Re: NFS File-server Oops
Strake wrote: Hi, I am running Debian 4.0 on a file-server, called bender, using NFS to serve the files, and a little while after I start to copy files to my server, my server's kernel oopses and the process on the client doing the copying hangs. Using stock kernel. The kernel according to `uname -r`: 2.6.18-4-amd64 Pid: 203, comm: pdflush Not tainted 2.6.18-4-amd64 #1 I'm not an expert at reading these outputs, but it would appear the problem is with pdflush. Searching for kernel oops and pdflush turned up this: http://lists.debian.org/debian-kernel/2007/02/msg00182.html It appears that other people have seen the same problem with the same kernel. -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Charging iPod / Listening to music
steve wrote: I am curious to know what computer, desktop or laptop provides power to accessories when the power supply is switched off... None of mine do? My Shuttle XPC SN95G5 does. -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
root (/) on software raid
I've managed to enable software raid on one of my systems. Four different arrays built as follows: /dev/md0: (/boot) - /dev/hda1 /dev/hdc1 /dev/md1: (/var/log) - /dev/hda6 /dev/hdc6 /dev/md2: (/) - /dev/hda7 /dev/hdc7 /dev/md3: (/opt/backup) - /dev/hda8 /dev/hdc8 Three of the four come up perfectly after reboot (md0, md1, md3). The problem is with md2. After every reboot it's degraded with only hdc7 as a member. I can manually re-add hda7 and it syncs without error and works fine until the next reboot. But I'd like to correct this so this isn't necessary. I'm using grub as my boot loader and the following is the entry being loaded on startup: title Debian GNU/Linux, kernel 2.6.11-1-386 root(hd0,0) kernel /vmlinuz-2.6.11-1-386 root=/dev/md2 ro initrd /initrd.img-2.6.11-1-386 savedefault boot The system is running sarge, with the exception of the kernel. I'd be happy to provide any further information that may be helpful. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: root (/) on software raid
On Tue, Aug 23, 2005 at 02:34:45PM -0400, Erik Karlin wrote: If you check the rootraiddoc.97.html, usually /usr/share/doc/mdadm/rootraiddoc.97.html or via http://alioth.debian.org/projects/rootraiddoc Down around section 8, just before the appendix, there is this little blurb: When using mdadm, mkinitrd will only detect disks in the array that are running at the time of execution. You should not install a new kernel while the array is degraded, otherwise, even if you do an mdadm --add, the next reboot will still be degraded! The array is started at boot time by script. You can see what is in the script of the initrd by mounting it, e.g. mount /boot/initrd.img-X.X.X /mnt -o loop cat /mnt/script And look for the array start line similar to mdadm -A /devfs/md/0 -R -u 23d8dd00:bc834589:0dab55b1:7bfcc1ec /dev/hda1 /dev/hdc1 That sounds like what you're seeing I think that was it, I recreated the initrd image once the second volume was resynced and it now boots fine. Thank you for that pointer. I did some searching on this prior to posting but didn't come across the document you referenced. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Difference between Gnome and Debian menus. Why ?
On Thu, Dec 23, 2004 at 09:47:11AM +, Jon Dowland wrote: On Wed, Dec 22, 2004 at 06:05:54PM -0200, Rogério Brito wrote: On Dec 22 2004, William Ballard wrote: You could file a bug and request an option to have Debian's menu be *the* Gnome menu. Now, that would be a good thing, IMO. Seeing the applications separately isn't that intuitive for the new users that I've been using as guinea pigs for deploying Open Source Software. Absolutely. The debian menu system's menu should be the primary menu for all menu-carrying apps imho. Any arguments against it are usually complaints about the menu system which should be fixed, not ignored. Having two menus in KDE/GNOME vs. the rest of the world is too confusing. So, why not file the bug reports to request these changes? -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Difference between Gnome and Debian menus. Why ?
On Thu, Dec 23, 2004 at 07:52:37PM +0100, Osamu Aoki wrote: On Thu, Dec 23, 2004 at 11:38:27AM -0700, Jamin W. Collins wrote: On Thu, Dec 23, 2004 at 09:47:11AM +, Jon Dowland wrote: On Wed, Dec 22, 2004 at 06:05:54PM -0200, Rogério Brito wrote: On Dec 22 2004, William Ballard wrote: You could file a bug and request an option to have Debian's menu be *the* Gnome menu. Now, that would be a good thing, IMO. Seeing the applications separately isn't that intuitive for the new users that I've been using as guinea pigs for deploying Open Source Software. Absolutely. The debian menu system's menu should be the primary menu for all menu-carrying apps imho. Any arguments against it are usually complaints about the menu system which should be fixed, not ignored. Having two menus in KDE/GNOME vs. the rest of the world is too confusing. So, why not file the bug reports to request these changes? Oh no. Aside from reasons already stated, not all programs are listed under Gnome menu. Gnome menu is a nice addition but it is not yet substitute of Debian menu. Perhaps I missed something but this didn't seem to be a suggestion to have the Gnome menu replace the Debian menu, but rather the Debian menu replace the Gnome menu. So, I don't see how your comment applies. Current set up is a good practical compromie. From a geek perspective yes. From an average user perspective, not really. It not very intuitive. Sure, once a user has had it explained to them, they know where to look, but I've seen a number that haven't found it initially. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian retail kit?
On Wed, Dec 08, 2004 at 04:57:24PM -0800, Paul Johnson wrote: I'm looking for suggestions on what people would expect in a retail kit for Debian, as I've been put in charge of pushing Debian in the shop I work at. Right now, since we're concerned only with i386, I'm thinking a Sarge CD set and a printed copy of the installation manual, all in a single binder. Sarge RC2 net-inst CDs have been selling with lukewarm reception at $9.50 for the CD by itself in an envelope, but I think I can do better than that. I haven't found anybody who sells boxed Debian sets, and haven't seen any discussion in the archives this decade about the topic. Target audience are more experienced Windows users looking for a way out. So if anybody has any suggestions for what a small shop can put together, or a distributor that sells Debian boxed sets, I'm looking to hear from you. I would highly avoid selling a non-release version. If you want to sell a version more current then woody now, I would look to Ubuntu or other similar Debian based distro. Selling a customer a non-release verion with no security support is not in either your or the customer's best interest. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: disappointment with linux icq clients
On Fri, Nov 26, 2004 at 04:58:49PM -0200, Frederico Rodrigues Abraham wrote: Hi... Does anybody else have the feeling that simply no linux icq client that is available nowadays fits your needs? No indication of what your needs are. licq, gaim, kopete, they all have this strange interface... Each of these were most likely coded to fit someone's needs. I used licq for quite some time before switching to Jabber. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Is life with 'udev' good?
On Fri, Nov 19, 2004 at 08:55:42PM -0600, Ron Johnson wrote:
On Fri, 2004-11-19 at 19:32 -0700, Jamin W. Collins wrote:
Don't know about a general concensus but I'm quite happy with
udev's operation and having consistent device names for my USB and
Firewire devices.
What kind of names? Can you give some examples?
$ cat /etc/udev/rules.d/cruzer.rules
BUS=usb, SYSFS{product}=Cruzer Mini, KERNEL=sd*, NAME=%k,
SYMLINK=cruzer%n
The above entry makes sure that anytime I connect my USB memory stick
the kernel device that it's assigned to is symlinked to /dev/cruzerX
where X is the partition of the devices partitions.
Then in my fstab I have something like the following:
/dev/cruzer1/mnt/cruzer autodefaults,user,noauto,noatime 0 0
No matter how many USB or Firewire devices I connect my Cruzer memory
stick is always mountable. I suppose if I ever connected multiple
Cruzer sticks I would have a problem, but I don't forsee myself doing
that anytime soon.
--
Jamin W. Collins
Remember, root always has a loaded gun. Don't run around with it unless
you absolutely need it. -- Vineet Kumar
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Is life with 'udev' good?
On Fri, Nov 19, 2004 at 02:43:31PM -0500, Christian Convey wrote: Hey guys, I'm considering installing the 'udev' package as part of my Sarge 2.6 installation. My motivation is that I'm often baffled when trying to figure out which USB device is associated with USB devices I plug in. Is there a general concensus about whether udev makes life better or worse? Don't know about a general concensus but I'm quite happy with udev's operation and having consistent device names for my USB and Firewire devices. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Is life with 'udev' good?
On Fri, Nov 19, 2004 at 03:41:32PM -0500, Williams, Allen wrote: What did you have to do to get it to work with the nvidia driver? as root: echo nvidia /etc/modules -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Dual Monitor Configuration - What Am I Missing?
On Thu, Oct 21, 2004 at 02:01:50PM -0700, Scarletdown wrote: I am trying to get my system set up to be able to output to the TV (an old Amiga monitor connected to a VCR). Yesterday, I went ahead and purchased an S-Video to composite adapter and ran a cable from the S-Video Out port on my video card (NVidia Geforce 5600FX) to the video in port on the VCR. When I rebooted, I was able to watch the boot process on the TV (kinda cool in a geeky way). But when the system tried to load the X-Server, the screens (both the TV and my primary monitor, which is a Trinitron Multiscan 17Se set to 1280x1024-24 Bit Color) flashed the NVidia logo a few times and then crashed back to the console. If it helps troubleshooting any, here are the relevant lines from my XF86Config-4 file... So, what do I have configured incorrectly here? It might also help to provide your /var/log/XFree86.* file(s). They should have at least some indication of why X failed to start. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: SSH Cracking Attempts
On Wed, Sep 29, 2004 at 04:10:58PM -0400, Nicolas wrote: So, my question is this. Is there a way to tell ssh to refuse connections from an ip address after a certain number of failed login attempts, or is snort the only way to do something like this? So far I've been taking the manual approach, blocking the ip address with my firewall after I see it hitting the logs, but that can give them about an hour to play before I notice it (e-mailed to me by logcheck). Any suggestions? If you dont have to much user who log in your server, you can allow only them from specific IP to log in. Or you can disable the password facility and only use keys (we do it this way at the job, It's also what I do at home). You'll want to be careful about how you disable password authentication and which versin of SSH you're using. Recent Debian ssh packages automatically enable the UsePAM directive when upgrading from older package versions (include the version found in woody currently). This can lead to password authentication being turned back on, even though the admin turned it off. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=250369 -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Does Unstable become Testing?
On Sat, Sep 25, 2004 at 08:55:41AM -0500, John Fleming wrote: So will there be some warning preceding the release such that those of us sitting on the fence will have a last minute chance to decide whether we want our sources.list to point to sarge or to testing? - John That can be decided now. If you want to follow sarge from it's current position of testing to stable just use sarge in your sources.list. If you want to continue using testing even after sarge becomes our stable release, use testing in your sources.list. -- Jamin W. Collins Facts do not cease to exist because they are ignored. -- Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Switch MDK - Debian, probably
On Wed, Sep 08, 2004 at 09:19:37PM +0200, Olaf Marzocchi wrote: - with MDK, to update the main packages I have to reinstall the OS every 6 or 12 months. I don't use linux as main OS (I use Mac OS X), so it's pretty annoying... the time I spend mantaining the OS is too much. What about Debian? Can I update the whole system (without adding packages, I mean, only updates) without the need of a complete reinstall? (MDK main OS installer is not really good in doing upgrades). I don't want to reconfigure everything every 6 months... Yes: apt-get -u dist-upgrade - In MDK, when I had to upgrade from KDE 3.1.x to KDE 3.2 (I had MDK9.2, I tried to update only what I needed), I had to force the urpmi system to uninstall all the kde packages (I had to force it because urpmi keeps track of every dependency: a wonderful system, it always worked beautifully except this time, not like the original rpm). Unfortunately, the process wasn't flawless. When I started kde 3.2, I found that kdm (login manager) lost every WM other than kde... OO.org never started anymore... and similar things. I heard a friend saying in Debian the process would have been as simple as a single cmdline. Is it true (I mean, *facts*, not it should be so, please...). Remember that a traditional update would be simple even under MDK, I take KDE 3.1.x - 3.2 as example because the packages number/names changed, urpmi couldn't cope with this. What about .deb? This point is important. I don't recall specifically having done this particular upgrade, but I've had very few problems with application upgrades using apt-get. - what about the kernel? Did the 2.4 - 2.6 change require a complete os install? No, I've upgraded two systems to the 2.6.x kernels without the need to reinstall the OS. In fact, it's very rare that you'll ever need to reinstall Debian short of either _wanting_ to or hardware failure. - kernels: are they patched? MDK ships a kernel heavily patched as standard, in my opinion this is really useful. If debian kernels are clean, can I take a MDK kernel (let's suppose I compile it, but what about taking the rpm with the kernel?) and use it in Debian? (I suppose yes, but who knows) Debian kernels are somewhat patched. I normally compile my own using the make-kpkg utility from kernel-package since I like to have a few other things such as ipsec (openswan), mppe, and i2c support. But make-kpkg makes this very easy. - rpm packages are everywhere... what about .deb? I'm able to compile apps, but, since having a package allow me to uninstall it cleanly and simply, I always prefer prepackaged apps. Will I be able to use rpms? You _could_ continue to use rpm packages via alien, but I wouldn't suggest it. In most cases you'll probably be able to find a deb (official or unofficial) for what you're looking to use. - will I be able to use MDK, SUSE and Fedora (the latter doesn't matter that much, I never seen them) configuration tools? AFAIK, Debian leave the user alone, there are no debian tool to configure the OS (I originally chose MDK due to this). Note: this point is a must. Without GUI tools to speed up system configuration, I won't choose Debian. In most cases, probably not. There are a number of GUI configuration tools available in Debian. More specific examples would probably be needed of what you'd like a GUI configuration tool for. If you need to know, I'd choose the testing branch, even if don't remember the kernel it ships... I hope 2.6. [update: no, it ships 2.4.something. What about 2.6? I want it] Testing currently has kernel packages for 2.6.7 and 2.4.27. Last thing: what about reiserfs4? will it be among the FS choices? if not I'll choose reiserfs, but is a rfs3.6-rfs4 upgrade possible without format? Not sure on this as there was a rather heated dispute a while back about the licensing. -- Jamin W. Collins Facts do not cease to exist because they are ignored. -- Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: VNC's XDMCP requests now fail?
On Sun, Sep 05, 2004 at 02:45:10PM -0700, Marc Wilson wrote: On Sat, Sep 04, 2004 at 08:49:34PM -0600, Jamin W. Collins wrote: Interesting, that was indeed it. Once I added a -fp to the Xvnc invocation, everything worked fine. Strange that has never been needed before. Any idea what changed to make this a new requirement? No idea. I went back as far as the versions I had in my cache, but didn't pursue it any farther than that. I suppose you could consider it a bug, since my xdm configuration as regards fonts is entirely stock, but based on d-u traffic I didn't think there were enough clued people running Xvnc in this manner to bother. Dunno if it'd be xdm's or vncserver's bug, though, which is another reason not to go there. Well, here shortly it should enter google's archives and be available for the next person to get hit by this. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: VNC's XDMCP requests now fail?
On Sat, Sep 04, 2004 at 02:51:48PM -0700, Marc Wilson wrote: On Fri, Sep 03, 2004 at 06:08:49PM -0600, Jamin W. Collins wrote: Sep 3 17:45:35 thor wdm: X Error of failed request: BadAlloc (insufficient resources for operation) Sep 3 17:45:35 thor wdm: Major opcode of failed request: 45 (X_OpenFont) Sure... if you'll note the actual error, wdm is trying to find the font you've configured to display in the chooser, and failing. Specify your fontpath *explicitly* in your Xvnc invocation, and you'll have no further problems. I merely pointed mine to the font server running on the local LAN. Or switch to a font that IS in the default fontpath Xvnc uses. Been there, done that, got the t-shirt several weeks ago. Interesting, that was indeed it. Once I added a -fp to the Xvnc invocation, everything worked fine. Strange that has never been needed before. Any idea what changed to make this a new requirement? -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
VNC's XDMCP requests now fail?
For a while now I've been tunneling VNC over SSH and everythings worked fine. I'd configured inetd to kick off a VNC session on the remote system and restricted to only allow connections from the loopback address. However, I noticed today that this was now broken on several (most) of my systems. The VNC session now immediately terminates on all but one system. After bit of searching I found the following errors in syslog: Sep 3 17:45:34 thor Xvnc[5787]: connect from localhost (127.0.0.1) Sep 3 17:45:35 thor wdm: Cannot open config file. Using builtin defaults Sep 3 17:45:35 thor wdm: X Error of failed request: BadAlloc (insufficient resources for operation) Sep 3 17:45:35 thor wdm: Major opcode of failed request: 45 (X_OpenFont) Sep 3 17:45:35 thor wdm: Serial number of failed request: 50 Sep 3 17:45:35 thor wdm: Current serial number in output stream: 51 Sep 3 17:45:35 thor wdm: Greet: guarenteed_read error, UNMANAGE DISPLAY Sep 3 17:45:35 thor wdm: Greet: pipe read error with /usr/bin/X11/wdmLogin These appear to kick off every time the VNC client attempted a connection and inetd attempted to start the VNC server. I can replicate the errors by running the following command on the remote system: Xvnc -once :1 -query localhost If the -query option is dropped, and thus VNC's XDMCP request, VNC works fine. I've tried switching display managers from wdm to xdm with no change in the error message. I've also tried both tight and real vnc on the remote system without any change in the error. Relevant portions of my configuration files: /etc/X11/wdm/wdm-config: # Don't listen for XDMCP !DisplayManager.requestPort:0 /etc/inetd.conf: vnc stream tcp nowait nobody /usr/sbin/tcpd /usr/bin/Xvnc -inetd -query localhost -once -geometry 1024x768 -depth 24 /etc/hosts.allow: Xvnc: LOCAL /etc/hosts.deny: ALL: ALL Any ideas? -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Home Debian Mirror
On Tue, Aug 17, 2004 at 02:04:05PM -0700, Paul Johnson wrote: Preston Boyington [EMAIL PROTECTED] writes: What is the best way to take the CDs and convert them into a working APT archive? Is this something I would use apt-mirror for, or is there another program? CD's come as-is, ready for apt. I believe he actually wanted to use the CDs to create a central internal repository that he could use for network installs rather than feeding each new machine the necessary CD(s). -- Jamin W. Collins Facts do not cease to exist because they are ignored. -- Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Home Debian Mirror
On Tue, Aug 17, 2004 at 04:37:18PM -0700, Paul Johnson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jamin W. Collins [EMAIL PROTECTED] writes: On Tue, Aug 17, 2004 at 02:04:05PM -0700, Paul Johnson wrote: Preston Boyington [EMAIL PROTECTED] writes: What is the best way to take the CDs and convert them into a working APT archive? Is this something I would use apt-mirror for, or is there another program? CD's come as-is, ready for apt. I believe he actually wanted to use the CDs to create a central internal repository that he could use for network installs rather than feeding each new machine the necessary CD(s). And that changes things how? 8:o) The CDROMs use basically the same filesystem layout. Just compare and contrast to a mirror... Having never used the CDs (I've always net installed) I honestly don't know. However, your initial response didn't seem to answer the question the poster was asking. I assume that each CD has a Packages listing for the files in it's subset of the archive. Checking a 3.0r1 CD seems to back this up since the Packages (non-compressed) listing on disk 1 is a mere 811K while the same branch on my local mirror is 6.3M. So at minimum these Package listings would need to be regenerated. This is probably what you mean by compare and contrast to a mirror. However, I do believe the original poster was looking for a tad more information than that. -- Jamin W. Collins It has always been Debian's philosophy in the past to stick to what makes sense, regardless of what crack the rest of the universe is smoking. -- Andrew Suffield -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Login Shell/Profile: Stop the Madness
On Sat, Jul 17, 2004 at 08:04:39PM +0200, Matthias Czapla wrote: On Sat, Jul 17, 2004 at 01:20:13PM -0400, Daniel B. wrote: But what the display manager should do at some point is start a login shell (that is, start the user's selected shell, with the standard login flag for shell programs so that shell reads the user's login-time files). That's what logging in to a virtual console does. Yes, and then this login shell keeps running and is the parent of any subsequently started processes. If the display manager just starts a login shell then this shell is unable to change the environment of it's parent (the dm) and if this shell exits everything will be exactly as it was before it has been run. So we would need to start a login shell that will become the parent process of everything else that you would like to have an initialized environment. I don't know how a dm works internally to say if this is possible though :| I'll admit I too don't know the internals of it but I would think that the solution proposed previously would work for at least some of the cases: Message-ID: [EMAIL PROTECTED] On Thu, Jun 17, 2004 at 07:58:56PM -0400, Michael B Allen wrote: There are no new files. The change is very simple. Ultimately the correct solution is to edit /etc/X11/Xsession.d/99xfree86-common_start to read: exec -l $SHELL -c $STARTUP If I'm understanding what I've looked at of the Xsession startup the value of $STARTUP could be set prior to calling Xsession and if so will be used as the command to start. If not it will fall back to using the user's personal startup config. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla/Firefox PostScript/default security problems
On Mon, Jul 12, 2004 at 09:33:52AM +0200, Magnus Therning wrote: On Sun, Jul 11, 2004 at 07:28:56PM -0500, Brad Sims wrote: On Saturday 10 July 2004 11:29 pm, Marc Wilson wrote: The numerous bugs that have been filed, and the way they've been dealt with, would seem to indicate that he's not interested in participating. Indeed, his entire argument consists of Me, Debian Developer. you, user. Me make decision; you no make decision. I will simply roll my own packages and he can go masturbate his ego in his own little corner of the net. Will you put those packages somewhere where others can reach them as well? If hosting for these packages is needed, I should be able to provide a repository for them. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: asterisk in console
On Sun, Jul 11, 2004 at 02:01:08PM -0500, Cheryl Homiak wrote: I'm trying to experiment with asterisk and installed the debian packages. However, this apparently starts asterisk as a daemon and then won't allow me to run it from the console since it's already running. Even if I stop the daemon, I can only run it from console as root; if I try to start it with the -c option as a user I get complaints about unable to open pid file: permission denied; unable to bind socket to /var/run/asterisk/asterisk.ctl: address already in use; unable to create event log: permission denied. What do I need to do to set this up so the user can run it in console, or am I going to have to go get the source and install it myself? I am blind and do need to run it from the console. thanks. The asterisk daemon runs as the user asterisk. This user isn't allowed an interactive login (shell set to /bin/false). However, when the daemon is running you can get an asterisk console by running: asterisk -r If you want to increase the verbosity of the console just add a few v's to that command like so: asterisk -vvvr With the current configuration I've found that I need to request the console as root, but that can be easily done using sudo. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] yahoo protocol switching
On Sat, Jul 10, 2004 at 05:49:42PM -0700, Karsten M. Self wrote: Problem I've had with GAIM on WinXP Pro (pardoning the OT issues) is that the installer fails with what appears to be a Win32/16 problem in the command interpreter. I found some MSFT KB articles referencing solutions in NT, but nothing appropriate to XP. Problem doesn't seem to exist with Psi on WinXP Pro. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla/Firefox PostScript/default security problems
On Wed, Jul 07, 2004 at 02:49:10AM -0400, Michael B Allen wrote: On Tue, 6 Jul 2004 23:19:14 -0600 Jamin W. Collins [EMAIL PROTECTED] wrote: Direct print is the only way I can get reliable output here (I have both options). Almost every time I use Xprint the last part of a line is missing between pages. I haven't been able to locate a cause for this. Is your paper definition correct? If it is set as A4 or something other than Letter that might account for the incorrect size. Yes, it set to letter (which is correct for the paper I'm using) on both the cups client and server machines. Any postscript printing works fine, but not xprinting. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla/Firefox PostScript/default security problems
On Wed, Jul 07, 2004 at 01:04:34PM -0400, Wayne Topa wrote: Jamin W. Collins([EMAIL PROTECTED]) is reported to have said: On Wed, Jul 07, 2004 at 02:49:10AM -0400, Michael B Allen wrote: On Tue, 6 Jul 2004 23:19:14 -0600 Jamin W. Collins [EMAIL PROTECTED] wrote: Direct print is the only way I can get reliable output here (I have both options). Almost every time I use Xprint the last part of a line is missing between pages. I haven't been able to locate a cause for this. Is your paper definition correct? If it is set as A4 or something other than Letter that might account for the incorrect size. Yes, it set to letter (which is correct for the paper I'm using) on both the cups client and server machines. Any postscript printing works fine, but not xprinting. And in /etc/Xprint/C/print/attributes/document ??? No, because I have my locale set (I thought) appropriately to LANG=en_US.UTF-8. Based on this, checking /etc/Xprint/en_US/print/attributes/document revealed: # US and some other countries use US-Letter as default paper size # (C-locale default is ISO-A4) *default-medium: na-letter Which would appear to be correct. For grins, I changed /etc/Xprint/C/print/attributes/document to *content-orientation: portrait *copy-count: 1 *default-medium: na-letter *default-printer-resolution: 300 on both the cups server and client machines, and restarted xprint (just to be safe). Test output from both the client and server itself still exhibit the exact same problem at the end of the pages. To be consistent with my testing I printed the same URL each time. In all cases except Postscript/Default the last few lines of the first page are truncated. You can see the truncation in this scan: http://gabfest.net/xprint-cutoff.png -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla/Firefox PostScript/default security problems
On Tue, Jul 06, 2004 at 09:15:36PM -0700, Marc Wilson wrote: Direct printing works for some people, and for others it doesn't. XPrint works for some people, and for others it doesn't. XPrint is *not* an arguably superior product, so why is that choice forced on people? Direct print is the only way I can get reliable output here (I have both options). Almost every time I use Xprint the last part of a line is missing between pages. I haven't been able to locate a cause for this. However, the same pages printed with the Postscript/Default are perfect. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Ugly firefox icon
On Tue, Jul 06, 2004 at 08:21:43PM -0300, Toshiro wrote: I'm using sid and I noticed that instead of the fine firefox default icon (the fox around the globe) I have a really ugly blue globe; am I the only one with this icon or is it part of debian? Anybody know how to get the original icon back? Short answer is you can't get the official icons in a non-official package. That is anything not packaged directly by the Firefox team. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] yahoo protocol switching
On Thu, Jun 24, 2004 at 11:58:46PM -0700, Paul Johnson wrote: Kirk Strauser [EMAIL PROTECTED] writes: On Thursday 2004-06-24 12:48 pm, Nori Heikkinen wrote: has anyone else been bitten by this, and found a workaround? I don't mean to sound like an ass, but that's what happens when you rely on the whims of a proprietary vendor. Set up a Jabber server and start migrating your friends to it. Or just grab a jabber client and use ursine.ca as a jabber server, my jabber server is open to the public. KDE users may find Kopete to be a very nice fit. I manage two public Jabber servers on hosted servers with good connections for anyone interested. They are gabfest.net and allchitchat.com. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] yahoo protocol switching
On Fri, Jun 25, 2004 at 02:04:10PM -0400, S.D.A. wrote: On Fri, Jun 25, 2004 at 01:21:41PM -0400 or thereabouts, Derrick 'dman' Hudson wrote: On Thu, Jun 24, 2004 at 01:37:59PM -0500, Kirk Strauser wrote: | Set up a Jabber server This is the easy part. | and start migrating your friends to it. This can be nearly impossible to do. (this is from someone who has had a jabber server running for ~2 years or so, but everyone I have regular (personal, not remote) contact with uses AIM if they use any IM at all) I think it's partially the feature set as well. I know with MSN and Yahoo that one can use VoIP, WebCams etc., can one do so with Jabber? I'm not completely sure of the specifics, but I know there was an article recently about myJabber adding SIP support. -- Jamin W. Collins It has always been Debian's philosophy in the past to stick to what makes sense, regardless of what crack the rest of the universe is smoking. -- Andrew Suffield -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Jabberd on Woody
On Fri, Jun 18, 2004 at 08:47:14PM -0500, Jacob S. wrote: Ok, so the xml config file doesn't look too bad. I've just about configured everything the way I want it (provided I really understood everything properly, of course. :-) However, I have a question or two. Do the Jabber packages in Woody come with the modules for acting as a gateway to aim/icq/etc.? No. /etc/jabber/jabber.xml said to be sure to read the documentation about each module before using it, but the documentation in /usr/share/doc/jabber didn't mention anything about them. That's because they were not packaged at that time. They've been packaged since and should make it into Sarge. Is the process for creating an openssl cert key for a Jabber server similar to creating one for a https server? IE: you can create your own, but clients will complain because it's not signed by a trusted certificate authority. Anything special to watch out for when creating your own? Pretty much. More details can be found here: http://jabberd.jabberstudio.org/1.4/doc/adminguide#security-ssl Is the Woody Jabber server recent enough, or are there some backports I should use? I would say no. There have been several enhancements to the package since the version in woody. I'm sure that some would still say that the version in testing and unstable is also not good enough due to some limitations of the implementation. However, I run 3 different servers using the Debian packages and have very few problems with them. There are not any official backports at this time. Though I have considered maintaining backports for the server and the transports. I put together a few backports for one of the servers I run. Let me know if you want them and I'll make them available. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Exim4 Config
On Sun, Apr 25, 2004 at 09:49:22AM +0100, David Cannings wrote: On Sunday 25 April 2004 04:06, Paul Johnson wrote: David Cannings [EMAIL PROTECTED] writes: Would it be best to write my own, simpler /etc/init.d/exim4? No, just edit /etc/exim4/conf.d instead, it's way simpler. So many folders, so many files! I quite like the monolithic configuration file and actually have one that I know works elsewhere so I chose that option when I installed. I can't seem to `dpkg-reconfigure exim4` for some reason, I must have broken something somewhere. IIRC, it's actually 'dpkg-reconfigure exim4-base' that will get the options your looking for. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Exim4 Config
On Sat, Apr 24, 2004 at 11:26:27PM +0100, David Cannings wrote: I'd like to write my own configuration file for Exim, without using the Debian way of editing template files or /etc/default/exim4. It seems even the file /etc/init.d/exim4 is set to regenerate the configuration file however, I don't want my custom one overwritten. How can I make my own configuration file and not have it overwritten by files generated from /etc/exim4/exim4.conf.template? Simply create an /etc/exim4/exim4.conf. If that file is present the debian scripts will not create their automated version. -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: 2.6.4 to 2.6.5 problems w/ a sarge machine
On Fri, Apr 23, 2004 at 05:13:33AM -0700, Richard Weil wrote: I guess there could be a bug. I'll file a bug report against 2.6.5, though it seems like such an odd problem. I've already filed a bug against 2.6.5[1] for something similar, it's since been reassigned to initrd-tools. [1] - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=245238 -- Jamin W. Collins It has always been Debian's philosophy in the past to stick to what makes sense, regardless of what crack the rest of the universe is smoking. -- Andrew Suffield -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: 2.6.4 to 2.6.5 problems w/ a sarge machine
On Thu, Apr 22, 2004 at 11:41:35AM -0700, Richard Weil wrote: A 2.6.4 versus 2.6.5 oddity in Sarge ... I have a hosted machine that is running Sarge with a self-compiled 2.6.4 kernel. The kernel matches the hardware very well, the kernel was made using kpkg, and update-grub is linked into the process of installing or uninstalling a kernel. The source for 2.6.5 came into Sarge a day or two ago. I complied it using the exact same configuration as 2.6.4. I checked all the new additions to the kernel and there was nothing important. The machine panics when I try to boot into the new kernel. I don't have full details because I don't have access to the machine, but I was told the machine had hung with an attempting to kill init! message. I recently ran into similar problems on one of my development boxes here locally[1]. I was originally running a custom 2.6.3 kernel and took it down to add more memory to it. I tried rebuilding the initrd image for my custom kernel a few different times and even forking a shell during the initrd load to manually try the steps. Eventually, I purged my custom kernel and went with the Debian 2.6.5-1-k7 image, then filed the bug report. [1] - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=245238 -- Jamin W. Collins Facts do not cease to exist because they are ignored. -- Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Antivirus (with exim+courier-imap+fetchmail)
On Thu, Apr 15, 2004 at 09:54:01PM +0100, Thomas Halahan wrote: My budget is small, maybe $100. The following is the only HOWTO I have found http://www.clues.ltd.uk/howto/debian-sa-fprot-HOWTO.txt My question is therefore, what sort of suggestions people have to apply antivirus scanning? Perhaps the following may be of use: http://www.flatmtn.com/computer/Linux-EximMailScanner.html -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: AMD vs. Intel
On Wed, Apr 07, 2004 at 12:44:48PM +0200, martin f krafft wrote: I am a huge fan of AMD, not only because their processors are cheaper. Recently, however, I have experienced random crashes on two machines that run AMDs. The crashes seem to be related to IO and happen usually when there is a lot of disk activity. The disks themselves are fine, though, and also the controller appears without problems. Thus I am logically considering chipset and processor. I can hardly imagine that this is a problem with AMD, but I would like to know from you success and failure stories of AMD processors and Linux. I run mostly AMD systems here. Currently have systems ranging from an older K6 233Mhz up through K7 XP2400+ several are subjected to near constant disk activity (one is in use as a PVR). Recently had a lockup of my K6 300Mhz, but that appears to be related to repeated lost interrupts to the main IDE drive (monitoring it now). Other than that recent lockup, I can't say I've had any problems with my AMD systems. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: VoIP with Debian - bad day ??
On Thu, Apr 01, 2004 at 01:57:37PM -0800, Alvin Oga wrote: On Thu, 1 Apr 2004, Paul Johnson wrote: Bryan Ladd [EMAIL PROTECTED] writes: Anyone here use asterisk or any other VoIP packages for Debian?? I'm sure there are. Try posting a question instead of just polling. http://www.catb.org/~esr/faqs/smart-questions.html yup... i agree about the troll question ... but i'll bite ... my impression .. asterisk more tailored as an (pbx) answering machine ... Really depends. You can use an Asterisk box to connect normally analog phones and use them to place VoIP calls. Additionally, the Asterisk could be used to connect to geographically divergent locations and provide local telco access from one to the other. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: webowner webgroup
On Tue, Mar 30, 2004 at 08:47:32PM -0700, Joseph wrote: I'm writing a script and I need to find out the webowner and webgroup of the apache server when I'm log-in as root. There are some scripts for this and more in the wwwconfig-common package. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: SER fails with Too much shared memory demanded
On Sat, Mar 27, 2004 at 04:58:49PM -, Martin Coggin wrote: I also found the following quote again on SER user mailing list but I have been unable to find out which version of Linux kernel Im using. Try uname -a from a command prompt. http://mail.iptel.org/pipermail/serusers/2004-February/005966.html in extract form Desc: ser won't run on linux kernels 2.4 (fails with EINVAL when intializing the shared memory) Until you've verified that you're indeed using a 2.4.x kernel, this is most likely your problem. I've got SER running just fine on at least 8 Debian boxes. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ut2004full installer
On Sat, Mar 27, 2004 at 08:15:27PM -0500, Thomas G wrote: Well since Atari or no one does Tech Support on UT I have a question... When I start installing ut2k4 everything goes ok. When it asked me for the second cdrom. I cant unmount the cdrom. I cant force unmount it and if I force lazy unmount it it unmounts but then the cdrom refuses to eject. Then if I use a paperclip to get the tray open and then I put the second CD there and I mount it but it still sees the contents as if its the first cd and I try to navigate it whatever im navigating with just locks up. I assume you're running the installer directly from the CD? If so, don't. Copy the installer to your HD, and run it from there. Can't unmount a CD that you're using, and if you're running the installer directly from the CD your using the CD. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Looking for nice, small display manager
On Sat, Mar 13, 2004 at 05:00:01PM +, W. Borgert wrote: I'm looking for a nice display manager (for XFCE4). My expectations are: * good looking :-) (GTK+ 2 preferred) * small (not hundreds of dependencies on GNOME, KDE, ...) * remote capable (XDMCP support) * system menu (reboot/shutdown) Considered so far: (snip) wdm - works, but I had problems with system menu, not sure about XDMCP What problems did you have with wdm? XDMCP works fine. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: USB NICs
On Mon, Mar 08, 2004 at 09:24:02AM -0500, stan wrote: I replaced a PCI NIC with a USB one this weekend, adn when the machine boots the networking does not come up any more. If I run the networking init.d script (stop and satr) then networking comes up fine. My suspcion is that the networking init script is run before the USB NIC is set up. dmesg is not a lot of help here, as this is all handled in the init scripts _after_ teh kernel finishes booting. There doesn't seem to be a log file that contains all the post kernel init boot stuff. Can I cause one to be created? Anyone have any sugestiosn as to how to start troubleshooting this? IIRC, by default networking starts at rcS.d/S40 and if you're using hotplug it's not started until rcX.d/S11 (where X is your runlevel). So, yes, networking does start before hotplug has a chance to load your USB modules. However, module-init-tools is run at rcS.d/S20 which is before networking. So, if you list the necessary modules for your USB NIC in /etc/modules, they should be loaded prior to network startup. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Maxtor IDE controller cards not working
On Sat, Mar 06, 2004 at 12:11:05AM -0500, Greg wrote: I have a pc that I use as a fileserver and I have recently migrated it from RH 7.3 to Debian Woody (with some stuff from Sarge and Sid in it). I have 6 disks in the pc as follows: System/OS /dev/hda4.2 GB data /dev/hdo40 GB data /dev/hdm80 GB data /dev/hde203 GB data /dev/hdi203 GB data /dev/hdk203 GB I want to set up 3 disk RAID 5 array using mdadm and software raid. RH had no problem with it so I know it can be done. The disks that I used are recognized on the box as hde, hdi, and hdk. I can use fdisk to recognize hde, but fdisk will not recognize any other disk on the pc. As you can see from the dmesg, I am using several Maxtor controller cards (they came with the hard drives) - 1 hard drive per bus, each set as a master. During startup the BIOS gives the appropriate messages on the hard drives and the dmesg seems to suggest that the system at least can identify them. However, I cannot use 4 of the 6 disks in my pc. If I can mount and format them then I am sure I can set up the RAID array on the 3 I use for a RAID array Have you checked to see if the device nodes exist in /dev? I suspect you will find that they don't. On the system I just checked it stops at /dev/hdh20, beyond that you would need to create them. Alternatively you could look into devfs. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber 2.0
On Thu, Mar 04, 2004 at 06:27:49PM -0500, Antonio Rodriguez wrote: Hi all out there. Has anyone compiled the debs of jabber 2.0 for debian? Yes, I've been working on creating debs for it. There are three currently uploaded to experimental: jabberd2-bdb, jabberd2-mysql, and jabberd2-pgsql. I believe they are waiting on processing at this point. Any issues I should be aware of? Well, first problem I ran into is that configuring the sources with different auth and storage options didn't result in different file generated, only different dependancies of the same binaries. This and some other issues have been discussed in the BTS here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=217352 In addition to these, I would like to automate the configuration of Jabberd2 prior to uploading it to unstable. However, I'm not aware of any tools to gracefully automate the configuration of the Jabberd2 XML files, at least none currently packaged for Debian that I could use via a Depends. So, I'm looking into the packaging of one of these. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian Slow?
On Fri, Mar 05, 2004 at 01:47:01PM +1300, [EMAIL PROTECTED] wrote: Given this, I'd expect the machines to be fairly similar in speed. While I know that gentoo does optimize stuff, and it does result in performance gains, It shouldn't make that much difference. Things like GNOME and Nautilus start somewhere in the order of 10 to 15 times faster than on my machine at home. Have you made sure you're using DMA access on all your drives? -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: phpgroupware doesn't install
On Sun, Feb 29, 2004 at 12:28:25PM -0500, Rich Johnson wrote: I just tried installing phpgroupware for the first time. apt-get/dpkg asked all the questions and apparently exited without setting things up. No phpgroupware was added to the MySQL databases. No references to /etc/phpgroupware/apache.conf were added to /etc/apache/httpd.conf What gives? Where did the process go off the rails? I couldn't find anything in the mailing archives. Which version did you install? Was it perhaps one of the stable versions (0.9.14-0.RC3.2.woody2 or 0.9.14-0.RC3.2.woody3)? The packaging has significantly changed (for the better) in the testing and unstable packages. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: DDs: etiquette of inquring about status of an ITP?
On Thu, Feb 26, 2004 at 04:01:18PM -0500, Chris Metzler wrote: Would it seem annoying to you to get such an inquiry after a couple of months? Six months after posting the ITP? A year? If there's some point when it isn't rude, is it more appropriate to do it by mailing to the WNPP bug, or by a private email to the DD? I see no problem with asking after a few weeks (or shorter). Just ask via an update to the ITP so that others can see the request and response (if any). -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: google or debian-user?
On Wed, Feb 25, 2004 at 02:39:29PM -0400, [EMAIL PROTECTED] wrote: It is a shame that at times debian-user can be so inhospitable to debain users. I think it is unfair to have a question answered with try google or some other variation of RTFM. If the answer is in a readily available manual or has already been asked numerous times and archived, should these not be used? Am I to understand that instead of using the debian-user list that I should use a search engine? IMO, yes. You should first try to answer the question yourself. I was under the impression that the debian-user list was a forum for debian users. I always search the debain-user archives before asking a question and if I don't find an answer then I ask my question. That's a good first step. Generally I do not do google searches unless I am seaking global information such as is my hardware supported under linux. You'd be suprised what google can turn up. Your question may not have been asked on the list, but may be documented somewhere else and google may just find it. However I will in the future assume that any problem I have is the same in all distros and I will do searches on the web first. I concede to RT proverbial FM even if that M is spread across all of the known Internet. Who knows? maybe I will never again need to post a message to the debain user list. One can only hope... That's a very good idea. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber server howto
On Wed, Feb 18, 2004 at 04:24:40PM -0500, Antonio Rodriguez wrote: On Fri, Feb 13, 2004 at 11:28:44AM -0700, Jamin W. Collins wrote: Shouldn't be, the destination server name is specified in the XML data sent to the server. Provide the server is configured to be responsible for the destination the client is attempting to reach it shouldn't matter how it gets there. It doesn't work. I am using redirection by gandi.net, and if I direct the query as indicated in the manual (hey, I read it!) [EMAIL PROTECTED]:~$ telnet domain-name.com 5222 Trying 80.67.173.5... Trying 62.80.122.203... telnet: Unable to connect to remote host: Connection refused This doesn't look like Jabber denying the connection but something else. You can configure the server to not allow inband registration (client's requesting and creating new accounts) and instead use some of the tools at jabberstudio.org to create the accounts. I got the create-user script (the only one I could find there) http://scriptrepo.jabberstudio.org/jodb_createuser/ Some comments aren't very encouraging. Do you know of any other script? Nope, only heard of their existence them, haven't used them. For now all configuration needs to be done manually in either /etc/jabber/jabber.xml or /etc/jabber/jabber.cfg depending on what you wish to change. All settings can be changed in /etc/jabber/jabber.xml, the /etc/jabber/jabber.cfg is simply for convience of setting the host and spool location without editing the xml directly. In several places they state that in the configuration files it should say jabber.domain-name.com In other places it is not clear. Does it mean that a whole subdomain must be established in the DNS records for jabber? Or just localhost everywhere should suffice (at least in my configuration case, w/o other servers communication)? Depends on whether you want external servers to be able to use the transports/services on your server. One example is jabber-muc, it's problably good to have a host name for it that is externally resolvable, unless you never plan to invite people on other jabber servers into a conference on your server. As for the transports to other IM networks, they work just fine with host names like aim.localhost, icq.localhost, etc. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Remote access PC support
On Tue, Feb 17, 2004 at 04:46:34PM -0800, Steve Lamb wrote: Paul Johnson wrote: Yup. If every other system you're supporting is Linux, then SSH is all you need (and it's X11 forwarding option is your friend). If not, you'll have to go with the much slower, much more insecure VNC. Much slower? Erm on the LAN I've switched to using VNC for access to my X desktop because X was slower than VNC. I can't imagine SSH+X would be faster than VNC. :P In my experience, it's not. Now less secure I'd grant you. Tunnel your VNC session over SSH. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Holy Shee-it
On Sun, Feb 15, 2004 at 11:42:57PM -0800, Paul Johnson wrote: On Sun, Feb 15, 2004 at 10:31:06PM -0500, Richard Hoskins wrote: I think it is probably more than 10% if you include Outlook. Outlook is a must-have for organizations that are already heavily invested in Exchange servers. Isn't koffice/kmail/knode more or less a drop-in replacement for MS Office/Outlook at this point? Nope. The closest replacement possible (TMK) is Evolution (with Exchange plugin) and Open Office. However, the Evolution plugin only works with Exchange 2000 or later. There are still shops running Exchange 5.5 that use specific Exchange functions not accessible from other MUAs. Sure there are better MTAs to use than Exchange, but when you're talking about a drop in replacement at the desktop level (Office/Outlook) you need to look at 100% (or near 100%) functionality replication. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber server howto
On Fri, Feb 13, 2004 at 06:20:23AM -0500, Antonio Rodriguez wrote: Hi Jamin, thanks for the reply. I am trying to set a jabber server that would allow several users to interact with one another. This should be easily done with the existing Debian packages for Jabber. I also want to be able to store the transcripts of the talks. This may be a bit more difficult. I'm assuming that you mean you want to log all traffic passing through the server, right? If so, you'll need something like BanderSnatch: http://www.jabberstudio.org/projects/bandersnatch/project/view.php From what I have seen, I need (and have installed) jabber, jabber-common, jabber-jud and jabber-muc. That should allow for your users to chat directly and in group chat rooms. The intention is for the users to run a client in their own machines that would connect to my server for this purpose. I remember seeing some clients for jabber for windows (gabber was the name of one if I am not mistaken), that will help, since I expect many of the users to be running windows. Another nice cross platform client is Psi: http://psi.affinix.com/ To give you a visual of what we are trying to do, please have a look at http://www.mathhomeworkcenter.com/sample1.html From that site, it appears you are looking for some whiteboarding functionality. The only Jabber client I'm aware of that provides whiteboarding is Cocinella: http://coccinella.sourceforge.net/ It is also cross platform, The UI could use some serious work, but the whiteboarding seems to work fairly well from the limited testing I've done. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber server howto
On Fri, Feb 13, 2004 at 12:48:35PM -0500, Antonio Rodriguez wrote: About jabberd: Would a FQDN accessed through redirection from the parking site be a problem to run the server? Shouldn't be, the destination server name is specified in the XML data sent to the server. Provide the server is configured to be responsible for the destination the client is attempting to reach it shouldn't matter how it gets there. I am thinking about redirecting the top domain name to a server here in my home, semistatic ip address, changes three times a year may be. Given the nature of the procedure, we don't expect many to try to intrude into the system. You can configure the server to not allow inband registration (client's requesting and creating new accounts) and instead use some of the tools at jabberstudio.org to create the accounts. You had mentioned a script that makes the installation process more less automatic, I don't remember seing any questions asked, was it then not interactive? No, the configuration of the Jabber server does not yet use DebConf. By default the server is set with a name of localhost which and is fully functional this way. Your JID would be [EMAIL PROTECTED] and many clients allow you to manually specify the IP or name of the server. For now all configuration needs to be done manually in either /etc/jabber/jabber.xml or /etc/jabber/jabber.cfg depending on what you wish to change. All settings can be changed in /etc/jabber/jabber.xml, the /etc/jabber/jabber.cfg is simply for convience of setting the host and spool location without editing the xml directly. -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber server howto
On Thu, Feb 12, 2004 at 09:22:48PM -0500, Antonio Rodriguez wrote: Does anyone know of a good jabber server installation how to in debian? I am trying to save some time. There's always the Admin Guide: http://jabberd.jabberstudio.org/1.4/doc/adminguide The Debian Jabber packages aren't very altered. The configuration is still the same as the upstream, we've just bundled an init script and a small config file that allows easily specifying a host name and spool location. Each of the transports/services include instructions on how to add them to the configuration in their README.Debian files. I hope to have this automated eventually (pointers, suggestions, and patches welcome). Is there something in particular that you are wanting to do? -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Multi-head (non-Xinerama) window managers
On Mon, Feb 02, 2004 at 11:33:01AM +0100, martin f krafft wrote: The scenario in which I would like to use dual head is a work screen and next to it a screen for GUI applications, like the browser, jpilot, and a mail reader. Furthermore, I want multiple desktops. Running X with Xinerama means that a desktop change affects both screens, which is not what I want. Thus, I want to go without Xinerama. I use a configuration much like this for my work system. Two 19 monitors side-by-side configured in Dual Head with an almost fullscreen VMware session on the right hand monitor with a single desktop and currently 6 desktops on the left. I'm using Blackbox for my window manager, it handles both screens automatically with a single invocation. So then I went back to WindowMaker and got it to run two separate instances on the two screens. Cut'n'paste works, dragging windows between the screen obviously doesn't, but that's fine by me. However, I am still not 100% happy, mainly because of the focus issue. I'll give you two examples, keep in mind that I run 'click' focus mode, that is, I have to click a window to give it the focus. I prefer this to sloppy focus where the window under the mouse cursor is focused. I use sloppy focus, so I wouldn't know how well click to focus would or wouldn't work. If I am working on screen 0 and would like to switch to a terminal or browser window on screen 1, I have to click on the window title bar -- clicking on the window does not transfer the focus. Within a single screen, however, this works as expected. Just configured both heads here for click to focus and clicking anywhere within a window on either screen switches focus to it. The second example is related to the previous one. I have Ctrl-Alt-T mapped to give me a terminal window. However, when the focus resides on screen 0 even though the mouse is on screen 1, pressing that keycombo will pop up the terminal on screen 0. Clicking on the background or application clip does not transfer the focus. Thus, in order to open a window on screen 1, one needs an already opened window to transfer the focus to screen 1 -- a bootstrap problem that can only be overcome with the window maker application menu -- which is painful. I only have bbkeys (a blackbox keygrabber) configured to run on the first head here. However, I do have a problem where if the focused window is on the second screen my keyboard shortcuts don't activate on the first screen. With sloppy focus if I move through an open application on screen 1 it switches focus and the key combos work. Likewise if I click on the root window of screen 1 the key combos work. There are like 40 or more window managers in Debian, and I really don't want to try one after the other. Thus, if you have already made experiences with window managers and multi-head configurations that don't use Xinerama, talk to me! I'm quite happy with Blackbox and Dual Head. I'd be happy to provide any help I can. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alternative to VMware?
On Mon, Feb 02, 2004 at 12:54:03PM -0600, Joel Konkle-Parker wrote: I really want to get a copy of VMware, but I don't have $140 for the student version. So I'm looking for alternatives. I just want some kind of sandbox where I can test out new software, distros, etc, without rebooting into a seperate partition. What do other people use? After reviewing a few of the possible alternatives, my wife and I decided that VMWare really was the best option and picked up two copies for our work machines. Bochs was just too slow to be useful for day to day work tasks. Win4Lin only works with Windows 95 and 98. This was much too restrictive. As it is we have our WinXP machines (needed for the company we work for) safely tucked away in a virtual machine. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sendmail vs Exim vs Others
On Thu, Jan 29, 2004 at 09:40:28PM -0800, Steve Lamb wrote: Adam Aube wrote: My personal preference is qmail. Not sure if it's available in the Debian archive or not, but you can check out www.qmail.org for more info - look for the links to netqmail. Probably not given the nature of its license. Yes and no. There is a qmail-src package which will create a debianized package for installation. qmail, by default, will not relay AT ALL, and I have found it very easy to install and setup. o.O I've had to work with QMail and I have to say that it is one big giant headache. It was last actively developed in a day and age when SMTP could be fairly open. To get any decent security requires that you need to patch in at least 6-7 different patches because the license forbids redistribution of modified source. The qmail-src package applies a few of the more useful patches when it creates the binary package. Compared to my work with Exim QMail is one big giant nightmare. QMail can be quite inflexible at times. I haven't worked with Exim on the same level as QMail yet, but do like what I've seen of Exim so far and am planning on migrating to it in place of QMail In short, QMail is the Windows of MTAs. Sure, you can get it to work but doing so is more trouble than its worth and maintaining it is even worse. That just FUD. It may not be the easiest MTA to work with but the above is just misleading and wrong. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Esta
On Fri, Jan 30, 2004 at 10:20:53AM -0600, Joel Konkle-Parker wrote: My Mozilla (1.0.0, 3.0r2) always reacts badly to these kinds of messages (with the subject full of s), freezing up, running the cpu to 100% for 30 secs or so, leaking memory, etc. Anyone have any advice? Try a more recent version of Mozilla? I realize that 1.0.0 is the current version in stable, but there are several enhancements to Mozilla in later versions. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: script to list installed packages
On Thu, Jan 29, 2004 at 12:27:11AM -0500, [EMAIL PROTECTED] wrote: I know that somewhere there is a command to list all installed packages Perhaps dpkg --get-selections would be a good starting point? -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT - nForce, GeForce and VMWare
On Tue, Jan 27, 2004 at 07:16:30AM -0600, Jeffrey L. Taylor wrote: In my experience, action games in VMware are problematic. I have used several versions of VMware. Sound has gotten a lot better, from unusable to minor stutters. Yes, it has improved quite a bit. I've actually managed frequently clear playback of audio from my VMWare installation. Within VMware, Freecell is about the only playable game. For things that need 3D acceleration, stick to native booting Windows. Wine quite possibly will be worse. YMMV I've been using WineX for about a year now. There are a number of games that it won't play or install, but ther are also quite a few that it plays quite well. However, if your goal is to play a game that uses DirectPlay multiplayer, don't bother. DirectPlay doesn't work at the moment. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: sound blaster live support
On Tue, Jan 27, 2004 at 11:16:12PM -0500, walt wrote: Any easy way to get a sound blaster live working in debian testing? I am googling right now but would really appreciate any hints. If you are using a 2.4.x Debian kernel image: apt-get install alsa-modules-`uname -r` alsa-base alsa-utils That should install the modules to match your running kernel and the necessary utilities. For configuration information you may want to read the following web site (long link): http://www.alsa-project.org/alsa-doc/doc-php/template.php?company=Creative+Labscard=Soundblaster+Livechip=EMU10K1module=emu10k1#modp If you're not using a 2.4.x kernel I would highly recommend upgrading to one before trying to get the SB Live working. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Unable to disable IDE DMA on boot
On Sat, Jan 24, 2004 at 03:40:18AM +0200, Johannes Lehtinen wrote: I have a problem disabling IDE DMA. I am trying to install Debian Sarge in to an old laptop and with DMA enabled (default) I keep getting DMA timeouts and retries from /dev/hda. The kernel image is 2.4.23-1-386 (2.4.23-1). I had a similar problem with another system and failed to find an option that I could use to disable it at startup. Wound up making a custom nodma kernel for it. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Benefit of source packages?
On Fri, Jan 23, 2004 at 03:30:13PM -0600, Joel Konkle-Parker wrote: I'm curious: what's the benefit of source packages? Do they allow you to optimize the package for your system like Gentoo does? Are there any other reasons why they're better than regular packages? I assume you mean Debian's existing source packages? If so, they are the source that the Debian packages for the various architectures are built from. Additionally, when the package is built it may use more recent versions of libraries (like those from unstable). These normally become dependancies of the resulting package. However, the package may work just fine with an older version of the libraries. So, if you want a newer version of a package, but it requires newer libraries that you don't want to mess with you can try grabbing and compiling the source package for the newer package version and compiling it with the older libraries. Then there's also the matter of licensing. Many of the software licenses require that the source for the distributed binary be made available. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Oldworld PPC install
On Sat, Jan 10, 2004 at 10:20:20AM -0500, Fraser Campbell wrote: I've recently come into possession of a Power Macintosh G3 in beige case. I understand this to be oldworld mac. It has a 6GB IDE disk, a 4.3 GB SCSI disk and 320 MB RAM. stable/main/disks-powerpc/current/powermac/images-1.44/boot-floppy-hfs.img was booted however after about 20 seconds the penguin in the middle of the screen gets covered by a red X and the floppy is no longer being read. I then tried the BootX installer (?) from stable/main/disks-powerpc/current/powermac/BootX_1.2.2.sit and was able to boot into the installer from MacOS (8.5). I have 10 years+ of Linux experience and 6+ years Debian experience but I am confused and scared when it comes to this MAC :-( I have no idea how to partition the disk, if I wipe out MacOS will I have any way of booting into the installer again? Now I am unsure how to proceed. This is actually more appropriate for the powerpc list, but to answer your question, no. If you remove the existing Mac OS on the older systems you will have no way of booting the system. The BootX boot loader hooks into the existing MacOS boot process. Additionally your kernels are actually stored on the file system of the existing MacOS install. I too found my first oldworld Mac installation to be a bit confusing. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: need advice on fixing my home lan
On Fri, Jan 09, 2004 at 08:52:53AM -0700, Paul E Condon wrote: I have a small LAN in my home. I need some advice on tuning it. I've started working on a project wherein I move large files (3GB) between two Debian boxes. This is a slow process. I would like to be sure that it goes as fast as is reasonable. I think all my LAN cards are claimed by their makers to be 10/100, but for some this might be marketing hype. All my cables are 'CAT5'. So, some questions: How do I determine whether my lan is passing data at 10 or 100 MHz? apt-get install mii-diag That should install the mii-diag and mii-tools applications that you can use to adjust and check the settings of most NICs. Beyond that you might want a monitoring application that will give you throughput on the interface(s). Right now I'm using an application called slurm I don't believe there's a debian package for it, but it is very easy to compile and use. Eventually I plan on replacing it with mrtg and snmp. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: urgent nvidia problem
On Wed, Jan 07, 2004 at 11:44:18AM -0500, Matt Price wrote: came into work today and the computer was crashed dead; don't know what the problem was (I'll attach the last few lines of /var/log/messages from yesterday; they don't tell me much, though). Rebooted and x wouldn't start because of a problem with nvidiactl (I have to use the nonfree drivers to make X work at all...). I've tried variouS things -- modprobe nvidia, insmod NVdriver, changing to an oder kernel, none of it matters. Gdm gives slightly different errors from those recorded in XFree86.log. Here's the last bit of gdm's log: Not loading .note.GNU-stack Not loading .note.GNU-stack Not loading .note.GNU-stack NV: could not open control device /dev/nvidiactl (No such device) Does /dev/nvidiactl exist? If so, what are the permissions on it? You'll also want to check for /dev/nvidia0 - /dev/nvidia3 and what their permissions are. (==) NVIDIA(0): RGB weight 888 (==) NVIDIA(0): Default visual is TrueColor (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0) (--) NVIDIA(0): Linear framebuffer at 0xD000 (--) NVIDIA(0): MMIO registers at 0xE400 (EE) NVIDIA(0): Failed to initialize the NVIDIA kernel module! Is the nvidia module loaded (should see it in the output of lsmod). Are you using the NVidia installation routines or the Debian binary and wrapper packages? If using the Debian packages, what are the versions of your nvidia-glx, nvidia-kernel-common, and nvidia-kernel packages (for the nvidia-kernel package the actual name should be something like nvidia-kernel-$kernelversion where $kernelversion is the output of uname -r). -- Jamin W. Collins Never underestimate the power of very stupid people in large groups. -- John Kenneth Galbraith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: urgent nvidia problem
On Wed, Jan 07, 2004 at 12:42:06PM -0500, Matt Price wrote: so, no, nvidia is not loaded. I tried modprobe nvidia, and got: modprobe: can't locate module nvidia I'm using the debian nvidia packages, which I compiled myself to fit a self-rolled kernel. nvidia-glx is 1.0.3123-4, nvidia-kernel-long-kernel-name is 1.0.3123-3 (odd that the package version is different) I guess these are somewhat old. Yea, current version in Debian is 1.0.4496 the procedure for compiling them has changed a bit, the nvidia-glx package no longer needs to be compiled and there is an nvidia-kernel-common package. You might try upgrading to the newer version. I just dpkg -i --reinstall 'ed them without a hitch, but nonetheless nvidia.o does not appear in /lib/modules/kernel-version/kernel/drivers/video. Should it? Yes. To find out where it's putting the nvidia.o file try a dpkg -L of the nvidia-kernel package name. It should list the entire contents of the package including the path to the nvidia.o or in that older version it might have a different module name (seem to recall the name did change). -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: spamassassin and bayes filter after sarge upgrade
On Tue, Jan 06, 2004 at 03:37:23PM +0800, Brian Walker wrote: Greetings all, my wonderfully smooth upgrade from woody to sarge last night is showing the first glitches. Spamassassin had been installed from source previously. The upgrade appeared fine, and indeed spam is being caught and marked. When I try to teach spamassassin based on missed spam I get the following error: [EMAIL PROTECTED]:~$ sa-learn --mbox --spam /home/brian/Mail/MissedSpam Cannot open bayes databases /home/brian/.spamassassin/bayes_*R/O: tie failed: Cannot open bayes databases /home/brian/.spamassassin/bayes_*R/O: tie failed: File exists Learned from 0 message(s) (10 message(s) examined). IIRC, the perl version change effected a change of the BDB version being used. The bayes_seen and bayes_toks files should be BDB files and should be able to be fixed by doing an db4.x_upgrade on them. This is noted in the README.Debian for the SA packages in Debian: There is a issue with DB_File that causes old Bayes databases and automatic whitelists to no longer be read with perl5.8. From the perl 5.8 changelog: * NOTE: DB_File now uses libdb4.0 (previously libdb2). Any DB_File databases created with earlier perl packages will need to be upgraded before being used with the current module with the db4.0_upgrade program (in the libdb4.0-util package, with HTML docs in db4.0-doc). The fix is to delete your automatic whitelist and bayes dbs from ~/.spamassassin/, or use the db4.0_upgrade program as explained above. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: nForce 2 and Woody: How can I achieve a network install and what functionality is available without the nVidia drivers?
On Tue, Jan 06, 2004 at 11:48:17AM +, Joseph Jones wrote: How can I do a network install of Woody on an nForce 2 motherboard? I'm guessing I would need to compile a kernel with the forcedeth patch. How do I do this, if possible, with the 2.4.20 kernel (which is what I generally use)? If not possible with the 2.4.20, could I have instructions for whichever kernel version it is possible with? Also, once I've got Debian installed, what functionality can I achieve without installing the god-forsaken drivers from nVidia? Obviously, with forcedeth in the kernel I can get network access, but what about sound and USB? I've read in the list about some guy using snd_intel8x0 (alsa driver), can this be used with esound and OSS? I use the snd_intel8x0 alsa driver on both of my SN41G2 systems (nforce2 based). Works in every way you'd expect ALSA drivers to work, so yes esound and OSS support works fine. I wasn't aware of the forcedeth patch, so I have no experience with it. However, I will be looking into trying it soon. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: PPTP setup
On Sat, Dec 27, 2003 at 11:10:50AM +0100, Lorenzo Rossi wrote: I would like to setup a pptp client to connect to a win2k pptp server in my office. I use Debian Testing, i have installed the following packages: kernel-patch-mppe pptp-linux APT installed also other packages, but I suppose it has installed all necessary packeages. I have seen in /usr/src/ kernel-patches/all/mppe/, 2 files: linux-2.2.19-openssl-0.9.5-mppe.patch.gz linux-2.4.20-openssl-0.9.6b-mppe.patch.gz But I do not know how to patch the kernel and obviously how to use this files... Easiest (debian) way I've found is using make-kpkg from the kernel-package package. Then I normally copy an existing config from a Debian kernel-image package for the same version of the kernel source I'm compiling into the the extracted kernel source as .config. Then do the following: export PATCH_THE_KERNEL=YES # case is important make-kpkg --initrd --append-to-version $VER --revision $REV \ --us --uc kernel_image kernel_headers modules_image Replace $VER with whatever you'd like to see after the kernel version and $REV with a number to indicate a packaging revision. This command will use the existing kernel configuration (.config) and prompt for any new entries. The targets kernel_image kernel_headers modules_image will result in a kernel image and headers package along with a package for any module source you have installed and extracted. You'll probably need the following packages beyond what you already have: kernel-package kernel-source-2.4.xx And to get an existing Debian kernel config: kernel-image-2.4.xx-1-yy xx above is the minor revision of the kernel (eg. 21, 22, or 23) and yy is the specific architecture (eg. 686, k7) As far as using PPTP once you have everything patched, I recommend following the instructions at this site: http://pptpclient.sourceforge.net/howto-debian.phtml And looking at the documentation there, it would appear that they have a repository with pre-patched packages. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X refuses to load nVidia module
On Mon, Dec 29, 2003 at 04:18:40PM -0500, Bradley M Alexander wrote: I just built a sid box from bare metal last week, and am having problems getting the nVidia drivers to load. I tried this under 2.4.23 and 2.6.0, and in both cases, the module refuses to load. At this point, I'm not sure if it is a module loader problem or within X. X itself is reporting: Almost every box I own uses the nvidia driver. How did you build the driver? Are you using the Debian packages? If so, which version of the packages (both nvidia-kernel and nvidia-glx)? This happens whether I load the module by hand or not. The nvidia devices (nvidia[0123] and nvidiactl) exist in /dev. When you load the module manually, does it show in the output of lsmod? -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X refuses to load nVidia module
On Mon, Dec 29, 2003 at 09:59:58PM -0500, Bradley Alexander wrote: On Mon, 29 Dec 2003 14:32:08 -0700 Jamin W. Collins [EMAIL PROTECTED] wrote: When you load the module manually, does it show in the output of lsmod? Yep. That was the first thing I checked when looking under 2.4.23. When I tried to load the module under 2.6.0, I got the message FATAL: Error inserting nvidia (/lib/modules/2.6.0/nvidia/nvidia.ko): Invalid module format However, the module was loaded under 2.4.23... When the module would load, but X would fail to start, I found I had differing versions of nvidia-kernel and nvidia-glx installed on the system. Are you sure you had the same version of both installed under 2.4.23? -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ipsec kernel patch 2.4.23
On Mon, Dec 22, 2003 at 12:27:44AM -, Antony Gelberg wrote: I'm trying to install freeswan on a woody box with 2.4.23 from backports.org. I apt-got kernel-patch-freeswan, did an export PATCH_THE_KERNEL=yes, and a make-kpkg. Try this: export PATCH_THE_KERNEL=YES ### Note the case is important! I had the same problem with patches not applying a while back. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt-get and Cache-Limit revisited
On Mon, Dec 22, 2003 at 01:19:27PM -0800, David G. Schlecht wrote: I've seen many posts regarding the apt-get problem of MMap ran out of room errors. I've tried all the suggestions and nothing works. I'm thinking there's something more going on with my system. I've done apt-get clean, reduced the sources.list to just a single line pointing to the debian server, removed all my files from the list and cache directories, added APT::Cache-Limit 1200; to my apt.conf file, and run apt-get update. And I still get a MMap ran out of room error. I've found that if the apt.conf file says: APT::Cache-Limit 1200; then the apt-config dump says: APT::Cache-Limit (with quotes and without a value), but if my apt.conf file says: APT::Cache-Limit 1200; then the apt-config dump says: APT::Cache-Limit 1200;. But, nothing helps, I still get a ran out of room message. Seems that the apt-conf doesn't know how to handle numeric values. This is how I've added it when needed: $ cat /etc/apt/apt.conf.d/11cache APT::Cache-Limit 12582912; As you can see, it recognizes it: ~$ apt-config dump APT ; APT::Architecture i386; APT::Default-Release stable; APT::Cache-Limit 12582912; ... Here's my version: $ apt-get --version apt 0.5.4 for linux i386 compiled on Aug 19 2001 01:02:26 Supported Modules: *Ver: Standard .deb *Pkg: Debian dpkg interface (Priority 30) S.L: 'deb' Standard Debian binary tree S.L: 'deb-src' Standard Debian source tree Idx: Debian Source Index Idx: Debian Package Index Idx: Debian dpkg status file -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [subscriptions] Re: apt-get and Cache-Limit revisited
On Mon, Dec 22, 2003 at 02:59:19PM -0800, David G. Schlecht wrote: It looks like my apt-get doesn't work like that. Jamin W. Collins wrote: Here's my version: $ apt-get --version apt 0.5.4 for linux i386 compiled on Aug 19 2001 01:02:26 Supported Modules: Ah, perhaps the crux -- I'm running apt-get version 0.3.11 compiled July 8 1999 I'd be glad to try a newer version, but how do I get one without apt-get? wget (or other web retrieval application) and dpkg. Assuming your on an x86 system the apt package you want is: ftp://ftp.debian.org/debian/pool/main/a/apt/apt_0.5.4_i386.deb It has the following dependancies that you may need to manually retrieve: Depends: libc6 (= 2.2.3-7), libstdc++2.10-glibc2.2 Once you have the files, use dpkg -i $filename to install them. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Proxy and Firewall Recommendations?
On Mon, Dec 15, 2003 at 01:09:08AM -0800, Scarletdown wrote: So, what I need to know is, what would be a suitable set of packages to download and install for routing and firewall services for him? Pretty much any iptables firewall package and squid. Since I want to have this system operational pretty fast, I need something that is fairly simple to configure. It should also be able to block outgoing stuff from his system that he doesn't want phoning home (that would include Windows Media Player, RealPlayer, and any spyware, for example.) This gets a little more tricky, but not impossible. Most iptables scripts define a set of trusted IPs and allow all their traffic out, but blocking specific outbound ports or destination IPs isn't very hard. I also need to make sure that the MMORPG that he has become addicted to can be played through the firewall. The game is Horizons, and it is played through Internet Excreter exclusively, unfortunately due to its use of ActiveX. Shouldn't be a problem. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Jabberd segfault
On Mon, Dec 15, 2003 at 11:14:23AM -0500, Matthew Daubenspeck wrote: I just upgraded my kernel to 2.4.23 (standard from kernel.org) and everything is working great except for jabberd. After the system restart, it will no longer starts and just segfaults. $ /etc/init.d/jabber start Starting jabberd: Failed Is all I get, and if I try to start it manually, I get: $ /usr/sbin/jabberd 20031215T16:18:46: [notice] (-internal): initializing server Segmentation fault There is nothing in the log other then the above line minus the Segmentation fault... Any ideas? Which jabber package version, which architecture, and which version of libc6? Recent libc6 versions have cause segfaults with older jabber packages ( 1.4.3-1). -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Jabberd segfault
On Mon, Dec 15, 2003 at 01:57:41PM -0500, Matthew Daubenspeck wrote: On Mon, Dec 15, 2003 at 10:36:26AM -0700, Jamin W. Collins wrote: Which jabber package version, which architecture, and which version of libc6? Recent libc6 versions have cause segfaults with older jabber packages ( 1.4.3-1). arch - ppc jabber - 1.4.2a-12 (I tried -13 as well with the same result) libc6 - 2.3.2.ds1-10 Yep, this is the _exact_ combination that first reported the problem to me. Upgrading the package or recompiling the source with the new libc6 corrects the problem. Bingo! Upgrading jabber to 1.4.3-1 solved the problem. Sorry, I didn't even notice that a new package was in the list.. I only uploaded it a few days ago, no problem. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux is not for consumers!
On Thu, Dec 11, 2003 at 08:21:20PM -0600, Terry Hancock wrote: The people who know a program best are the ones who work on its internals. No one else can write documentation like the guy who built the thing in the first place. Failing that, you can have someone step in and write it, yes. But it'll never be as accurate as you'd like, nor as up-to-date. This is something of a fallacy. In some cases the guy that wrote it is the _last_ guy you want documenting it. They may take certain items/features for granted because they have become second nature to them. A good compromise is a collaborative effort where a user (or someone else) of the software/product creates the documentation using the original author as a resource. This way you get an outside point of view. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Can't get jabber to start on installation
On Fri, Dec 12, 2003 at 03:02:24AM -0800, Doctorcam wrote: I have managed to get jabber installed, and running it from the command line seems to work fine, following the instructions in the 1.4.x administration guide. Which user did you run it as? Trying to get dpkg to finish installing it, however, or (more specifically) trying to run it from the init.d script is another matter. Did the initial installation fail? Which package version are you using? The script runs fine up to this portion, and then dies: if pidof $DAEMON /dev/null 21; then echo $NAME. That is checking to see that the server actually started and stayed running. In fact, it does (I stuck a few echos into the script to track progress), though sometimes it endures past the failure of the script, and sometimes it does not. This is a known problem with the 1.4.x jabber daemon. If it exits for a bad configuration it will frequently leave it's PID file behind and then fail to start again because the PID file exists. The init script in the Debian package works around this by checking for a stale PID file (and removing it if it can) before launching the daemon. When it has endured, the owner was daemon, and permissions were 755 /var/run/jabber has jabber:nogroup permissions 755 Looks like you're using the -12 package? Can you provide the output of the following commands: ls -ld /var/run/jabber ls -l /var/run/jabber -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Can't get jabber to start on installation
On Fri, Dec 12, 2003 at 01:59:27PM -0800, Doctorcam wrote: * Jamin W. Collins ([EMAIL PROTECTED]) wrote: On Fri, Dec 12, 2003 at 03:02:24AM -0800, Doctorcam wrote: I have managed to get jabber installed, and running it from the command line seems to work fine, following the instructions in the 1.4.x administration guide. Which user did you run it as? Thank you for your quick response. In between my question and your response, I tried something else: purged jabber and the other transports and started again. I think the problem was that I had installed a number of other things at the same time, and that interfered. Even when the other packages were installed and configured, it wouldn't budge. So it is now working OK. Sort of. Two problems emerge. (1) When I tried to log on from my Mac laptop, I got as far as the 'reg1' request, to which I got no response. Are you manually sending the XML? Have you tried using a Jabber client to register with the server? Did you set the host name of the server? /var/log/jabber/error.log has an entry citing 'Invalid entry', but the appropriate file appeared in /var/lib/jabber/... Interestingly, it included the information that would have been sent to the Mac (in response to the request), though the comparable record from logging onto the server machine did not. Never seen anything like this. (2) I have tried to set up AIM, following the Debian instructions, but the logon using gaim hangs. The progress bar immediately goes to half-way and sits there. I've never used gaim, all testing has been done using Psi as the Jabber client. I do have the AIM transport installed and working here using the instructions in the package. I notice that in jabber.xml, the formatting of the transports is different from the Debian instructions, but being a True Believer (TM) :-) I did as I was told, commenting out the existing entries. The existing entries should have already been commented out. The aim transport uses ip127.0.0.1/ip. Should it be this way, or do I put in my current (dynamic, but listed on dyndns) ip number? Provided the aim transport is running on the same host as the Jabber server, yes. The jabber-aim package is configured to create a seperate instance of jabber that runs only the AIM transport. This seperate instance then connects to your main Jabber process and provides the AIM functionality. So, the main jabber.xml has to be configured to listen for and accept connection from an external process. The instructions for this are provided in the jabber-aim package's documentation. If you'd like, I could take a look at your configuration file (just e-mail it to me privately). -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber
On Wed, Dec 10, 2003 at 11:49:05PM -0500, Matt Giddings wrote: Anybody out there successful in getting jabberd running on their system? Several of us, myself included. I'm not able to connect remotly, only using localhost as my hostname. Have you configured your Jabber server's hostname? In the debian package you can do this in /etc/jabber/jabber.cfg. I have opened ports 5222 5223 That'll cover clear and ssl client connections to the server, but will not allow other servers to connect to your server, for that you'll need 5269. and followed the quick start steps from the documentation (http://jabberd.jabberstudio.org/1.4/doc/adminguide). I make it partially way through Checkpoint #2, my login dies on step 4 of CP#4. Have you tried launching the server from the init script (after configuring your desired hostname in /etc/jabber/jabber.cfg) and using a normal client to register a user? -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber
On Fri, Dec 12, 2003 at 12:55:09AM +0800, Stephen Liu wrote: Hi Matt, I am interested to know the major application of a jabber server http://www.jabberdoc.org/FrontPage does not provide much information. Could you please shed me some light. Thanks in advance. The following link may help: http://jabbermanual.jabberstudio.org/test/about/overview.html -- Jamin W. Collins This is the typical unix way of doing things: you string together lots of very specific tools to accomplish larger tasks. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber
On Thu, Dec 11, 2003 at 05:59:05PM +, Ken Gilmour wrote: On Thu, 2003-12-11 at 16:25, Jamin W. Collins wrote: and followed the quick start steps from the documentation (http://jabberd.jabberstudio.org/1.4/doc/adminguide). I make it partially way through Checkpoint #2, my login dies on step 4 of CP#4. Is there a way to get Jabber server via apt does anyone know? I can't see it in apt-cache search jabber $ apt-cache search ^jabber jabber-common - Jabber server and transport (common files) jabber-jit - Jabber ICQ Transport psi - Jabber client using Qt jabber - Daemon for the jabber.org Open Source Instant Messenger jabber-aim - Provides AIM messenger transport for Jabber IM server jabber-dev - Daemon for the jabber.org Open Source Instant Messenger jabber-jud - Provides User Directory support for the Jabber IM server jabber-msn - Provides the MSN transport for the Jabber IM server jabber-muc - Multi User Chat module for the Jabber IM Server jabber-yahoo - Provides Yahoo messenger transport for Jabber IM server webmin-jabber - jabber server control module for webmin In woody, the jabber package was in non-US/main. Since woody's release this has changed and the package is now in main. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: jabber
On Fri, Dec 12, 2003 at 01:33:15AM +0800, Stephen Liu wrote: Hi Jamin, Thanks for your advice. I went through the document unfortunately some of the link 'Section , ???Introduction???' die. Yea, not sure what those are intended to link to, but the rest of the document should give you a brief idea of the purpose for Jabber. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: configuring jabber conferences (was: Re: jabber)
On Thu, Dec 11, 2003 at 06:34:27PM -0500, Paul Smith wrote: Sorry to hijack this thread, but I've searched high and low and I can't find what I'm looking for. So, I set up a Jabber server on my Debian box and it's all good. I put in mu-conference in there, and that's good too. Now, how do I create rooms?!?! Using GAIM I can create a temporary room, but I want to create permanent rooms. From the MU-C web site I painstakingly translated and entered the appropriate XML to create a single permanent room and it does work (although I haven't been able to change some of the details like the join/leave banners, etc.) But surely, SURELY there must be some kind of user interface somewhere for managing Jabber servers and performing this kind of operation that doesn't involve typing raw XML into a telnet session!! The MUC upstream source includes two scripts that can help with this: roomname.pl:Takes a list of jids and returns the sha1 hash. Used as the room filename for the spool roommaker.pl: Allows you to create predefined persistent rooms without first starting the service These are not currently included with the package. I will add them to the documentation area of the next udpate, or if there's enough interest I could put them in a jabber-muc-util package. The author talks a bit about these here: http://mailman.jabber.org/pipermail/jadmin/2003-February/008996.html I've looked all over the jabber site and other sites, mailing list archives, etc. and I can't find anything! What's the deal? How do you all manage this stuff on your servers? I don't use conferencing very often, and when I have it's been a dynamic room on my server. -- Jamin W. Collins Remember, root always has a loaded gun. Don't run around with it unless you absolutely need it. -- Vineet Kumar -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: Voicemail/fax software
On Wed, Dec 10, 2003 at 03:43:26PM -0500, Mark Roach wrote: My needs aren't terribly complex, I need to be able to - connect it to my phone switch - have multiple voice and fax mailboxes - configure the voicemail over the handset and I would like to also be able to have faxes sent to an appropriate email recipient. I have googled around, but there doesn't appear to be many mature choices. The Bayonne project seems like it wants to be what I need at some point, but doesn't seem ready yet. Have you looked at Asterisk? http://asterisk.org/ http://digium.com/ -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: Voicemail/fax software
On Wed, Dec 10, 2003 at 05:36:50PM -0500, Mark Roach wrote: On Wed, 2003-12-10 at 17:09, Jamin W. Collins wrote: Might want to double check their mailing list I seem to recall that they had a fax solution that would pass fax calls off to another extension and/or hylafax. Thanks for the tip, I'll check it out. Although I talked to one of the sales guys at digium, and he said that they had a fax solution, but it didn't work all the time... so that may be a bust too... I'll have to keep looking. Does it need to detect the fax or just route it? I've had it configured here in the past to pass calls on to a modem (connected as a station) running under hylafax. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mixing woody and sarge
On Tue, Dec 09, 2003 at 09:59:44PM +0530, Rajkumar S wrote: What will happen when I add testing lines also in sources.list of a stable (woody) box and apt-get a package available in testing? For example ulogd. If you don't set a Default Release or any other pinning configuration, the next time you do an upgrade many of your installed packages will be upgraded to the version available in testing, provided they don't require the installation of any additional packages or the removal of any. However, if a dist-upgrade is used instead your system will be wholely upgraded to testing. After that will the box be stable (woody), with just that package (and dependencies) from testing? What happens when a security updates comes in security.debian.org for that package? Will it gets installed or will it take the updates from testing? This would depend on how the package versions compared. What happens when I do apt-get upgrade, will all packages get updated to testing? what about apt-get dist-upgrade? See above. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Pinning question
On Fri, Dec 05, 2003 at 04:28:28PM -0500, Paul Morgan wrote: On Thu, 04 Dec 2003 23:48:39 -0800, Paul Johnson wrote: Don't pin between stable and anything newer, or you'll end up just having some serious cascading dependencies that will result in you running testing or unstable in the end anyway. See also: apt-pinning considered harmful unless you *really* know what it's going to do. Use http://www.apt-get.org/ to find good backports for woody instead. I thought pinning seemed dicey when I first read up on it. Whenever I think of the word pinning, I have this persistent visual of how entymologists store insects. Blanket statements like the above aren't usually incorrect. I using pinning in one form or another on the majority of my systems, quite a few of which are stable. Pinning is a very useful feature, you just need to be aware of what it is doing. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: APT::Default-Release doesn't seem to affect upgrades
On Wed, Dec 03, 2003 at 11:39:09PM -0800, Ross Boylan wrote: We have a winner. Every single package that I checked that apt-get -s said was from unstable had the same version in testing and unstable. Which entry is first in your sources.list (unstable or testing)? -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: APT::Default-Release doesn't seem to affect upgrades
On Thu, Dec 04, 2003 at 08:10:27PM +, Carlos Sousa wrote: On Thu, 4 Dec 2003 12:57:23 -0700 Jamin W. Collins wrote: On Wed, Dec 03, 2003 at 11:39:09PM -0800, Ross Boylan wrote: We have a winner. Every single package that I checked that apt-get -s said was from unstable had the same version in testing and unstable. Which entry is first in your sources.list (unstable or testing)? apt-get will use the unstable source even if it comes last, e.g. when downloading from the testing source fails, for some reason. I saw it happen before my eyes once. Humbling experience... Right, when downloading from the testing source fails. However, it will use the first matching source with the desired version top to bottom. So, even if testing would have worked fine, if unstable is listed first it will always pull from unstable for packages who have the same version in both testing and unstable. -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: APT::Default-Release doesn't seem to affect upgrades
On Tue, Dec 02, 2003 at 10:57:31PM -0800, Ross Boylan wrote: Don't think so. apt-cache policy shows one unstable entry, priority 50. Maybe there is an upgrade that depends on an uninstalled package that is only in unstable. And then the presence of that package pulls in others? Some results: apt-get upgrade does nothing apt-get -t unstable upgrade pulls in lots apt-get dist-upgrade wants to upgrade gaim gedit ghex gnome-session gnomeicu grip libdate-calc-perl libfnlib0 libgnomedb-dev libgnomedb0 libgtk2.0-0 libhtml-format-perl libmail-mbox-messageparser-perl libofx0c102 libqt2 libxft2 libxine1 pan and install quite a few new packages. apt-get -t unstable dist-upgrade is massive This is all as expected. With the first you've asked apt to _upgrade_ your system. The man page states the following for _upgrade_: under no circumstances are currently installed packages removed, or packages not already installed retrieved and installed With the second you've changed your default release to unstable, thereby increasing it's priority for this run to 990. As a result, it's going to attempt to upgrade everything that's already installed to the version available in unstable. With the last command you've asked apt to _dist-upgrade_ which is defined in the man page as: in addition to performing the function of upgrade, also intelligently handles changing dependencies with new versions of packages; apt-get has a smart conflict resolution system, and it will attempt to upgrade the most important packages at the expense of less important ones if necessary. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: APT::Default-Release doesn't seem to affect upgrades
On Sun, Nov 30, 2003 at 12:25:06AM -0800, Ross Boylan wrote: I have a testing system (with a few items built from unstable sources), and just added unstable to my apt sources.list. As recommended in the HOWTO, I put APT::Default-Release testing; in apt.conf. I do not have an apt/preferences file. When I tried apt-get upgrade (or dist-upgrade) it wanted to upgrade a bunch of packages, all from unstable. I tried commenting out unstable from my sources.list, and apt-get upgrade becomes a no-op. I expected apt would not use unstable unless I explicitly told it to. What am I missing? You've probably installed one (or more) package(s) with version(s) newer than that available in testing but older than that available in unstable. With a Default-Release of testing you've set the priority for testing packages to 990, and all other releases that your system knows about to 500. The problem is that installed versions that only exist in /var/lib/dpkg/status get a priority of 100. This means that if the installed version is newer than what is in testing and older than what is in unstable, unstable is seen as a desirable upgrade. Now, when you preform and apt-get upgrade it will pull the newer version of the package from unstable _if_ there are new packages that are needed that wouldn't be upgraded of their own accord. Many of the packages to be upgraded were from mozilla, which is something I did build from source. So I could sort of see this drawing from unstable, even if I don't understand why. But others are definitely not like that, e..g, gnome-pim, install-doc, openuniverse. As a side mystery, # apt-show-versions -a gnome-pin Perhaps you mean gnome-pim? To get a better idea of why a package is being upgraded from a specific release take a look at the output of apt-cache policy $package. -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: nvidia vs ati
On Sun, Nov 30, 2003 at 10:12:07AM +0100, Elie De Brauwer wrote: On Sunday 30 November 2003 00:15, Frank Thomas wrote: Gerard Ceraso wrote: I am planning on getting a new video card. I have a nvidia geforce2 right now and it works great under linux. I have not had any problems. I have noticed that some of the Ati cards seem to have a bit better performance in some of the tests on the hardware review sites. I was wondering how the Ati support for linux is. My system is currently has an Asus A7N8X Deluxe with an AMD 2400+ and 1G of 3200 ram. Let's do some benchmarking I have a PIV 2.8 ghz (800 mhz fsb), 1 gig ddr 400 ram and an geforce fx 5900 ultra with 256 meg ram. When running X at 1600x1200 resolution I get about 4800 fps in glxgears AMD Athlon(tm) XP 2400+ 512 Megs DDR 2100 GeForce FX 5200 128Meg X = 1280x1024x16 With the above and a few nominal applications running I get just over 3700 FPS from glxgears. -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Possible LKM Trojan , Need Help
On Sat, Nov 29, 2003 at 05:49:31AM -0500, Thomas H. George wrote: chkrootkit reported possible LKM Trojan. 4 processes hidden for ps command. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=217278 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=219730 Before reformating the hard drive and reinstalling Debian Why are you doing this? -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Replacing Exim with self compiled qmail
On Fri, Nov 28, 2003 at 08:47:28PM +1100, Andre Marenke wrote: Is there a way to uninstall Exim and have dpkg recognize my qmail installation as the system mailer so that I can install other packages that rely on an mta? apt-cache show equivs -- Jamin W. Collins To be nobody but yourself when the whole world is trying it's best night and day to make you everybody else is to fight the hardest battle any human being will fight. -- E.E. Cummings -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: installed qmail, apt-get always trying to re-install MTA
On Tue, Nov 25, 2003 at 02:28:40PM +1100, Ross Tsolakidis wrote: I recently installed qmail, I removed Exim just to be sure. Everything is now running fine. The problem I'm having is every time I try and install another package, it always attempts to re-install exim. For example, apt-get install apache, it will attempt to install Exim etc.. Is there anyway of stopping this ? How did you install qmail? Sounds like you did so manually and have not informed the package database that you have an MTA installed. So, when you install something that depends on an MTA, it tries to pull in the default MTA. -- Jamin W. Collins Facts do not cease to exist because they are ignored. --Aldous Huxley, Proper Studies, 1927 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian and Ultra Ata 133 Controller
On Fri, Nov 21, 2003 at 04:06:53PM +, Stefan Lemsitzer wrote: I want to install Debian on a Harddisc, which is plugged into an Ata 133 Controller. Unfortunately the installation program doesn't recognize the disc and asks for a floppy with additional drivers in directory boot... Any hints? Which boot images are using? -- Jamin W. Collins Linux is not The Answer. Yes is the answer. Linux is The Question. - Neo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
