvDaOtxg98g09qquHRH57rbBcyVE0bJCLC+l9qJiYzPjSxmD3RHvOKtZlL P7WQVXUVi0M= =WzBR -END PGP SIGNATURE-
Re: Xlib, or I'm an idiot...again
-BEGIN PGP SIGNED MESSAGE- Dwarf, Does your .xsession-errors file contain any information, after you try executing these programs from the menu? Regards, Jor-el Human cardiac catheterization was introduced by Werner Forssman in 1929. Ignoring his department chief, and tying his assistant to an operating table to prevent her interference, he placed a ureteral catheter into a vein in his arm, advanced it to the right atrium [of his heart], and walked upstairs to the x-ray department where he took the confirmatory x-ray film. In 1956, Dr. Forssman was awarded the Nobel Prize. On Sun, 1 Aug 1999, Dale Scheetz wrote: On Sat, 31 Jul 1999, Jor-el wrote: -BEGIN PGP SIGNED MESSAGE- [ list changed to debian-user since this definitely belongs there ] Dale, You need to read the xauth man page. You probably tried to start the programs in question after su'ing to an id which was not the id used to enter your X session. Right? Right on the nose! I had su'd to root to install the new mozilla package to try it, and hadn't dropped back. I don't think I've had this experience before with root. What happened to my super user capabilities? Now the newly compiled xcircuit can be run from the prompt, and mozilla comes up to the splash screen before it reports a segfault ;-( Neither of the new programs will run from the menu. Is menu broken, or has it just changed and my package and mozilla haven't caught up? Thanks for the pointers, -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBN6ThNPrE9j2ZpWNBAQFnYwP/UYJJFof9cg8sXoVrp2Kf+BzdU+rhBbzj GVgUCMkJet4DPOZNiLd5ZUDICyLiTNlhDoNpZ+itnobid3eQVV7s2NvqD9/3U6fy 5EKP7oUpeeqoelfgZqAlZvs8JFGl5HO6jT+vTpaJj1T51ns5WW5fiOb9YvKBtelF +uufFgx4W94= =T02F -END PGP SIGNATURE-
Re: Xlib, or I'm an idiot...again
-BEGIN PGP SIGNED MESSAGE- [ list changed to debian-user since this definitely belongs there ] Dale, You need to read the xauth man page. You probably tried to start the programs in question after su'ing to an id which was not the id used to enter your X session. Right? Regards, Jorel The Macintosh is Xerox technology at its best. On Sat, 31 Jul 1999, Dale Scheetz wrote: I upgraded my system to potato a week or so ago, and since then mozilla will not load. I just built the newest version of xcircuit, and it refuses to load for the same reasons given by mozilla. Xlib: connection to :0.0 refused by server Xlib: Client is not authorised to connect to server Error: Can't open display :0.0 I also can't seem to ftp into this machine any more (and I did it all the time before the upgrade), are all these issues related? What do I do to fix this? Waiting is, Dwarf -- _-_-_-_-_- Author of The Debian Linux User's Guide _-_-_-_-_-_- aka Dale Scheetz Phone: 1 (850) 656-9769 Flexible Software 11000 McCrackin Road e-mail: [EMAIL PROTECTED] Tallahassee, FL 32308 _-_-_-_-_-_- See www.linuxpress.com for more details _-_-_-_-_-_-_- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBN6OfhfrE9j2ZpWNBAQFphgP+KJi5KFKd/zL5Ltysb6mepCXWmRCd63v3 ZS36Wt/qwPMJLCjk8dhPXuBwal4iX1tN+ZFljU8lecruxkAcsVW9qTltFjPA9IB5 KvRPBDLWwhFeA+X/udsjehB557r/xuUbn3MMtNsSpc7XOFMzdpxYM3OVBOWwmFJm d0AIMBfaKfo= =O5/K -END PGP SIGNATURE-
Re: Init takes long time and what is lspci?
-BEGIN PGP SIGNED MESSAGE- Jonas, 'lspci' is not needed and will not work on the 2.0.x kernels. They need a file in the /proc heirarchy which was introduced in the 2.2.x kernels. Cant answer your INIT question. Regards, Jor-el Human cardiac catheterization was introduced by Werner Forssman in 1929. Ignoring his department chief, and tying his assistant to an operating table to prevent her interference, he placed a ureteral catheter into a vein in his arm, advanced it to the right atrium [of his heart], and walked upstairs to the x-ray department where he took the confirmatory x-ray film. In 1956, Dr. Forssman was awarded the Nobel Prize. On 1 Aug 1999, Jonas Steverud wrote: Debian Potato. Last queation first: what's lspci? I get the message that lspci is not found so no PCI conflicts are calculated when I boot. I presume it's in pciutils but who do I know if I need it or not? I bought the computer in november '96 and I really don't know/remember what's in it. (I gave up keeping track of all new standards when SVGA meant better then 640x480x16 and the i486 came.) When I boot, init prints out INIT 2.74 (?) and then it stops for 10-20 seconds and then continues with NET4: Unix domain sockets 1.0 for Linux NET4.0. which takes another 5-10 seconds. Anyone who knows why this happens? Is it trying to get some information from somewhere or what? Before I changed to the 2.2.10-kernel (see other thread) this did not happen. Or is it perfectly normal? -- ( Jonas Steverud @ www.dtek.chalmers.se/~d4jonas/ !Wei Wu Wei) ( U2MoL, Roleplaying, LaTeX, Emacs/Gnus, SCWM, etc. ! To Do Without Do ) -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -BEGIN PGP SIGNATURE- Version: 2.6.3a Charset: noconv iQCVAwUBN6RPxvrE9j2ZpWNBAQFsbwQAwo8kz62RAhgvBSQ3SptxOlVtXGhWCusR 0opRaAqUaybZEAHXhuYE3K6+d4RRzvP9esmcmvDINXLAq0OntKjE15N9ckVBe8xY TmihHYrbS6QI/+zgofe8Xw+s6Ti0VHbItE1+HIGgp5u9MCv0iy2yWLtfuRMJvROJ nhXbdvmC19Q= =Mijy -END PGP SIGNATURE-
Re: Exim config question
Gary, The problem is with exim which thinks that mach1 is your entire domain name. Here is a snippet from the ROUTERS CONFIGURATION section of exim.conf : # # Lets deliver mail to remote users on the megadodo.umb domain # qualify : driver = domainlist route_list = ^[A-Za-z0-9]*$ $domain.megadodo.umb local_net : driver = domainlist domains = *.megadodo.umb transport = local_net_smtp route_list = * $domain byname host_find_failed = fail_soft self = send The 'qualify' router takes a single component domain name and replaces it with single component.megadodo.umb . It does no deliver since there is no transport specified, so exim takes this converted domain down to the 'local_net' router. The 'local_net' router uses the 'domains' keyword to filter out all non megadodo.umb hostnames. This is essential because if I want to send mail to [EMAIL PROTECTED], I do _not_ want the 'local_net' router acting on it. Note the 'route_list' keyword. It tells the router to verify the hostname via DNS. Since my DNS server only provides lookups for my local network (*.megadodo.umb), it will fail on all non-local addresses. The 'host-find-failed' option now tells exim to pass it onto the next router, rather than failing the routing process. What you do in the next router is whatever you need to do to deliver internet mail. HTH, Jor-el The price of greatness is responsibility. On 25 Jul 1999, Gary L. Hennigan wrote: I'm in the process of getting a little home network set up. One of the hosts, call it mach2, won't have a connection to the internet, just to other hosts on my home network. The other host, call it mach1, will occasionally connect to the internet via dialup. I have mach1 all set up but can't seem to get mach2 to forward all the mail to mach1. In fact I can't get mach2 to deliver mail directly to mach1 at all. What I want to do is set mach2 so that it uses mach1 as a smarthost. I think I got that right in my exim.conf (configuration 2 from the debian installation) but I can't seem to make it work. When I send email to, for example, [EMAIL PROTECTED] I get this in my exim mainlog: 1999-07-25 10:38:36 118R4n-B0-00 == [EMAIL PROTECTED] routing defer (-32): retry time not reached or 1999-07-25 10:39:47 118R4n-B0-00 == [EMAIL PROTECTED] R=smarthost defer (-1): I get the same thing if I send the email to [EMAIL PROTECTED] Also, something's not quite right with mach1. If I connect manually, eg., telnet mach1 smtp and type rcpt to: [EMAIL PROTECTED] I get: 550 relaying to [EMAIL PROTECTED] prohibited by administrator but if I do rcpt to: [EMAIL PROTECTED] it works. Any ideas? Thanks, Gary -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Problem with mkfs
Hi, My /dev/hdb was partitioned into /dev/hdb1 and /dev/hdb2 primary partitions. I used fdisk to repartition /dev/hdb1. What I did was (1) delete /dev/hdb1 (2) Create an extended partition replacing the space occupied by /dev/hdb1 (3) Create two logical partitions in that space : /dev/hdb5 and /dev/hdb6. Here is what I now get : /etc#fdisk /dev/hdb Command (m for help): p Disk /dev/hdb: 255 heads, 63 sectors, 525 cylinders Units = cylinders of 16065 * 512 bytes Device Boot Start End Blocks Id System /dev/hdb11 200 1606468+ 5 Extended /dev/hdb2 201 525 2610562+ 83 Linux native /dev/hdb51 70 562212 83 Linux native /dev/hdb6 71 200 1044193+ 83 Linux native However, when I now try to create an ext2 filesystem on these newly created partitions, I get : /etc#mkfs -t ext2 /dev/hdb5 mke2fs 1.12, 9-Jul-98 for EXT2 FS 0.5b, 95/08/09 /dev/hdb5: Invalid argument passed to ext2 library while setting up superblock What is going on here? I get the same results with /dev/hdb6. I read and reread the mkfs and the mke2fs manpages until I became blue in the face, but I cant find any syntax error in my invocation. TIA, Jor-el PS. The system is running Slink. Except for 75% of the women, everyone in the whole world wants to have sex. - Ellyn Mustard
xkbcomp error
Hi, Can anyone explain why I am getting this error during 'startx' and what I can do to fix it? System: `/usr/X11R6/lib/X11/xkb/xkbcomp -w 1 -R/usr/X11R6/lib/X11/xkb -xkm -m us -em1 The XKEYBOARD keymap compiler (xkbcomp) reports: -emp -eml Errorsfrom xkbcomp are not fatal to the X server keymap/xfree86 /var/tmp/xfree86.xkm' TIA, Jor-el
Re: Exim SMTP
Ali, Try removing the specification of the smtp server in your Communicator preferences - leave it blank. If this works, good. For an explanation of this phenomenon, see the reply I received just yesterday on this list for my message with the subject Exim, Pine, and smtp-server. Regards, Jor-el On Thu, 8 Jul 1999, Ali Onur Uyar wrote: I am using Debian Slink with Exim, configured for an Internet Site, to receive and send mail using SMTP. Everthing is configured using default configuration paraemeters. That is relaying mail through smtp for nonlocal domains is disabled. Now the question is wether Exim rejects smtp requests from localhost. I have installed netscape communicator and as the Outgoing Mail (SMTP) server I am trying to setup localhost. When I try to send mails a dialog appears, informing me that the smtp server does not accept the connection or sth. like that. What might be causing this behaviour? Is it because of the config defaults of exim? TIA -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Exim, Pine, and smtp-server
Hi, I use exim as the MTA and the relay_domains option is not set (commented out). When the smtp-server parameter is configured in pine to be that for the machine running the MTA, I get an error whenever I send mail to non-local addresses. The error says that the administrator has set up the MTA so that relaying mail for anything other than non-local domains is prohibited. However, when the smtp-server parameter is not set in pine, I can send mail just fine. Is this expected behaviour or a bug in pine? TIA, Jor-el
Rewriting message headers for a smarthost relay using exim
Hi, I connect to the internet via a dialup connection, so while installing exim, I selected the smarthost relay option (which is the recommended option, I believe) during the configuration. I want my MUA to spool up mail into the mail queue, and when connected, the SMTP server that I run should deliver the mail to my ISP's mail server for delivery across the internet. My local domain name (on my private network) is megadodo.umb . My ISP connects me to an ibm.net domain, and their SMTP server refuses to accept mail with an envelope FROM field having a non ibm.net address (DNS lookup fails for them - only natural). Hence, my desire to rewrite this header field to my ISP mail account address - lets call it [EMAIL PROTECTED] . For this purpose, I added the following lines in exim.conf : [EMAIL PROTECTED]${lookup{$1}lsearch{/etc/email-addresses}\ [EMAIL PROTECTED] F And this works correctly and rewrites the correct header field. All is well, except for the fact that the envelope now gets rewritten for local deliveries too. Thus, if local user [EMAIL PROTECTED] is mapped to [EMAIL PROTECTED], when I send mail from [EMAIL PROTECTED] to [EMAIL PROTECTED], the mail showing up in [EMAIL PROTECTED] 's mailbox says that it was sent by [EMAIL PROTECTED] . This is most certainly NOT what I want. Is there a way to force header rewriting only for non-local deliveries? Thanks in advance, Jor-el
Re: A dumb, somewhat off-topic question...
Hi, Actually, in the Open Source book by O' Reilly, Linus himself answers this question : all other variants of UNIX were actually derived from the ATT and (eventually) BSD source code. Linux was written from scratch and is not based off of the ATT or BSD sources. So Linux is really a UNIX like OS (because it does have the same architecture) but is not a UNIX variant. Regards, Jor-el On Tue, 8 Jun 1999, Jakob 'sparky' Kaivo wrote: On Wed, 9 Jun 1999, Alisdair McDiarmid wrote: Mark Wright writes: Did someone register FreeBSD? If you check out FreeBSD.org, they say FreeBSD is an advanced BSD UNIX operating system. They don't need anyone's permission to call FreeBSD UNIX. They aren't selling it. I don't think that's anything to do with it. BSD UNIX *is* a UNIX operating system (Berkeley Standard Derivation or somthing), so it is within its rights to call itself UNIX. Not really. It /is/ BSD, which is directly derived from ATT UNIX. However, UNIX is a trademark which an operating system must be branded with. It involves testing with The Open Group and paying a lot of money to them for the right to use the name. I tend to refer to *BSD, as well as commercial Unices, as GNU-like operating systems, while GNU/Linux and (of course) GNU/HURD are GNU operating systems. Noone owns a trademark on the term GNU, and anyone but RMS and the FSF would have a rather difficult time trying to get one, so it should be quite all right to call such things GNU-like operating systems, rather than labelling them UNIX-like. -- Jakob 'sparky' Kaivo - [EMAIL PROTECTED] - http://www.ndn.net/ As time goes on, my signature gets shorter and shorter... - me -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Recovering the original boot sector
Hi, When I purchased my PC, it came with Win95 preloaded. After installing linux on it, lilo did back up the original bootsector and stored it in /boot, but over the past two years, and after a couple of Linux reinstalls, I lost the original bootsector (blush). Now, I am planning to reorganize my Win95 partition again, and I need to recover the original bootsector. The CompUSA support people that I talked to said that there is no way I could recover just the bootsector - they could recover to the original configuration shipped. But this is useless to me. They did say that 'fdisk /MBR' would recreate the original bootsector. So here is what I propose to do : 1. Take backups of th existing bootsector. Create Linux bootdisk. 2. Run 'fdisk /MBR' from within Win95. This should wipe out my existing partition table in the MBR (first 512 bytes). 3. Reboot using my Linux boot disk. Now create a copy of the old (backed up) bootsector (file A). 4. do 'dd if=/dev/hda of=/boot/B bs=446 count=1' 5. do 'dd if=/boot/B of=/boot/A bs=446 count=1' This should give me the original bootsector with the current partition tables. Am I right? Or am I shooting myself in the foot? Better ideas are welcome. Jor-el
HAVE_WAITPID
Hi, The source code for fvwm95 has the following code in fvwm/misc.h : #ifdef HAVE_WAITPID #define ReapChildren() while ((waitpid(-1, NULL, WNOHANG)) 0); #else #define ReapChildren() while ((wait3(NULL, WNOHANG, NULL)) 0); #endif Now linux (atleast, for 2.0.36) does have the waitpid function. However, because the HAVE_WAITPID macro is not defined in the standard includes, the ReapChildren macro is defined as the call to wait3 (which is also defined in Linux). What is the status of the HAVE_WAITPID macro? Is it the standard way to recognize whether the waitpid call is defined for the OS, or is this some arbitrary test that the fvwm95 author came up with? If it is the former, shouldnt it be added to features.h or something like that? Thanks in advance, Jor-el
Re: Direct serial connection
On Sun, 25 Apr 1999, Ralf G. R. Bergs wrote: On Sun, 25 Apr 1999 11:39:53 -0500 (CDT), Jor-el wrote: Is this possible? Yes. See /usr/src/linux/Documentation/serial-console.txt. I've just done it (after having some trouble due to an incorrect /dev/console device). You need a 2.2.x kernel for this to work or a patched 2.0.x one. Basically you use a kernel command-line parameter of console=ttyS0 to make a serial terminal (box A in your setup) hooked to /dev/ttyS0 on machine B the display for machine B. You also need a getty running on /dev/ttyS0. Ralf, Do you know where to get hold of the patch for the 2.0.x kernels? Also, the kernel command line parameter that you described - is it for box A or box B? I need the monitor to be a console for Box A as well as Box B (preferably on different ttys). Jor-el
Re: Direct serial connection
On Sun, 25 Apr 1999, Robert Vollmert wrote: Hi, On Sun, Apr 25, 1999 at 11:39:53AM -0500, Jor-el wrote: The what I would like to do is to use the monitor that I have as a console for both machines A and B. Unfortunately, I cant accomplish this via a CPU switch, thanks to the non-standard monitor cable (Aptiva S90) that I have. I was thinking that maybe I could hook up B as using a serial console, and connect the serial port to the serial port of A, and then somehow get the stuff displayed on a separate window / tty on A. I'd suggest creating a small network using PPP over your serial ports (you would need a null-modem cable for this). You can then use telnet on A to work on B, and can even run X apps on B. There is a section on this in the PPP HOWTO (packages doc-linux-text or doc-linux-html). Robert, Thanks for the answer, but you misunderstood my requirement entirely. The two machines are connected via ethernet already, and telnet works perfectly. But this doesnt make the monitor of machine A act as a _console_ for machine B. For instance, how would you monitor the bootup messages of machine B using the setup you described? I think the answer that Ralf Bergs has provided is more suitable. Jor-el
Direct serial connection
Hi, I have the following setup : --- | | | | |A| | B| | | | | --- | | --- | | | Monitor | | | --- The what I would like to do is to use the monitor that I have as a console for both machines A and B. Unfortunately, I cant accomplish this via a CPU switch, thanks to the non-standard monitor cable (Aptiva S90) that I have. I was thinking that maybe I could hook up B as using a serial console, and connect the serial port to the serial port of A, and then somehow get the stuff displayed on a separate window / tty on A. Is this possible? Does anyone have any clues on how to accomplish this? Ideas / suggestions / alternate ideas are welcome. Thanks, Jor-el
kernels 2.0.x, 2.2.x and networking
Hi, I just shot myself in the foot, when trying to reach for the stars... I was trying out the 2.2.5 kernel on my Slink install. Before I saw the page on running 2.2 by Anthony Towns (did I get the name right?), I had noticed that when bring up my 2.2 kernel, my route using lo was not coming up. And I couldnt bring it up manually either. The messages that I was getting indicated some kind of syntax error - which was strange, since I was booting off of the same rootdisk used by my 2.0.36 Slink install. Anyway, when I discovered Anthony's page, I went and downloaded the new netbase and installed it. This of course required the new libc6 (otherwise it wouldnt configure), and which required apt version 0.10 (and which I couldnt find - what I found was apt version 0.3x [x = something or the other]). Anyway, everything is installed, and now my 2.2 kernel still doesnt produce a route for lo, but worse (oh horrors!) my 2.0.36 kernel now produces the following 'ifconfig' output : eth0 Link encap:Ethernet HWaddr 00:20:78:15:E3:9E unix addr: Bcast: Mask: UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:2 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Interrupt:10 Base address:0x1200 loLink encap:Local Loopback unspec addr:00-00-00-00-00-00-00-00-00-00-00-3B-01-40-02-00 Bcast:00-00-7F-FF-FF-FF-00-00-00-00-00-3B-01-40-02-00 Mask:00-00-FF-00-00-00-00-00-00-00-00-3B-01-40-00-00 UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 This is what a working 'ifconfig' setup looks like (on my Redhat system - same machine) : loLink encap:Local Loopback inet addr:127.0.0.1 Bcast:127.255.255.255 Mask:255.0.0.0 UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 TX packets:0 errors:0 dropped:0 overruns:0 eth0 Link encap:10Mbps Ethernet HWaddr 00:20:78:15:E3:9E inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:5 errors:0 dropped:0 overruns:0 TX packets:5 errors:0 dropped:0 overruns:0 Interrupt:10 Base address:0x1200 Save me! Jor-el
Re: boot failed due to superblock error and FAT BREAD Failure
Li, I think the message is saying that you have partitioned /dev/hda so that /dev/hda7 ends on a block that lies beyond the end of the hard disk. Things should probably work out correctly if you repartition your hard disk to fix this error. Regards, Jor-el On Mon, 5 Apr 1999, J. Lee wrote: Hi, I browsed the recent posts and didn't quite find an answer for my problem. Can anyone help me out? Thanks. My hardrive is partitioned into four parts, namely /dev/hda1 MSDOS /dev/hda2 NTFS /dev/hda3 linux swap /dev/hda4 linux I can boot from the floppy disk but has problem in booting from hard disk. Error message is: Partition check: hda: hda1 hda2 hda5 hda3 hda4 attempt to access beyond end of device 03:07: rw=0, want=2, limit=0 ext2-fs: unable to read superblock attempt to access beyond end of device 03:07: rw=0, want=2, limit=0 minix-fs: unable to read superblock attempt to access beyond end of device 03:07: rw=0, want=1, limit=0 FAT bread failed attempt to access beyond end of device 03:07: rw=0, want=33, limit=0 isofs_read_super: bread failed, dev 03:07 iso_blknum 16 block 32 kernel panic: VFS: unable to mount root fs on 03:07 It is appreciate if anyone can give me some suggestions on fix it. Li _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Compiling fvwm95 on Slink
Hi, I am having problems compiling fvwm95 on Slink. I installed the required development libraries from the oldlibs packages. Here is an example of the problem : debian$gcc -o FvwmAudio FvwmAudio.o -L../../libs -L/usr/i486-linuxlibc1/lib -L/usr/i486-linuxchecker/lib -lfvwm95 -lXpm -lXext -lX11 /usr/lib/crt1.o(.text+0xe): undefined reference to `__libc_init_first' /usr/lib/crt1.o(.text+0x18): undefined reference to `_environ' Quite obviously, the above error is due to the crt1.o object being picked up from /usr/lib, rather than /usr/i486-linuxlibc1/lib . However, when I try to fix this problem by using the -nostdlib flag, I get : debian$gcc -o FvwmAudio FvwmAudio.o -L../../libs -L/usr/i486-linuxlibc1/lib -L/usr/i486-linuxchecker/lib -lfvwm95 -lXpm -lXext -lX11 -nostdlib ld: warning: cannot find entry symbol _start; defaulting to 08048630 How can I get round this problem? Jor-el
Sending mail on a dialup connection
Hi, When one has a remote connection, one can use fetchmail to download mail in a batch. I am looking for the opposite - sending composed mail from a designated mail folder as soon as the dialip link is up. So far, none of the MUA's I looked at seem to have this capabilty. Of course, I maybe looking in the wrong place - maybe MTA's are what I should be looking at? Jor-el
Re: Strange fetchmail behaviour
Mark, I dont have a solution for you, but another symptom. Sometimes when my link goes down in the middle of a download, I get messages that are tagged as seen on the server, but are not flushed. In order to get this mail on subsequent downloads, I would have to do 'fetchmail -a'. During the period when I had the Palm Pilot packages (from source - not Debian) installed on my system, fetchmail would give me a core dump every time I tried to download such mail. When I finally removed the Palm Pilot libraries, the problem went away. So... maybe the problem is with some exotic libraries present on your friends system? Regards, Jor-el On Mon, 29 Mar 1999, Mark Phillips wrote: A friend of mine has a laptop, which I installed debian hamm on for him. He says that every time someone sends him a microsoft word document via email (these are huge files), fetchmail seems to choke on it. The strange thing is that I cannot replicate the same problem on my computer --- fetchmail takes a while to download, but it doesn't choke. I can't work out the problem so I was hoping someone here could help --- or give me some ideas. Here's an example of the problem output: genoa:/home/bill# fetchmail fetchmail: 2 messages for bill at adam.ist.flinders.edu.au. fetchmail: reading message 1 of 2 (1367 bytes) . flushed fetchmail: reading message 2 of 2 (281432 bytes) ... ... ... .. (log message incomplete) fetchmail: terminated with signal 13 You have new mail in /var/spool/mail/bill Any ideas? Mark. _/\___/~~\ /~~\_/~~\__/~~\__Mark_Phillips /~~\_/[EMAIL PROTECTED] /~~\HE___/~~\__/~~\APTAIN_ /~~\__/~~\ __ They told me I was gullible ... and I believed them! -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null