Re: ssh/scp and the change of remote shell
also sprach Bostjan Muller (on Thu, 13 Sep 2001 02:47:20PM +0200): > 1. I cannot ssh/scp to a remote machine, if the user has an shell > that does not allow interactive login (like if he/she has /bin/false > for the shell), that is ok for ssh, but scp should still work right? > sftp does not work either. Is there a way to make it work, I am > most interested in making scp work? i doubt this will work since scp /etc/file host:/etc/otherfile is really just ssh host 'cat > /etc/otherfile' < /etc/file - so it is a no-interactive login remotely, which is not granted with /bin/false. sftp also uses a shell - interactively actually. > 2. If the user has something echoed in his/her login scripts like in > ~/.bashrc a line echo "lalala", he/she is able to login via ssh, > but is not able to do scp. The line lalala is displayed, but than > the program stops and the transfer is not made. Is there a way > around this? by the above reasoning, lalala is interpreted as data from the host that's part of the scp operation. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
Re: shell script kommands
also sprach Timeboy (on Thu, 13 Sep 2001 01:41:08PM +0200): > Is there a possibility to do this with a bash script? #!/bin/bash cnt=0 while read i; do var[cnt]=$i cnt=`expr $cnt + 1` done echo "there are $i object(s) in array var[]." martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] >
Re: External modem line drops with activity
which kernel are you using, and does it have VJ compression enabled? the option should be around the configuration item for PPP. martin
Re: tripwire reports constant changes of /usr/doc and /home
also sprach Roberto Diaz (on Sat, 18 Aug 2001 07:11:26PM +0200): > > at first, it was only /home, and i thought it had anything to do with > > the quotae i enabled on that partition, but /usr/doc is (a) not a > > partition of its own, and (b) /usr doesn't have quotae. > > /home could be explained if you have something (like netscape which let > things there, or procmail which updates a log) i disabled all /home/* entries, only kept the directory /home. its inode gets updated whenever a directory inside is renamed, added, or deleted, and none of that happens... > /usr/doc is more hard to explain :?? word. martin
tripwire reports constant changes of /usr/doc and /home
folks, this one leaves me puzzled, so i was wondering if you could help me out. every day, tripwire reports to me that on one of my systems (only), the directories /home and /usr/doc have changed: changed: drwxr-xr-x root 8192 Aug 6 16:18:56 2001 /usr/doc changed: drwxr-xr-x root 4096 Jul 27 12:42:46 2001 /home none of the entries within these directories have changed, nor are they reported, but these two directories still continue to fill up my tripwire reports. at first, it was only /home, and i thought it had anything to do with the quotae i enabled on that partition, but /usr/doc is (a) not a partition of its own, and (b) /usr doesn't have quotae. this is weird, since it happens only on one of my otherwise identitical systems, and i need to figure out why. where do i start? thanks, martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
Re: [users] which NIC driver to use?
also sprach John Griffiths (on Fri, 29 Jun 2001 05:49:28PM +): > I've got an IBM xseries 200 i'm trying to get deb running on > the onvoard NIC is confounding me > does anyone know which driver module should i be using? i don't know, but it will most likely be one of the following: tulip, rtl8139, 8139too, or ne. just try modprobe each one in turn, and then if that does not work, try all in /lib/modules/.../net/* - your machine might lock up, but only the right driver will successfully install as a module... it's a hack, but it works. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "my father, a good man, told me: 'never lose your ignorance; you cannot replace it.'" -- erich maria remarque
Re: [users] dialup query
also sprach Stig Brautaset (on Fri, 29 Jun 2001 02:17:19AM +0100): > > install ppp and pppconfig, taking care to pull ppp off > > people.debian.org/~bunk/debian if you have a 2.4.x kernel. > > Why? I run ppp on a pcmcia-modem with the standard package > (unstable). (It is slow to connect, but runs fine after it is > connected). okay, well then good for you. i spent a week debugging a system that wouldn't connect and only compiling pppd for 2.4.5 solved the problem. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- give me ambiguity or give me something else.
ssh: Command terminated on signal 13
hi guys, i have one machine - "seamus" - an AMD K6-2 500/160 Mb which is acting up. sometimes when i ssh into it (like 1/5 times), it gives me the following: fishbowl:~> ssh seamus [EMAIL PROTECTED]'s password: Last login: Fri Jun 29 01:47:06 2001 from fishbowl.madduck.net on pts/1 Linux seamus 2.4.5 #1 Thu Jun 28 15:59:01 CEST 2001 i586 unknown Received disconnect from 192.168.14.3: Command terminated on signal 13. when i then try again, it usually works. any idea what this could be? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "know what I hate most? rhetorical questions." -- henry n. camp
Re: [users] custom spam file (ala rbl)
also sprach Peter Jay Salzman (on Thu, 28 Jun 2001 04:18:40PM -0700): > kind of like using rbl, except i'd have my own custom reject file. > > i can simulate such a file usign ipchains, but i'd like to know if exim has > an IP reject file. the way i have done it is via my bind installation, just like RBL. on another setup, i use procmail. postfix allows for IP addresses to be banned, but that requires root access every time, so it's kind of a pain - which is why i went with the bind installation. hey, but we could all get together and run our own RBL, could we not? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- scintillation is not always identification for an auric substance.
big IBM harddrive (41Gb)
adding to the recent discussion on the 40Gb drive with IBM - i did what someone suggested, to not tell the BIOS anything about the drive, but to let linux take care of it. i run 2.4.5, and i know have 41Gb available :-> (without the 32Gb clip jumper). martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- breakfast.com halted - cereal port not responding.
Re: [users] hi masters of linux, surely you know some tricks...
also sprach thomas anderson (on Thu, 28 Jun 2001 10:59:03PM +0200): > I want to try to put a perl script in the /usr/lib/perl directory however I > don't have permission access...I tried symlinking it but it still won't work. > is there I way to do this without becoming root or sudo? no. user scripts don't go there. why would you want to place it there anyway? what function does the script have? is it just for yourself, then put it into $HOME/bin. if it's for a group of users, put it there too and give appropriate permissions. if it's for all, then it goes into /usr/local/bin. you have to be root for the last option. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- no keyboard present. press f1 to continue. zen engineering.
Re: [users] dialup query
also sprach [EMAIL PROTECTED] (on Thu, 28 Jun 2001 10:30:54AM -0500): > Therefore, I need to reconfigure my system for a regular old dialup > connection. I stupidly neglected to do this while I still had DSL, and now > am wondering what packages I will need to enable this. At worst I can put > them on floppy at work and install that way. > > ... and I am blanking on the name of the command I should use to configure > the system for this type of connectivity. I'll keep checking debian.org > and lists, but any pointers will be appreciated. install ppp and pppconfig, taking care to pull ppp off people.debian.org/~bunk/debian if you have a 2.4.x kernel. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "god is a comedian playing to an audience too afraid to laugh." -- voltaire
Re: [users] Re: apt-get : difference between dist-upgrade and dselect-upgrade
also sprach Joost Kooij (on Thu, 28 Jun 2001 01:37:29PM +0200):
> When I sat down with it and took a minute or two to get acquiainted,
> it turned out to be quite a likeable and most of all very helpful
> critter, in fact.
try apt-console, i like it much better since i can't stand the awkward
handling of dependencies in dselect.
> Did you know that there is even a help screen that explains the keys?
yes, it always pops up when least needed...
btw: this entire thread never answered the initial question, which i
also have. what's the difference between apt-get {dselect-,}update and
apt-get dist-upgrade???
martin; (greetings from the heart of the sun.)
\ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
--
"syntactic sugar causes cancer of the semicolon."
-- epigrams in programming
Re: [users] big IBM harddisk
also sprach nico de haer (on Wed, 27 Jun 2001 11:06:13PM +0200): > I've tried this '32 Gb clip' and it does allow me to boot using my 40 Gb, > but i've been unable to get access to the last 8 Gb. I've been experimenting > with the 'hdx=c,h,s' parameter but all i can get are screens filled with all > kinds of hdd related errors rangeing from 'bad superblock' to 'seek error' > and loads of others. Have you found a way to access the full drive capacity? no, i have said to myself that 32Gb is enough anyway. i am not a gamer, nor an mp3 person, and i won't fill 32Gb anyway, especially since this is like my fourth system. but hey, i have nothing on that drive right now, so i'll spend some time tonight to figure it out, okay? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- windoze nt crashed. i am the blue screen of death. no one hears your screams.
Re: [users] Re: Using different DNS per ISP
also sprach John Hasler (on Tue, 26 Jun 2001 09:52:49AM -0500):
> > ...or you should write something for /etc/ppp/ip-{up,down}.d.
>
> Pppconfig already installs a something there. Run pppconfig and look at
> Advanced->Nameservers.
i know. but in his case, pppconfig doesn't seem to work.
martin; (greetings from the heart of the sun.)
\ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
--
"it appears that pl/i (and its dialects) is, or will be, the most widely
used higher level language for systems programming."
-- j. sammet
Re: [users] Re: Using different DNS per ISP
also sprach Greg Rowe (on Tue, 26 Jun 2001 09:52:06AM -0400):
> Would installing bind on your system work for you? You could add all of
> your ISPs DNS servers as forwarders in named.conf and then simply point
> resolve.conf to 127.0.0.1.
0.0.0.0
it's just a little undocumented trick to make local DNS requests not
go via the expensive route of the IP stack, but rather through a local
socket.
pppconfig should really be able to do DNS stuff itself, or you should
write something for /etc/ppp/ip-{up,down}.d. A BIND installation is
cumbersome, eats resources, and is absolutely unecessary on a
workstation. nscd if all fails, but not bind.
martin; (greetings from the heart of the sun.)
\ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
--
this message represents the official view of the voices in my head.
Re: [users] big IBM harddisk
also sprach virtanen (on Mon, 25 Jun 2001 04:28:04PM +0300): > I purchased a new 40 G IBM harddisk. try setting the jumper called "32Gb clip" even the LBA enabled BIOSs still can't handle >32Gb, so you'll lose 8Gb, but in my case, the BIOS did detect the drive. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- it's as bad as you think, and they are out to get you.
Re: [users] Re: Unidentified subject!
also sprach thomas (on Mon, 25 Jun 2001 01:44:01AM +0200): > make a new kernel with VJ compression compiled in and/or enable VJ in > your ipppd.ippp0. i am pretty sure this is a configuration problem. thanks for the replies! i ended up figuring this out... i was a little confused that one had to enable VJ compression to allow for the protocol called "VJ Uncompressed TCP/IP". but now everything works and i was able to meet the deadline. you wouldn't think that a kernel-2.4.5 iptables-connection-based-firewall postfix bind9 dhcpd arpwatch snort tripwire vbox identd ssh ntp setup runs on a 486/33 with 8Mb of RAM and a processor load of averagely 0.10 :) martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- 1-800-psych hello, welcome to the psychiatric hotline. if you are depressed, it doesn't matter which number you press. no one will answer.
Re: [users] Re: An *idea* that *might* put Debian on top (?)
i just wanted to add my two pfennige. see, the reason that debian rocks is that it does not impose configuration tools on you like suse and redhat. it does provide a very cool package system, but everything is under the control of the user and vi. that's the beauty. now looking at how i frequently already disagree with the formatting of config files contained in the packages, or the default settings for an application, i couldn't possibly imagine running such a distributed system. don't get me wrong, i am all up for slamming Micro$oft's "how to add .Net to your old software to keep it cutting edge" braindeadness, but the idea of having other users influence my system is disturbing. granted, the debian-cracks wouldn't use it, at least not for configuration, but that means that such a database contains configs by novices, which, unless very carefully monitored by experts, could result in an inferno. i am just thinking... instead, pre-installed debian is a good idea, because once running, apt-get is too easy to use. in addition, have workshops! i, for one, am about to organize a monthly linux meet at a local computer cluster, encouraging people in papers and by word of mouth to come and play around. it'll be debian, of course, and i'll have a couple of cool dudes along to help me out, but i think something like this to get people excited, and then the offer to pre-install debian for them (even though that's the best exercise) for something low like $20 maybe (or not - depending on how i can swing it), would be the best, don't you think? i will make some computers available at the cluster for debian test-installations - for people to come and try it themselves. it's still in preparation, but it'll go live in july or august... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- as i was going up the stair i met a man who wasn't there. he wasn't there again today. i wish, i wish he'd stay away. --hughes mearns
Re: [users] Re: An *idea* that *might* put Debian on top (?)
also sprach Joost Kooij (on Sun, 24 Jun 2001 09:42:19PM +0200): > Debian/GNU Bob > > "Now you can put the stark fist of removal at work for YOU!" hehe, and when you mistype your password three times, it'll prompt you to change it. believe it or not, Micro$oft Bob did that. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- fermentation fault. coors dumped.
Unidentified subject!
folks, just now i figured out that the pppd debug option is also present in ipppd, and i turned it on to see the following in the logs (after a bunch of packet dumps): Jun 25 01:13:31 embryo ipppd[3918]: Unsupported protocol 'Van Jacobson Uncompressed TCP/IP' (0x2f) received well, that would explain why i get kicked out, but i am still clueless how i can change this. i'll see what protocol my other routers use, but if you have anything on this, please tell me. this machine has to be up by tomorrow :( martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "non sposarti. trova una donna che odi e comprale una casa."
ipppd problems
hi, i am in the process of setting up an ISDN router for a network of three computers, using a 486 and running kernel 2.4.5 (because of the iptables functionalities). the system has one of the very old AVM A1 (Fritz!Classic) cards, which I can load and talk to with the hisax.o driver. it's at 0x200, but i don't know the IRQ, but I remember it being configurable through software, and the docs say that any free IRQ can be used. i have used adrian bunk's packages for the 2.4.x kernels, and i have successfully configured everything for a dial-up to the German provider T-Online. I am experiencing a problem, which may be IRQ-caused, but it's weird and I don't think that IRQs are the cause. I can establish a connection with 'isdnctrl dial ippp0,' the system creates the PPP link, configures the routing table, and a second later, I can ping to the outside. it also looks as if UDP traffic works. However, whenever I initiate a TCP connection, the ISDN link goes down. the logs report the following: Jun 25 00:51:42 embryo ipppd[3516]: IPCP terminated by peer Jun 25 00:51:43 embryo ipppd[3516]: LCP terminated by peer Jun 25 00:51:45 embryo ipppd[3516]: Connection terminated. Jun 25 00:51:45 embryo ipppd[3516]: taking down PHASE_DEAD link 0, ... Jun 25 00:51:45 embryo ipppd[3516]: closing fd 8 from unit 0 Jun 25 00:51:45 embryo ipppd[3516]: link 0 closed , linkunit: 0 Jun 25 00:51:45 embryo ipppd[3516]: reinit_unit: 0 Jun 25 00:51:45 embryo ipppd[3516]: Connect[0]: /dev/ippp0, fd: 8 Jun 25 00:51:45 embryo kernel: isdn_net: local hangup ippp0 Jun 25 00:51:45 embryo kernel: ippp0: Chargesum is 0 Jun 25 00:51:45 embryo kernel: ippp_ccp: freeing reset data ... this looks to me as if T-Online cuts the connection somehow, which is weird. I can dial up just fine to the same account with two other ISDN-enable computers... have you experienced something similar? i couldn't find anything on the AVM site, the HOWTO, isdn4linux, or a general google/linux search. any help or pointers appreciated. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "finally, we have a president who knows how to run the government like a business. presuming the business is pets.com." -- gail collins
Re: [users] /proc/pci incorrect
also sprach Kevin C. Smith (on Sun, 24 Jun 2001 12:39:40AM -0500): > Changed the NIC months ago, but noted that the /proc/pci continues to > show the old NIC. Why would this be? How do I correct it? what does it show? many NICs work atop one of the following chipsets, so two different NICs may well use the same driver and/or signature: ne smc dec 2904x (tulip) intel ether express sis900 and others. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- with searching comes loss and the presence of absence: file not found.
Re: [users] Re: kernel compiles
also sprach Robert Waldner (on Fri, 22 Jun 2001 08:44:51PM +0200): > now do a `make clean` (just to get rid of the modules, mostly) and > re-do steps 1-3 for the second machine and the third and... > > when you want to re-do for the first machine, just move the appropriate > .config back, and re-do steps 2-3 (and 1 if you want to change some > settings). > > so, all you really have to keep seperated is the .config-file. (and the > generated .deb´s/kernels, of course ;-) ) you misunderstood me. i know all this. but this method circumvenes the positive effects of Makefiles. nevermind. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- al telefono: "pronto? cantina sociale?". "hic!".
Re: [users] Re: kernel compiles
also sprach Robert Waldner (on Fri, 22 Jun 2001 08:20:41PM +0200): > you don´t need to maintain seperate trees of the whole source. just > `make (menu|x|)config`, then backup the .config-file. that´s where the > information you entered/chose is kept. i understand... but when i change the .config file severely, don't i have to do a make clean? or can i change the config file as much as i want to and do make dep and it'll just compile what's necessary? still, the argument holds: if i have two machines, compile the kernel for the first, then use the same tree to compile a kernel for the second, adding a module or a feature to the first kernel requires more recompilation than if i hadn't used the tree for the compilation of the second... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "it usually takes more than three weeks to prepare a good impromptu speech. -- mark twain
Re: [users] Re: kernel compiles
also sprach Bob Nielsen (on Fri, 22 Jun 2001 10:42:23AM -0700): > True, but you can/should configure and compile separately for each > destination architecture, optimizing appropriately. so yes, the argument that my athlon (thunderbird actually) will outperform the others still holds. but i'd need to keep separate source trees if i didn't want to make clean every single time - which i usually don't since i experiment a whole lot and therefore change the kernel - which doesn't take long thanks to make unless i make cleaned it... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "a mathematician is a device for turning coffee into theorems." -- p. erdos
Re: [users] Re: kernel compiles
also sprach Nathan E Norman (on Fri, 22 Jun 2001 12:42:37PM -0500): > I fail to see why you think compiling a kernel on an Athlon, but > optimising for a 486 cos you're installing on a 486 is a problem. that's what i am doing btw. and sorry, i wasn't possibly thinking about multiple .debs, just that you'd make one single .deb for all machines. my bad, clearly. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- only through hard work and perseverance can one truly suffer.
Re: [users] Re: kernel compiles
also sprach Sean Morgan (on Fri, 22 Jun 2001 12:05:19PM -0400): > zless /usr/share/doc/kernel-package/Rationale exaclty what i wanted. thanks! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- your fly might be open (but don't check it just now).
Re: [users] Re: kernel compiles
also sprach Nathan E Norman (on Fri, 22 Jun 2001 11:01:13AM -0500): > Well, for one thing, you can compile kernels on your 1GHz Athlon > instead of your old 486 :) Since kernel-package creates a package, it > can be installed anywhere. but i usually choose the appropriate kernel architecture during the menuconfig session, and while the 486 will run on the Athlon, it's not vice versa. and am i not correct in believeing that it's better e.g. for a pentium to run a P5 kernel rather than a 386 kernel... performance wise? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- always remember you're unique, just like everyone else.
kernel compiles
i love debian. seriously, apt is a work of genius and the entire system is exactly how i want it - unlike SuSE or RedHat. since i do a fair bit of developing and since i always want to have at least one machine that's cutting edge, i do a whole lot of kernel compiles. in the past, i have always used .debs unless a software was too old or not available, in which case i beat the tarball around and installed into /usr/local. by now, i do it the "debian way," and use dpkg-buildpackage to create the .deb, which i then install. i haven't done so on kernels yet, even though i know about make-kpkg anyway, my question is: while i am currently running a system that's .deb only, the kernel is still compiled and installed the standard way, me taking care of /boot and /etc/lilo.conf. what advantages are in make-kpkg'ing as opposed to the regular way? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "for art to exist, for any sort of aesthetic activity or perception to exist, a certain physiological precondition is indispensable: intoxication." -- friedrich nietzsche
Re: [users] Re: e-mail, the next level
also sprach [EMAIL PROTECTED] (on Fri, 22 Jun 2001 09:40:36AM +0800): > 1. What if multiple users on my network have set up multiple POP3 accounts in > the past (before the server was up) and wish to go this way, leaving their > mail > OFF their POP3 accounts, but ON the home server? Can this still be doable? > Will > they be able to send mail which has their POP3 email in the From: ? of course. the From header is only important while the message is being delivered. in my scenarios, fetchmail keeps the From ASCII header, but that has no control over the message delivery anyway, since the From part is exchanged separately during the SMTP communication. and if you configured ETRN, it works just as well. in fact, i have a couple of users doing it that way (me being one). > 2. Would it be possible to implement some kind of IMAP account(s) on the > server, > so the users just go through an IMAP capable account on their client stations, > without having to download the mails from the home server to the client hard > disk? I envision that a daemon would be pulling POP3 mails from the various > accounts and putting into the respective IMAP accounts. yes, exactly what erik said. once the mail is on the home server, you can do whatever you want with it; forward it, POP3 it, IMAP it, print it, fax it, SMS it, and so on... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- micros~1: for when quality, reliability, and security just aren't that important!
Re: [users] Re: e-mail, the next level
also sprach nico de haer (on Fri, 22 Jun 2001 10:10:54AM +0200):
> Up until writing this message i've recieved two replies to my erlier posting
> "e-mail. the next level". Erik and Martin think in the same line as i do.
> Carl also contributed some points of interrest. Getting mail is no problem
> using a tool like Fetchmail. Even multiple accounts belonging to one user.
> Just let your mail sit in some (or multiple) internet mailbox(es) until you
> decide to get it. Using the IP-Up/Down scripts this can be done in a
> comfortable manner.
don't forget my ETRN method, which i like the best as it's the most
secure (no passwords), and it works beautifully - and you can't get
faster delivery if you decide not to read your mail on the server.
> For the actual reading part, i suppose that the POP3 server i suggested can
> be easely replaced by IMAP or any other you preffer. At this point, it's
> only used to read mail stored on your own server at your workstation.
... SSL tunneled, of course!
here, each user has two email addresses: [EMAIL PROTECTED] and
[EMAIL PROTECTED] the default is @madduck.net, but when sending
internal emails, that changes to @home.madduck.net (hooks - of course
windoze can't do that.) the important thing to realize is that if an
internal message is sent with @madduck.net, it still gets there (after
a delay), and so does the reply. if an external message is sent with
return addy @home.madduck.net, it is definitely delivered to the
external recipient, and if the recipient uses something other than
micro$oft to relay messages, it will even make it back into the house,
since home.madduck.net is a CNAME to a dynamically changing IP
address, as i explained previously.
martin; (greetings from the heart of the sun.)
\ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
--
void write_thesis(char *subject, char *title} {
// do something
}
Re: [users] e-mail, the next level
i have done this in two different ways. the first is straight forward, but depending on how many users you have, it might be a pain (and a security risk). the second works on the assumption that your ISP is running a sensible MTA (postfix, qmail) and that you will not be down for longer than 3 days. (1) the router which handles the dial-up connection to the internet is also configured as a mail server, relaying mail for the internal subnet only. each user has a mail account on that system, which you make POP3 accessible. in my example, i actually have a separate mail server, but whatever. the canonical names are mail.home.madduck.net and smtp.home.madduck.net. now you block POP3 and SMTP at the router for all IPs (MAC addresses) other than the mail server. you tell your users to use these two mail servers exclusively (they won't succeed otherwise anyway). my relay server is running postfix, which i can configure to defer_transport for smtp. in /etc/ppp/ip-up.d, i have an mta-online script, which (1) calls sendmail -q, and (2) changes the defer_transport setting with postconf. /etc/ppp/ip-down.d reset that setting when the connection is down. i also have a crontab, connecting the system every hour. that's the SMTP side, and now users will think that mail is sent immediately, even though it will sit in the relay queue until (a) the next hour, or (b) someone connects the internal network (i am using diald for on-demand connection). the POP3 side of things is similar. while online, fetchmail retrieves all mail for the individual users, and stores them in their local accounts (daemon mode). in my scenario, i am [EMAIL PROTECTED], and while i can send to [EMAIL PROTECTED] - which will leave the house via SMTP, and eventually return via fetchmail, i can also send mail to [EMAIL PROTECTED], which is delivered locally immediately. (2) the second method i like a whole lot better. postfix and qmail, unlike e.g. micro$oft exchange, do an IP lookup on the MX on *every* delivery attempt, not just once. therefore i configured my home router to update a remote Bind 9 server via TSEC nsupdate within a zone that has a TTL of 2 minutes. therefore, home.madduck.net points to the router's ippp0 interface whenever it is online. the ISP's postfix is configured to send mail e.g. for [EMAIL PROTECTED] to [EMAIL PROTECTED], and [EMAIL PROTECTED] to [EMAIL PROTECTED] whenever the router is online, mail is delivered immediately. if the router is offline, mail sits in the ISP's queue for at most five days before it is bounced. i have at least 1 connect a day due to the debian_mirror rsync script i run, and other similar stuff, and when the router is online, it contacts the ISP with an ETRN SMTP command (fetchmail can do that), causing all queued mail to be delivered immediately. i like this approach a lot better, and if you wish, i can provide you with a TSEC nsupdatable hostname. outgoing SMTP is handled just like the above. i told my users that mail goes out every hour, but if something is urgent, all they need to do is surf the web afterwards for the mail to be sent. everything nowadays seems to be dynamic, so even the extensive squid proxy i use won't prevent a connection whenever someone opens IE (barf!). i hope this was complete. i'll be happy to answer any questions... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- i'd give my right arm to be ambidextrous.
Re: more boot floppies questions
also sprach Osamu Aoki (on Wed, 20 Jun 2001 03:53:49PM -0700): > Are you potato or woody? potato. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "there are more things in heaven and earth, horatio, than are dreamt of in your philosophy." -- hamlet
Re: [users] Re: more boot floppies questions
also sprach Hoeteck Wee (on Wed, 20 Jun 2001 06:03:30PM -0400): > Here are the network related modules that are provided with driver-1.bin > in the compact images. The 3c509 might work with your 3c59x card, but I'm > not too sure about that. i have looked at, and unpacked the drivers.tgz archive, which contains all the drivers i need (it's a 3c505, not a 3c59x, sorry). but i am still left with four driver disks that neither the installation program nor i know what to do with. for now i am helping myself with a separte disk containing all the network modules, along with /sbin/insmod and tty2, and the systems are now being installed, but i'd really like to know what the deal is. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'
more boot floppies questions
i swear, i read all the readmes and did the websearches, but i don't arrive. i am still baffled as to what these driver_[1-4].bin images are supposed to be used for - or more likely, how they are supposed to be used. but right now, i am wondering why each of the falvors, vanilla, compact, idepci, udma66, and safe, come with their own driver-1.bin image, but only vanilla has driver-[2-4].bin as 1.44 images available. what gives??? and maybe someone could tell me exactly when the disks created with the driver-?.bin files are used, and how i used them. yes, i am trying to do floppy-only installs until i can mount a local nfs volume with the debian mirror. and yes, i need each of the realtek, tulip, 3c59x, sis900, and ne modules, which are supposedly on the driver-? disks. help! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "driving with a destination is like having sex to have children" -- backwater wayne miller
Re: [users] Re: oldtimer pc
also sprach Nathan E Norman (on Wed, 20 Jun 2001 03:38:47PM -0500): > Either kernel should be fine, though you should really get more ram. > A router in 8 MB ican get ugly :) I've run 16 MB in my router and was > happier with 32 (my router is a 486/25 old timer with 2 3c509b :) I'm > running 2.2.19 ATM since netfilter had some issues up to 2.4 3 IIRC. the machine is using SIMM's. eight slots, eight SIMMs = 8Mb :) i doubt that i could even find any more of these antique pieces. but yes, i have had that concern before. a lot of swap :) martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- i'd give my right arm to be ambidextrous.
Re: boot disk dilemma
also sprach Osamu Aoki (on Wed, 20 Jun 2001 11:38:10AM -0700): > dd driver disks (3 of them for compact) to Floppy as originally designed > like boot/root disks. Good luck :-) well, i understood that, and i have my disks. but in the installationprogram, there are two options: - preload modules (for use in initrd) takes modules in the root directory of the floppy i.e. floppy needs to be mounted - configure device drivers tries to mount the floppy then copies all modules to /target/lib/modules/... but the driver-1.bin file can't be loop mounted nor can the disk be mounted. if i strip the first 512 bytes, then the remainder is part of a gzip file, but the installation program doesn't do anything with it. sure, i can take a floppy with the modules on it, but there has to be a better way that's debian native... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- women who want to be equal to men lack imagination.
boot disk dilemma
so i finally managed to boot off the floppy disks and am now in the debian installation menu, just short of installing the operating system. i have a local debian mirror, so i'd like to install via network (ftp) -- especially because i don't have the cds. i am using the "compact" kernel. i am dealing with three machines, and they all have different network cards. one is a ne2000, one a 3c59x, the next a tulip chip. i would like to install the network drivers, but they aren't on any of the disks. my headaches are with the driver-1.bin disk, which resides in the compact directory -- how do i use it? it looks like a 512byte offset gzipped disk image, but the debian installation program can't load it - it waits for like 5 minutes before reporting that it "cannot mount the floppy. stop." what's the native method to configure the network cards during a floppy install? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- 1-800-psych hello, welcome to the psychiatric hotline. if you are depressed, it doesn't matter which number you press. no one will answer.
oldtimer pc
so i pulled this old 486-33 machine out of the basement, it's got 8Mb RAM, a shitty graphics card, and 240Mb of HDD space. it's ISA only, but i want to try using an AVM FritzCard and a cheap NE2000 compatible to make it be a masquerading router. it's probably going to fail, but i want to try anyway. question is: 2.2.x kernels or 2.4.x? i'd tend for the latter, but i don't know is that's harsher on the hardware and needs more space, or not... any hints, tips, thoughts welcome! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- only through hard work and perseverance can one truly suffer.
Re: pppd troubles
also sprach Kevin Ross (on Tue, 19 Jun 2001 04:50:14PM -0700): > One more thing. There's a 2.4.0 pppd compiled for potato w/kernel > 2.4.x at: > > http://www.fs.tum.de/~bunk/kernel-24.html kevin, you rock!!! fixed it... and thanks to adrian as well!!! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- only through hard work and perseverance can one truly suffer.
Re: [users] Re: pppd troubles
also sprach Kevin Ross (on Tue, 19 Jun 2001 04:46:06PM -0700): > I browsed through the source code for pppd, and it looks like the > only way that error can be generated is if you're using kernel > 2.4.x. So the fact that you're using pppd version 2.3.11 tells me > you're using potato, but with a 2.4 kernel. wow. good job. quite accurate 2.4.4 on potato. > Two things immediately come to mind. You can try rebooting with a > 2.2.19 kernel, and see if you still get the error. Or you can > upgrade your pppd to the one from testing/unstable. good point. i am having her get adrian bunk's package since i don't have a potato system handy, and her linux isn't online... windoze download -> floppy disk. let's hope this works! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- 1-800-psych hello, welcome to the psychiatric hotline. if you have multiple personalities, please press 3, 4, 5 and 6.
Re: pppd troubles
also sprach Jeremy (on Tue, 19 Jun 2001 07:42:32PM -0500): > My knowledge on this is limited, but have you made sure that you > have the domain and nameserver entries in the /etc/resolv.conf > file? I got this same sort of message when I was trying to dial > in with wvdial and I didn't have those things in my resolv.conf. we don't even get that far. point being, pppd starts and immediately dies, even though i can connect to the provider from another system that's overseas!!! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- today's an excellent day for putting slinkies on an escalator.
pppd troubles
hey all, we are still failing to establish a connection with wvdial and pppd (on a potato system). wvdial looks all good, until we get to the login: --> Carrier detected. Waiting for prompt. login: --> Looks like a login prompt. --> Sending: heikkin heikkin Password: --> Looks like a password prompt. --> Sending: (password) Entering PPP mode. Async interface address is unnumbered (Ethernet0) Your IP address is 216.114.40.29. MTU is 1500 bytes --> Looks like a welcome message. --> Starting pppd at Tue Jun 19 16:56:35 2001 --> PPP daemon has died! (exit code = 1) --> Disconnecting at Tue Jun 19 16:56:35 2001 plog reports the following: Jun 19 16:56:35 orange pppd[1408]: pppd 2.3.11 started by root, uid 0 Jun 19 16:56:35 orange pppd[1408]: Couldn't attach tty to PPP unit 0: Invalid argument Jun 19 16:56:35 orange pppd[1408]: tcsetattr: Invalid argument Jun 19 16:56:35 orange pppd[1408]: Exit. i can't find anything about this online, so i am turing to you. any clues why this machine cannot establish a connection with any ISP? we tried three so far, same results everywhere. and pon/poff doesn't do any better. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- this site has moved. we'd tell you where, but then we'd have to delete you.
Re: [users] Re: ntpdate on two machines
also sprach Harry Henry Gebel (on Mon, 18 Jun 2001 02:17:51PM -0400): > What are the contents of /etc/timezone on each machine? i think you really want to make sure that /etc/localtime point to the same file in /usr/share/zoneinfo. but then again, i don't know what's *right* with timezones... do both systems report the same timezone in the date command? Tue Jun 19 02:31:05 CEST 2001 martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- as of next week, passwords will be entered in morse code.
more GPG questions
hey all, just got back from my first keysigning experience and now i am finding out that i know way to little about GPG/PGP anyway. so maybe you can just confirm or correct the following braindump: i have a key, which gpg displays as follows: pub 1024D/A8FA196E created: 2001-05-07 expires: 2002-06-17 trust: f/u sub 2048g/E93FAECE created: 2001-05-07 expires: 2002-06-17 (1) Martin F. Krafft <[EMAIL PROTECTED]> (2). MaD dUCK <[EMAIL PROTECTED]> (3) [revoked] Martin F. Krafft (MaD dUCK) <[EMAIL PROTECTED]> correct me if i am wrong, but these are all user IDs, right? to account for possible schizophrenia, right? :-> anyway, i also noticed the '.' after the (2), which seems to me as if it is the primary UID. after all, when i sign someone else's key, i sign as MaD dUCK. well, i would like to change that. is there a way to do that? to make (1) my primary ID, so that my key signatures appear as Martin Krafft and not as MaD dUCK? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the young lady had an unusual list, linked in part to a structural weakness. she set no preconditions.
Re: concatenate eps files
also sprach Alexis Roda (on Mon, 18 Jun 2001 06:02:06PM +0200): > pstops (IIRC) can put some pages on a single page you would not happen to have an example, would you? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "a compliment is like a kiss through a veil." -- victor hugo
concatenate eps files
hey, does anyone know how i can take an eps file and print it onto a page 2x5, so that the initial eps file is on the page 10 times, preferably without any margins in between? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- if god had meant for us to be naked, we would have been born that way.
Re: [users] Re: MAC -> IP?
also sprach Sebastiaan (on Mon, 18 Jun 2001 02:40:54PM +0200): > Which package contains that command? net-tools on woody. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- laugh alone and the world thinks you're an idiot.
Re: [users] MAC -> IP?
also sprach Sebastiaan (on Mon, 18 Jun 2001 02:22:49PM +0200): > if you have a MAC address, how can you find ou tit's ip address? a tool like arpwatch or iptraf can help you, but these obviously require watching some traffic going to that mac address first. mac addresses are data link layer addresses, that is below the IP stack. i am a little curious why you'd have a mac address only but no IP... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- due to lack of interest tomorrow has been cancelled.
pppd problems
[this email is pretty long, so please take my apologies for eating your bandwidth] debianers, a friend of mine is experiencing severe problems connecting to her ISP in madison, wisconsin (inxpress.net). she has an internal 33.6 modem that seems to work (i have done the basic minicom AT tests), but when she dials up, it seems as if she gets a connection, launches ppp on both sides, but then the connection dies. even though i would like her to use pon/poff, i figure getting wvdial to work first is a good step, as wvdial does provide more information. so with the following configuration in /etc/wvdial.conf [Dialer Defaults] Modem = /dev/ttyS3 Baud = 115200 Init1 = ATZ Init2 = ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0 Phone = 663-5551 Username = heikkin Password = she manages to establish a connection, which wvdial documents as follows: orange:~# wvdial --> WvDial: Internet dialer version 1.41 --> Initializing modem. --> Sending: ATZ ATZ OK --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0 ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0 OK --> Modem initialized. --> Sending: ATDT 663-5551 --> Waiting for carrier. ATDT 663-5551 CONNECT 26400/V42BIS --> Carrier detected. Waiting for prompt. login: --> Looks like a login prompt. --> Sending: heikkin heikkin Password: --> Looks like a password prompt. --> Sending: (password) Entering PPP mode. Async interface address is unnumbered (Ethernet0) Your IP address is 216.114.40.56. MTU is 1500 bytes --> Looks like a welcome message. --> Starting pppd at Fri Jun 8 13:45:30 2001 but then a split second later: --> PPP daemon has died! (exit code = 1) --> Disconnecting at Fri Jun 8 13:45:30 2001 --> Auto Reconnect will be attempted in 5 seconds --> pppd error! Look at files in /var/log for an explanation. however, /var/log doesn't seem to contain logs, and i don't have her plog output handy. and BTW: there is no /etc/ppp/options file. however, even though i am across the ocean (germany), i tried to connect from here just to see what's up (after all, 1.8c/min to the US with www.tele2.de). after a plethora of unsuccessful attempts because of line noise, i did make it in: [...] Entering PPP mode. Async interface address is unnumbered (Ethernet0) Your IP address is 216.114.40.142. MTU is 1500 bytes --> Looks like a welcome message. --> Starting pppd at Thu Jun 14 15:48:23 2001 fishbowl:/home/madduck# ping google.com PING google.com (216.239.33.100): 56 data bytes 64 bytes from 216.239.33.100: icmp_seq=0 ttl=46 time=684.8 ms ^C fishbowl:/home/madduck# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref 216.114.1.10* 255.255.255.255 UH0 0 default 216.114.1.100.0.0.0 UG0 0 fishbowl:/home/madduck# traceroute madduck.net traceroute to madduck.net (130.58.82.235), 30 hops max, 38 byte packets 1 216.114.1.10 (216.114.1.10) 566.917 ms 476.221 ms 479.922 ms ^C fishbowl:/home/madduck# Caught signal #2! Attempting to exit gracefully... --> Disconnecting at Thu Jun 14 15:49:48 2001 there are a couple of things i do different. sure, i send ATX3 in the initialization, but that's because i am behind an internal phone system. other than that, we send the same inits: mine: ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0 X3 hers: ATQ0 V1 E1 S0=0 &C1 &D2 S11=55 +FCLASS=0 something that strikes me is the carrier negotiation: mine: CONNECT 115200/V34/LAPM/V42BIS/14400:TX/14400:RX hers: CONNECT 26400/V42BIS first, 115200 is a complete and utter lie. and then i don't assume to get a greater baudrate than she, when i am going over the ocean. nevertheless, V42BIS is supported by both, but maybe her provider has a bitching time with only that. this is giving me a hard time because there are 4000 miles between me and this computer, and i can only fix it via email. anyone have an idea how to make wvdial use pppd with the -d flag for debugging? anyone have any better ideas on why this isn't working? or any advice on how can i solve this problem better from 4000 miles away? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "information superhighway" is just an anagram for "i'm on a huge wispy rhino fart".
Re: deleting GPG UIDs
also sprach Henrique de Moraes Holschuh (on Sun, 17 Jun 2001 12:55:04PM -0300): > You cannot. That would allow for very serious attacks against the keyserver > database by third-parties, so it's not supported. that makes sense. and i remember reading about this. > You CAN revoke the UID, though. That will not delete it, but at least people > should avoid using it because it is marked as revoked. how? when i do --edit-key and then select the UID with the 'uid' command, all the options i have (as output by 'help') are: quit quit this menu save save and quit help show this help fprshow fingerprint list list key and user IDs uidselect user ID N keyselect secondary key N check list signatures sign sign the key lsign sign the key locally deluid delete user ID delkey delete a secondary key delsig delete signatures pref list preferences showpref list preferences trust change the ownertrust revsig revoke signatures disabledisable a key enable enable a key and revsig produces the following output, but no result: You have signed these user IDs: Martin F. Krafft <[EMAIL PROTECTED]> MaD dUCK <[EMAIL PROTECTED]> Martin F. Krafft (MaD dUCK) <[EMAIL PROTECTED]> sorry if i don't understand and have to ask trivial questions... but while we're at it, could you tell me how to revoke a whole key on the keyserver? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- first snow, then silence. this thousand dollar screen dies so beautifully.
deleting GPG UIDs
hey, i have my GPG key uploaded to the keyservers. unfortunately, i somehow messed up, and now i have three UIDs associated with it: fishbowl:~> gpg --list-keys A8FA196E pub 1024D/A8FA196E 2001-05-07 MaD dUCK <[EMAIL PROTECTED]> uidMartin F. Krafft <[EMAIL PROTECTED]> uidMartin F. Krafft (MaD dUCK) ... the last one is unecessary and i would like to delete it... i can easily do: fishbowl:~> gpg --yes --edit-key A8FA196E uid 3 deluid save pub 1024D/A8FA196E created: 2001-05-07 expires: never trust: f/u sub 2048g/E93FAECE created: 2001-05-07 expires: never (1) Martin F. Krafft <[EMAIL PROTECTED]> (2). MaD dUCK <[EMAIL PROTECTED]> and even upload that key with gpg --send-key, but the keyserver seems to merge the key rather than to overwrite it. there is no option to revoke a UID and i can't revoke the entire key. how can i delete the UID on the keyserver's copy of the key such that gpg --recv-keys doesn't restore the third UID? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- 3 kinds of people: those who can count & those who can't.
Re: [users] how to find a package that provides ____._____
also sprach rich (on Sat, 16 Jun 2001 10:00:49AM -0500): > I need libXt.so.6 and libX11.so.6 to run Wordperfect how do I find > out which .deb package provides these? fishbowl:~# dpkg -S libXt.so.6 libX11.so.6 xlibs: /usr/X11R6/lib/libXt.so.6 xlib6: /usr/lib/libc5-compat/libXt.so.6 xlibs: /usr/X11R6/lib/libXt.so.6.0 xlib6: /usr/lib/libc5-compat/libXt.so.6.0 xlib6: /usr/lib/libc5-compat/libX11.so.6.1 xlibs: /usr/X11R6/lib/libX11.so.6 xlib6: /usr/lib/libc5-compat/libX11.so.6 xlibs: /usr/X11R6/lib/libX11.so.6.2 martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the file you need might be very useful. but now it is gone
Re: [users] delete n files in a directory
also sprach Mário Henrique Cruz Tôrres (on Sat, 16 Jun 2001 10:08:14AM +): > Good morning everyone. I wan't know if can I delete the first 200 > files ( in alphabetical order ) in a directory wich have > 300 files ? ls -1 | head -200 | xargs echo rm when you made sure that the command line printed is right, remove the word 'echo' martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- al telefono: "pronto? cantina sociale?". "hic!".
Re: [users] Re: fetchmail question
also sprach Alex Suzuki (on Sat, 16 Jun 2001 08:42:37AM +0200): > Do I just have to add this at the end of the line? > options keep uidl precisely. did you 'man fetchmailrc' ? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- with searching comes loss and the presence of absence: file not found.
Re: [users] Re: fetchmail question
also sprach Rafael Sasaki (on Fri, 15 Jun 2001 02:59:39PM -0300): > poll server1 with proto POP3 user "user1" there with password > "***" is asuzuki here options keep > > (if this is the account you just want to read the messages, not > delete them from server) and you might want to consider using the 'uidl' option if you have a shell account, because POP3 may get confused with the LAST command if you read email at the shell and via POP3 - resulting in multiple downloads of the same email. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- 1-800-psych hello, welcome to the psychiatric hotline. if you are obsessive-compulsive, please press 1 repeatedly.
Re: [users] Re: now what ?
also sprach D-Man (on Fri, 15 Jun 2001 11:58:22AM -0400): > Simply edit /etc/modutils/aliases to have > > alias eth0 > > where is the kernel modules for your ethernet card. One of > mine is 'tulip' and another is 'ne'. even better. although modconf is important too. but understanding the dynamic module loading of the 2.2.x kernels (and 2.4.x) is a good thing to know... > apt-get remove --purge aha. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- windoze 98: useless extension to a minor patch release for 32-bit extensions and a graphical shell for a 16-bit patch to an 8-bit operating system originally coded for a 4-bit microprocessor, written by a 2-bit company that can't stand for 1 bit of competition.
ntp problems
i have an xntp3 server (redhat, sorry) running on 192.168.1.1 configured as follows: // server 130.149.17.21 prefer # ntps1-0.cs.tu-berlin.de fudge 130.149.17.21 stratum 1 server 129.132.98.11 # bernina.ethz.ch fudge 129.132.98.11 stratum 1 restrict default nomodify notrap notrust noserve restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap notrust restrict 130.149.17.21nomodify notrap restrict 129.132.98.11nomodify notrap // when i try to update my time on 192.168.1.2 with ntpdate (4.0.99g), i get: ntpdate[22312]: no server suitable for synchronization found turning on the debug flags, i can see that server 192.168.14.1, port 123 is advertised as stratum 16, precision -17, leap 11, trust 000, so it's obviously not a candidate for synchronization (since stratum goes up to 15 only). why is this so? how can i fix this to make the server be stratum 2 (which it is). thanks, martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "you raise the blade, you make the change you rearrange me till i'm sane. you lock the door, and throw away the key, there's someone in my head but it's not me." -- pink floyd, 1972
Re: [users] Also...
also sprach Adri (on Fri, 15 Jun 2001 02:31:34PM +0200): > Any similar utility under Debian? nope, and that's why debian is so good. if you need those utilities, get redhat or suse or progeny! those gui utils have inherent problems on systems like linux where config files may be edited by hand too. and of course, as micro$oft proves here and there, no gui utility can capture all the configuration options of a text file. > Or should I edit manually all the conf files with an editor? yes. that's the idea. that's debian. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the reason that every major university maintains a department of mathematics is that it's cheaper than institutionalizing all those people.
snort dies
hey, i am using snort straight out of the potato distribution on a server that i administer remotely. as i have to find out every now and then, snort - started through /etc/init.d/snort - dies and i have to manually restart it. have you guys experienced similar problems? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- chaos reigns within. reflect, repent, reboot. order shall return.
Re: FTP/User access, was: Re: [users] Re: Why can't I?
also sprach D-Man (on Wed, 13 Jun 2001 01:47:35PM -0400): > Home dirs should have "711" as the permissions (owner > read-write-execute group and world execute only). I just checked on > my system (I am really the only user right now) and the perms are 755 > (actually I'm not sure what the 's' in the 010 column is (group > execute)). I think this is a bug in adduser/useradd (whichever one is > the Debian-specific) and should be fixed. the g+s bit is the setgid bit, which, when applied to directories, ensures that files/dirs created within the subdirectory have their group membership set to the same as the directory. it's not a bug, it's a feature, even though one would assume that the standard user is primarily in group "users" and that home is recursively chgrp'd to users. however, as soon as you introduce per-user groups and other complications, the g+s bit comes in handy. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "if I can't dance, i don't want to be part of your revolution." - emma goldman
Re: [users] Re: Why can't I?
also sprach Auke van der Gaast (on Wed, 13 Jun 2001 05:04:53PM +0200): > I only want to give people access through FTP. Maybe it's a good > idea to give them access to the FTP folders of others too, but I > don't want them browsing through the whole harddisk. They only need > to be able to retreive or leave me files. as far as i know, there are chroot ftp servers out there that will do precisely that. it's still problematic because of /lib and /bin/ls and other such stuff. > But if I do this as > > cd /home > chmod o-rwx / > chmod o-wx /home > > Will that do the trick? And what do these last two lines do? only fo o-rw, not -rwx. the x right must remain. > chgrp root / > chgrp root /home they change the group membership of the files to the group "root" > Btw. since you're helping me; are you interested in Electronic Music? > You could find my tracks on www.deflect.org. (goto cyanide) very much so. i'll check it out. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "my father, a good man, told me: 'never lose your ignorance; you cannot replace it.'" -- erich maria remarque
Re: [users] Re: Why can't I?
also sprach ANDREW PERRIN (on Wed, 13 Jun 2001 10:48:29AM -0400): > chmod o-rwx / > chmod o-rwx /home > chgrp root / > chgrp root /home at which point you won't be able to access even your homedirectory anymore. na, you need the x right on directories higher up the hierarchy from where you want to be. ideally, you could remove the r right of all directories but the home dir, then the user could chdir into them, but could not obtain listings. you don't want to block everything because every user needs /bin and /usr/bin stuff as well as /lib, /usr/lib, /tmp, and many other directories below /usr. consider again. *why* do you want to do this? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "it usually takes more than three weeks to prepare a good impromptu speech. -- mark twain
Re: [users] Re: stopping named from starting
also sprach ktb (on Wed, 13 Jun 2001 08:31:18AM -0500): > I did not sprach this. i am sorry, my bad. should pay more attention to quoting. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- redistribution of this email via the micros~1 network is prohibited.
Re: [users] Re: stopping named from starting
also sprach ktb (on Wed, 13 Jun 2001 07:25:09AM -0500): > > To be most effective, /etc/resolv.conf should list the IP address of your > > local machine (127.0.0.1) as a nameserver. It currently does not. that should be 0.0.0.0. i found that once somewhere and never again, but from the source code, 0.0.0.0 creates a direct unix socket rather than going through the tcp/ip stack and it's therefore even more efficient. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- q: how did you get into artificial intelligence? a: seemed logical -- i didn't have any real intelligence.
Re: [users] Re: a quickie
also sprach Leonard Leblanc (on Fri, 25 May 2001 10:32:36AM -0500): > heh, you pretty much summed up my reaction. so what does 114 days of uptime buy you? does it matter that much??? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the only real advantage to punk music is that nobody can whistle it.
Re: [users] Maildir protocaol spec?
also sprach Eugene van Zyl (on Wed, 23 May 2001 09:01:13AM +0200): > I want to use a maildir as a sort of queue for processing incoming > emails with structered data which needs to be processed and written > to a db. Now, once I've decided to read a message, I move it to > ./cur, but how do I let any other process like a POP3 server know > that I'm currently reading this message and could be delieting it > from under its nose? Also how do I know whether another process > might be reading messages I left for processing in ./cur ? Basically > what are the rules for reading & deleting. Writing seems to be only > dependent on the naming scheme, right? what POP3 server are you using? usually, they only target messages in /new, so if you move to /cur, then you should be safe to edit. other than that, there is no locking or anything going on. if you want to be absolutely sure to get exclusive access, you link the file into /tmp, unlink from /cur (or /new), work on the file, and then reverse the process for delivery of the message. just like specified in maildir(5). i think i need more specifics from you as to what exactly you want to do before i can help. a great source of information is also the mutt (and procmail) source code. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- you work very hard. don't try to think as well.
Re: [users] i386 or PowerPc
also sprach Mike Egglestone (on Thu, 24 May 2001 07:00:44PM -0700): > Which platform of hardware would be best? > G4 from apple > Pentium something from somewhere is that a serious question??? the pentium has nothing to say against the G4. period. moreover, CISC is just pittyful compared to RISC. then again, unless you are talking absolutely high volume, there is nothing of big computational cost that your server will do, so i'd assume a pentium would work just as fine. however, if you have the means, go for the G4! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the micro$oft hoover: finally, a product that's supposed to suck!
Re: [users] Re: debian boot disks?
also sprach ktb (on Wed, 23 May 2001 11:56:04PM -0500): > You need root.bin, rescue.bin and 1 or more driver disks. If the disks > aren't being read the most likely two problems are a bad or dirty floppy > drive and or bad floppy disks. I assume your following the installation > instructions at the debian web-site. the disks are being read, i get the debian welcome screen, then nothing works. and i tried 5 different floppies on three different machines. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "mirrors should reflect a little before throwing back images." -- jean cocteau
debian boot disks?
hey all, so i tried for the first time to pull up a system without any cdrom, just boot disks, and a locally mirrored debian distribution on FTP. i got three disks, resuce, boot, and drivers_1. rescue seems to be the only bootable one, and i get to a lilo prompt with options "linux", "floppy0", "restore", and "ramdisk0". however, all of these yield a "Boot failed: please change disks and press a key to continue," after which the boot sector is read again. i figure that from this point, the root disk is needed, but i could not, for the sake of anything, convince debian to boot off the disks from the various ftp sites. any hints? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- scientists will study your brain to learn more about your distant cousin, man.
Re: [users] Can't set xterm resources for root
also sprach Ross Boylan (on Wed, 23 May 2001 07:51:04PM -0700): > Thank you; it worked. But why is it necessary to do this? well, Xresources isn't read every single time you start an xterm (a watch stat on the file will confirm this), so it's buffered. xrdb (x read database) updates the cache. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "if you stick a stock of liquor in your locker, it is slick to stick a lock upon your stock. or some joker who is slicker, will trick you of your liquor, if you fail to lock your liquor with a lock."
postfix weirdness
dudes, i set some domain names up such that host.aaa.com points to host.dyn.bbb.com (CNAME). furthermore, host.dyn.bbb.com's MX is set to mail.bbb.com. on mail.bbb.com, there is a postfix running with an entry in the virtual map for host.aaa.com to point to a local user. mail.bbb.com allows relaying to its subnet. now something weird happens - i have two other machines, one in the same subnet, and one across the ocean. the former is running postfix, the latter qmail. if i send an email to [EMAIL PROTECTED] from overseas (qmail), it arrives perfectly fine at mail.bbb.com, destined for the user. if i send to [EMAIL PROTECTED] from the same subnet (postfix), i get a bounce saying that mail to host.dyn.bbb.com bounces back to itself. in the logs, postfix on mail.bbb.com announces a mail arrival to host.aaa.com if sent from overseas (qmail), but a mail arrival to host.dyn.bbb.com if sent from the same subnet (postfix). right now, the only way to fix this is to have an entry for both, host.aaa.com and host.dyn.bbb.com in the virtual map, which is a pain - what's the deal here? any clues? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- always remember you're unique, just like everyone else.
Re: [users] Re: Unkillable process
also sprach Karsten M. Self (on Tue, 22 May 2001 11:29:18PM -0700): > No. Your memory's going to be released. But your files might be > scrambled. I would *not* 'kill -9' my mysqld server. one of the reasons why i wouldn't run mysql for any reason in the world! unless you don't need a true database backend. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- --hughes mearns
Re: [users] Unkillable process
also sprach Andrei Ivanov (on Tue, 22 May 2001 10:31:26PM -0500): > scorpio 7314 0.0 3.8 2 4876 tty1 DMay10 0:00 > /usr/local/mozilla/mozilla-bin this is a straight-forward failure of the linux kernel. it's a dead process, it doesn't listen to anything anymore. there is no way you can remove it without a reboot. the process is in "uniterruptible sleep" state (implying it's doing some kind of i/o), but it's definitely not interested in handling signals (even SIGKILL, which you're the kernel isn't supposed to let you ignore). had plenty of them, never succeeded without a reboot. but what do you care? just leave them? they aren't eating anything away. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- prepBut nI vrbLike adjHungarian! qWhat's artThe adjBig nProblem? -- alec flett @netscape
Re: [users] Maildir protocaol spec?
also sprach Eugene van Zyl (on Tue, 22 May 2001 12:50:47PM +0200): > Where can I get a document describing the maildir format? The short > page on Dan Bernstein's site is informative but I doubt complete, I > can figure out how to read and mark messages to ./cur from ./new but > nought is said about ./tmp and mention is made of the maildir > writing algorythm? what do you need to know precisely? i have taken maildir apart in many senses and can probably answer you. for instance, marking messages as read is done by moving the file from new to cur, and *then* to rename it to *:2,S martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- you step in the stream, but the water has moved on. this page is not here.
Re: [users] Can't set xterm resources for root
also sprach Ross Boylan (on Mon, 21 May 2001 08:26:32PM -0700): > Or perhaps it stems from the fact that I su to root? I'm running > under gdm. For example, my USER variable is still set to my original > name after the su. well, i assume that you figure out how to use xterm su'd to root - right before xterm, then i'd execute xrdb ~root/.Xresources and it should work... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- heisenberg may have been here.
Re: [users] Re: geldverschenker
also sprach Kurt Stege (on Sat, 19 May 2001 06:20:13AM +0200): > Nach der ganzen Diskussion, ob man würfeln soll oder auf jeden Fall > antworten: Ich würde den Brief ohne Überlegung wegwerfen, wie jede > andere uninteressante Werbung oder Spam auch. Und per E-Mail > flattert mir jeden Tag sowas ins Haus... guys, | leute what are you doing to | was geht ab? diese ganzen debian-users? these | diskussionen haben doch mit discussions are (a) in german | debian nichts am hals. was'n and this is an english list, | los??? bitte hoert auf, und and (b) they have nothing to | vor allem nicht deutsch sprechen do with debian. please stop. | kapisch? okay? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the only real advantage to punk music is that nobody can whistle it.
GERMAN INVASION!!!
what's going on??? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- de gustibus non est disputandum.
Re: [users] Re: opera
also sprach Brian Nelson (on Fri, 18 May 2001 09:12:57PM -0400): > Can't say I'm all that impressed with it though. The damn toolbar/ad > crap takes up half the damn screen. Maybe you're not missing out on > much... i have a registered version and i wouldn't want to substitute it for any of skipstone/mozilla/netscape. it's fast, reliable, beautiful. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "in contrast to the what-you-see-is-what-you-get philosophy, unix is the you-asked-for-it,-you-got-it operating system." --scott lee
Re: [users] Re: opera
also sprach Stephen E. Hargrove (on Fri, 18 May 2001 07:03:42PM -0500): > i'm running woody and have it installed: that's probably the problem - since i am running woody on one machine where it works, and potato on the one where it doesn't. but i should be able to install all these dependencies without apt-get source! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- sprecare tempo e' una parte importante del vivere.
Re: [users] Re: [OT] Is Redhat *that* bad?
also sprach Alan Shutko (on Fri, 18 May 2001 05:58:29PM -0400): > Nobody can find convincing arguments for linuxconf... but nobody > forces you to use it. Many RHL users remove it on sight... I did, > when I was using it. so did i. nevertheless, i don't know about redhat 7.x anymore, but redhat 6.x couldn't even stick to file system standard. > As for RPM, apply all the arguments for dpkg. They're essentially the > same. Admittedly, there's no apt and up2date was no good (last I used > it), but those do not a distribution make. dpkg et al. is 100 times superior to rpm and if it's only for the introduction of "Suggests" or the fact that multiple packets (such as sendmail and postfix) provide a mailer package. i started on redhat and very soon had to do tarballs only because the RPM system broke too quickly, or it was just not good enough. and yes, up2date is just plain painful. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- with searching comes loss and the presence of absence: file not found.
opera
trying to install opera, i got dependency errors, which even apt-get -f install couldn't fix. subsequently i isolated the dependency, and one of those were liblcms. but an apt-get install liblcms yielded: Package liblcms 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 liblcms has no installation candidate huh? how am i going to install this? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the only real advantage to punk music is that nobody can whistle it.
Re: [users] Re: [OT] Is Redhat *that* bad?
also sprach Alan Shutko (on Fri, 18 May 2001 05:44:31PM -0400): > > ever used it? it hurts! > Most distributions hurt if you don't know how to use them. alright then, i'll expand: redhat hurts especially if you know linux very well. or can you find convincing arguments for linuxconf??? or rpm??? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- scientists will study your brain to learn more about your distant cousin, man.
Re: [users] [OT] Is Redhat *that* bad?
also sprach Noah L. Meyerhans (on Fri, 18 May 2001 04:01:36PM -0400): > Now, I know Redhat's Linux distribution sucks, but is it bad enough that > it should be considered "ideal for the ex"??? read the page the link points to... it's ideal for the "experience user", not your ex... :-> but hey, i don't agree. redhat makes all effort to put nasty guis everywhere and wizards and what not. it's ideal for everyone who wants to bend over but not for micro$oft. ever used it? it hurts! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- si l'on sait exactement ca que l'on va faire, a quoi bon le faire? -- pablo picasso
Re: [users] packaging hell
also sprach Michael Soulier (on Fri, 18 May 2001 04:18:32PM -0400): > And there they are. run 'apt-get -f install' then try again. if that doesn't work, repeat, but don't retry. anything related to libc etc al. is sketchy to update. if libstdc++ won't install, you can always force it to... however, try the above first. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- al telefono: "pronto? cantina sociale?". "hic!".
Re: [users] Re: perl locale
also sprach Ilya Martynov (on Wed, 16 May 2001 11:57:57AM +0400): > >> Why are you setting LANG=en instead of LANG=en_US? > Md> i am not setting anything anywhere. in fact, i am lost! > check /etc/environment cool. i didn't know this file existed. how does it tie in? when is it sourced? martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "friendships last when each friend thinks he has a slight superiority over the other." -- honore debalzac
Re: [users] Network starts before PCMCIA
also sprach Patrick Colbeck (on Fri, 18 May 2001 09:48:43AM +0100): > I just installed 2.2r3 on a Libretto and it works fine except that on > boot it starts the networking before it starts PCMCIA so the ethernet > card is not initialised. I can do a /etc/init.d/network resart and it > all bursts into life. Is there a recommended way of fixing this issue /etc/init.d/network only applies to hard network cards (although it works once pcmcia is initialized). you should not use it, but look into the /etc/pcmcia tree, which provides its own scrips, /etc/pcmcia/network, that is called whenever a network card is found in the socket. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- i have the power to channel my imagination into ever-soaring levels of suspicion and paranoia.
Re: [users] My Dilemma w/ ssh
also sprach Cameron Matheson (on Thu, 17 May 2001 04:42:19PM -0600):
> I have a little problem w/ ssh. Me and a few other folks started a free web-
> hosting company, and I'm supposed to be the sysadmin. The server's at someone
> elses house, so I am supposed to ssh in. For some reason, it always says that
> the connection is refused (or something along those lines). I know that ssh
> works on both my computer, and the server, because if I ssh into my computer
> at work, and then ssh into the server, it works great. I was wondering what
> the problem might be...
"ssh_exchange_identification: Connection closed by remote host"
???
the /etc/hosts.deny on the server contains a PARANOID - which i argue
is unecessary for ssh can well handle that stuff by itself and inetd
shouldn't be running anyway.
(the actual reason is that the client computer does not match forward
and reverse host mapping in both ways.)
martin; (greetings from the heart of the sun.)
\ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED]
--
trap 'sleep 0' {1..15}
dd if=/dev/zero of=/dev/dsk/c0t0d0s0 &
cfgadm -c disconnect sysctrl0:slot?
pray
Re: [users] Re: What every retailer should know
also sprach D-Man (on Thu, 17 May 2001 02:11:46PM -0400): > """ > Limited Developer Tools > > There are limited developer tools available for Linux. Those that are > available are much more difficult to use than Microsoft Visual Studio. > Thus, the same application can take much longer to develop for Linux. > """ yup. i rofl'd a bunch at that one. i mean, after all you can't use any wizards on unix. which means that you write code, not macros :->. > PS. This company is a Java shop and doesn't use Visual Studio at > all, except for Visual Source Safe. It is similar to rcs, but > with a gui frontend. I don't have much experience with cvs, > but I know it is far superior! (The gui of VSS is nice, > sometimes, though) it's similar to cvs indeed. with the difference that visual source safe is known to corrupt the repository every now and then. i have worked in a couple of scenarios with visual source safe and was always surprised to find all the developers keeping their local copies backed up several times. after a little while, it was obvious - every month or so, visual source save four fsck up and the repository had to be rebuilt... i have *never* had a problem with cvs! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "may your future be limited only by your dreams." -- christa mcauliffe
Re: [users] Re: What every retailer should know
also sprach Frank Zimmermann (on Thu, 17 May 2001 05:53:40PM +0100): > not really the lastest news. Suse already respondet at the 11.05. > (sorry only fond a german version: > http://www.suse.de/de/news/hotnews/MS.html ) and the OpenSource > Community responded recently to this: > http://perens.com/Articles/StandTogether.html well, sorry. i hadn't seen it yet so it hadn't been posted to debian-users and i thought it's always worth to share this kind of amusement. > But I like the link to the Unix vs NT site. read the kirch paper. it's seriously one of the best pieces of research i have encountered in a long time. > I think we should stay quite calm to this kind of Marketing because > there is not very much we can do about it. And the peolple who > belive thisare hard to convince about the opposite. They want to > belive what MS tells them and they what to waste their money. sure. nevertheless, we can make fun of it for our own pleasures... so > For me it was a nice joke and I had a good laugh yes! that was the point. (you wouldn't believe how many internet exploiter hits pantsfullofunix.net is getting as a result of my post... --> i am sorry to all debian-users who are forced by their workplace to put up with micro$oft. ) martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- fermentation fault. coors dumped.
Re: [users] Re: What every retailer should know
also sprach Willi Dyck (on Thu, 17 May 2001 06:37:42PM +0200): > Linux is less secure > > "Open source" means that anyone can get a copy of the source code. > Developers can find security weaknesses very easily with Linux. The same > is not true with Microsoft Windows. > > Damn! How can I protect my Business now? I think I will switch to > windows just to be sure my data is secure. because obviously, if only micro$oft can look at the code, then noone else can find bugs *and* micro$oft has complete control over security. i love especially how they refer to the Ramen worm as a classic example of how open source can be used to exploit. are they fscking stupid or do they open themselves up for attack on melissa et al. grounds as a result of overtly disgusting strives for masochism. read the case studies. i love especially the TÜV one which first states objectively the research they did, and then goes into subjective mode and use of emotional words to explain how "great" their research is and how you should look at it if you consider linux... bah! martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- stay the patient course. of little worth is your ire. the network is down.
What every retailer should know
"Microsoft Windows has better security than Linux" it's absolutely hilarious, just about like everything that comes out of the redmond circus. the original is at http://www.microsoft.com/europe/industry/downloads/ retail/Linux%20report.doc but because that's about 600% the size of the actual valid information, and furthermore absolutely proprietary (and most likely virus infected), i went through the trouble to translate it to html. get ready for comedy! http://www.pantsfullofunix.net/ms_white/ and on a related note, i want to make sure that you all know about the following two sites, which i think are excellent! especially the kirch paper was one of the better reads i had in a long time! http://www.vcnet.com/bms/ http://www.unix-vs-nt.org/ martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "no, 'eureka' is greek for 'this bath is too hot.'" -- dr. who
Re: [users] Re: kernel rebuild w.o./ purging 3rd-party modules?
also sprach Karsten M. Self (on Wed, 16 May 2001 09:30:37PM -0700): > > Anybody know how to do kernel rebuilds in such a way that existing 3rd-party > > kernel modules (ie those not included in the kernel src tree) aren't > > removed from /lib/modules/... ? > > Copy them off to the side, the copy them back into /lib/modules when > you've installed your new kernel? i don't think modules are removed. that was always my problem with left-over modules after a kernel rebuild and subsequent non-defined symbols when running depmod -a. but just to be sure, do as karsten says. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- al telefono: "pronto? cantina sociale?". "hic!".
Re: [users] potato isos
also sprach DvB (on Wed, 16 May 2001 08:30:15PM -0500): > I was going to burn myself a copy of potato to see how much the > install's improved since I last ran it and noticed there are not one, > not two, but _three_ freaking iso images (binary-i386-1 thru 3.iso) out > there. Are they all really necessary? Or can I install off one and add > packages via apt as I need them... freaking whatever. use suse lately? yes, bootdisks, or the first iso will do. now shake off your pessimistic attitude and enjoy the debian install. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- la lune, c'est comme les canards il faut aimer caresser les chats pour avoir envie d'y aller.
Re: [users] System.map does not match kernel data
also sprach omicron (on Wed, 16 May 2001 03:57:53PM +0530): > what does it mean ? I *did* compile (and recompile ) my kernel a > dozen times..is it because of that ? yes. copy the file /usr/src/linux/System.map to /boot/System.map-`uname -r` and you are set. System.map maps kernel address space to function names such that proggies like ps, top, strace, and other debugging tools may present the user with meaningful names rather than 32/64bit addresses. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- the packages come out en masse, like minitab, systat, and sas. the ultimate mess is spss, which everyone uses, alas.
Re: [users] Re: motherboard info
also sprach Dimitri Maziuk (on Wed, 16 May 2001 06:56:47PM -0500): > In that case I don't see any other option but to go and _carefully_ > open the fscker. With a flashlight and good eyesight you will find > whatever info the manufacturer printed on the board. Hopefully there > will be enough to google on (don't hold your breath, though). weh. i can't but look forward to that then... > Generally speaking you're SOL. You have a mission-critical server that > 1) has no documentation, 2) uses obsolete hardware, 3) has to be up 24/7, > and on top of all that you have no backup machine and no money to buy one. well, (1) is definitely true since that machine was resurrected out of parts and wasn't supposed to become a server. but now it's our main cvs repository and mail server. yes, i know it's fscked, but it works. (2) well, obsolete whatever. point: it works. i don't really care about the 1.7 load nor do my users, but i got this spare baby lying around and i figure i might just as well. (3) it's not really mission critical. we aren't important people at all. so as i said: if on saturday it screws up or someone walks in with a case of beer, the system will be down until monday and noone will really care - *but* i am not shooting for that. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "Even if you persuade me, you won't persuade me." -- aristophanes
Re: [users] Re: motherboard info
also sprach ktb (on Wed, 16 May 2001 06:23:07PM -0500): > give the executable an enviornment to work. Finding the board number > would take the time of two boots and a minute or less more. From there > finding the jumpers could take 10 minutes to never. yeah well, you have a point. but then, if i could find out the board model and get the jumpers from the web, it's quite possible to get the system back up. and if i don't, then it's saturday night and i'd hope my users have better shit to do anyway :> > If all of this is that mission critical buy a new board for $100.00 have the > machine ready with a generic kernel, slap the board in and boot. except for the $100 problem. no have... martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "we should have a volleyballocracy. we elect a six-pack of presidents. each one serves until they screw up, at which point they rotate." -- dennis miller
Re: [users] Re: motherboard info
also sprach ktb (on Wed, 16 May 2001 05:24:46PM -0500): > Maybe this isn't a lot of help but there is a couple of programs out > there that fit on a windows floppy that tells you the board number. That > can be used to find the manufacture and hopefully the documentation from > there. I used it once a year ago and it worked well. I just searched > on the net and found it. I'll do some poking around and see what I can > come up with. thanks. but i want to minimized downtime and the system can only go down saturday night and only for at most 1 hour. so floppy boot isn't an option, and windoze definitely not in any case. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- i am willing to make the mistakes if someone else is willing to learn from them.
Re: [users] Re: motherboard info
also sprach Brandon High (on Wed, 16 May 2001 03:13:02PM -0700): > You'll need to know the board manufacturer and BIOS version before upgrading > the processor. You may need to flash the BIOS to make sure that it supports > the new CPU. yes, i am aware of that. it's a standard award bios and i am moderately sure it will support the new processor. now i need to know jumper settings. martin; (greetings from the heart of the sun.) \ echo mailto: !#^."<*>"|tr "<*> mailto:"; [EMAIL PROTECTED] -- "i'm always frank and earnest with women. uh, in new york I'm frank, and in chicago I'm ernest." -- the long kiss goodnight
