firewall -- best practices
Hi, I have a couple of old machines that I will be installing Debian on them. I would like to dedicate one of the machines to a firewall, and the other machine to a mail server. I have a dsl line with a static IP (with the router acting as a firewall) and several other debian machines that will sit behind my firewall along with my mail server. I have several questions regarding this: 1. Is it best to not have the firewall doing anything else, i.e. acting as a web and/or mail server, and instead use a different machine for the mail server? 2. Occasionally, I would like to ssh into my network from work. Is it best to only open up the port on the firewall or do some port forwarding so that ssh connections automatically go to a different (non-firewall) machine? 3. I have been perusing different howtos on various networking setups mail server, etc. but am always looking for a must read site, book, etc. Anyone have any good suggestions? Thanks, John 3. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: /etc/modules, /etc/modules.conf and others
On Thursday 20 February 2003 8:58 pm, J.F.Gratton wrote: > On Thu, 2003-02-20 at 22:27, Matthew Weier O'Phinney wrote: > > /etc/modules is used by Debian to manually add modules the > > user/sysadmin wishes to load at boot time. After adding them, you > > then run 'update-modules' as root, which reads this, as well as > > some files in /etc/modutils/, to *create* /etc/modules.conf. (This > > script is run as part of /etc/init.d/modutils at boot time, in case > > you're wondering why you haven't run it yourself and things still > > worked.) > > > > There's a lot more to it, and this answer is definitely a bit > > simplistic, but it's the basic idea -- and it's why you don't need > > to (and *SHOULDN'T*) mess with /etc/modules.conf in Debian. > > > > -- > > Matthew Weier O'Phinney > > [EMAIL PROTECTED] > > Hi Matthew, > > Thx for the answer ! > > So you're telling me that I should run update-modules whenever I > modify /etc/modules ? Since I've never done that after a modif. , I > guess I was simply lucky to keep my system up and (at least looking > to be) properly running ? > > Jeff The easiest thing to do is use the program modconf. It is a menu driven program that allows you to select the modules you want. It will install them into the kernel for you and update the necessary files so that upon reboots, the module(s) are loaded for you. You don't have to use update-modules if you use modconf. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
How to burn a CD -- Helpful page
Hi, I found the following website very helpful when it came to burning a couple of CDs: http://www.cpqlinux.com/cdrw.html John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: dpkg error -subprocess paste killed by signal
On Friday 21 February 2003 9:16 pm, Carla Schroder wrote: > howdy, > > I'm running Libranet 2.7, which is woody 3.0 & lots of nice extras. > My troubles started with upgrading KDE 3.05 to 3.1. I didn't want to, > I wanted Quanta, and it needed KDE 3.1. So I says well, OK. > > I started with > > apt-get upgrade kdelibs4 > > which went fine. Then: > > apt-get upgrade kdelibs-bin > > That's when the trouble started. libarts1 and libartsc0 won't > install. I tried apt-get -f install, it failed too. As did > dpkg -i /var/cache/apt/archives/libarts1_1.1.0-0woody4_i386.deb > > So I says to myself, I says self, uninstall everything and start > over. Well, if I try apt-get remove this same darned unmet > dependency pops up: > > Sorry, but the following packages have unmet dependencies: > arts: Depends: libartsc0 (>= 1.1.0-0woody4) but it is not going to > be installed > Depends: libarts1 (>= 1.1.0-0woody4) but 1.0.5a-0woody1 is to > be installed > > I won't bore you with ranting about how installing a single KDE app > leads to this huge ugly winderz-type interdepency mess. I didn't want > to upgrade, I just wanted Quanta. My source is > deb http://download.kde.org/stable/3.1/Debian stable main > > So- any bright ideas? > > Thanks! > Carla > > > # apt-get -f install > Reading Package Lists... Done > Building Dependency Tree... Done > Correcting dependencies... Done > The following extra packages will be installed: > libarts1 libartsc0 > The following NEW packages will be installed: > libartsc0 > 1 packages upgraded, 1 newly installed, 0 to remove and 172 not > upgraded. 8 packages not fully installed or removed. > Need to get 889kB of archives. After unpacking 946kB will be used. > Do you want to continue? [Y/n] y > Get:1 http://download.kde.org stable/main libarts1 1.1.0-0woody4 > [846kB] Get:2 http://download.kde.org stable/main libartsc0 > 1.1.0-0woody4 [43.2kB] Fetched 889kB in 21s (40.6kB/s) > (Reading database ... 147934 files and directories currently > installed.) Preparing to replace libarts1 1.0.5a-0woody1 (using > .../libarts1_1.1.0-0woody4_i386.deb) ... > Unpacking replacement libarts1 ... > dpkg: error processing > /var/cache/apt/archives/libarts1_1.1.0-0woody4_i386.deb (--unpack): > trying to overwrite `/usr/bin/artscat', which is also in package > libarts1-bin dpkg-deb: subprocess paste killed by signal (Broken > pipe) > Unpacking libartsc0 (from .../libartsc0_1.1.0-0woody4_i386.deb) ... > dpkg: error processing > /var/cache/apt/archives/libartsc0_1.1.0-0woody4_i386.deb (--unpack): > trying to overwrite `/usr/bin/artsdsp', which is also in package > libarts1-bin dpkg-deb: subprocess paste killed by signal (Broken > pipe) > Errors were encountered while processing: > /var/cache/apt/archives/libarts1_1.1.0-0woody4_i386.deb > /var/cache/apt/archives/libartsc0_1.1.0-0woody4_i386.deb > E: Sub-process /usr/bin/dpkg returned an error code (1) > > -- > ~ > Carla Schroder > www.tuxcomputing.com > this message brought to you > by Libranet 2.7 and Kmail > ~ Yes, you should use dpkg --purge --force package_to_remove. You can't reliably do an update for kde from the official kde packages to the unofficial kde packages found on at kde.org. The packager Ralph Nolden recommends that you remove stuff before installing the new stuff. Familiarize yourself with dpkg, especially the force, since it will remove stuff even if there are dependencies. You may have to do it package by package in the worst case, but hopefully, you can remove the few offending ones, and the use apt-get --purge remove (the other packages) which should help automate things. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: wtf? (long and frustrated)
On Tuesday 11 March 2003 10:48 am, Glenn English wrote: > Many thanks to all of you for the compassion and the information. The > little sucker is making pictures now - I learned a lot about X in the > past couple days. > > Next step is getting it to realize there's an Ethernet connector in > it's (PCMCIA) NIC. But I expect that to be a pretty straightforward > task now that I can actually use the machine and some of the > configuration tools. > > I've got to agree with David Krider's complaints about the Debian > install process. This isn't rocket science; it's just a hardware > prober. If the 'lesser' distros can install in one try (and without > rebooting several times), surely the mighty Debian can. This install > process seems to be a test that must be passed in order to join the > Debian community. > You might want to install the package discover. I have it installed on my laptop, and it was able to find a lot of the hardware I have (sound, NIC, etc). John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 'gcc logjam'
On Thursday 13 March 2003 8:14 pm, Matt Price wrote: > hi everyone, > > I've noticed frequent references in this list to the 'gcc logjam'. > Now, I know gcc is the C compiler, and that more or less everything > in Debian is dependent on it... ut I don't know much else. Can > anyone tell me what the _issues_ are with gcc, and what the nature of > the 'logjam' is? I just want to understand batter what it's all > about. thanks as always, > matt The default compiler for woody is gcc-2.95.4. They are moving to gcc-3.2 in unstable(using currently) and testing. The change in compiler versions exposes problems that need to be reconciled. In some cases, different hardware platforms that Debian supports have "issues" with the new compilers (compiler may have a bug(s) that only effect certain platforms and code doesn't compile correctly). Or in some cases software doesn't "like" the new compiler (won't compile or compile correctly). Finally, in the case of the c++ compiler, there was an ABI (application binary interface??) change going from 2.95.4 to 3.2 which means that all the c++ code needs to be recompiled. It would be great if all the software compiled cleanly, but that doesn't happen. So volunteers must resolve the problems which requires a lot of time and effort. The sheer volume of software and architectures that Debian supports is a big stress test for some of the fundamental software pieces such as compilers, X11, the kernel, glibc, etc. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RAID & EXT2-fs error
On Tuesday 21 January 2003 11:31 pm, Michael Kahle wrote: > > I am trying to install a Debian testing box with root on RAID > > 1 following the instructions given at > > http://karaolides.com/computing/HOWTO/lvmraid/ > > I am working on setting up a system with this configuration. I > decided to work step by step with this howto to get my system > running. Thanks for the link. > > > Everything > > works fine until I add the initial install disk into the RAID > > array (http://karaolides.com/computing/HOWTO/lvmraid/node26.html). > > I'm sorry I can't help you with your specific problem, I was hoping > that I could by working through this howto. But I'm afraid that I am > stuck as well. > > When I ran vgcreate it seamed to work fine. However it lists the MAX > LV Size as being 255.99GB. This does not make sense with my > configuration. I have 5 SCSI disk partitions in a RAID5 array. Each > partition is 36446.22mb. My math shows I should have a MAX LV size of > being 145784.88mb. By running vgdisplay I show that my VG size is > 169.71GB. What am I missing? How can your MAX LV Size be > 255.99gb?!? > > As soon as I get this ironed out I will report back with my findings. > I can't wait to get this thing running. Neat stuff. > > Michael Hi, Someone posted a howto they wrote that detailed Root-on-LVM-on-RAID HOWTO to debian-user. I am copying & pasting their post: I've written a doc on how to install a Debian with root file system over LVM and RAID. If it's of any interest it can be found at this URL: http://www.midhgard.it/docs/index_en.html Any suggestion/criticism/correction is welcome. Massimiliano This may be of interest. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Upgrading to KDE 3.1
On Tuesday 28 January 2003 12:39 pm, Jeetu Golani wrote: > Hey ppl, > > I have KDE 3.0.4 running at the moment, had installed the debs of > this. I wanna upgrade to KDE 3.1. I've read that an apt-get upgrade > doesn't do the trick and causes problems. Would appreciate if someone > here could tell me what's the right way to upgrade. > > Thanks Put this deb line in your /etc/apt/sources.list deb http://devel-home.kde.org/~nolden/kde stable main 1. Log out of KDE and use a plain ol' console window. 2. as root /etc/init.d/kdm stop 3. as root apt-get update 4. as root apt-get upgrade -s (tests to see what would be upgraded) 5. as root apt-get --purge remove libqt3 (this should remove all the other kde packages that depend on qt. doc stuff doesn't and will not be removed). 6. as root apt-get upgrade -s 7. See if #6 caused any problems with the removal of libqt3. 8. as root apt-get install arts kdebase kdelibs kdeaddons kdeaddons kdeadmin kdegraphics kdeedu kdenetwork qt3 etc (rest of kde packages) 9. If you run into a bind you can use as root dpkg --purge --force-all any conflicting kde packages that weren't removed with apt-get --purge remove libqt3. I had to do this because I didn't remove kde 3.0.5a first. I have kde3.1 installed on both a x86 and powerpc computers using the debs provided by Ralph and others. The new version of kde is just great. Note that for non-officially sanctioned deb packages such as available at Ralph Nolden's site it is best to remove all of kde and qt first before doing an install of kde3.1. He has a web page that contains some more detailed information: http://devel-home.kde.org/~nolden/kde/README He has koffice and other packages as well. All of the kde3.1 is compiled with gcc-2.95 (stable's compiler). Once the gcc-3.2 transition. Official debian packages will go into SID and be migrated to testing. If you run stable or testing Ralphs debs will work. Holler if you need help. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
compiling modules-scyld-source
Hi, I have a cheap RLT 8139 card that works with kernel-2.2.20 and the rtl8139 module, but it won't with the 2.4.20 kernel and the 8139too module (locks up the machine when I do an insmod 8139too). I saw that there is a modules-scyld-source (which I downloaded) that contains the rtl8139.c and a makefile. The makefile in question has some magic for determining the location of modversions.h and other needed include files that need to be modified. I am unsure of what is the proper way to setup the compilation enviroment. I have downloaded the kernel-headers-2.4.20 and the kernel-source-2.4.20 and need to figure how the Debian way for compiling the contents of modules-scyld-source. Do I need to recompile the kernel or can I just get away with compiling the modules that I am interested in. What is the proper way to set up the symlinks for /usr/src/linux? Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
compiling for different architectures
Hi, I have downloaded the source for kde 3.0.4 (debian version of things) and am in the process of compiling it on a powerpc sarge box. The precompiled debs are available only for i386. I use dpkg-buildpackage to build things and for the most part things go well. The only caveat is that in one particular application, kcmlilo, it should only be compiled for an i386 box. The debian/control file indicates this and doesn't compile, but the installation scripts, specifically dh_install fails because the kcmlilo.install file has some files listed to be installed but since the application is never built, the files don't exist. When I use dh_install -Xlilo, I get past the install problem, but dh_gencontrol has a problem, since it sees the i386 in the Architecture field of the control file. Not knowing what the proper way to handle this, I just edited the control file and removed all references to kcmlilo and dpkg-buildpackage went without a hitch. Now for the question, for multi-architecture builds, is it proper to edit the control file and delete entries. This just doesn't seem right, but I couldn't see any other way to get a successful build. This is the first time I have compiled from the debian source and want to find out the correct way to do things. Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: OT: Politics of Java
On Thursday 21 November 2002 10:42 am, Kent West wrote: > Derrick 'dman' Hudson wrote: > >On Wed, Nov 20, 2002 at 10:04:30PM -0800, Sean 'Shaleh' Perry wrote: > >| On Wednesday 20 November 2002 21:06, Kent West wrote: > >| > I'm just curious; do other folks (particularly real developers, > >| > not just tinkerer-wanna-be's like myself) have a similar problem > >| > with Java, or have I just been channeling too much RMS lately? > > > >I would avoid java if I wasn't require to use it in class and at > > work. Since I am required, use this apt sources line and install > > the 'j2sdk1.4' package. > >deb ftp://ftp.tux.org/pub/java/debian/ sid main non-free > > > >| If you are looking for an easy, fun, quick to learn language give > >| Python a try. > > > >I agree here. In fact, python is older, more cross-platform and > > more OO than java is. All it lacks is the marketing budget of Sun. > > > >-D > > Shaleh was the one to plug python first. But what dman says here gets > my attention. Do others concur that python is more cross-platform and > more OO than java? What are the disadvantages with python as opposed > to java? Why would someone pick java over python then? Is it only > because of marketing, as dman says? > > Kent I write c++ code for my job and don't have any experience with either java or python (other than nominally playing around with python). Python is developed in c, so if an architecture has a compliant c compiler, more than likely, you can get the python interpreter compiled and run your python scripts. Porting java to a new architecture is probably harder as evidence by the lack of willing partners to get the blackdown port of the latest (1.4??) running on the powerpc chip. Reasons for choosing one over the other: 1. Performance 2. Development enviroment -- toolchain support. 3. Availability of qualified developers that know the language -- certainly more java developers than python. 4. Notion of higher (or better) cross-platform support for the big 3 in OSes (Windows, UNIX/Linux, MacOS(X)) -- companies actually provide the support for these enviroments as opposed to volunteers in the open source community. 5. Library support (or the availability) for the given application. 6. Size of application, how well does the language scale for large code bases. Scripting languages don't as a rule scale -- you don't write tens of thousands lines of code in a scripting language -- although I have heard that python does scale much better than tcl for example. You will always hear someone say, but ". . oh I have written X number of lines of code in blah, and it was just fine. . ." Is that typical?? 7. Higher ups dictating the tool for the job without understanding all of the requirements -- pros, cons of using one tool over the other. 8. Marketing hype (or lack of). I am not a language expert to know if python is more or less OO than java. This depends on your definition of OO (which even defies an absolute definition from the language gurus). I would certainly think that python would work better for gluing different applications together than java, since scripting languages do a good job in this department. I also think that python would win for rapid prototyping some code idea. It might also be more fun to develop in python because of the rapid feedback from working in an interpretative enviroment. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
new installation -- install lilo after making boot floppy
Hi, I just installed GNU/Linux using the compact boot floppies onto my boss's windows 98 computer. He has 2 hard drives, the first, hda devoted to windows 98, and the second drive hdb devoted to linux. At the end of the installation, when it says make the system bootable, I decided to go ahead and make a boot floppy instead of overwriting the MBR with lilo (if that is the right terminology). I was afraid of making his machine unbootable with windows 98 and the boot floppy approach seemed to be the safe route. After making the boot floppy, I rebooted and finished the installation (I should note that booting from the floppy was painfully slow, is this normal?), and everything seemed fine. However, I would like to go ahead and use lilo and install things in the MBR of the first hard drive hda (now used by windows 98) and let lilo control the booting of either windows or linux. Is there an easy way to get lilo set up so that it can boot up either windows 98 or linux? Do I need to run just lilo or do I also have to run mbr-activate? I have read through the installation guide and have taken a look at the Manual.txt.gz in the lilo documentation. It seems that the install program makes setting up lilo fairly easy (during installation) and wondered if I should go back to the installer to make the system bootable, or do it from the command line after editing /etc/lilo.conf. Note, I have installed the compact flavor (i386) on a single drive system that was just running Debian and on a mac 68K and powerpc, but don't have any experience with lilo in a dual boot situation. I don't want to render my bosses windows 98 setup unusable if I can certainly help it. Any guidance would be very much appreciated. Thanks, John Schmidt [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: pon - Modem keeps redialing
Try this for the #modeminit: # modeminit '' 'ATZ' OK ATZ OK ATM0 the ATM0 turns off the sound completely. > # modeminit > '' ATZ I am not sure about the redialing thing. Don't know if it has anything to do with some ip6 hosts lookup??? Just taking a wild guess. I hope someone will correct me if I am wrong, but I added in my /etc/hosts file: 127.0.0.1 localhost 192.168.1.2 golden 192.168.1.1 retriever 192.168.1.3 lab # The following lines are desirable for IPv6 capable hosts # (added automatically by netbase upgrade) ::1 ip6-localhost ip6-loopback fe00::0 ip6-localnet ff00::0 ip6-mcastprefix ff02::1 ip6-allnodes ff02::2 ip6-allrouters ff02::3 ip6-allhosts :::127.0.0.1localhost :::192.168.1.2 golden :::192.168.1.1 retriever :::192.168.1.3 lab John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Help with login/environment: bash_profile, bashrc, xdm
Hi, I don't run either gdm or xdm, but do run kdm. For kdm, there is a file in /etc/kde3/kdm/Xsession. At the top of this file, it had: #!/bin/sh I changed this to #!/bin/bash --login and now when I fire up an xterm equilavents inside of kde, all the shells are login shells and my ~/.bash_profile is processed with my defined enviroment variables. You may be able to get away with editing /etc/X11/Xsession and doing the above change. There may be problems with making all shells start as login shells. I don't pretend to know what they may be and various pros/cons, but the above change works for me and gets my ~/.bash_profile read when I fire up kde. John On Monday 09 December 2002 10:25 pm, Paul Mackinney wrote: > Hi. I use the blackbox window manager, but I'm having trouble with my > default environment. > > I was using gdm and all was fine, except that >- My customizations to the login window kept getting bonked. >- If you ctrl-alt-f1 to get a non-X terminal, it keeps coming back >every n seconds without being asked. Especially annoying if your X >configuration is bad... I'm just not a GNOME fan. > > So I switched to xdm >- and my .bash_profile stopped getting processed. So that after >opening an xterm, none of my custom environment variables were >active, just the custom aliases in .bashrc. >- I tried putting 'source ./.bash_profile' at the top of my > xsession file, but no luck. > > Can someone tell me, or point me to the right docs, to get xdm to run > .bash_profile so that all term windows inherit the environment? > > And a more philosophical question: Anyone care to provide a 50 word > essay on the proper use of .bash_profile and .bashrc? I have an > environment that generally works, but it's based on random examples, > trial, and error rather than consistent application of fundamental > principles. > > TIA, Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
finding network activity
Hi, I have demand dialing turned on and would like to determine why ppp is starting up at 15 minute intervals. I am looking at syslog and not seeing anything that would initiate outside traffic. I also have tcpdump -i any turned on and not seeing anything either. I removed ntpdate and exim, since earlier I was seeing exim in the syslog. However, with exim and ntpdate gone, something is still causing pppd to start up. In addition, my web browser was on my home directory. Just for testing sake, I don't have my web browser running now. Are there any other network monitoring tools that may tell me what is initiating tcp connections and forcing pppd to start up. Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Fwd: Re: finding network activity
Hi, Replying to my own post, I turned on tcpdump -vvv -i any and saw that a windows95 box that is on my small internal network is the culprit. It is generating stuff like this: 19:39:24.082631 lab.netbios-dgm > 192.168.1.255.netbios-dgm: >>> NBT UDP PACKET(138) Res=0x1102 ID=0x668 IP=192 (0xc0).168 (0xa8).1 (0x1).3 (0x3) Port=138 (0x8a) Length=194 (0xc2) Res2=0x0 SourceName=DEFAULT NameType=0x00 (Workstation) DestName= WARNING: Short packet. Try increasing the snap length (ttl 128, id 12553, len 236) 19:39:24.082703 lab.netbios-ns > 192.168.1.255.netbios-ns: [udp sum ok] >>> NBT UDP PACKET(137): QUERY; REQUEST; BROADCAST TrnID=0x66C OpCode=0 NmFlags=0x11 Rcode=0 QueryCount=1 AnswerCount=0 AuthorityCount=0 AddressRecCount=0 QuestionRecords: Name=WORKGROUP NameType=0x1B (Domain Controller) QuestionType=0x20 QuestionClass=0x1 (ttl 128, id 12809, len 78) 19:39:24.834084 lab.netbios-ns > 192.168.1.255.netbios-ns: [udp sum ok] >>> NBT UDP PACKET(137): QUERY; REQUEST; BROADCAST TrnID=0x66C OpCode=0 NmFlags=0x11 Rcode=0 QueryCount=1 AnswerCount=0 AuthorityCount=0 AddressRecCount=0 QuestionRecords: Name=WORKGROUP NameType=0x1B (Domain Controller) QuestionType=0x20 QuestionClass=0x1 (ttl 128, id 13065, len 78) 19:39:25.585801 lab.netbios-ns > 192.168.1.255.netbios-ns: [udp sum ok] >>> NBT UDP PACKET(137): QUERY; REQUEST; BROADCAST TrnID=0x66C OpCode=0 NmFlags=0x11 Rcode=0 QueryCount=1 AnswerCount=0 AuthorityCount=0 AddressRecCount=0 QuestionRecords: Name=WORKGROUP NameType=0x1B (Domain Controller) QuestionType=0x20 QuestionClass=0x1 (ttl 128, id 13321, len 78) 19:39:26.341866 arp who-has golden tell lab 19:39:26.342067 arp reply golden is-at 0:5:9a:20:75:65 (0:40:f4:44:ed:2b) 19:39:26.342313 lab.netbios-ns > 209.210.176.9.domain: [udp sum ok] 1644+ A? WORKGROUP. . (34) (ttl 128, id 13577, len 62) 19:39:26.342585 67.2.152.75.netbios-ns > 209.210.176.9.domain: [udp sum ok] 1644+ A? WORKGROUP. . (34) (ttl 127, id 13577, len 62) 19:39:26.343701 67.2.152.75.33184 > 209.210.176.8.domain: [udp sum ok] 14911+ PTR? 75.152.2.67.in-addr.arpa. [|domain] (DF) (ttl 64, id 36174, len 70) 19:39:27.320572 67.2.152.75.33185 > 209.210.176.8.domain: [udp sum ok] 51493+ PTR? 75.152.2.67.in-addr.arpa. [|domain] (DF) (ttl 64, id 36272, len 70) The timestamp for when this query is starting up coincides to when my modem starts to dial. Is there any way to block this query. I am running ipmasq with a 2.4.19 kernel which uses iptables. I just have the default setup. I am ignorant about the rules that are used, but I guess this is a good time to learn about them and how to add one. Thanks, John Schmidt On Sunday 15 December 2002 12:45 pm, John Schmidt wrote: > Hi, > > I have demand dialing turned on and would like to determine why ppp > is starting up at 15 minute intervals. I am looking at syslog and > not seeing anything that would initiate outside traffic. I also have > tcpdump -i any turned on and not seeing anything either. I removed > ntpdate and exim, since earlier I was seeing exim in the syslog. > However, with exim and ntpdate gone, something is still causing pppd > to start up. In addition, my web browser was on my home directory. > Just for testing sake, I don't have my web browser running now. > > Are there any other network monitoring tools that may tell me what is > initiating tcp connections and forcing pppd to start up. > > Thanks, > > John Schmidt --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: intel 7.0 compilers; rpm malfunctioning
On Thursday 19 December 2002 01:08 am, Oleg wrote: > > P.P.S To preemptively answer the inevitable question why I want Intel > compilers in addition to GCC: I'm interested in high-performance > numeric computing, looking at the code Intel compilers generate, and > comparing different compilers' performance. Our university funded DOE ASCI center went through a similar comparison with a large C++ code. We actually had several Intel engineers bang away on the compilers trying to tweak compiler flags and the code itself. Even turning on special undocumented flags, and tweaking the code, the performance was no better than using gcc. In fact the optimizations they did to the code also helped gcc. We didn't find that the Intel compilers would yield any benefit. Your mileage will vary though. Good luck John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: dial on demand ... dials by itself !!
On Friday 03 January 2003 01:54 pm, Daves Debian wrote: > I have enabled dial on demand, it works great except sometimes > it dials up by itself. I assume by some program trying to access and > unknown ip address. > > How do I find out what programms are using my ppp link ? > ps ax | grep ppp > tells me ppp is active, but what caused it to activate and what > program is sending all this data ???!!! > > Many thanks for any help > Dave Dave, I had this problem too. I have a windows box that is connected to my little LAN and it was emanating a Windows "who-has" lookup about every 15 minutes forcing a dial-up. I discovered this by using tcpdump, e.g. tcpdump -vvv -i any and comparing the output of this with the logs in /var/log/syslog which would record the time of any dial-up. I posted this to debian-user, you can search the archives under the subject "finding network activity" John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ppp won't dial out
On Friday 30 August 2002 07:50 pm, Jorge Santos wrote: > Hello, the phone compay provides me with an answering machine of > sorts, such that when I have a message the phone, instead of giving > me the usual dial tone gives me something like a busy tone, well, it > is broken now, and I got this busy tone at all times, it shouldn't be > such big of a deal tho, because I can still dial out. > > Now, the problem is the modem refuses to dial out because it doesn't > get a dial tone. Any suggestions on how to make it dial out even > when there's no dial tone? > > Greetings, > > Jorge Santos If you edit /etc/chatscripts/provider and look where the aborts are: jas@golden:~$ more /etc/chatscripts/provider # This chatfile was generated by pppconfig 2.0.10. # Please do not delete any of the comments. Pppconfig needs them. # # ispauth PAP # abortstring ABORT BUSY ABORT 'NO CARRIER' ABORT VOICE ABORT 'NO DIALTONE' ABORT 'NO DIAL TONE' ABORT 'NO ANSWER' ABORT DELAYED # modeminit You should be able to remove the 'NO DIALTONE' from the above. Of course this assumes you are using pon/poff. I assume that other dialers would have similar configurations. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
gcc-3.2 on a testing box
Hi, I am running a testing box and would like to try out gcc-3.2 which is in unstable. I have created an apt-preferences file with the appropriate information and updated my sources.list to include unstable. As a check I ran: golden:/home/jas# apt-get install -s -t unstable gcc-3.2 Reading Package Lists... Done Building Dependency Tree... Done The following extra packages will be installed: binutils cpp-3.2 gcc-3.2-base libc6 libc6-dev libgcc1 locales The following packages will be REMOVED: modconf modutils The following NEW packages will be installed: cpp-3.2 gcc-3.2 gcc-3.2-base 5 packages upgraded, 3 newly installed, 2 to remove and 120 not upgraded. Remv modconf (0.2.43 Debian:testing) Remv modutils (2.4.19-3 Debian:unstable) Inst libc6-dev (2.2.5-14 Debian:unstable) [] Inst locales (2.2.5-14 Debian:unstable) [] Inst libc6 (2.2.5-14 Debian:unstable) Conf libc6 (2.2.5-14 Debian:unstable) Inst binutils (2.13.90.0.4-1 Debian:unstable) Inst gcc-3.2-base (1:3.2.1-0pre1 Debian:unstable) Inst cpp-3.2 (1:3.2.1-0pre1 Debian:unstable) Inst libgcc1 (1:3.2.1-0pre1 Debian:unstable) Inst gcc-3.2 (1:3.2.1-0pre1 Debian:unstable) Conf libc6-dev (2.2.5-14 Debian:unstable) Conf locales (2.2.5-14 Debian:unstable) Conf binutils (2.13.90.0.4-1 Debian:unstable) Conf gcc-3.2-base (1:3.2.1-0pre1 Debian:unstable) Conf cpp-3.2 (1:3.2.1-0pre1 Debian:unstable) Conf libgcc1 (1:3.2.1-0pre1 Debian:unstable) Conf gcc-3.2 (1:3.2.1-0pre1 Debian:unstable) I am nervous about the removal of modutils and modconf and the update of binutils libc6 libc6-dev libgcc1 locales. Will this mess up my current testing development enviroment or are the dependencies such that things will just work? As an aside, I am developing some code that may have triggered a g++ bug in 3.0.4 on a powerpc box and want to see if things are fixed in 3.2. I depend on this box to get my daily work done but don't want to mess it up. Thanks, John Schmidt [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: gcc-3.2 on a testing box
I installed the modutils from unstable and modconf from testing. Is the likelihood that modutils from unstable should be okay for a testing distribution? The kernel is based on the 2.2compact series from woody. On a different testing box, I am actually building gcc-3.2 from the debian source. Thanks for your input. John On Friday 06 September 2002 11:49 am, Sean 'Shaleh' Perry wrote: > On Friday 06 September 2002 06:48, John Schmidt wrote: > > Hi, > > > > I am nervous about the removal of modutils and modconf and the > > update of binutils libc6 libc6-dev libgcc1 locales. Will this mess > > up my current testing development enviroment or are the > > dependencies such that things will just work? > > no modutils/modconf means any kernel modules will become unhappy, > probably a bad thing. > > binutils should be safe to upgrade and is definately required. > > What you could do is compile the new gcc locally so it does not > require the libc upgrade. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
demand dialing with pppconfig
Hi, I use pppconfig to set up my modem for dialout. Everything works well if I use pon/poff from the command line. However, I would like to use demand dialing. I turned on demand dialing via pppconfig and also did a pon to start up pppd. I looked at /var/log/syslog to verify that pppd was on -- it was. An excerpt from my syslog shows that both my local and remote IP were set immediately without my modem dialing out: Sep 22 18:51:02 golden modprobe: modprobe: Can't locate module char-major-108 Sep 22 18:51:02 golden pppd[11431]: pppd 2.4.1 started by root, uid 0 Sep 22 18:51:02 golden pppd[11431]: Using interface ppp0 Sep 22 18:51:02 golden pppd[11431]: Couldn't set pass-filter in kernel: Invalid argument Sep 22 18:51:02 golden pppd[11431]: Cannot determine ethernet address for proxy ARP Sep 22 18:51:02 golden pppd[11431]: local IP address 10.64.64.64 Sep 22 18:51:02 golden pppd[11431]: remote IP address 10.112.112.112 Sep 22 18:51:26 golden kernel: Packet log: output DENY ppp0 PROTO=6 10.64.64.64:3817 128.110.42.7:22 L=60 S=0x00 I=50598 F=0x T=64 SYN (#5) Sep 22 18:51:29 golden kernel: Packet log: output DENY ppp0 PROTO=6 10.64.64.64:3817 128.110.42.7:22 L=60 S=0x00 I=50599 F=0x T=64 SYN (#5) Sep 22 18:51:35 golden kernel: Packet log: output DENY ppp0 PROTO=6 10.64.64.64:3817 128.110.42.7:22 L=60 S=0x00 I=50600 F=0x T=64 SYN (#5) Sep 22 18:51:47 golden kernel: Packet log: output DENY ppp0 PROTO=6 10.64.64.64:3817 128.110.42.7:22 L=60 S=0x00 I=50601 F=0x T=64 SYN (#5) I also have ipmasq installed on this machine. I only use the firewalling rules that come as default. In the above snippet, I tried to access a machine at 128.110.42.7, but am denied. Or I should say that my modem is not dialing out and setting up "legitimate" local and remote IP that I would normally see with pppd running via manual pon/poff sequences. A couple of questions: 1. Should the local and remote IPs be set as they are when pppd is started with demand dialing in effect? 2. Is ipmasq somehow messing things up with demand dialing such that some configureation needs to be adjusted for the two to play along together? 3. Should I also set: /proc/sys/net/ipv4/ip_forward and /proc/sys/net/ipv4/ip_dynaddr to 1? Another thing I observed when I was playing around with pppconfig, that when I went to change a setting back (turn off demand dialing), the file in /etc/ppp/peers/provider was not edited correctly, in that the changes were at the bottom of the file. A quick edit of the file got things working with the manual pon/poff sequence. Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: gcc 3.2 & kernels compiled with 2.95
No personal experience here, but I read on the linux kernel mailing list that if you need to compile both the "main" kernel and any modules with the same compiler. The c++ ABI has changed from 2.95 to 3.x series, which means any c++ application that you compile with a 3.x compiler will not be able to work with a c++ library that is compiled with 2.95 (and visa versa). I believe there may have been an ABI change from 3.0.4 and 3.2. I am pretty sure that any c or fortran code should be okay with respect to using code from either compiler version with the exception of the kernel and modules as stated above. If any of this is wrong please correct me. John Schmidt On Friday 27 September 2002 12:12 am, Andy Saxena wrote: > Hi, > > I don't quite understand the complexity of the process to move Debian > to gcc 3.2. It seems there's an issue about programs compiled with > 3.2 not being able to work with libraries compiled with 2.95. Without > going into too much technical details, could somebody tell me if this > is correct? > > I was just wondering if this transition is going to cause trouble for > those of us who compile our own kernels using gcc 2.95. > > Thanks, > Andy -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: KDE2 helloworld.cpp
On Tuesday 01 October 2002 02:54 am, John Batistic wrote: > INCLUDES= -I/usr/share/qt/include -I/etc/kde2/include Are you absolutely positive that the qt includes are in /usr/share/qt/include?? On my system, I have them in /usr/include/qt: jas@golden:~$ locate qlabel.h /usr/include/qt/qlabel.h jas@golden:~$ locate qlabel.h /usr/include/qt/qlabel.h jas@golden:~$ locate qstring.h /usr/include/qt/qstring.h John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Debian and Dell?
On Wednesday 15 December 2004 08:30 pm, Ed Sutherland wrote: > Hi All, > > I just purchased a Dell for my home office and am interested in using > Debian on a partition. As my only Linux experience comes from a Mac, I > have some questions: > > 1) Will I be able to easily dual-boot Windows or Linux using yaboot, or > will I need to go through some BIOS mumbo-jumbo? > > 2) Does Debian support the Dell flatscreen monitors? > > 3) Does the i86 side of Debian better support Web graphics and animation > formats -- such as shockwave? > > Thanks for the answers. > > Ed You might want to check out: http://linux.dell.com/ They do have a section about Debian (under Distributions). Answers: 1. Should be able to using either lilo or grub bootloaders (same concept as yaboot) to select which OS you want to boot. 2. More than likely, but you should probably check the above website. 3. Yes but that has more to do with outside parties that provide the non-free binaries such as shockwave. Get the latest Sarge installer such as the netinst CD image for i386 found on: http://www.debian.org/devel/debian-installer/ After installation, submit an installer report even if you don't have any problems. Take a look at the documentation on the debian-installer web page for help. Specific installer problems can be directed to the debian-boot mailing list. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Simple Email Server?
On Wednesday 15 December 2004 09:44 am, David Baron wrote: > On Wednesday 15 December 2004 16:48, > > [EMAIL PROTECTED] wrote: > > > What need I set to get exim4 to receive messages to my domain (dynamic) > > > directly? > > > > Basically just 'apt-get install exim4'. > > > > Have a look at this article: > > http://www.trekweb.com/~jasonb/articles/exim4_courier/exim4.html > > (It's a bit of text but really very simple) > > I have exim4 working fine. I now get email through fetchmail from > providers, use procmail to check with spamassassin and clamav, have system > filter, etc., etc. Outgoing mail is handled by exim4 smarthosting to my > provider (or directly to the provider from kmail). > > The article is quite good. A static domain is simple. > > I have a dynamic IP. This is what I need to configure. You might want to look at a dynamic dns outfit such as no-ip.com, dyndns.org, etc. You can sign up for a domain and use something like ddclient (in Debian) to send your new ip number to the dns company you are using. You should be able to configure exim to use your domain name as registered even with a dynamic ip. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
module-assistant no GUI on startup
Hi, I have just upgraded module-assistant to version 0.7 and the default GUI does not start when I do: sudo module-assistant Does anyone else see this? I looked at the changelog.gz in the documentation and didn't see anything that relevant to this. Thanks, John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: module-assistant no GUI on startup
On Tuesday 04 January 2005 09:37 am, Hanspeter Kunz wrote: > On Tue, 2005-01-04 at 09:30 -0700, John Schmidt wrote: > > Hi, > > > > I have just upgraded module-assistant to version 0.7 and the default GUI > > does not start when I do: > > > > sudo module-assistant > > > > Does anyone else see this? I looked at the changelog.gz in the > > documentation and didn't see anything that relevant to this. > > yes, I noticed this too. Use > > module-assistant select > > cheers, > Hp. Thanks for the pointer. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Debian on an old PC
On Thursday 06 January 2005 01:42 pm, Vegard Lundby Rekaa wrote: > > I would be surprised if that old machine could *boot a cdrom*. Have > > > > you tried *that*? > > > > ;-0 > > It can't boot from cdrom, it has one (4X) but the system doesn't know of > it untill an OS is loaded (i.e. win98se). I'm forced to install linux with > floppy disks. Try using this: From Debian-boot mailing list: Smart Boot Manager (http://btmgr.sourceforge.net/) can be started from floppy and allows some older computers to boot from CD. That's how I loaded my firewall/router machine with Sarge. Might be worth a try. --Don John
Re: Debian on an old PC
On Thursday 06 January 2005 02:37 pm, David Jardine wrote: > On Thu, Jan 06, 2005 at 09:42:58PM +0100, Vegard Lundby Rekaa wrote: > > > I would be surprised if that old machine could *boot a cdrom*. Have > > > > > > you tried *that*? > > > > > > ;-0 > > > > It can't boot from cdrom, it has one (4X) but the system doesn't know of > > it untill an OS is loaded (i.e. win98se). I'm forced to install linux > > with floppy disks. > > Can't you change the boot order in the BIOS setup to look for > cdrom first? Debian install CDs are bootable, surely? > > David > > -- > David Jardine > > "Running Debian GNU/Linux and > loving every minute of it." -Sacher M. Old pcs often can't boot from a CD even if they have one. You might be able to flash the BIOS to upgrade it, but that assumes there is an update out there (highly unlikely). John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Starting wlan0 instead of eth0
On Thursday 26 May 2005 01:39 am, Jason Edson wrote: > I'm new to this part of my debian system. Instead of having eth0 load > up when my system starts, I want to run a script that enables my > wireless card(wlan0) that uses ndiswrapper to load. My script looks > like this: > > #!/bin/bash > echo "Starting wireless networking." > ifconfig eth0 down > modprobe ndiswrapper > sleep 1 > iwconfig wlan0 mode Managed > iwconfig wlan0 essid cocksmoker > iwconfig wlan0 key blahblahblahmynetkey > dhcpcd -t 10 -d -G 192.168.1.1 wlan0 > echo "Wireless networking started.hopefully!" > > I don't know how to achive this so any pointers or links for some > reading would be very appreciated. Thanks for your time. Take a look at the man page for interfaces. I think what you want to do is to make sure you have a line in the /etc/network/interfaces that looks something like this: auto wlan0 This should be near the top of the file. In addition, you can then do something like this for your iwconfig stuff show above (this is also in the file /etc/network/interfaces) : iface wlan0 inet dhcp pre-up modprobe ndiswrapper pre-up iwconfig wlan0 essid cocksmoker pre-up iwconfig wlan0 key blahblahblahmynetkey post-down rmmod ndiswrapper John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: adduser problem
On Sunday 10 October 2004 09:45 pm, Jim Hall wrote: > Greetings, > > I volunteer in my church's computer lab (hence my email name) and > advocate Linux where and whenever possible. We have a sarge system as a > test system. A very knowledgeable Debian expert (call him Buddy) helped > me set it up. I started getting used to it, especially apt, because I've > been using RedHat since I started using Linux. > > About a year ago, my local LUG set up a small weekly newspaper with > Linux. The workstations are dual boot W98/RedHat 7.3 (don't ask, long > story), the server is Debian (named debbie). Buddy set it up. He is > very, very good, but had to do some customization (maybe a lot). After > doing updates, upgrades, and dist-upgrades on the lab test system with > no apparent ill effects, I thought it was time to do the same to the > newspapers server. In the process I broke 'adduser'. So far, the > solution appears to involve LDAP. None of us seems to have much > experience with that. Buddy has since moved and isn't responding to emails. > > The newspaper has been working on a series of articles about the > switchover from Windoze to Linux. Right now they're not very happy with > Debian. Is there anyone who offers paid, secure support for Debian? > Donation to the Debian project is also acceptable. Remote access can be > provided. > > I don't know if this is the correct thing to do, but I'm totally lost > and don't know what else to do. I haven't included any specifics because > I don't know what would be needed or if this list is the right place to > post this request. > > Thanks, > Jim Jim, There is a Debian package call cpu which can do adduser type of operations using a LDAP backend. Check out apt-cache show cpu. For example to add a user to a LDAP system, you can do something like this as superuser (root): #> cpu-useradd testuser John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Keyed SSH login problem
On Monday 11 October 2004 10:43 am, Stephen Tait wrote: > I'm having a great deal of difficulty setting up two computers to log into > one another for automated backup purposes. For the moment, I'm just trying > to get one machine to log into the other non-interactively, and since it's > over the internet I was going to use SSH. > > Generated a v2 DSA public/private keypair on host1 under > /home/sync/.ssh/sync-host1 and sync-host1.pub > SCP'd the public key over to host2 > Added the .pub to ~/.ssh/authorized_keys2 (and authorized_keys FWIW) > > Now when I try and SSH from host1 with it (please note, names have been > changed to protect the innocent); I just went through this yesterday, and here is my recipe. On machine 1: 1. Create your public/private key (I used dsa): ssh-keygen -t dsa 2. Copy the contents of ~/.ssh/id_dsa.pub to machine 2 using ssh-copy-id: ssh-copy-id -i ~/.ssh/id_dsa.pub machine2 ssh-copy-id copies the id_dsa.pub file located in your .ssh/ to machine2 using ssh. It sets up the authorized_key file and permissions appropriately. Once it is copied over, then you should be able to ssh into machine2 from machine1 without typing in a password. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: trouble with pam-ldap
On Tuesday 12 October 2004 02:09 pm, Jeremy Brown wrote: > I'm trying to get a Debian sarge machine to authenticate against an > OpenLDAP server (running on the same box) with no success. > Take a look at: http://people.debian.org/~torsten/ldapnss.html John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: routing help please...
On Wednesday 13 October 2004 05:45 am, Sebastiaan wrote: > Hi, > > On Wed, 13 Oct 2004, Mark Maas wrote: > > Thanks for reading! > > > > I hope someone can help me with a routing issue: > > > > I've attached a situation scetch. > > > > The thing is, my road warriors connect via a pptp connection > > to my VPN server via GW2. > > This fails because the default gateway (GW) on the VPN is > > GW1, so all request attempts end up beeing sent through GW2 > > but answered by my VPN server to GW1. > > > > But as soon as I put GW2 as default gateway, My local lan's > > do not get routed as they should anymore... > > > > So all PPP connections should have GW2 as the default > > gateway, and set up routes so data between the local lans > > and ppp connections should go through GW1. > > > > But how? > > Thanks for any pointers... > > > > Mark > > Hmm, usually a host only recognizes replies from the same remote host e.g. > nic. You can probably do something with the package 'iproute', but that's > as far as I can point you. > > Greetz, > Sebas > > Mark, In the package vpnc, there is an option to specify custom routes (taken from the man page): Custom route setting By default, the default route is deleted after connection and replaced with the new one (going trough the VPN tunnel device). However, some people wish to limit the target address range to few IP ranges. This can be done using the config directive Target networks in the config file. For example: Target networks 123.234.210.0/24 10.1.0.0/16 In the file vpnc-connect they use ip directives to specify particular routes. You may want to install this package and take a look at how they specify the routing. Mimic what they do in vpnc-connect as far as routing goes and see if that doesn't solve your problem. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Problem with DSL and /etc/resolv.conf
On Friday 08 October 2004 04:21 pm, Andrew Carter wrote: > I'm having a problem with my /etc/resolv.conf file and I'm hoping > someone might be able to explain what is going on. My home network > has an ActionTec wireless DSL modem/router running as the DHCP. Qwest > is my DSL provider. I have three Macintoshes, an XP box, and a > Debian/FreeBSD dual boot box. The Debian and FreeBSD box both have > the same problem. The Macs and XP don't. > > When the computer boots up and gets DHCP information, it puts two > addresses in the /etc/resolv.conf file for nameservers. One is the > router address (192.168.0.1) and the other is one of the two ISP DNS > servers. This causes any internet traffic to go unresolved. If I > change the entry in /etc/resolv.conf that contains the router address > to the other ISP DNS server, everything works just great. However, > the next time I boot up the machine, it goes back to the router+ISP > address. > > Since Mac OSX is based on FreeBSD, I looked at what its > /etc/resolv.conf file contains. It has the same thing but works just > fine. > > So, I have three questions: > > 1. Should my router's ip address be in the resolv.conf file? > 2. Shouldn't my router just forward DNS requests to the ISP DNS servers? > 3. How can I keep the resolv.conf file from changing each time I reboot? > 4. Why does this fail for Linux/FreeBSD but not for OS X or XP? > > Thanks, > Andrew I found that dhcp3-client would not overwrite whatever was in my resolv.conf whereas dhcp-client would overwrite my entries in resolv.conf. I suggest using dhcp3-client and editing the entries in resolv.conf to whatever works and then rebooting to see if those changes persist. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: WiFi
On Monday 25 October 2004 03:43 pm, Alvin Oga wrote: > On Mon, 25 Oct 2004, Gilbert, Joseph wrote: > > http://www.linuxant.com/driverloader/ > > > > I haven't had a chance to work with it yet. Has anyone else done > > anything with this driver wrapper? > > http://ndiswrapper.sourceforge.net does the same thing and is free/GPL'd > whereas, linuxant is you have to pay or else > > you'd need to know make/install or apt-get > - or - > pay linuxant $20 for not wanting to do apt-get :-) > > - problem is bug fixes and if it doesn't work "right" ... > oh boy .. now what .. humm .. file a bug report, try to patch it, > buy a another wifi card that is supported natively on linux > > c ya > alvin Hi, I found that this was a better source for ndiswrapper than the ndiswrapper.sourceforge.net. deb http://rigtorp.se/debian/ unstable/ John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Building binaries for older versions of libc6
On Wednesday 27 October 2004 09:47 am, Karl E. Jorgensen wrote: > On Wed, Oct 20, 2004 at 02:58:16PM +0100, Ben Hutchings wrote: > > I have a C++ program which requires g++ 3.4 to build due to parser bugs > > in older versions of g++. I'm currently building on a development > > machine running mostly woody with some packages from sarge, including > > g++-3.4 of course. This requires version 2.3.2 of libc6 itself, and any > > binaries I build with it appear to require version 2.3. Now, while I > > prefer to do development on Debian, I need to build binaries that will > > run on Red Hat Enterprise Linux 2.1, which has version 2.2.4 of libc6. > > > > I'm wondering whether it's possible to build binaries with g++ 3.4 that > > will require only libc6 2.2, and if so, how. If I remove g++-3.4, > > downgrade to woody and then build and install g++ from source, is that > > likely to work? > > You're probably better off doing builds in a chroot environment - that > will allow you to play with the environment in a safe with without > destabilising your installation. > > Have a look at the pbuilder package - altough it is targetted towards > building debian packages (which you probably want to do anyway), it is > handy to keep multiple chroot environments, e.g. one for woody, one for > sarge and one for whatever.. > > Hope this helps In addition, you can install dchroot and switch between various chroot enviroments. If you modify in your chroot enviroment, the ~/.bash_profile (make sure the debian_chroot is defined) and put a line describing the chroot in /etc/debian_chroot, you can have your prompt tell you what chroot enviroment you are in. Handy if you have multiple chroot enviroments. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: List of packages
On Wednesday 27 October 2004 01:36 pm, Jim Hall wrote: > on Sarge, is there a way to list every installed package? I don't think > I need things like libs, just the package names. I need to compare two > systems. > > Jim dpkg --get-selections > machine1.pkgs John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: different networks, same computer (map mapping ?)
On Monday 01 November 2004 03:19 pm, H. S. wrote: > I would like a laptop to work in two kinds of networks automatically if > possible. At home, I am running a DHCP server and if the laptop is > connected to my switch(CAT5 cable to eth0) and booted up, it looks for > and gets an IP address (it is running a dhcp client). Now when that > laptop is taken to the university, the user needs to change the > /etc/network/interfaces file to give the machine a static address. I am > looking for a way that this choice between dhcp/static happens > automatically. > > I have been reading manpages of interfaces and learned we can map a > physical device as logical devices and make it work in different modes. > I am looking for examples where this is already done, the documentation > in man interfaces is, well, not very clear about all the nuts and > bolts, or so I believe. All help is appreciated. > > Thanks, > ->HS I have a similar setup. I used ifplugd and guessnet. You have to modify slightly /etc/network/interfaces. I supposed you don't need ifplugd, but it is nice to be able to start up my laptop quickly without having any type of network cable installed. Once you plug a cable in, ifplugd will configure things for you. Anyway, here are some snippets from my /etc/network/interfaces file: # The primary network interface # turn off since ifplugd is controlling things #auto eth0 mapping eth0 script /usr/sbin/guessnet-ifupdown map default: none map timeout: 3 map verbose: true iface work inet dhcp test-peer address x.x.x.x MAC.address iface home inet static address 192.168.1.5 netmask 255.255.255.0 network 192.168.1.0 broadcast 192.168.1.255 gateway 192.168.1.1 test-peer address 192.168.1.1 MAC.address Note: x.x.x.x is the IP address of a known computer on the network that should be always there. The MAC.address is the hardware MAC address of this computer. Pretty simple setup but works nicely. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sarge install problems with Adaptec 7902 controller
On Thursday 04 November 2004 09:41 am, Raj Kiran Grandhi wrote: > Hi, > > I am trying to install sarge on a HP Workstation with a Seagate SCSI > disk on an Adaptec 7902 Ultra 320 SCSI adapter. However, the installer > fails to detect the hard disk at all. There was no such problem when > installing Redhat 9. Redhat loads the module aic79xx.o before the > install program begins. During sarge install, lsmod shows that the > module aic79xx.o is loaded, but still the hard disk is not detected. I > am stuck as to what is to be done. Any pointers would be greatly > helpful. > > Thanks and regards, > > Raj Kiran I believe there is a bug in the aic79xx.o module for kernel-image-2.6.8-1. I assumed you tried to use linux26 during the sarge install. You can use the 2.4 series kernel during the install. At the boot: just hit the enter key or manually type linux and it will install a 2.4.27 kernel. 2.6.9 kernel was just uploaded into unstable. You can try installing that kernel once your initial install with a 2.4 kernel is done. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: moving from woody to sarge
On Thursday 11 November 2004 10:19 am, linux wrote: > sorry, should have explicitly stated that one reason for the move from > woody to sarge is that the former doesn't recognise my Ethernet card so I > cannot do installations over the network > - Original Message - > From: "Jerome BENOIT" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Sent: Thursday, November 11, 2004 5:14 PM > Subject: Re: moving from woody to sarge > > >A sound idea is to read > > > > http://www.debian.org/doc/manuals/reference/ch-woody.en.html > > > > first. > > > > hth, > > Jerome > > > > [EMAIL PROTECTED] wrote: > >> okay, I'm going to take the plunge and forget about using ''woody'' & > >> trying to upgrade the kernel (too many dependencies issues) and go str > >> for ''sarge''... so, I presume I might as well just wipe the current > >> (non-WinXP!) partitions and start from scratch -- does that make sense? > > > > -- > > Dr. Jerome BENOIT > > room A2-26 > > Complexo Interdisciplinar da U. L. > > Av. Prof. Gama Pinto, 2 > > P-1649-003 Lisboa, Portugal > > email: [EMAIL PROTECTED] or [EMAIL PROTECTED] It is probably easiest to just get somehow a netinstall CD for Sarge either a pre-rc2 or a daily build if pre-rc2 fails in someway for you. I have used the pre-rc2 to install on new and old hardware quite recently and found it quite nice and easy to setup. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Error inserting ndiswrapper
On Wednesday 17 November 2004 10:59 pm, nornagon wrote: > I'm using the stock debian kernel 2.6.7-1-k7, to get that out of the way. > > I need to use ndiswrapper, since my card isn't yet supported by prism54. > > Okay, so I compiled ndiswrapper. Then, I modprobe it. > > Output: > $ sudo modprobe ndiswrapper > FATAL: Error inserting ndiswrapper > (/lib/modules/2.6.9-1-k7/misc/ndiswrapper.ko): Unknown symbol in > module, or unknown parameter (see dmesg) > > Relevant bits of dmesg: > ndiswrapper: disagrees about version of symbol per_cpu__softnet_data > ndiswrapper: Unknown symbol per_cpu__softnet_data > ndiswrapper: disagrees about version of symbol unregister_netdev > ndiswrapper: Unknown symbol unregister_netdev > ndiswrapper: disagrees about version of symbol eth_type_trans > ndiswrapper: Unknown symbol eth_type_trans > ndiswrapper: disagrees about version of symbol skb_over_panic > ndiswrapper: Unknown symbol skb_over_panic > ndiswrapper: disagrees about version of symbol register_netdev > ndiswrapper: Unknown symbol register_netdev > ndiswrapper: disagrees about version of symbol alloc_skb > ndiswrapper: Unknown symbol alloc_skb > ndiswrapper: disagrees about version of symbol netif_rx > ndiswrapper: Unknown symbol netif_rx > ndiswrapper: disagrees about version of symbol __netdev_watchdog_up > ndiswrapper: Unknown symbol __netdev_watchdog_up > ndiswrapper: disagrees about version of symbol skb_copy_and_csum_dev > ndiswrapper: Unknown symbol skb_copy_and_csum_dev > ndiswrapper: disagrees about version of symbol alloc_etherdev > ndiswrapper: Unknown symbol alloc_etherdev > ndiswrapper: disagrees about version of symbol __kfree_skb > ndiswrapper: Unknown symbol __kfree_skb > > Any more necessary information I'd be more than happy to provide. > > Thanks in advance. > > P.S. Sorry if this is a double post, wasn't sure if it went through. > > -- > - nornagon > http://www.nornrock.com Did you do this: 2. Install your windows driver -- Download the Windows XP drivers, unpack it and locate the .inf for your card. Run ndiswrapper -i to install the driver > ndiswrapper -i This copies all necessary files to /etc/ndiswrapper and creates the config files for your card. After installing you can run > ndiswrapper -l to see the status of your installed drivers. If you have installed the correct driver you should see something like this: Installed ndis drivers: bcmwl5 present Where "present" means that you have a PCI-device present that can be used with the driver bcmwl5. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Newbie looking for some answers please...........
On Thursday 18 November 2004 10:17 am, Chad wrote: > I just installed Debian 3.0 r3. I'm a newbie and looking for some > anwers to some of my questions...if someone can anwser one, some, or > all Please > > 1. I know that apt-get is the main utility to add and remove programs > (in Debian anyways), also to veiw what is installed on your OS. But > what about other packages or applications that are not installed > through apt-get. Is there a another utility to tell you want is all > installed on your OS, or to keep track of all > software/packages/applications installed? > > 2. How do you check for all running services and how to start/stop > system services that are unused? > > 3. How do you check for all open ports and what programs are using the > ports. > > 4. What is the common folder Where most software/packages/applications > installed into? > > 5. Anyone has a good site for descriptions of the configuration files > on a linux system. For Example XF86Config-4. I have no idea of what > configuration files do what or where they are located. > > 6. Where is the boot files? So I can control or know what programs > start at boot. Chad, You might want to install debian-reference-en and quick-reference-en. Then point your web browser to /usr/share/doc/Debian and you should see folders for quick-reference and reference. They should provide you with answers to your questions plus provide you with more tips and other bits of useful information. apt-get install debian-reference-en quick-reference-en will get you going. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: alternatives to NIS and NFS
On Monday 02 August 2004 04:24 pm, Paul William wrote: > Hi, > > I am in charge of a small office network. The server is running Debian > stable with some testing packages and the desktops are running mandrake > 10.0. > > Currently we are using NIS for authentication and NFS to share the home > directories. > > I have been having some hassles with NIS and would like to upgrade to a > more modern system. > > Are there any alternatives to NFS and NIS? > > As long as it not to complex to setup and is fairly easy to administer > its fine. X is not on the server so all admin takes place over > ssh.Security is an issue. > > It is safe to assume that there will not be any windows clients on the > network, ever :) There is one osx ibook being used but it does not need > to 'login' to the network. > > Thanks very much, > > Paul Paul, I have a ldap based authentication system and still use NFS for sharing /home. I used http://people.debian.org/~torsten/ldapnss.html to guide me through the setup. I then use cpu (in Debian) to do the passwd and user account creation. Everything I needed was packaged in Debian. There are some ldap utilities that will convert an exisitng /etc/passwd to what is needed for ldap. I didn't have many users and found it easier to just use cpu to add my users and passwords. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to upgrade to 2.4.23 (or .18) kernel
On Friday 23 January 2004 02:01 pm, brfg3 at yahoo wrote: > I have Debian3 rc2. After installing from the CD, the command uname -r > shows a 2.2 kernel. I want to use a 2.4 kernel for USB and IPTables > support. I made two attemps to install a debian kernel image from two > different tutorials and both times kernel did not find the standard netgear > 311 nic card. It works in 2.2, though. I changed the nic to a linksys, and > then a dlink (both cards are detected by suse 7.0 - 9.0 and Red Hat 7.1 - > 9.0), and still got the same result. This leads me to the conclusion that > there must be something I missed. Is there some other undocumented or > little-known step to making a the standard Debian 2.4.18 kernel detect my > NIC? ? Is there some way to install a 2.4.18 kernel while installing from > the cd, and maybe avoid this hassle? > > TIA ! > > > > You may want to try to install discover. apt-get install discover apt-cache show discover will provide you with some info about what it does hardware dection wise. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Configuring Courier
On Wednesday 01 September 2004 01:36 pm, Upayavira wrote: > I want to set up a Debian based mail server, with SMTP sending, IMAP and > webmail. I am trying to do this with the Courier package. Actually, what > I want, at the moment, is a self contained mail system, it doesn't have > to deliver to the Internet. I'm using Sarge. > > I am having trouble getting mailboxes configured. When I send myself a > mail (mail -s "hiya" upayavira), it creates a file called Maildir in my > home directory. It looks rather like a 'mbox' file. If I delete that > file, and create a proper Maildir with maildirmake ~/Maildir, then try > sending again, I get "./Maildir: No such file or directory: No such file > or directory" showing in my /var/log/syslog file. > > Similarly, when I have a ~/Maildir directory and log into the courier > webmail system, I get "Internal error (module sqconfig.c, line 77) - > contact system administrator". When I do it with an mbox file called > ~/Maildir, I get "Unable to open the maildir for this account -- the > maildir doesn't exist or has incorrect ownership or permissions". > > I suspect I'm very close, but I can't figure it out. > > So, can anyone either (a) help me fix the above or (b) tell me where I > can get help fixing the above or (c) recommend a SMTP/IMAP/Webmail setup > that is relatively easy to configure, and that can be installed with > apt-get from Sarge? > > Thanks in advance for any help. > > Regards, > > Upayavira This may be helpful to you: http://talk.trekweb.com/~jasonb/articles/exim_maildir_imap.html John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
dpkg-dist
Hi, I have some configuration files in my /etc/pam.d that have the dpkg-dist suffix. Is is safe to copy the dpkg-dist version over the current version? Once I copy over the dpkg-dist version and rename the current version, would this confuse the package manager? Thanks, John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: creating swap partition
On Tuesday 24 February 2004 07:49 pm, xucaen wrote: > Hi all, quick question here. how big should a swap partition be relative to > the size of the drive? > > another quickie.. does each hard drive need it's own swap partition? > > thanks! > > Jim A good rule of thumb is that the size of swap should be twice the size of RAM. Some don't like to make swap that big. It really depends on the applications that you will run, if they are memory intensive and there is the possibility of using all the RAM, having more swap space is better than having less. However, with cheap memory prices, it is probably better to buy more RAM if you are doing memory intensive computations. You don't have to put swap on each drive, however there are some advantages to splitting the swap up over several drives. I haven't experimented with this to find out, but that is what I have read. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ext3 why?
On Tuesday 24 February 2004 08:10 pm, xucaen wrote: > On Tue, Feb 24, 2004 at 06:43:02PM -0800, Alvin Oga wrote: > > On Tue, 24 Feb 2004, xucaen wrote: > > > hi all, just looking for information here; why use ext3 journaling fs > > > instead of ext2? > > > > copy all your important *.mp3 and *.mpeg files onto > > your 100GB disk partitioned as ext2 .. as if that was > > your only copy of it > > > > now power off your system ... and reboot > > > > and if you're smart, you'd answer "y".. please check the fs when > > it reboots from a power failure/reset switch > > > > > > than do the same test with ext3, reiserfs, jfs, etc > > where "journeling" is supposedly working right > > > > c ya > > alvin > > I don't get it are you saying ext3 recovers better or worse? > > jim ext3 is a journaling file system, ext2 is not. If you shut off the power to a machine that is running ext2, you will have to go thru the "equivalent" of scan disk to get things back in order. With a ext3 system, the journal takes care of this for you. Greatly simplified explanation, but that is the gist. ext2 will take long time to boot up after power cycling a machine. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: nss-ldap and different shells?
On Monday 23 February 2004 06:51 pm, Kirk Strauser wrote: > Sorry for the strange subject, but I wasn't sure how to make this question > pithy. > > I have several hosts that authenticate off the same OpenLDAP server using > nss-ldap and pam-ldap. The problem is that not every shell that users may > want is installed on every single host. Some of the hosts aren't > Linux-based, and have the shells in various directories other than /bin. > > Short of installing every shell on every host and coming up with a > canonical list of paths (which is a highly unpleasant prospect when looking > at networks with lots of machines from different vendors), is there a way > to get nss_ldap to "rewrite" the 'loginShell' attribute returned by the > LDAP server to something reasonable on the local server? > > For example, if the user's configured shell is /bin/bash, and bash lives in > /usr/local/bin/bash on the local system, then I'd like a re-writing rule > similar to: > > map loginShells: /bin/bash => /usr/local/bin/bash > > Likewise, if the user wants /bin/zsh, but it isn't installed, then I'd like > to substitute bash with a rule like: > > map loginShells: /bin/zsh => /bin/bash > > Even better would be an "alternatives-like" list that could be shared > across machines, like: > > when loginShells == /bin/bash try: > /bin/bash > /usr/local/bin/bash > /bin/sh > > when loginShells == /bin/zsh try: > /bin/zsh > /usr/local/bin/zsh > /usr/ksh > /usr/local/bin/ksh > /bin/sh > > so that, given a particular value of 'loginShells', the first available > entry is executed. I'm reasonably sure that this ideal solution doesn't > exist, but I've been pleasantly surprised before. > > How have people in this situation managed this problem? Check out this link: www.cit.gu.edu.au/~anthony/info/apps/LDAP_unix.notes John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Install debian on raid system
On Friday 27 February 2004 08:21 am, [EMAIL PROTECTED] wrote: > Hello everybody, > > I'm trying to install debian on a raid system (Promise FastTrak s150 sx > with 3 SATA drives), I downloaded the netinstall iso and tried installing. > Unfortunately the drivers for the raid card aren't there and I don't seem > te be able to find them anywhere on the net (except for RH and SUSE). > Could anyone please help as I'm completely stuck right now? > Thanks! > > Ron > > PS. Could you please cc any replies to me? You might want to check out this site: http://oregonstate.edu/~kveton/debian/ It has instructions for making a custom boot cd with the drivers you need. You might also get lucky with the iso that is shown on this page as well. A kernel config is shown which might be useful thing to check out. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: A Newbie LVM Question
On Thursday 04 March 2004 10:38 am, stan wrote: > On Thu, Mar 04, 2004 at 05:09:02PM +0200, Alexei Chetroi wrote: > > On Thu, Mar 04, 2004 at 08:13:18AM -0500, stan wrote: > > > Date: Thu, 4 Mar 2004 08:13:18 -0500 > > > From: stan <[EMAIL PROTECTED]> > > > To: [EMAIL PROTECTED] > > > Subject: Re: A Newbie LVM Question > > > > > > On Wed, Mar 03, 2004 at 02:42:30PM -0500, stan wrote: > > > > On Wed, Mar 03, 2004 at 09:37:06AM +0100, Erich Waelde wrote: > > > > Content-Description: message body text > > > > > > I made a little more progress on this last night. I was able to > > > actually create a working lvm, and format it as XFS (I do think I left > > > some rements of a bad atempt laying aroud BTW, is it safe to just > > > rmmive these traces?). > > > > > > > > > In any case, when I rebooted that machine the new lvm parition did not > > > mount (yes I put it in /etc/fstab)/ Atempts to mount it by had result > > > in a message about it not being "active". > > > > Do you have script /etc/init.d/lvm ? 1st you must run vgscan to scan > > volume groups and after that vgchange -a y > > I do, but I don'r seem to have any links to it from the various /etc/rc.d > directories. > > How can I create these (In a Debian sort of way?). I know I can create them > by hand, but thre must be a more "Debian" way fo doing it, right? > You need to have lvm10 and lvm-common installed and if using a standard debian kernel, need to have lvm-mod module running. I am pretty sure that installing lvm10 and/or lvm-common should have set up the /etc/init.d/lvm for you. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
static ip to dhcp conversion -- getting a hostname
Hi,
My university is switching everyone over from a static ip to one assigned via
dhcp. In addition, they are specifying the hostname for each of these
addresses. Unfortunately, we don't get an option to choose a hostname. It
seems that the current mechanism within Debian is to specify a hostname
regardless of how a ip is assigned. I would prefer to assign my own
host-name, but don't have that luxury.
I am using dhcp3-client to pull the ip number and other assorted information.
However, I can't get a hostname returned from the dhcp server. I have set
the option in dhclient.conf to request host-name (along with some other
options). I have turned on the debugging flags in the
dhclient-{enter,exit}-hooks.d/ to see what info is gathered from the server.
Hostname is not one of the pieces of information that is returned.
I am assuming that the dhcp server is actually sending out hostnames, but am
not 100% confident. I can take the returned address and do a nslookup to
actually get the hostname, but this seems like a kludge. Are there any known
problems with getting a hostname from a dhcp server that I am overlooking?
Michael Ash had asked a similar question back in Oct 2003 (debian-user
archives), and someone responded, but it didn't appear like there was any
definite solution.
Thanks for any help.
John
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: static ip to dhcp conversion -- getting a hostname
On Friday 05 March 2004 09:00 am, Jonathan Schmitt wrote: > >My university is switching everyone over from a static ip to one assigned > > via dhcp. In addition, they are specifying the hostname for each of > > these addresses. Unfortunately, we don't get an option to choose a > > hostname. It seems that the current mechanism within Debian is to > > specify a hostname regardless of how a ip is assigned. > > Hi, > sorry, I can't solve Your problem, but I would like to ask You if You're > sure it is a good idea to dynamically change the hostname? > A long time ago, I was much younger and more inexperienced, I decided, the > hostname assigned during installation wasn't what I expected, so I simply > changed it. The result was different programs not working properly anymore > (I think, You can still find my cry for help on debian-kde mailing list as > kdm was one of those packages). > It was potato with a self compiled kde by then, and I never ever played > with my hostname after that day. So, are You sure, that is a wise idea? js Unfortunately, it is not my decision to make. If it were, I would not have the dhcp server assign hostnames. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: LDAP client configuration question
On Monday 08 March 2004 05:23 pm, Kamaraju Kusumanchi wrote: > I am relatively newbie in LDAP area, so please excuse me if the question > is not interesting. > > We have an LDAP server up and running redhat 8. I know that the ldap > server is functioning properly because I can access it from other redhat > machines. But I am not able to access it from a client runnind Debian > testing. > > The configuration file I am using is as follows (I obtained this > configuration file from the redhat machines). I would like to know what > I have to do/read to set up this LDAP client. I have read the > LDAP-HOWTO.html. From what I understood, it discusses only server setup > but no discussion about client set up... I also browsed google but could > not get anything useful (or may be wrong keywords).. > > host k2.mae.cornell.edu > > base o=cttg,c=US > > ssl no > > pam_password md5 > > > regards > raju You need to install libnss-ldap, libpam-ldap, and nscd. You should probably take a look at: http://people.debian.org/~torsten/ldapnss.html for more info. He talks both about the server and some client info. Actually, where he talks about doing libpam-ldap and libnss-ldap are relevant for the client. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: NFS Groups
On Saturday 13 March 2004 10:07 am, Michael Satterwhite wrote:
> I want to thank those who helped me with the userid. The other half of
> this, however, is the group id:
>
> Assume that a directory on the server is owned by root:users. The group id
> number for users is different on the server and the clients. It's not
> possible to have the directory owned by the same user as on the client as
> there are many of those, so group permissions need to control whether the
> directory can be written.
>
> How can I set up this type of directory so that it can be written by the
> client machines?
>
> thanks in advance
I used ldap to solve this problem. I use it to essentially replace NIS. On
my server, I use the package cpu (does the equivalent of user{add,del}, and
is packaged in debian) to assign users and groups. I used the page:
http://people.debian.org/~torsten/ldapnss.html
for guidance on how to set up ldap on both my server and clients. It has
worked well.
John
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 8139too Module
On Tuesday 27 January 2004 07:24 pm, Timothy Paling wrote: > On Tue, 2004-01-27 at 19:36, Timothy Paling wrote: > Producing this listing shows that 8139too is not in the list. > > modprobe simple reports that it cannot find the module. > > How do I proceed further? I've located the source for the module in the > Kernel source however, I'm unsure as to how to go about compiling this is > introducing it into my kernel. > > Many Thanks > > Tim Paling > > Original Request: > > I am somewhat of a newbie to Debian and have managed to sucessfully > > install > > > 3.0r2 on my Compaq 2104EA laptop. > > > > I am, however, having some issues with getting my Belkin F5D5010 Ethernet > > CardBus card working. The Cardbus itself is working (I have verified this > > by > > > looking at /var/log/syslog when inserting the card). > > > > I am aware that I need to install the 8139too module, however, I'm > > completely unsure on how to do this. > > Suggestion: > If you are using the 2.4.18-bf2.4 kernel, I think that support for that > card is compiled in, and you dont have to insert any modules. If you are > using some other kernel, you will need to make sure that you've got that > module compiled (or compiled into your kernel). To see if you have that > module, type: > > ls /lib/modules/`uname -r`/kernel/drivers/net/ > > and look through the files listed for one called 8139too.o > > to install a kernel module, you must switch to root, then type: modprobe > > > for example: modprobe 8139too > > you may need to insert the mii module as well. > > good luck! > > -davidc > Tim, You might want to install discover (apt-get install discover). It does hardware detection for you and would probably recognize your ethernet hardware. Or you can as root, do modconf and search for the /kernel/drivers/net. Once in there, you should hopefully see the 8139too module. You can then install the module. Once installed via modconf, everything is set up so that future reboots will load up the 8139too module automatically. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Cloning
On Thursday 29 January 2004 08:00 am, Pedro Hernandez wrote: > Hello all! > > I'm about to install Debian on 12 computers (i*86). They will use the > same setup regarding software, but the hardware differs somewhat > between them. > > I would like to know if there is possible to, for example, install and > configure one box and then somhow "copy" the install to the others. Or > is the debian installer maybe scriptable? > > Well, I have never done anything similar before- suggestions, anyone? > > > Thanks, > --ph Look into replicator. Not used it, but did apt-cache show replicator and it may be what you want. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Font Mystery
On Wednesday 04 February 2004 04:43 pm, Hervé Piedvache wrote: > Steven, > > Did you also loosed the Fixed[Sony] and Fixed[Jis] fonts ?? > > Since I have updated from Woody to SID ... I have the same trouble of you, > and I also loosed those fonts but only under KDE ... I can choose them with > xemacs ... but impossible under KDE ??! > My font under Konsole is really ugly ... and I have also loosed the > Helvetica font :o(( > > Could it be in relation with the bug detailed here ? > http://lists.debian.org/debian-qt-kde/2004/debian-qt-kde-200402/msg00027.ht >ml > > I have reproduced also this trouble under GNOME in my 3 PC configurations > under SID after the migration from Woody ... :o( > > And no solution of the moment ... including testing the x-font guide from > here : > http://egads.ertius.org/~rob/font_guide.txt > Try reconfiguring fontconfig (dpkg-reconfigure fontconfig) and choose enable bitmapped fonts to see if it will help the kde lousy font problem. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Home Mail Server
On Thursday 19 February 2004 04:49 am, Peter A. Cole wrote: > - Original Message - > From: "Antonio Rodriguez" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Thursday, February 19, 2004 9:37 PM > Subject: Re: Home Mail Server > > > On Thu, Feb 19, 2004 at 09:23:22PM +1000, Peter A. Cole wrote: > > > Hi all, > > > > > > I'm trying to set up a home mail server here (have had previous posts > > re: > > > Exim) and I've discovered I now need to concentrate on sendmail due to > > > wanting to utilise the benefits of mimedefang. > > > > man procmail, man procmailrc, man procmailex > > > > mail is delivered to your machine by fetchmail, filtered through the > > procmail recipes (very simple to create) which will make local copies, > > redirect them to whereever you want (exim), etc. Your choice. > > HTH > > AR > > Sorry, forgot to say I haven't look at procmailex yet, but I have > definitely looked through the other two. I shall have a look at procmailex > as well. > > Thanks, something I missed already :-) > > Pete This article may be of interest for those wanting to set up a home mail server: http://talk.trekweb.com/~jasonb/articles/exim_maildir_imap.shtml John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: apt-get update error
On Thursday 19 February 2004 10:21 am, Deboo wrote: > I'm using debian woody. After reading an article in LG on apt (Issue 86 - > Debian APT, part2), I wanted to try using some package from testing and I > did as per the artile, putting 2 new lines for testing and unstable in > sources.list. But after I apt-get update, at the end, I get this output: > > Reading Package Lists... Error! > E: Dynamic MMap ran out of room > E: Dynamic MMap ran out of room > E: Error occured while processing magnus (NewVersion1) > E: Problem with MergeList > /var/lib/apt/lists/ftp.iitm.ac.in_debian_dists_stable_main_binary-i386_Pack >ages E: The package lists or status file could not be parsed or opened. > > I tried deleting this file and re-running apt-get update but still the > same. What am I doing wrong? The file seems to be fine if I open it with a > E: Error occured while processing magnus (NewVersion1) > E: Problem with MergeList > /var/lib/apt/lists/ftp.iitm.ac.in_debian_dists_stable_main_binary-i386_Pack >ages E: The package lists or status file could not be parsed or opened. > > I tried deleting this file and re-running apt-get update but still the > same. What am I doing wrong? The file seems to be fine if I open it with a > pager/editor. > > Regards, > Deboo Create a file /etc/apt/apt.conf as root and insert the following: APT::Cache-Limit 25165824; Note: don't forget the ';' at the end. After creating this file, rerun apt-get update and hopefully things should now work. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 2.6.4 kernel install wants to remove current kernel
On Friday 09 April 2004 12:41 pm, Nori Heikkinen wrote: > of course, i've totally failed to show the slickness of apt to my > coworker, because as soon as i booted up with 2.6.4, my mouse (neither > USB nor PS2) didn't work, and i wasn't online. (right now, i've > reverted to 2.4 to type this :-P). surely the 2.6 kernel comes with > USB support compiled in? am i just going to have to suck it up and > roll my own kernel? > > thanks again, > > Make sure you have the modules mousedev and psmouse loaded. I believe I had to use modconf to select psmouse from kernel/drivers/input/mouse whereas mousedev was already loaded for me using discover. Once psmouse was loaded then my mouse was active. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sparse matrix computation package
On Wednesday 07 March 2007 02:53, Chris Bannister wrote: > On Mon, Mar 05, 2007 at 05:59:18PM -0500, Kamaraju S Kusumanchi wrote: > > Micha Feigin wrote: > > > I am trying to find a package for performing sparse matrix > > > computations. All I could find under debian is superlu and libufsparse. > > > Both of which seem to only solve sparse linear systems but not perform > > > sparse matrix-vector computations (I need to implement so iterative and > > > multigrid methods for block tridiagonal matrices, so banded matrices as > > > appear in lapack too many extra zero lines). > > > > What a coincidence. I am also looking for similar thing - iterative > > solver > > Ask on the debian-science list. > > -- Look at petsc and hypre both available in debian. In our project we have used both to solve big sparse systems of equations on over 1000 processors. John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Is that all there is to it??
On Tuesday 13 March 2007 14:19, Jim Hyslop wrote: > OK, so I installed Sarge on my machine. The other day, I decided to > upgrade to Etch. I modified the sources file, changing "stable" to > "etch", and ran `apt-get --ignore-hold dist-upgrade`. After sorting out > one minor issue with inetd, apg-get reports everything's OK, and the > /etc/debian_version file reports '4.0'. > > U... is that it?!? Is it really that simple to upgrade? > > -- Yep, pretty impressive! John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
internal lan configuration and web address issues
Hi, I have a home DSL connection with a static IP assigned. The DSL modem is fed into one a ipcop firewall with port forwarding for port 25 and 80 directed to my webserver/mailserver in my DMZ zone. Machines in the green zone of my LAN cannot access my registered domain name, i.e. www.domain.org instead it returns the web interface to my DSL modem. Specifying the 192.168.1.100 gets me to my webserver. Outside of my LAN, accessing my registered hostname returns me to my webserver and not my DSL modem. I have registered my domain with no-ip.com using their basic host registration services. With this service, my host type is DNS Host (A) and is specified on no-ip.com 's management web front end. With this set up, I assumed that I don't have to do DNS on my end, since they are taking care of it. Is this an incorrect assumption on my part? My DNS Text Record is set as follows: "v = spf1 a mx ptr" I am not doing DNS on my machines. I tried editing my /etc/hosts file to specify the 192.168.1.100 for the registered hostname, but it instead returns the registered IP address when I ping my hostname or use lynx to access it from an internal LAN machine. Outside of my LAN, I can access the webserver just fine using the registered hostname. Is there a way to configure things so that machines on my LAN can access the web server using the registered name. Would this require that I do DNS on my machines? Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: internal lan configuration and web address issues
On Friday 16 March 2007 14:18, John Schmidt wrote: > Hi, > > I have a home DSL connection with a static IP assigned. The DSL modem is > fed into one a ipcop firewall with port forwarding for port 25 and 80 > directed to my webserver/mailserver in my DMZ zone. Machines in the green > zone of my LAN cannot access my registered domain name, i.e. www.domain.org > instead it returns the web interface to my DSL modem. Specifying the > 192.168.1.100 gets me to my webserver. Outside of my LAN, accessing my > registered hostname returns me to my webserver and not my DSL modem. > > > I tried editing my /etc/hosts file to specify the 192.168.1.100 for the > registered hostname, but it instead returns the registered IP address when > I ping my hostname or use lynx to access it from an internal LAN machine. I had mistyped my registered hostname in the /etc/hosts file. The correct entry allows my internal lan machines to access the webserver. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Speeding up boot time
On Wednesday 21 March 2007 18:39, Michael Pobega wrote: > I'm looking to speed up my Debian Etch boot speed, but I have no idea > where to start. > > Currently I have everything default, but I'd like to remove a few > things (Like networking, because networking tries to connect to my > Ethernet and my wireless drivers are loaded separately). > > Are there any easy tools to look through my startup programs, or will > I have to sort through everything manually? > > And if I remove networking, how do I go about using modprobe to start > up my Ethernet in case I ever want to use it/need it at another time > (i.e. When there is a new Debian kernel release I have to reinstall my > wireless before I can do anything)? You should take a look at installing ifplugd (apt-cache show ifplugd) ifupdown, and guessnet. This should allow you to boot normally without having the network timeout if you do not have a cable plugged in for your ethernet connection. John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Speeding up boot time
On Wednesday 21 March 2007 19:45, Michael Pobega wrote: > On Wed, Mar 21, 2007 at 07:32:26PM -0600, John Schmidt wrote: > > On Wednesday 21 March 2007 18:39, Michael Pobega wrote: > > > I'm looking to speed up my Debian Etch boot speed, but I have no idea > > > where to start. > > > > > > Currently I have everything default, but I'd like to remove a few > > > things (Like networking, because networking tries to connect to my > > > Ethernet and my wireless drivers are loaded separately). > > > > > > Are there any easy tools to look through my startup programs, or will > > > I have to sort through everything manually? > > > > > > And if I remove networking, how do I go about using modprobe to start > > > up my Ethernet in case I ever want to use it/need it at another time > > > (i.e. When there is a new Debian kernel release I have to reinstall my > > > wireless before I can do anything)? > > > > You should take a look at installing ifplugd (apt-cache show ifplugd) > > ifupdown, and guessnet. This should allow you to boot normally without > > having the network timeout if you do not have a cable plugged in for your > > ethernet connection. > > > > John Schmidt > > Do you recommend any of those, or would I be safe with just ifplugd > and ifupdown in case that fails? I think ifplugd and ifupdown would be sufficient for your needs. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Speeding up boot time
On Wednesday 21 March 2007 20:19, Michael Pobega wrote: > On Wed, Mar 21, 2007 at 09:05:55PM -0500, Ron Johnson wrote: > > -BEGIN PGP SIGNED MESSAGE- > > Hash: SHA1 > > > > On 03/21/07 20:44, Michael Pobega wrote: > > > On Wed, Mar 21, 2007 at 08:12:37PM -0500, Ron Johnson wrote: > > >> On 03/21/07 19:39, Michael Pobega wrote: > > >>> Are there any easy tools to look through my startup programs, or will > > >>> I have to sort through everything manually? > > >> > > >> It's not as much as you think. > > >> > > >>> And if I remove networking, how do I go about using modprobe to start > > >>> up my Ethernet in case I ever want to use it/need it at another time > > >>> (i.e. When there is a new Debian kernel release I have to reinstall > > >>> my wireless before I can do anything)? > > >> > > >> Run "ls -1 /etc/init.d" and post it here. That will tell us what > > >> you can deinstall. > > > > My first pass at things you *might* be able to remove are: > > apache2 > > avahi-daemon > > bittorrent > > clamav-freshclam > > hyperestraier > > mysql* > > nfs-common > > And to remove these I just remove them from /etc/init.d? If you don't use them, then I would do this: sudo aptitude purge apache2 avahi-daemon bittorrent clamav-freshclam hyperestraier mysql* nfs-common > > > BTW, where's your MTA? I don't see exim4 or postfix. > > I just use Google's SMTP server, I have no need to have my own > outgoing server. > > What are the advantages to running an MTA? What are the disadvantages? advantages: running a local mta for reporting issues with your system, i.e. logcheck, rkhunter and having the results mailed to you. It can be configured to just do local delivery and with exim4 and the debconf questions that are asked, it is very easy to configure for local delivery. disadvantages: one more thing to worry about, but with a local delivery configuration I can't see any disadvantages. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Speeding up boot time
On Wednesday 21 March 2007 20:24, Michael Pobega wrote: > Also, I forgot to ask. Would it be safe to remove networking from > /etc/init.d, and just switch to using ifupdown? I would just install ifplugd and configure it and see if it doesn't help alievate the time delay that occurs if you don't have a network cable plugged in. I wouldn't remove networking from /etc/init.d unless ifplugd does NOT solve your problem. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: inserting mouse module in bootup
On Monday 27 November 2006 08:03, Sam Rosenfeld wrote: > I would like to insert a mouse module somewhere in the kernel boot up. I > am currently typing "modprobe mousedev" to enable the mouse and therefore > X, too. Is there a config file for this? > > Thanks. > > Sam Install the program modconf (aptitude install modconf). The start it up as root and goto the menu item /kernel/drivers/input and select mousedev. Exit out of program and the appropriate files should be updated with the mousedev module. You then should be able to access your mouse without having to do the modprobe thing. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Replacing Gnome with KDE
On Monday 27 November 2006 17:55, Arlie Stephens wrote: > I made the mistake of selecting 'workstation' when installing etch. It > managed to do the X installation correctly, without needing to be > rescued manually, which is way better than I've seen from earlier > debians. That's the good news. The _bad_ news is that it installed > gnome, and apparantly only gnome. > > I used aptitude to select any packages that looked like they might be > part of kde, and remove any packages that looked like they might be > part of the guts of gnome. The result is a mess. I appear to be > running kdm (according to ps), but the result has the look and feel of > nothing much. If it's kde, it's sure changed - a lot. > > I'm not sure whether the problem is that I'm missing a few pieces of > kde, still have gnome bits running that shouldn't be, or simply that > the various package installation scripts emphatically failed to do the > right thing. At this point, the system is only usable if I bypass X > entirely - there's no way to get a shell window inside X. There's also > no control center, or any of the other things that ought to be on the > icon/menu bar that normally loves at the bottom of the screen. > > Any suggestions for how to fix this mess? At the moment, the best > thing I can think of would be to reinstall, with tasksel/kde-desktop. > I'd prefer something a little less drastic. > > -- > Arlie > > (Arlie Stephens [EMAIL PROTECTED]) The easiest thing to do is to get a listing of the packages installed: dpkg --get-selections Then you can search for gnome packages, i.e dpkg --get-selections | grep gnome Once you get a listing of the packages, then you can start removing them, i.e. aptitude purge gnome_package_name For instance, it looks like you probably have gnome-core installed, so doing this would remove quite a bit of gnome: aptitude purge gnome-core Keep doing dpkg --get-selections | grep gnome and the aptitude purge to remove packages. Also, you may want to make sure you get rid of gdm if you can, i.e. aptitude purge gdm Then you can install kde, i.e. aptitude install kde kdm (just to be sure) It "should" drag in a bunch of packages. If it doesn't then you can drag in the meta packages for the different areas, like kdenetwork, kdebase, kdeutils, etc. aptitutude install kdebase kdenetwork kdeutils You can use the command "apt-cache show kde" to get a listing of what packages are connected to the package "kde". Note this is all command line stuff. I am sure you can do the same thing with the aptitude front-end, I just don't use aptitude that way so I can't offer any advice. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bash script question
On Thursday 07 December 2006 11:16, Nate Bargmann wrote: > Since there is a lot of knowledge on this list, I thought I'd aske > here. > > This may be trivial, but I'm not even sure how to search for what I > want to do. > > I have a directory of files that are created daily using > filename-`date +%Y%m%d`.tar.gz so I have a directory with files whose > names advance from filename-20061201.tar.gz to filename-20061202.tar.gz > to filename-20061203.tar.gz and so on. Based on the date in the > filename, I would like to delete any than are X days older than today's > date. So, I'm not interested in the actual created/modified date, just > the numeric string in the name. This will require some debugging on your part, but hopefully this will be pretty straightforward old_date=20061201 for file_name in *tar.gz; do file_to_remove_date=echo $file_name | tr -d [:alpha:] | tr -d [=-=]; if test $file_to_remove_date -le $old_date; then echo "rm $file_name"; fi done There are certainly better ways to do this. You can do a man tr to see what the above is doing. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
recommended network/server layout for website, email, and backup hosting
Hi, I would like to host several low traffic web sites at my home with some older computers (400 Mhz P2) that I have laying around. I would like to get some recommendations on effective ways of setting up my set of computers that would provide a web server, and email server and back up servers. A big reason for doing this is to learn about more what all is involved and don't mind digging into details but would like to ensure that I reasonably aware of what I am getting into and potential pitfalls and security issues. I have a static IP with an IPCOP firewall (with 3 NICs), and a internal LAN with several machines running debian behind the firewall. Nothing is hanging off my DMZ right now. I block everything coming into my firewall except ssh traffic. These web sites would be publically accessible with low traffic volumes. In addition, I forsee email hosting for each of the domains. I would not have that many email accounts (not more than 10-20). I figured that exim with the ability to do multiple hosting would suffice. I would probably set up a couple of mailing lists as well using something like mailman. I would like to set up my email server with imap, and pop cabilities for both the publically accessible domains and my own personal email access. I would like to have a couple of machines set up in my LAN that would be able to provide two levelsof backups for my configurations, both internal LAN backups and DMZ level backups (web server and email server). Initially, I was thinking that I would put two machines in my DMZ zone, one acting as a web server and one acting as an email server. My two backup machines would be in my LAN along with my fileserver and another development machine. Regarding server security (email and web server), I have the following questions? 1. Because the machines are slow, would it be better to have the two machines do some sort of load balancing or would it be better to have a separation of responsibilities? 2. Would it be better (security wise) to have my email server located in my LAN and not in my DMZ zone and just tunnel port 25 traffic through? 3. I know nothing about DNS, and figured that I would let someone like no-ip.com provide this service for me. Or would it be fairly straightforward to do my own DNS hosting and combine two of my machines for doing primary and secondary DNS with other responsibilities, i.e. email/DNS on one machine, DNS/web server on another? Is it possible to have my DNS machines inside my LAN, or is it necessary to have both primary and secondary DNS machines in my DMZ for better security. 4. For imap and pop stuff can the imap server be inside my LAN and access be tunneled through as needed. 5. Should any server, i.e. mail, imap,pop, web be located in the DMZ zone so if they are hacked, my internal LAN machines are safer? 6. Are there some suggested or best practices for having my machines in the DMZ access my back up servers? Thanks, John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
chroot testing of apache installation with multiple fqdn
Hi, I am testing out the installation of some software that involves specifying hostname information for the install. I have set up a minimal chroot environment and downloaded all of the debian packages and the other software that is needed to build it. I am using this on my laptop that is brought back and forth from home to work. While at work, I have a static routable address assigned to me via dhcp. At home, the laptop sits on my private lan with a 192.168.1.*. statically assigned. One of the pieces of software I install is apache and a routeable address/hostname is required as part of the install procedure for apache and some of the other software. If I do the install testing from home with my 192.168.1.* address in my chroot environment, things work fine. However, if I am at work testing the install done at home, then the apache web server won't be able to start. Likewise, installs done at work and then tested at home have issues when starting apache. Is there a way to set up my chroot environment (or perhaps it is an apache issue) that allows me the freedom to move the machine from one network to another and just always go to the localhost instead of the actual machine name/routeable address while still allow apache to start up regardless of where the initial apache installation was done? Thanks for any insight into this matter. John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Intel Deluxe PC Camera
On Sunday 04 November 2007, Marc Shapiro wrote: > > It looks like the driver module that I need is spca501. There is a > source package for Etch: > > spca5xx-source > > What is the best way to compile this? I generally just use stock > kernels and the modules that come with them so I am unfamiliar with > compiling modules separately for a kernel that I already have. Once I > do have the module compiled, will udev/hotplug load the driver at > boot-up, or should I add it to /etc/modules? > > -- > Marc Shapiro > [EMAIL PROTECTED] Marc, I use module-assitant and modconf (apt-get install module-assistant modconf) to manage modules that I have to compile. Do this: sudo m-a It will put up a dialogue box, and then you need to first prepare the build by downloading the kernel headers. Then choose select and check the spca5xx-source, and it should download it for you. Then build it and install it. These are all menu selectable items in module assitant. Then I use modconf to select the modules so load. With those two toosl, dealing with modules is really quite straightforward and painless. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Intel Deluxe PC Camera
On Sunday 04 November 2007, Marc Shapiro wrote: > > Well, I tried to follow the above instructions, but... > > I didn't have the kernel headers for my 2.6.16 kernel and they do not > seem to be available in the Etch repository. I took this as a sign that > it was time to upgrade my kernel and upgraded to 2.6.22-3-k7 from > backports. Then I reran module-assistant. > > This time, m-a found and downloaded the kernel headers. I then selected > the module to compile. The gspca-source was not listed, so I used the > spca5xx-source. When I tried to build the sources, however, the build > failed. The logfile contained nothing but the date and time, so I have > no specific errors to report. It may be that the problem is using > spca5xx-source instead of gspca-source, but that was not listed as an > option. If I download it manually, might m-a find it and allow me to > use it. > > Any ideas on what I should try next? > > -- > Marc Shapiro > [EMAIL PROTECTED] Marc, You have a couple of options, one is to upgrade your etch to the current 2.6.18-5 kernel and then build the spca module. I built the spca module from source with 2.6.18 kernel. The spca5 source doesn't build with the newer kernels 2.6.22. You need to use the gspca-source. I am running testing/unstable, and gspca-source is available in testing/unstable repositories. If you grab the 2.6.22 from backports, they might also have that source package. Otherwise you could just grab it from a debian repository or use apt-pinning to pin a lower priority for lenny/testing and and then grab the gspca source. I am using the gspca module for my web cam, so do know that the module builds with 2.6.22. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ssh port not opening
On Monday 05 November 2007, John O Laoi wrote: > On 11/5/07, Andrew Sackville-West <[EMAIL PROTECTED]> wrote: > > ssh-agent is *not* the program that allows ssh connections. That is > > sshd. It should be started with /etc/ini.d/ssh start as root. Is there > > no output when you do that? Anything in the logs? > > > > > However, I cannot ssh into my host. > > > The contents of /etc/default/ssh is > > > > that's because sshd is not running and needs to be. > > But the file /etc/init.d/sshd does not exist. > The only such file in /etc/init.d is ssh. > > > How do I check the logs? > > John Do this: dpkg --get-selections | grep ssh and send the results. I am wondering if you have the openssh-server installed. You can get everything by just doing aptitude install ssh. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ssh port not opening
On Monday 05 November 2007, John O Laoi wrote: > # dpkg --get-selections | grep ssh > openssh-client install > openssh-server deinstall > # > > Thanks to everyone for your help. > > Maybe I should remove ssh and reinstall? > aptitude install openssh-server This should solve your problems. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Driver loaded, now, how do I access the camera? (WAS: Re: Intel Deluxe PC Camera )
On Tuesday 06 November 2007, Marc Shapiro wrote: > > So my drivers seem to be loaded. Now, how do I access the camera, so > that I can capture an image to be processed? I am currently thinking of > using opencv through either C/C++ or Python. How would I access the > camera to generate an image (probably jpeg) that I could then save to > disk or, preferably, pipe to opencv? > > -- > Marc Shapiro > [EMAIL PROTECTED] I played around with camstream which is a debian package and it might work for your needs. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: how manage pasword with ldap
On Friday 19 January 2007 01:54, abdelkader belahcene wrote: > Hi, > I am looking for a simple example or doc, for "authentification via ldap". > > I have a lab with 20 machines where the students can use anyone at > anytime, so I want to centralise the authentification ( login and > password). something like perhaps NIS ?? > thanks for replay > > bela Take a look at this: http://people.debian.org/~torsten/ldapnss.html I then used the package cpu to manage adding users and groups to my ldap server. I am sure there are better tools such as luma, etc. However, cpu (debian package) was all I needed for my simple set up (much like what you are looking to do but with fewer machines). John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ldap + pam howto?
On Tuesday 30 January 2007 10:04, Rakotomandimby Mihamina wrote: > Hi, > I am using Testing, and I want to setup the debian way an LDAP + pam > authentication system for system users. > Would you know a recent howto talking about that? > I dont need generic howto, I am interested in the debian specific way. > > Thanks a lot! See the following web page: http://people.debian.org/~torsten/ldapnss.html John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Fonts problem on Debian Etch
On Wednesday 31 January 2007 18:00, David Shultz wrote: > I've installed debian etch(daily built). And i gotta say > it's impressive. Now i have a problem. The problem is > related to the display of fonts on screen. Fonts that > are displayed on screen are partly blurry and partly ok. > Even the cursor is like that. Is it possible to fix this? > > Thanks. > > --David David, Please check to see if you have the package fontconfig-config installed. If so, then do this as root: dpkg-reconfigure fontconfig-config I have a laptop, so my settings may be different, but playing around with fontconfig-config dramatically improved my fonts under kde. First screen shows: Native Autohinter None I selected Autohinter Next screen shows enable subpixel screen rendering: Automatic Always Never I selected: Automatic Next screen shows enable bitmapped fonts by default: Yes No I selected: No I then logged out, and restarted the xserver just to be sure, and my fonts were significantly improved. John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: backup/restore
On Thursday 15 February 2007 14:28, Casey T. Deccio wrote: > I'm looking for a solution to temporarily backup then restore a Debian > install--preserving the filesystem contents and attributes. The caveat > is that the capacity of the drive I'll be restoring to is smaller (all > other hardare is unchanged). I wasn't sure if there was a way to do > this with dd because of the smaller drive on the restore. Initially, I > used 'rsync -avH --delete --no-numeric-ids src server:dst' to send > everything to another server and back. The number of files (including > links, etc.) seemed to check out, but disk usage (using 'du -cs /') was > different before and after (block size on the filesystem was the same as > before). Plus I was not sure if extended attributes were transfered > with the files. You might want to consider mondo or partimage for this task. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Tunnel iceweasel?
On Wednesday 26 March 2008, Rich Healey wrote: > Joost Witteveen wrote: > > On 24/03/2008, Douglas A. Tutty <[EMAIL PROTECTED]> wrote: > >> On Mon, Mar 24, 2008 at 11:46:56AM +0100, Joost Witteveen wrote: > >> > On 23/03/2008, Rich Healey <[EMAIL PROTECTED]> wrote: > > I'm trying to tunnel an iceweasel instance via ssh from one > > >> > > of my boxes at my house to remember the name of an add-on i > >> > > installed. > >> > > > >> > > The problem is that i create a ssh session (ssh -XC > >> > > ssh.psychotik.info), login and run iceweasel at the bash > >> > > prompt, which takes forever, but then finally *opens a > >> > > local iceweasel!!!* > >> > > >> > I suppose that iceweasel -P uniqueprofilename would do what you want? > >> > > >> > Also, it's *much* faster use vnc (tunnel through ssh): on the remote > >> > host, start: vnc4server on your localhost, start (and login to) ssh > >> > -L 5900:server:5901 server > >> > > >> > and then on the localhost (different window) vncviewer localhost:5900 > >> > > >> > The 5901 portnumer is assuming the vncserver opens a X11 screen on > >> > :1. When I start epiphany diretly over X11, it takes about 30 min to > >> > show a page; when I do it using VNC as above, it takes seconds. > >> > >> I run iceweasel over ssh all the time, however, I don't have it > >> installed locally so there's no local version to run. It may take a > >> few seconds to give the initial window, but then it displays as fast as > >> the box can swap. The network is 100 MB/s ethernet, the box I'm sitting > >> at is a P-II with 64 MB ram, the box I'm sshing into to run iceweasel is > >> an AMD Athlon64 with 1 GB ram. It doesn't even take 30 minutes to show > >> a page when I ssh from my 486 with 32 MB ram so something is wrong > >> there. > >> > >> Why would VNC be faster if both are encrypted? > > > > No, over a 100Mb/s ethernet, running iceweasel over VNC probably > > wouldn't be much faster than directly over ssh (and running over an > > ssh-tunneled VNC connection would of course be slower than straigt > > VNC). > > > > But the OP complained iceweasel was very slow. So I suppose he didn't > > run it over a direct 100Mb/s connection, but over something slower, > > probably with larger ping times, ping times of 10-30 ms are enough to > > make it slow, and with slow, I mean that it can take over 20 min for > > iceweasel to even start showing the home page. > > I notice that when that happens, starting iceweasel on the remote site > > on a VNC X server an watching the output via a VNC viewer is a lot > > faster. And a lot here means just a couple of seconds to show the home > > page, instead of 20 min. > > As the OP reported using ssh, I assumed he didn't want to connect > > unencrypted (somethign VNC as far as I know does), so I suggested > > using an ssh tunnel. > > Hi, the issue here isn't the speed, and besides, i prefer to have it > directly connected to my Xserver, rather than runnign in VNC. > > The point here isn't eh startup time though, it's that it starts a local > iceweasel! > > In trying to build FF from source on my new 64 bit machine i > accidentally wound up with a ff3 beta, but running that now also opens > iceweasel. > > Somehow the binary has managed to associate EVERYTHING with itself. > > The real thing that does my head in is when i launch FF on another box.. > it still creates a local iceweasel? this should happen AFAIK.. my > starting a command on that box via should not be able to cause commands > to be run on my local? > > Does this constitute a security issue? i'll see if i can get a PoC > during the week, even if one couldn't get arbitrary code, one could > still point the new iceweasel on the host machine to a site witha FF > exploit. > > Now that i think of it.. it would be simple enough to create a free > shellserver with busybox aliased to a malicious FireFox call in the > system bashrc.. that'd probably do it. > > I'll look into it. I noticed the same slowness, tunneling via ssh via very fast connections. However, if you use the iceweasel -no-remote it seems to really help. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
dma errors
Hi,
I have two IDE drives connected together using lvm. The drives are quite new,
the motherboard is quite old. I put a single ext3 partition on top of lvm
which is used for temporary large data sets. I get the following error
messages from dmesg:
EXT3 FS on dm-0, internal journal
EXT3-fs: mounted filesystem with ordered data mode.
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x84 { DriveStatusError BadCRC }
ide: failed opcode was: unknown
hde: dma_timer_expiry: dma status == 0x20
hde: DMA timeout retry
PDC202XX: Primary channel reset.
PDC202XX: Secondary channel reset.
hde: timeout waiting for DMA
hde: dma_timer_expiry: dma status == 0x21
hde: DMA timeout error
hde: dma timeout error: status=0x58 { DriveReady SeekComplete DataRequest }
ide: failed opcode was: unknown
hde: dma_timer_expiry: dma status == 0x21
hde: DMA timeout error
hde: dma timeout error: status=0x58 { DriveReady SeekComplete DataRequest }
ide: failed opcode was: unknown
hde: dma_timer_expiry: dma status == 0x21
hde: DMA timeout error
hde: dma timeout error: status=0x58 { DriveReady SeekComplete DataRequest }
ide: failed opcode was: unknown
What would be the source of these errors and how should I go about fixing
them?
Thanks,
John
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Pumping Gas in Oregon (WAS: Re: Osama Bin Laden Take Over List!)
On Tuesday 22 August 2006 17:35, Paul Johnson wrote: > On Tuesday 22 August 2006 12:25, Joey Hess wrote: > > Paul Johnson wrote: > > > On Tuesday 22 August 2006 07:30, Marcelo Chiapparini wrote: > > > > Hi! > > > > > > > > It could be possible to use this list for what was created: Debian > > > > related questions *only*? > > > > > > Sorry, this list was made for Debian users. Go read lists.debian.org > > > again. > > > > You seem to have missed the word "Support" in the mailing list > > description[1]. It's hard to support people when I'm busy downloading the > > incessant nonsense about Oregon that you manage to turn any thread you > > touch into. > > debian-user: Help and discussion among users of Debian > > I'm not seeing "support" there. Perhaps you're speaking a different > version of English than the rest of us learned. See this link: http://lists.debian.org/debian-user/ debian-user mailing list Help and discussion among users of Debian Support for Debian users who speak English. (High-volume mailing list.) This list is not moderated; posting is allowed by anyone. Posting address: debian-user@lists.debian.org Read the line that says, "Support for Debian users who speak English." How about getting a clue! John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: fonts messed up
On Wednesday 23 August 2006 07:41, Lorenzo Bettini wrote: > Hi > > after a recent upgrade (unstable) all my fonts are almost complitely > messed up. > > I cannot use KDE since almost everything text is missing, and also gnome > is barely usable (most text disappear from menus and terminal in > general)... > > any clue please? > > Go into the Control Center, select Appearance & Themes, select Fonts. Then turn off and then on the Anti-aliasing button. Log out and log back in. See if that doesn't fix things. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: LDAP howto?
On Tuesday 03 October 2006 17:28, Ishwar Rattan wrote: > Pointers to good LDAP-howto for server coniguration > details. > > -ishwar This is where I started: http://people.debian.org/~torsten/ldapnss.html John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: automounting usb-storage 2
On Wednesday 04 October 2006 13:37, Bernd Kloss wrote: > Hello, > I read the thread contributed by Florian e.al.. > I have the same/similar problem (etch/KDE3.5.4) > Plugging in USB-device the automounter pops up with tree options: > open in new window > play with kaffeine > do nothing > > Choosing new window I get an empty window /system/media/sda1 and the > konqueror-Message: > > A security policy in place prevents this sender from sending this message > to this recipient, see message bus configuration file > (rejected message had interface "org.freedesktop.Hal.Device.Volume" > member "mount" error name "(unset)" destination "org.freedesktop.Hal") > > dpkg -l udev\* hal\* dbus\* pmount ==> > > > +++---= >=== ii dbus > 0.92-2 simple interprocess messaging system un dbus-1 > (no description) ii dbus-1-utils > 0.92-2 simple interprocess messaging system > (utilities) un dbus-qt-1 (no > description) un dbus-qt-1c2(no > description) ii hal 0.5.7.1-2Hardware > Abstraction Layer un hal-device-manager (no > description) ii pmount 0.9.13-1+b1 mount > removable devices as normal user ii udev 0.100-1 >/dev/ and hotplug management daemon > > What can I do to let every user read its USB-stick? > Thank You! > > Make sure you have your users in the plugdev and powerdev groups. In addition, I had to edit the file /etc/dbus-1/system.d/hal.conf. My version is attached. John
Re: wired or not wired, that's the question
On Wednesday 04 October 2006 16:16, wimpunk wrote:
> Hi,
>
> I'm running debian testing on my laptop and I have a question about
> choosing a configuration.
> I use my laptop on wired and wireless environments. At this moment I
> switch manually but I want to get it done automaticly: if there's a
> wire plugged in (with signal) try to use it and if no signal, try to
> get wireless up. Has anyone a suggestion on how to do it?
>
> Tnx,
>
> wimpunk.
Use a combination of ifplugd and guessnet.
ifplugd will check to see if you are ethernet cable is plugged in or wlan is
up.
guessnet when appropriately configured via /etc/network/interfaces will choose
which network to associate with.
Be sure to read /usr/share/doc/{guessnet,ifplugd}/README.gz
John
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to PIN a package?
On Monday 16 October 2006 00:26, Mirto Silvio Busico wrote: > Hi all, > I didn't get any answer. > > Is it a really stupid question? (or I missed the replay?) > > Please help me: I need to PIN a package, but the documentation doesn't > help. > > Regards > Mirto > > Mirto Silvio Busico wrote: > > Hi all, > > I had to downgrade openoffice to version 2.0.3. > > Now I need to avoid that this package will b e upgraded again. > > > > Following the manuals I put in apt files (in /etc/apt): > > > > apt.conf: > > - > >--- APT::Authentication::TrustCDROM "true"; > > AcquireProxy "false"; > > APT::Default-Release "testing"; > > > > - > >--- > > > > preferences: > > - > >--- > > > > Package: openoffice.org* > > Pin: version 2.0.3* > > Pin-Priority: 10001 > > > > > > - > >--- > > > > But apt-get update; apt-get -s dist-upgrade tell me that apt still want > > to upgrade openoffice to the 2.0.4~rc3-1 version. > > > > What I'm doing wrong, or what is missing? > > > > Mirto > > -- > > __ Try doing this as root: aptitude hold openoffice.org Then try and do an aptitude upgrade to see if the above took effect (it should). John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: routing only certain traffic through vpn?
On Thursday 19 October 2006 12:31, Matt Price wrote: > On 10/19/06, Jacob S <[EMAIL PROTECTED]> wrote: > > -BEGIN PGP SIGNED MESSAGE- > > Hash: SHA1 > > > > On Thu, 19 Oct 2006 11:03:20 -0400 > > > > "Matt Price" <[EMAIL PROTECTED]> wrote: > > > hi, > > > > > > i'm wondering whether it's possible to route only certain internet > > > traffic through a vpn, or to exclude certain ip addresses/ranges from > > > the vpn. > > > > > > my situation is as follows: I work mostly from home and rely on the > > > university's vpn to be able to access online journals. ths works > > > fine., but when I'm connected to the vpn I can't send mail from my > > > home email account (postfix doesn't work properly). I'm wondering > > > whether I could contact my smtp host from outside of the vpn somehow. > > > > > > has anyone tried this and/or any suggestions? > > > > This sounds like you don't have your routing setup properly. I use a > > vpn regularly for work and only traffic going to their range of ip > > addresses goes through the vpn. > > > > What does "route -n" show on your computer? And how do you connect to > > the internet? > > to answer both of your questions: > > The vpn server runs openvpn, which I also use on my computer as a > client. this vpn sends all internet traffic through itself; I imagine > but don't know for sure that this is done with the redirect-gateway > directive as described in the openvpn howto: > http://openvpn.net/howto.html#redirect > > when I'm connected, route -n shows: > > n$ route -n > Kernel IP routing table > Destination Gateway Genmask Flags Metric RefUse > Iface 128.100.56.140 192.168.2.1 255.255.255.255 UGH 0 0 > 0 eth1 142.150.248.1 142.150.248.165 255.255.255.255 UGH 0 0 > 0 tun0 142.150.248.165 0.0.0.0 255.255.255.255 UH0 0 > 0 tun0 192.168.70.00.0.0.0 255.255.255.0 U 0 0 >0 vmnet1 192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 > 0 eth1 172.16.137.00.0.0.0 255.255.255.0 U 0 0 >0 vmnet8 0.0.0.0 142.150.248.165 0.0.0.0 UG0 > 00 tun0 > > (vmware server is up, I guess that's what the vmnet1 is about) > > this is all uninterpretable to me so help welcome... > > thanks, > > matt > With the vpnc client which is probably not what your are using, you can specify target networks in the config file located in /etc/vpnc/example.conf. Perhaps the openvpn client would have something similar where you can route only a certain range of traffic through that tunnel. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: IBM eServer x series 206 RH-to-Debian migration
On Thursday 09 November 2006 12:34, Lee Whalen wrote: >Greetings all, long-time Debian user, first-time poster here. So, > I've got a predicament. I need to get Debian Etch on a box that is > already an in-production HeadRat Enterprise server (RHEL 3.0 Typhoon 6), > running a few production apps (Apache with a handful of VHosts, postfix, > a ticketing system we use called GForge and it's attendant Postgres > database, and a few NFS mounts). The company I work for no longer wants > to pay the HeadRat extortion fee just to be able to download package > updates and whatnot, so I am tasked with migrating all of our HeadRat > servers over to Debian. > > >For those of you who managed to read this far (go YOU!) does that > sound feasible, or am I doing far, far too much work? > > Many thanks for your assistance! > --Lee Lee, You might want to take a look at: http://twiki.iwethey.org/Main/DebianChrootInstall for doing a chroot install. I have never done this, but it might be a useful method for getting Debian installed with minimal interruption to your current system. John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: rsync question
On Friday 09 November 2007, John O Laoi wrote: > > Also, I get confused sometimes on the effects of a trailing slash on > > source and target arguments. Check for a ~/Documents/Documents/ > > directory or something. > > You are all correct. > There is a ~/Documents/Documents/. > > I must be using it incorrectly - I'll be more careful in future. > John Here is what I do with rsync: rsync -aPn directory machineB:~/directory The 'n' option just does a dry run. After you are convinced things are working then remove the 'n' option. If you put a trailing slash on the first directory, i.e. rsync -aPn directory/ machineB:~/directory you will get a new directory created on machineB, i.e. ~/directory/directory John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Kppp
On Sunday 11 November 2007, Darko wrote: > Charlie wrote: > > On Sun, 11 Nov 2007, Darko shared this with us all: > >> --} Andrew Sackville-West wrote: > >> --} > On Sat, Nov 10, 2007 at 06:43:28PM +0100, Darko wrote: > >> In a terminal as root do: > >> > >> adduser dialout > >> > >> Consider using pon maybe? > >> > >> You might like to install pon and pppconfig to make it work and add > >> yourself to the dip group. As above, but put in dip instead of dialout. > >> > >> HTH > >> Charlie > > Thanks Charli but kppp still not working does some one have an idea? > > Darko Did you log out and then log back in? When adding yourself to a new group, it is easiest to just log out and log back in so the change will take effect. There are other ways to do it, but that is the easiest. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
kppp groups
If you are trying to use the kppp program, you need to be part of the "dip" group: -rwsr-xr-- 1 root dip 1211176 2007-10-15 07:16 /usr/bin/kppp John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: No sound with video
On Thursday 15 November 2007, Ralph Katz wrote: > On 11/15/2007 11:50 AM, Tom Ashley wrote: > > I'm stumped and need help with configuring my computer for sound with > > video. I can play CD's and listen to audio streams but get no sound > > with video feeds (CNN, MSNBC, You Tube, etc.). I've searched the list > > archives and Google for the past 3 days finding similar problems but > > none of the solutions help in my situation. > > > > I'm running Debian Etch on a P4 with 2GB RAM; sound card is C-Media > > Electronics Inc CM8738 (rev 10). > > > > > > The permissions are okay; my user name is in the audio group. > > > > If I understand correctly, ALSA is correctly configured and activated. > > Per cat /proc/modules: > > > > > > I'd appreciate if anyone can offer a solution or point me in the right > > direction. > > > > Thanks in advance. > > Hi Tom -- Yes, it's frustrating. Search for my posts here in 2007 for > my struggles and solutions with sound on etch. > > You have sound, you say. But not with video. > > Which players work/don't work? Sound with flash running thru a browser > is different from sound from an mpeg4 file playing on mplayer. > > If you use mplayer, what's in .mplayer/config? > > This list is the right place. Just provide some more detail, and the > experts will jump in after we mere mortals swing and miss. :-P > > Good luck, > Ralph Hi, If you are using xine as your engine, and trying to play mp3s, then you need to make sure you have the following installed: libxine1-ffmpeg Once you install that, then go into your home directory and remove the .xine/. I was having problems listening to mp3 streams and doing the above resolved my problems. I note that using the mplayer engine in kmplayer worked fine however, if switching to the xine engine for kmplayer or amarok, mp3s wouldn't play without the addition of libxine1-ffmpeg library. The above assumes you are using the xine engine for your sound. If you are using other engines, this fix shouldn't resolve your problem. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
PII fast enough for firewall
Hi, I have a 15K Mbs connection (up/down) to my house (fiber to the home). I have a Buffalo router that connects to my WAN and then one of the LAN ports on this router connects to my IPCOP firewall that is running on a PII -- 400 MHz box with 64 MB of RAM. When I do a speed test from my box behind my IPCOP firewall, I get about 10K Mbs up/down. If I move the connection to one of the Buffalo router LAN connections, I get the advertised 15K Mbs up/down speed. So routing traffic thru the IPCOP firewall slows things down quite a bit. Is this to be expected? I was thinking of changing the firewall to a debian box running shorewall, and was wondering if I could tweak the firewall/router to not slow things down appreciably like the ipcop box is doing. Thanks, John Schmidt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
