Re: Suspend on old MacBook with fresh Bullseye installation
Hi Stefan, > > I also used hibernate as well as the LXDE controls or closing the lid - > > the result is always the same: > > > > 1. Black screen > > 2. Network disabled (ssh session freezes, no ping from other host) > > 3. Fan keeps on moving silently > > 4. Can't wake up to normal operation neither by > > - opening the lid > > - pressing any key > > - pressing power button > > Hmm... hibernate "should" work reasonably well, regardless of the > underlying firmware. Are you booting via EFI or via the "legacy BIOS" > (which at some point was referred to as BootCamp)? You may want to try > the other one, just to see if it helps. I used the installer default which is EFI. I admit I have no idea how to convince that MacBook to use something else. Its the first time I've got my hand on any Apple hardware - so no idea how I can switch to something else from hardware side (nor whether I need to change something on the Debian installation. > How did you initiate the hibernate? sudo systemctl hibernate (as well as what lxde offers in its menu - no idea how that is called in English locale). > What does `cat /sys/power/disk` say? $ cat /sys/power/disk [platform] shutdown reboot suspend test_resume > Try changing it, e.g. via `echo shutdown >/sys/power/disk` # echo shutdown >/sys/power/disk # cat /sys/power/disk platform [shutdown] reboot suspend test_resume > (that won't help for suspend but it might help for hibernation since > step 3 above suggests that the machine didn't *really* turn itself off). Yes, this is what I suspect. The issue is in the process of suspending / hibernating and not in the process of waking up again. Unfortunately the suggested change does not help either. > > I was wondering about the black screen and suspected that the video > > driver might be responsible for the problem. Despite I'm perfectly fine > > with the nuoveau driver I simply tried nvidia-legacy-340xx-driver (only > > in unstable) hoping that this might solve the suspend problem. > > FWIW, I'd expect `nouveau` to behave a bit better w.r.t suspend/hibernate. OK, thanks for confirming and for your hints in general Andreas. -- http://fam-tille.de
Suspend on old MacBook with fresh Bullseye installation
Hi, I've got some MacBook from 2008 to install Debian on it and hand it over to some user who needs a decent office machine. I've installed Bullseye RC1 which worked nicely (except that I had to manually add WLAN driver broadcom-sta-dkms later). The only real flaw is that neither suspend nor hibernate are working. I think the problem is independent from the desktop choice (I've installed lxde) since I can reproduce the issue also at command line: sudo systemctl suspend I also used hibernate as well as the LXDE controls or closing the lid - the result is always the same: 1. Black screen 2. Network disabled (ssh session freezes, no ping from other host) 3. Fan keeps on moving silently 4. Can't wake up to normal operation neither by - opening the lid - pressing any key - pressing power button So I had to press power button for some seconds to power down the hard way. I checked /var/log/syslog which has something like: May 6 07:38:57 noel systemd[1]: Reached target Sleep. May 6 07:38:57 noel systemd[1]: Starting Hibernate... May 6 07:38:58 noel systemd-sleep[6542]: Suspending system... May 6 08:00:47 noel kernel: [0.00] microcode: microcode updated early to revision 0xa0b, date = 2010-09-28 May 6 08:00:47 noel kernel: [0.00] Linux version 5.10.0-6-amd64 (debian-ker...@lists.debian.org) (gcc-10 (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.3 5.2) #1 SMP Debian 5.10.28-1 (2021-04-09) -- May 6 08:34:20 noel systemd[1]: Reached target Sleep. May 6 08:34:20 noel systemd[1]: Starting Suspend... May 6 08:34:20 noel systemd-sleep[2478]: Suspending system... May 6 08:34:20 noel kernel: [ 2020.808214] PM: suspend entry (deep) May 6 08:36:00 noel kernel: [0.00] microcode: microcode updated early to revision 0xa0b, date = 2010-09-28 May 6 08:36:00 noel kernel: [0.00] Linux version 5.10.0-6-amd64 (debian-ker...@lists.debian.org) (gcc-10 (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.3 5.2) #1 SMP Debian 5.10.28-1 (2021-04-09) I did some research and found a hint to ArchLinux Wiki gefunden[1]. In the first place /proc/acpi/wakeup was just empty. I followed the recommendation and created /etc/systemd/system/suspend-fix.service and so I got: $ cat /proc/acpi/wakeup Device S-state Status Sysfs node ADP1 S3*disabled platform:ACPI0003:00 LID0 S3*enabled platform:PNP0C0D:00 ECS3*disabled platform:PNP0C09:00 OHC1 S3*enabled pci::00:04.0 EHC1 S3*enabled pci::00:04.1 OHC2 S3*enabled pci::00:06.0 EHC2 S3*enabled pci::00:06.1 GIGE S5*disabled pci::00:0a.0 Unfortunately this does not change anything. I was wondering about the black screen and suspected that the video driver might be responsible for the problem. Despite I'm perfectly fine with the nuoveau driver I simply tried nvidia-legacy-340xx-driver (only in unstable) hoping that this might solve the suspend problem. I was able to install the driver but it had really strange colors. For instance the Bluetooth symbol was brown and the desktop icon for power off was blue (instead of red). Besides the strange colors there is also no change in the broken suspend process so I switched back to nuoveau driver. Is there any idea how to get suspend working reliably on this otherwise pretty decent machine? Kind regards Andreas. PS: I'd be happy if you could CC me but I can read the web archive of this mailing list. [1] https://wiki.archlinux.org/title/Mac#Wake_Up_After_Suspend -- http://fam-tille.de
Yet another list statistics for debian-user
Hi, as you can read in my lightning talk at DebConf http://people.debian.org/~tille/talks/200808_lightning/ I did some investigation on who is frequently posting on our mailing lists. I now created graphs until end of last year and write a short summary for those lists I regard worth a comment. I'm not CCed to all of this list so if you want to discuss something please keep me in CC. If you want to discuss the results in general just write to debian-project. All graphs and the code that was used to create the graphs are available at http://people.debian.org/~tille/liststats/ If you are interested in a mailing list which was not analysed, just tell me. I was running the scripts on those lists I personally had some interest and those with more than 1000 subscribers. I plan to clean up code and write some doc about it but this will not happen in the next couple of weeks. The graph for this specific list is --- -- http://people.debian.org/~tille/liststats/authorstat_user.pdf Who are the Johnson brothers??? ;-) Kind regards Andreas -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Bug#465608: "education" has made my debian mentally challenged
On Tue, 19 Feb 2008, Tyler MacDonald wrote: No such luck :-( A (hopefully final) fix for the problem was tested yesterday by Fabrice Lorrain <[EMAIL PROTECTED]> successfully. The packages versioned 0.826 just hit incoming and should be propagated to unstable soon. Sorry for the inconvience and thanks for testing and patience Andreas. PS: I don't think that debian-user is an apropriate list to discuss problems of packages in unstable but once you are CCing this list I'm answering there as well. Please try to avoid flooding more than one list with the same mail if possible. -- http://fam-tille.de -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [RFC] Suggestion for package ordering
On Wed, 29 Dec 1999, Tomasz Wegrzanowski wrote: > I dont think it would. This would only make stats of how often packages > are upgraded. If some package changes 5x more often than other it will > reach much better downloads-count, althru isnt used more often Hmmm, you are right. I was ignoring the fact that many people upgrade much more frequently than me. Anyway this statistic could be a first guess which packages are important, because I would think that often upgraded packages will be used often. Just an idea. May be that one of the administrators could check if it is reasonable. By the way: I would consider it to be important, that we try to get packages which "Depends" or "Suggest" another package should be on one CD with this package if this is possible in any way. Kind regards Andreas. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [RFC] Suggestion for package ordering
On Sat, 25 Dec 1999, Nick Moffitt wrote: > Quoting Martijn van Oosterhout: > > First, there is a package popularity contest package. This needs to > > be improved to the point where normal people can install it. > > I don't get it. apt-get install popularity-contest and then > asnwering "yes" to "Do you want to participate?" was remarkably > straightforward. You're saying that this isn't easy enough? Is this popularity-contest really necessary. What about a script which analyses FTP-servers log-files and counting how often a package was downloaded? Running this script on the main mirrors would make a reasonable statistic. Kind regards Andreas. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Wie Reply-To Tag in E-Mail einfuegen
Hallo, ich benutze Pine als Mail-Client und frage mich, wie ich ihm beibringen kann, daß er ein "Reply-To"-Tag einfügt. Da sich meine Mailadresse kürzlich geändert hat, möchte ich bei Mails, die ich noch von der alten Adresse schicke, dieses einfügen. Viele Grüße Andreas.
FreeCiv problem
Hello, Now after the GTK+ client for FreeCiv 1.8.1 is available I managed to start the civclient successfully. The xaw client did only core dumps on my box at home. After reading quite a lot of documentation a fairly managed the first steps. The only (and I think very basic thing) I didn't found out is, how to continue with a saved game. I started the civserver with "-f civgame-xxx.sav" which seemed to be successfully: ~> civserver -f civgame-3000.sav This is the server for Freeciv version 1.8.1 You can learn a lot about Freeciv at http://www.freeciv.org 1: Loading saved game: civgame-3000.sav 1: Loading rulesets 1: Now accepting new client connections Get a list of the available commands with 'help'. > But when trying to connect with civclient-gtk the log says the following: You were rejected from the game: Sorry you can't join. The game is full. Lost connection to server! But it is the game I left before. There are only two AI plaiers except than mine. What do I have to to to restart from a saved state. Moreover, is there a way to reload a game without leaving the server? Any hint to a part of documentation would be suffice as well. Kind regards Andreas.
Re: bash manual/info lacks examples - addendum
Thank you very much Mirek. Yes, I wanted to express, that the use of documentation is limited if there are no examples in it. In the case of `let` it was only my personal problem (which I would be able to solve by searching for my own old example at home ... or by asking you over the list :) ). I wanted to point out that fact which is a cruxial one if we want to build a user-friendly OS (if man pages could be user friendly at all. I consider them to be useful, but others don't). Kind regards Andreas.
Re: bash manual/info lacks examples
On Thu, 5 Aug 1999, Mirek Kwasniak wrote: > 1) For builtins bash has also help: > >$ help let | less > > 2) Usage of man (my pager is `less') > > > 3) Usage of info Please don't understand me wrong. I *found* the text where the description of let is documented. But what do I have to type if I want to increase a shell variable? The syntax of "arithmetic expression" remains unclear and an example, how to do a = $b + $c * $d and a = ($b + $c) * $d (because this demonstrates also braces in expresions which might be misinterpreted by the shell) would be really helpful. Kind regards Andreas.
bash manual/info lacks examples
Hello, every time I read the bash manual to get help I'm missing examples for the basic usage of a builtin. For instance I managed to write a simple shell script with a loop which increased a variable i = 0; while [ $i -lt $MAX ] ; do echo $i let ... done But I havn't my small loop script handy this time and wanted to know the syntax of let using `man bash` or `info bash`. I'm I stupid or what is the reason that I can't find out, how this expression after let has to be formed? It is the same with the other bash features. The manual is IMHO very abstract and gives less practical help. So I would like to suggest to beg the programmers to add some examples for better reading. By the way could someone please enlighten my how to form the expression. Kind regards Andreas.
Intellimouse
Hello, sorry for posting to this list, but I have to know quickly, if it is possible to use the wheel of the MS-Intellimouse as third button to get the selections of gpm and X. Kind regards Andreas.
Re: UNEXPECTED_DATA_AFTER_ADDRESS@.SYNTAX-ERROR
On 3 Aug 1999, Martin Bialasinski wrote: > * "Andreas" == Andreas Tille <[EMAIL PROTECTED]> wrote: > > Andreas> [EMAIL PROTECTED] > > Check the original mail. Does it have a sparce Comma after the last > address in Cc? I also remember (long time I used pine at home), that > debian-user adds a (slightly malformed I believe) header that > causes this. This is the header, if I save the mail into a file: >From [EMAIL PROTECTED] Tue Aug 3 12:20:30 1999 Date: 03 Aug 1999 12:00:19 +0200 From: Martin Bialasinski <[EMAIL PROTECTED]> To: Debian Userslist Subject: Re: [EMAIL PROTECTED] Resent-Date: 3 Aug 1999 10:00:49 - Resent-From: debian-user@lists.debian.org Resent-cc: recipient.list.not.shown:;@rki.de so there is noextra character after the list address. This error occures in debian-user *and* debian-devel. I remember that I had exactly the same problem at the university after an upgrade to a newer pine version on our HP-mailserver some years ago. The administrator was able to fix it, but I have no clue how. > Maybe you can use procmail to correct this. How to do that. I never fidled around with such stuff. Kind regards Andreas.
UNEXPECTED_DATA_AFTER_ADDRESS@.SYNTAX-ERROR
Hello, I use pine and fetchmail both from the latest potato. If I want to Reply to all recipients? I get at first the addres from the original sender, and as first CC the list. But the But the next (last) entry in the CC-list is [EMAIL PROTECTED] What can I do to avoid this problem? Kind regards Andreas.
Re: Matrox Millenium G200LE
On 3 Aug 1999, Martin Bialasinski wrote: > Andreas> is the Matrox Millenium G200LE supported by the current > Andreas> 1) XF86 drivers > Andreas> 2) SvgaTextmode > Andreas> 3) SvgaLib > > I have a G200 (don't know what the LE is suppose to be). So for the > above items: LE means somewhat "light". There is no video extension and it is "limited" to 8MB RAM. > 1) yes > 2) I think no, but I use the linux 2.2 framebuffer to get high > resolution and refresh rate at the console Is this the prefered method. I'm a little bit slow in following new things and I'm not really sure, how to implement that. > 3) Don't know, I don't use this. OK, it's from less importance since I'm using boxes with >= 200 MHz. Kind regards Andreas.
Apt faster with FTP or HTTP?
Hello, I'm sitting behind a quite slow modem. Coudl I gain a little more speed with http or ftp addresses if I have the choice. Kind regards Andreas.
Matrox Millenium G200LE
Hello, is the Matrox Millenium G200LE supported by the current 1) XF86 drivers 2) SvgaTextmode 3) SvgaLib My company has brought some new boxes and I wonder if I should take one from the new boxes in favour of my good working one (with Matrox Millenium II). Kind regards Andreas.
How to access different trays of a printer
Hello, I'm using magicfilter from potato. The target printer is a compaq Pagemarq 20. This printer has three different trays (loaded with two different kinds of A4 paper and the third with A3. There is no special driver contained in magicfilter (ghostscript). So I'm using hplj4m as postscript printer. Now I wonder how I could use the different trays. Kind regards Andreas.
smbfs vis ncpfs
Hello, because I couldn't solve my ncpfs-problem to let users mount their Novell-drives at request themselves via an /etc/fstab-entry I wonder if this is possible with smbfs. Did I understand it right that it is possible with smbfs to mount Novell and Win drives. I didn't used Samba yet and I'm very confused about the readmes of smbfs about new smbmount and old smbmount and what I have to use now. If this is of interest: I run an up to date potato system with kernel 2.2.10. So the first question is if I should install the smbfs or smbfsx package. Kind regards Andreas.
Re: Proxy server with Apache
On 22 Jul 1999, Ramin Motakef wrote: > > % squidclient http://www.netscape.com/ > test > > > You have to set permissions in /etc/squid.conf. Look for something > like: > http_access allow "some acl" > To get started: "http_access allow all" as last entry. OK, this test seems to work so far. Do you have any example how to define an acl? > The name of the Host and the Port of the proxy_server eg.: > proxy 3128. > It's probably a good idea to write an proxy autoconfiguration file if > you have a lot of clients. > If you are interested i can send you an example. I wrote such a file for another proxy server and it worked so far that I got no errors. Anyway I didn't know any method to check, whether the pages are *really* cached, i.e. if the PAC file works. I tried to merge this PAC file with the example in the squid FAQ, for the proxy I installed here and here I have the possibility to check whether the pages are cached or not. I do an `ls -lR /var/spool/squid` and compare the output with a former one. Unfortunately there aren't any differences so far. So I suspect that something is wrong with my PAC fiel and a working example would be welcome. Is there any method to debug such things straightforeward? Kind regards Andreas.
Re: FreeCiv
Hello,
I did further investigations, why freeciv diesn't work on my
machine at home. I traced down the source and found the
exact position in the code, where the problem occures. I
appended a little patch which includes two "printf"-lines
into xmain.c of FreeCiv-1.7.2. It is exactly the same problem
with 1.8 (it is in the same statement of xmain.c also in 1.8),
which I compiled myself from the potato source package.
Does anybody have an idea what to do or do I have to
subscribe to the freeciv mailing list?
Kind regards
Andreas.
*** xmain.c Tue Jul 20 20:58:52 1999
--- xmain.c.ori Thu Dec 3 11:48:51 1998
***
*** 595,601
left_column_form,
NULL);
- fprintf(stderr, "%s(%i)\n", __FILE__, __LINE__);
unit_pix_canvas = XtVaCreateManagedWidget("unitpixcanvas",
pixcommWidgetClass,
left_column_form,
--- 595,600
***
*** 603,609
XtNheight, NORMAL_TILE_HEIGHT,
NULL);
- fprintf(stderr, "%s(%i)\n", __FILE__, __LINE__);
unit_below_canvas[0] = XtVaCreateManagedWidget("unitbelowcanvas0",
pixcommWidgetClass,
left_column_form,
--- 602,607
Re: Proxy server with Apache
On Wed, 21 Jul 1999, Shao Zhang wrote: > apache is a really a web server. AFAIK, it has very limited features to > be a proxy server. > If you are using debian, make sure the module is installed. > > For a better proxy server, I suggest you to try out squid. There is a > debian package for it. OK, thanks for this hint. Now I installed the potato package of squid and tried to work out how to use it. The FAQ says: % squidclient http://www.netscape.com/ > test Unfortunately I get a HTTP/1.0 403 Forbidden message with each server I try, also with http://localhost/. To make it clear, I append the file "test". The second recommended test is % squid -k check What should I expect from this call? It runs without any message. Assuming that I would be able to install squid successfully I would like to know, what entry I have to set into my netscape preferences for the proxy. Is it the http address of the host running Apache and Squid or what else? Kind regards Andreas. HTTP/1.0 403 Forbidden Server: Squid/2.2.STABLE3 Mime-Version: 1.0 Date: Thu, 22 Jul 1999 09:29:00 GMT Content-Type: text/html Content-Length: 680 Expires: Thu, 22 Jul 1999 09:29:00 GMT X-Squid-Error: ERR_ACCESS_DENIED 0 X-Cache: MISS from wr-linux01.rki.de Proxy-Connection: close ERROR: The requested URL could not be retrieved ERROR The requested URL could not be retrieved While trying to retrieve the URL: http://localhost/";>http://localhost/ The following error was encountered: Access Denied. Access control configuration prevents your request from being allowed at this time. Please contact your service provider if you feel this is incorrect. Generated Thu, 22 Jul 1999 09:29:00 GMT by wr-linux01.rki.de (http://squid.nlanr.net/Squid/";>Squid/2.2.STABLE3)
Re: ncpfs and /etc/fstab
On Tue, 20 Jul 1999, Alex V. Toropov wrote: > >I played around something with ncpfs. I managed to get > >an entry in /etc/fstab so that root is able to mount a > >netware drive using mount. > > Can You explain what kind of entry You've use ? SERVER/N_user /home/L_user/novell ncp defaults,mode=644,uid=L_user,gid=L_group,volume=VOL1,multiple,passwdfile=PATH_TO_PASSWDFILE Kind regards Andreas.
Re: 3Com 509B Ethernet card working in Windows but not in Linux
On Wed, 21 Jul 1999, Evan Van Dyke wrote: > 1) insmoded the 3c509.o module -OR- compiled it hard into the kernel? > 2) if so, what does it say when the module is loaded/on boot up? I had used several network adapters before. Someone required to be compiled hard into the kernel and someone as module to work without further investigation. The 3Com cards seem to tend to work better as module for me. Just try to compile it the other way you've done. Kind regards Andreas.
Proxy server with Apache
Hello, I want to set up my Apache seerver as Proxy host. I'm really unexperienced in this topic and at first I want to describe my environment: There are different departments in our institute. The main department has a fast internet connection and uses a firewall and a proxy server. I work in a department with a slow connection (modem line) to the main department. We are sitting behind the firewall. My idea was to speed up the internet connection using a local proxy server in our house to serve visited sites for all user via local net. I tried it setting up the Apache server and commented out the line LoadModule proxy_module /usr/lib/apache/1.3/libproxy.so in /etc/apache/httpd.conf. After that I did /etc/init.d/apache restart You guess it or not, it doesn't work. Could someone enlighten my why not. Kind regards Andreas.
ncpfs and /etc/fstab
Hello, I played around something with ncpfs. I managed to get an entry in /etc/fstab so that root is able to mount a netware drive using mount. Now I have some questions: 1) If I want to enable a user to mount a device I usually add the option "user" to the mount flag. Unfortunately this conflicts with the user option of ncpmount. What do I have to do to enable users to mount and unmount their netwere drives themselves. 2) In the boot process the /etc/fstab is read completely and also the ncp entry. a) If I would want the netware drive to be mounted on startup then it fails with: ncpmount: Cannot assign requested address when trying to find SERVER I suspect there is a deamon to be started first to enable this service. How to do that? b) If I don't want a device to be mounted at startup I generally add "noauto" to the mount options. Infortunately I get an error message with this option that ncpmount doesn't understand this option. How to awoid the atempt to mount a netware drive? Kind regards Andreas.
Re: E2fs failures
On Mon, 19 Jul 1999 egm2@jps.net wrote: > Which kernel version are you using. I understand that there has been a > file corruption problem in some of the newer kernel versions. In the last test I was using 2.2.3 but the first problems occured while using 2.0.34. Kind regards Andreas.
E2fs failures
Hello, recently I detected problems with 2GB harddisk in an old 486 machine. The disk worked for two years very well. I tried several checks under DOS including a check routine from Quantum (it's a Quantum Bigfoot disk) and Calibrat (from Norton Utilities). No DOS check program detected any error. Repartinioning with Linux fdisk and creating e2fs worked well but after writing data to the disk and unmounting leads to more errors than I could count. E2fsck is unable to repair the problems. Any idea what could cause this strange problem? Kind regards Andreas.
Re: FreeCiv
On Fri, 16 Jul 1999 12:52:31 +0200, Nils Rennebarth wrote: > > I tried several times to get freeciv working but failed each time > > because civclient stops with SIGSEGV. > Upgrade your xpm library and file a bug report against civclient to make > it > need xpm >= 3.4j ~> dpkg --status xpm4g Package: xpm4g Status: install ok installed Priority: optional Section: x11 Installed-Size: 92 Maintainer: Michael Alan Dorman <[EMAIL PROTECTED]> Source: xpm Version: 3.4j-0.6 Depends: libc6 Conflicts: xpm4.7 (<= 3.4j-0), xpm4 (<= 3.4j-0) Description: X Pixmap libraries (for libc6) - runtime XPM (X PixMap) is a format for storing and retrieving pixmaps to and from files. This package includes a runtime library for including, storing, reading and writing the XPM format. So this should be OK. I use at home a clean (off-line) Slink System, with only the GTK+ related packages updated, which shouldn't conflict with FreeCiv. Stangely enough, I run FreeCiv with comparable configuration without any problem. Anyway the strace-log seems to show that a graphics-related library causes the problem. Should I try the xpm library from potato? Kind regards Andreas.
FreeCiv
Hello,
I tried several times to get freeciv working but failed each time
because civclient stops with SIGSEGV. Are there any known
problems? I successfully started civserver and it seems to
do reasonable things. But regardingless in which state the
server is civclient breaks. If it is helpful I have an
`strace civclient` handy and I append a short part (beginning
and end to the end of this mail).
I tried civserver and civclient without any options and tried
also with several options which might help. I should note
that it is a standalone machine (OK, local net with two
boxes, but the other one was switched off.)
Kind regards
Andreas.
~> strace civclient
execve("/usr/games/civclient", ["civclient"], [/* 28 vars */]) = 0
brk(0) = 0x80c389c
open("/etc/ld.so.preload", O_RDONLY)= -1 ENOENT (No such file or directory)
open("~/usr/lib/libSM.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY) = 3
fstat(3, {st_mode=S_IFREG|0644, st_size=15559, ...}) = 0
mmap(0, 15559, PROT_READ, MAP_PRIVATE, 3, 0) = 0x4000c000
close(3)= 0
... (deleted)
write(3, "H\0\16\0K\6\300\3L\6\300\3\10\0\10"..., 2048) = 2048
write(3, "H\0\16\0{\6\300\3|\6\300\3\10\0\10"..., 2048) = 2048
write(3, "H\0\16\0\253\6\300\3\254\6\300\3"..., 2040) = 2040
write(3, "H\0\16\0\334\6\300\3\335\6\300\3"..., 2040) = 2040
writev(3, [{"H\0\16\0\r\7\300\3\16\7\300\3\10"..., 2020}, {"Grey", 4}], 2) =
2024
read(3, 0xbfffe184, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\325\231\16\0\0\0\0\276\276\276"..., 32) = 32
writev(3, [{"5\30\4\0=\7\300\3&\0\0\0\2\0\2\000"..., 436}, {"lightblue", 9},
{"\0\0\0", 3}], 3) = 448
read(3, 0xbfffe174, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\325\255\16\0\0\0\0\346\330\255"..., 32) = 32
write(3, "5\30\4\0I\7\300\3&\0\0\0\2\0\2\000"..., 368) = 368
read(3, 0xb57c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\276\16\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "7\30\t\0S\7\300\3&[EMAIL PROTECTED]"..., 328) = 328
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\315\16\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0[\7\300\0035\30\4\0\\\7\300"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\332\16\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0b\7\300\0035\30\4\0c\7\300"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\347\16\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0i\7\300\0035\30\4\0j\7\300"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\364\16\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0p\7\300\0035\30\4\0q\7\300"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\1\17\0\0\0\0&\0\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0w\7\300\0035\30\4\0x\7\300"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\16\17\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0~\7\300\0035\30\4\0\177\7"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30\33\17\0\0\0\0&\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0\205\7\300\0035\30\4\0\206"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\30(\17\0\0\0\0&\0\0\0\0\0\0\0"..., 32) = 32
write(3, "<\30\2\0\214\7\300\0035\30\4\0\215"..., 256) = 256
read(3, 0xb56c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\0305\17\0\0\0\0&\0\0\0\0\0\0\0"..., 32) = 32
writev(3, [{"<\30\2\0\223\7\300\0035\30\4\0\224"..., 268}, {"Gray", 4}], 2) =
272
read(3, 0xbfffe17c, 32) = -1 EAGAIN (Resource temporarily
unavailable)
select(4, [3], NULL, NULL, NULL)= 1 (in [3])
read(3, "\1\325C\17\0\0\0\0\276\276\276\0"..., 32) = 32
writev(3, [{"U\30\5\0#\0\0\0\5\0\4\0", 12}, {"Black", 5}, {"\0\0\0", 3}], 3) =
20
read(3, 0xbfffe17c, 32)
Re: [RfS] Request for Submissions: Upcoming Debian Logo
On Thu, 18 Feb 1999, Steve Greenland wrote: > Question: does anybody else think that the "debian" with the line back > to the dot in the 'i' that Rob Malda did for Slashdot is best logo yet > done? Not me and so I wouldn't answer this posting, but please check your mail agent. I've got your mail seven times!! Regards Andreas.
Re: [RfS] Request for Submissions: Upcoming Debian Logo
On Thu, 18 Feb 1999, Oliver Elphick wrote: > No, no! Think of Mars. Everybody immediately recognises that word, in > that colour and typestyle: they think of chocolate -- not of a planet, a > Roman god or a piece of music. Written in ordinary type, it could be > any of the four, or maybe something else. > > Logo recognition is created by persistent and expensive advertising, or the > equivalent in news coverage. Since we can't afford the one, and see no > prospect of the other, it seems perverse to throw away one of the best > aids to recognition. I think it would be to expensive to include so much chocolate bars into the envelopes of our CDs to attract so much people that it has the same effect. Also you have to find a way to beam chocolate bars via FTP or HTTP. Regards Andreas.
Re: Debian Logo *Idea* Contest! (Delay for painting contest required)
On Thu, 18 Feb 1999, Chris Waters wrote: > Actually, I tried to start with that. Look in the debian-devel archives > for a thread entitled "Mascot for Debian" (or something like that). I'm really happy to know, that my image of the Debian project is right and there are everytime people who think over importand things :). > > Idea no. 1: Orca for Debian > Um, this is actually idea number 24 or so. :-) Moreover I'm happy to lose the rule of a contest manager :) > We batted around a bunch of mascot ideas. I believe dolphins were > suggested, but not Orca. OTOH, several people were objecting to the use > of carnivores, which would disqualify Orca. :-) Hoops, thats an interesting point of view. We should try to feed Keiko with sea gras ... may be it works :) > After watching the lengthy debate that ensued following my lone innocent > post, I decided that we might as well allow the artists the freedom to > propose ideas through their art, rather than trying to force them to > render a pre-selected image. This follows the traditional Debian cry: > "show me the code!" Those who actually *do* the work are in charge. > :-) OK, but defining what we understand under the term logo would help the artists and would save the time of creating Debian "signatures". By the way we should really save these creative works to have it handy for the web pages in addition to the logo! > Here's a scattering of other proposals that came out of the earlier > thread: > > 1. Dragon > 2. Octopus > 3. Monkey > 4. Flying pig ^^ Borland ?? > 5. Ant > 6. Bee > 7. Dolphin > 8. Monkey ^^ This would be OK together with 3. and 10 further monkeys :) > 9. Eagle If there are many vegetarians in the Debian project the bee will rule :) Did you ever noticed that flesh eating animals are potential mascots for different purposes ... I'm sure this was haevily discussed in your mascot thread ;-) > And that's just what I can recall without browsing the thread. A number > of those had, er, philosophical arguments behind them, much as you had > with your proposal. So the remaining question for me is, why we can't bring it to an end. The Debian leader ship election was comming to an end. Set up new voting pages with the most discussed animals, lets choose the mascot and give the artists a chance to work to this topic. > My suggestion: if you like the idea of Orca, create a logo and submit > it. Note that the fans of the octopus and the bee already have logo > submissions in the Gimp contest. I'm really sure that my orca wouldn't be the winner :). I could spend my time in more succesful things. But I'm happy to bring up this discussion once again so a call for decision was repeated and may be this helps. Kind regards Andreas.
Re: Debian Logo *Idea* Contest! (Delay for painting contest required)
Hello, to support my idea of "Orca for Debian" I would like to point to some very nice pictures which can be found under http://members.aol.com/stejacorca/index.html (Disappointer: I do not have anything in common with the author but like this site since I was searching for "real" information about orca whales after my son was looking Free Willy.) I was told that the decision for a logo idea has to be illustrated with a kind of realisation. I would like to compensate my lack of ability to paint with these real images (which can be found in the picture gallery). I think many people would prefer a "living" logo in favour of a technical one -- this is the strategy of the Linux, the GNU and the Suse logo and of many others. And it works. Kind regards Andreas.
Re: [RfS] Request for Submissions: Upcoming Debian Logo
On Thu, 18 Feb 1999, Chris Waters wrote: > Um, IBM's logo is just letters. Sun's logo is, well, just letters, > technically. I wouldn't rule out a logo based on letters, but it should > *be* a logo, which most of the entries in the GIMP contest aren't. The >>should *be* a logo<< question in my opinion requires a detailed concept >>what is a Debian logo<<. That's why i started the Debian logo *idea* contest. Please vote for a "letter-shaped" logo and tell us your reasons for it to compete against my idea of Orca for Debian Sorry for the low level fighting style. I'm really embarassed about the logo contest topic. Unfortunately our world is ruled by advertisement. I'm really against this fact, but this doesn't help. We want to attract people, so we have to cope with this fact. Kind regards Andreas.
Debian Logo *Idea* Contest! (Delay for painting contest required)
Hello Debian deveolopers and friends, this day is the first I started thinking about the importance of a Debian logo. Someone will say it is to late for a maintainer who maintaines seven packages, but I concentraited my brain on other topics. Some alarming postings lead me to the right path now. I wrote my ideas down in reply to a mail of Joey and repeat them here with some advancements to call for a new contest, which has to be run before we start to paint logos, the Debian Logo Idea Contest Wy do we need such a contest. The mails of Joey have shown, that there is the problem of "no-unique-concept" for the logo. Friends we run one system, we develop one system, we work after the same concept all day, why don't we use this well working style to represent our project to the community? So let us define first, *what* we want to show on our logo. This is always the first step in software related things. Wy did we step over it?? To start the contest I want to bring in the first idea, which I want to offer all the engaged and well gifted painters in our community which could take the idea and realize it much more better than me, because I'm not sooo good in painting. Idea no. 1: Orca for Debian Why do I think that an orca whale is exactly what we need for our logo: - often logos are animals - let us choose an animal which stands for freedom - there are many animals which fit this fact, but many of them are occupied by governments of countries - we shouldn't stik on any sign of a certain country - lets take an orca whale - everybody knows the "Free Willy" story ... I don't want to say anything about the quality of this film, but the general idea is good - it has something in common with our ideas: * we want to bring software to freedom --> its a *process to get free* * Debian is big, as well as an orca whale * it has big tooth to fight against commercial software developers - it could be a strong logo (may be it could have something in common with the "world-idea" of Joey, because orcas live in all oceans of our world which may or may not be expressed in the futur logo, thats a design question which I don't want to deal with - we can brofit very much from the popularity of the whales in our day! ... we want to have popularity and we need to have popularity - a good designer cold paint a *funny* orca -- as funny as the penguin and other ones - an orca can be painted with less colors ... no problem with old fashioned displays and even on black and white it is fairly well - the simple shape could in any way easy painted is well scalable (I think also for the prefered bitmap concentrated idea!!) and looks well on every output medium - its not so easy to forget such a logo if it is well painted and we want to attract people OK, lets assume these ideas are good or someone has a better, its just a suggestion. *NOW* its time to make a logo contest for the best/funnyest/... orca which stands for Debian. Once more: let us delay the painting work. Its time for thinking first. If we could cope with this (or any better idea) then we have one aim -- as well as we have one aim in developing the operating system. Kind regards Andreas.
Re: [RfS] Request for Submissions: Upcoming Debian Logo
On Thu, 18 Feb 1999, Martin Schulze wrote: > I'd like *everybody* think about a possible logo. Please notice the > following simple rules: > > . The logo must NOT be linux-centric. Thus a penguin in any form is >inacceptable. We have Debian GNU/Hurd and people are working on >something similar to Debian Solaris. Please accept that. > > . A logo is NOT a set of letters. Thus the string "Debian" rendered >very nice is ... a picture but not a logo. Forget that. > > . A logo must be very recognizable. You notice a cameleon and >immediately you know that this is SuSE, you see a stylized red hat >and for sure, that's RedHat. Got it? > > . A logo needs to be simple, no details please. People need to be >able to recognize a logo, keeping it in mind. They shall never >lose it since it would represent the best set of free software. > > . A logo is simple. A drawing is not a logo, nor is a painting or an >image. So put away all your van Gogh albums, their contents are >useless as logos. > > . A logo needs to look good in colour (e.g. for a web page or a >T-Shirt or a sweatshirt or similar, rendering here is fine) *AND* >in black/white (as letterhead for button or similar). > > . A logo must be scalable. You know, on a web page the logo is small >and you don't want such a small logo on your T-Shirt, right? > > . A new logo must be new. Don't come up with a logo that already >exists. So put away the Universal Studios logo, just because >Debian is the Universal Operating System doesn't mean we may use >their logo. (although I believe that using a world or universe >within the logo would be an interesting idea.) > > . The strings "Debian GNU", "Debian", "Debian GNU/Hurd", "Debian >GNU/Linux" and/or "The universal operating system" will be added if >needed. Please don't include any of these strings in your >submission. > > Please keep in mind that a set of letters is NOT a logo and that both > a penguin and "Linux" are not acceptbable. > > If you lack some impressions please go to the railroad station, > airport, harbour, tram station or shopping mall and realize how other > logos look like. > > After that please draw your ideas on paper, digitalize and submit > them. It is the first time for me that I deal with the logo question, because it was of lower importance for me. Joeys mail lead me to another opinion. Please, please guys stick letter for letter onto this rules for a logo Joey mentioned in his mail. All rules are the result of a long thinking of a wise person! We can't change our Logo after each release so let us find a good clue. In my opinion the first step is thinking. Let us discuss the motive of the logo at first and *after* this lets start drawing (whatever drawing program you will take is the next question)! For example follow these idea: - often logos are animals - let us chose an animal which stands for freedom - there are many animals which fit this fact, but many of them are occupied by governments of countries - we shouldn't stik on any sign of a certain country - lets take an orca - everybody knows the "Free Willy" story ... I don't want to say anything about the quality of this film, but the general idea is good - it has something in common with our ideas: * we want to bring software to freedom --> its a process to get free * Debian is big, as well as an orca whale * ... fill in some more ideas here - it could be a strong logo (may be it could have something in common with the "world-idea" of Joey, because orcas live in all oceans of our world which may or may not be expressed in the futur logo - its a design question - we can brofit very much from the popularity of the whales in our day! ... we want to have popularity - a good designer cold paint a *funny* orca -- as funny as the penguin and other ones - an orca can be painted with less colors ... no problem with old fashioned displays and even on black and white it is fairly well OK, lets assume these ideas are good or someone has a better, its just a suggestion. *NOW* its time to make a logo contest for the best/funnyest/... orca which stands for Debian. PLEASE, PLEASE friends think first, before grabing your mouse to paint!! Kind regards Andreas.
Re: Installation with BT-930
On Mon, 29 Sep 1997, Jeff Noxon wrote: > > The last words were: > > > > Started kswapd v 1.4.2.2 > > FDC 0 is a post-1991 > > md driver 0.35 MAX_MD_DEV=4, MAX_REAL=8 > > > > It seems that after this the SCSI controller is to be initialized. > > I don't follow you here. The kernel actually stopped after initalizing > the md driver? Which image was this, exactly? Hmm, I don't know anything about the md driver, but anyway I used the latest disk images from ftp://ftp.de.debian.org/debian/bo/disks-i386/1997-08-01/special/v30 The README claims: Linux 2.0.30 added support for Buslogic's FlashPoint SCSI adapters, syn flooding protection and the new ISDN drivers. Unfortunately this kernel version tends to hang machines in some cases, especially when they are tight on RAM. So keep a 2.0.29 kernel somewhere! > You _should_ be able to use the 2.0.30 "special" images. If those don't > work (please confirm), let me know and I will compile a custom kernel for > you to put on the rescue disk. OK, as I said, they stop after the md driver. May be I've found a solution: I booted DOS and tried loadlin using a kernel I compiled on another box. While writing this I compile a second try because I forgot to include RAM disk support ... all who tried that before know the result. But before this kernel got his panic it detected the available SCSI devices and the buslogic controller and so I'm very hopefully that this will work. If not I will give a further report. (But it will take me one hour because the box which compiles the image is s slow :-(((). Thanks for your help Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Installation with BT-930
On Sun, 28 Sep 1997, Egon Schmid wrote: > Can you boot from CD-ROM? A WD-7000 isn't on your hardware list. Which > debian-version is it? Yes, as I said I insert the CD-ROM and the installation programm comes up without any problem -- so the controller works! No a WD-7000 isn't on my list because I havn't such a thing inside my box. What kind of controller is it? Thought it should be any emulation of the internel ZIP-drive??? It is Debian GNU/Linux 1.3.1 from the German manufacturer J.F. Lehmann. I installed two other boxes (non SCSI) without trouble from this CD. Greetings Andreas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Installation with BT-930
Hello, first my configuration K6-200, BT-930, ASUS TX97-XE, Matrox-Millenium, IBM SCSI-DCAS 34330, Plextor SCSI-CDROM, IOMEGA-ZIP SCSI intern, 64MB SDRAM all new assembled. I tried to start with the Debian CD (German manufacturer J.F. Lehmanns) I seted up the Buslogic BIOS to boot from CD: it works perfectly and my box booted from CD. I've got the installation menue. But when I try to partition a harddisk it failed ans didn't found any harddisk. ALso trying to mount any /dev/sd? at tty2 failed with: Kernel could not recognize ... as a blockdevice (try insmod driver) ... or things like that. I have to say that the harddisk works with the factory installed DOS without any floppy or CD in the drives perfectly without any problem. That means the drive seems to be all right. While booting I've got: Warning : Unknown PCI device (8096:71xx). Please read include/linux/pci.h (xx aus 00, 10, 11, 12, 13, sowie zus"atzlich (102b:51b)) but this happens on some boxes as I heard and seems not to be connected with my problem. Well I tried the 2.0.30 special installation diskettes which should work with Buslogic Flashpoint controllers but should crash often. Yes I veryfied that they crash ... :-((( The last words were: Started kswapd v 1.4.2.2 FDC 0 is a post-1991 md driver 0.35 MAX_MD_DEV=4, MAX_REAL=8 It seems that after this the SCSI controller is to be initialized. Using the standard diskettes the next lines are: Failed initialization of WD-7000 SCSI card! PPA: unable to initialise controller at 0x378, error 2 scsi : 0 hosts scsi : detected total. but I think the error message is connected with failing to initialize the parallel port zipdrive which I havn't because using an internal SCSI Zipdrive. Or is this the point??? But the CD is read very well!?!?!? Any hints Greetings Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: The Gimp?
On Mon, 28 Jul 1997, Remco van de Meent wrote: > : I did so but without LinuxThreads installed. I hadn't any trouble so far. > > Maybe you installed pthreads instead? No. I only found oldmitpthreads but it is only to support gnats and I havn't installed both. > LinuxThreads provides an almost kernel-level of threading. > I don't know if Gimp really needs it, but I saw it mentioned in the > configure-script. Hmmm, I compiled Gimp 0.99.10 on my Linux box and a HP Workstation (same sources - no trouble so far). When tha next version is out I will have a closer look and try to install LinuxThreads before. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: The Gimp?
On Sat, 26 Jul 1997, Remco van de Meent wrote: > Another option is to get the .tar.gz file from the source distribution and > compile Gimp your self. I did, using just libc5 (with LinuxThreads > installed). I did so but without LinuxThreads installed. I hadn't any trouble so far. It is possible that I would run in trouble because of the lack of LinuxThreads? What kind of program/package is LinuxThreads and why could this be neccessary for "The Gimp". Greetings Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: FIC PA2011 Board (Was: Debian for PPC Macs)
On Thu, 17 Jul 1997, Rick Hawkins wrote: > it looks like waiting for the hot 603 is worth it--if you can wait. What kind of board is this and how long I have to wait? Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
FIC PA2011 Board (Was: Debian for PPC Macs)
On Wed, 16 Jul 1997, Rick Hawkins wrote: > I'm leaving in a few days for a workshop, so I'll buy second week of > august. If the "Hot 603" motherboard is out, and reasonably priced, What about this motherboard? I read a lot about board but never heard about it. > i'll try for one of those with a k6. If not, I think it's a FIC 2011 > with a K6-200. I've found I can get that in a full tower with 3gb, 32mb I decided, that FIC PA2011 is the only one which fits the following: 1) 1MB Cache (supports more than 64MB memory) 2) supports SDRAM chips (which I consider to be the better ones in terms of flexibility or the possibility of overclocking) Is there anybody who successfully used this board with Linux? My local dealer never heard from this board and may be he has trouble to get it? It is worth enforcing him to take this board? Thanks for advance Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian for PPC Macs
> The new PowerPC chips (the next generation, called "G3" or 740/750 or > Arthur or Mach V...) are, well, FAST. I haven't run any linux > benchmarks on them, but I intend to once the "Money-making OS" is > running on them officially. > > So, basically, the decision is similar to decing between Alpha and Intel > linux... the Alpha is FAST, but there are still more linux pieces out > there built and tested on intel boxes. Personally, I have a PPro box, > it works, and I will probably end up with a CHRP box running some PPC > Linux as well (when we ship the darn thing!). My conclusion is if I buy my new boc next week I should buy a Intel, if I buy next year PPC or Alpha. Right? Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Debian for PPC Macs
Hello, I want to buy a new box and wonder about some alternatives to the Intel (compatible ... I think of K6) architecture. I've got the following information from Johannes Ramm-Ericson: > I just came across the following information: > * > > This announcement is from: > > Apple PR > > Apple Computer, Inc. today announced that it has reduced prices made > available to resellers on select Power Macintosh 6500 and Macintosh > Performa 6400 series computers. Effective July 12, 1997, this move cuts > prices on these popular computers by up to U.S. $300. The price cuts are > consistent > with Apple's strategy of providing the industry's most > powerful and easy-to-use computers at competitive prices. > > ... (some stuff deleted -- ask me if you are interested in the complete text) > > The PowerMac 6500 series does not yet fully support MkLinux > but there are people who have reported being able to install it. > > Also, some have installed Linux for PMacs (www.linuxppc.org) > on the 6500 series. What about Debian on Macs? If I'm right, in the unstable times of bo there was a binary-m68k - tree. This tree now has moved to unstable/hamm. Does this mean that Linux for Macs is only unstable?? I havn't compared the contents of the m68k tree with the i386 tree. Is the amount of packages the same? I need a very stable system on my new box because I will use it far distant from the internet (the prices of phone calls are quite high in Germany). I can't relay on the good support of the mailing lists and can't load some packages via ftp. Is there anybody who would recommend to by a PowerMac if it is the better (faster) hardware or shoul I stick to the Intel compatible? Thanks for advance Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
ftape question
Hello, I have a Iomega 250 floppy tape drive. I compiled ftape into the Kernel (2.0.30). But what to do to get a device /dev/ftape ??? According to the CDROM-HOWTO I have to create this entry via mknod. But what are the correct parameters. I didn't find any information neither in the HOWTO nor in the kernel package. Futhermore in Ftape-howto I found, that there is a newer version than 2.08 (which is shipped within the kernel-package) with the version number 2.09. My search on sunsite showed me Version 3.03a from May 23, 1997. Any recommendations about this drivers? Any help to this questions? Greetings Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Updatedb with Cron Daemon
Hello, this is my /etc/updatedb.conf : # This file sets environment variables which are used by updatedb # filesystems which are pruned from updatedb database PRUNEFS="afs proc" export PRUNEFS # paths which are pruned from updatedb database PRUNEPATHS="/tmp /usr/tmp /var/tmp /m /home/e2od5/[NOQ]" export PRUNEPATHS # netpaths which are added NETPATHS="" export NETPATHS This should mean that no files above /m would included when running updatedb, shouldn't it? Especially not /m/ra which is one directory in /m (all directories in /m are mounted filesystems, /m/ra is one of some NFS filesystems). Nevertheless root recives the following mail every day (because updatedb is called inside cron.daily): te: Fri, 13 Jun 1997 07:00:14 +0200 (CEST) From: Cron Daemon <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Subject: Cron <[EMAIL PROTECTED]> run-parts /etc/cron.daily find: /m/ra: Permission denied What's the matter here? Including /m/ra explicitely in PRUNEPATH does help nothing. Any recommendations? Have a nice weekend everybody Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Re: InfoMagic's new LDR
On Tue, 13 May 1997, Portuesi Simone wrote: > I'd like just to add one thing to it, it'll be confortable (at least for > me) that the mirroring can be done from another computer. I'll explain > better: you che the Packages file on a PC connected to internet, bring > it to a PC not connected to the net, the script will determine the files ^^ Sorry, I didn't followed this discussion in the past. What script would do such a job. I would like to have a automagical mechanism to update my box at home. This time I do it manually (using MC and compare each directory with the directories on a ZIP disk I updated on my net-connected PC. Thanks for advise Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Backspace in rxvt
On Wed, 7 May 1997, Karl M. Hegbloom wrote: > > "Jens" == Jens B Jorgensen <[EMAIL PROTECTED]> writes: > > Jens> If typing Backspace on the shell command line prints "^H" > Jens> rather than doing a backspace, type the command "stty erase > Jens> ^h" (make sure you type "^" and "h", not CTRL-h. Note that > Jens> this only changes the current shell (terminal) so you'll > Jens> probably want to put it in your .profile. > > I have the same problem--- in an `rxvt`, <-Backspace prints a '~', so > I have to use C-h to backwards-delete-char. `stty` doesn't effect it > at all, as far as I can tell. I think it's a bug in `rxvt`. No matter if I follow the hint of Jens (stty erase ^h) or not does the job I expect of (delete under cursor) does: ~ It is really a bug of rxvt? (As I mentioned xterm works OK) Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Problem with svgalib
Hello, I installed the tmview_96.05-1.deb package which I got from bo/binary/tex some days ago. At home it works fine at home but in the university it produces only a very strange vertical pattern. There aren't any error messages. The demos of svgalib work well. Does anybody have a clue Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Backspace in rxvt
Hello, I use rxvt instead of xterm because I read it is smaller and is able to do all things I want it to do. Additionally it has colors by default (nice to have a colored mc :-)). But what to do to make work as BackSpace and work as Delete??? In xterm this works as I want it to work, so the general keybindings seem to be OK. Thanks for any help Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Help installing dosemu
On Mon, 5 May 1997, Rick Jones wrote: > Have you looked in the /etc/dosemu directory? > > lrwxrwxrwx 1 root root 32 May 1 15:13 conf -> > ../../usr/lib/dosemu/config.dist > -rw-r--r-- 1 root root4 Apr 16 17:32 users I had to `dpkg --purge dosemu' before installing the new one. Now I found it, thanks. But another problem apeared: I can't lredir any directory. Dosemu refuses any command lredir X: LINUX\FS\ I remembered that I have done it successfully using Dosemu 0.64.0. Unfortunately I've lost all my configuration files caused by a disc-crash (and the backup was out of date :-(). (With 0.64.0 I used MSDOS, but now I tried fdos. Could this be the reason?) Any ideas Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Help installing dosemu
On Mon, 5 May 1997, Rick Jones wrote: > I have 0.66.2-1 from frozen and it has the config files. dpkg -c dosemu_0.66.2-1.deb | grep etc drwxr-xr-x root/root 0 Apr 16 23:32 1997 etc/ drwxr-xr-x root/root 0 Apr 16 23:32 1997 etc/dosemu/ -rw-r--r-- root/root 4 Apr 16 23:32 1997 etc/dosemu/users Is there something I'm doing wrong??? Andreas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Help installing dosemu
On Sun, 4 May 1997, Rick Jones wrote: > FreeDOS is the fdos package. It's in dselect under fdos. Incase you > aren't aware of it the package list in dselect is searchable (using a / > character). If I'm right in dosemu_64.0.deb was in /etc/dosemu a file fdos.conf which contained useful initialisation for dosemu with fdos. But in dosemu_66.*.deb *all* configuration files are missing :-) (but they are mentioned in the docs). Is there any failure while assembling the package??? Andreas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Error Message of kerneld?
Hello, when starting my Debian box I get the following error message: Loadin modules /etc/init.d/kerneld: cd: /proc/sys: No such file or directory It is the same for different kernel versions (2.0.29 from HD or 2.0.25 from an old floppy). What consequences could happen? I get some warnings about unresolved symbols while `depmod -a'. Is there any connection? If not I ask the depmod-problem more detailed in another question. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: NFS and debian packaging system
On Thu, 17 Apr 1997, Portuesi Simone wrote: > The problem is that I would like to reduce workarounds as much as > possible as I'me not the only one to administrate it. Probably the best > solution is to un-mount the shared partitions when removing packages, > but: What parts do you want to share??? I try to share /usr parts of /var /home (in the future) How to unmount /usr while installing/uninstalling?? > I don't know how dpkg and dselect are going to react to this. If I would be dpkg I would complain about a unmounted /usr ;-). > Yep, I think I can, I just wanted to make installation as clean as > possible as University installation with many people administrating or > working at root level (S.O. and network) experimenting tend to enhance In my opinion a System can't only be as clean as possible if only one adminitrator (may be two if the main administrator is absend) does the job. I think your idea will lead to a confusing system. > over time all the initial installation flaws, thats why we chose debian > (as well with others considarations), it's packaging system permits to > renew and maintain installations at lower risks. I think that's the right choice :-). Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Emacs keys differences between console and X
On Thu, 17 Apr 1997, David Pfitzner wrote: > > On Thu, 17 Apr 1997, Andreas Tille wrote: > > I had a problem like this, and I think the main thing I had to do > to get the behaviour I wanted was to uncomment "XkbDisable" > in my /etc/X11/XF86Config. > (I had guessed wrong when the setup had asked about xkb...) Hmmm, yes now works like expected but German Umlauts disapeared. So that couldn't be a soulution for me because I need the Umlauts much more than the Meta key moreover I don't know how to write Umlauts in this case but I know a method to get Meta if isn't working :-). But thanks for this answer. By the way my XF86config looks like XkbLayout "de" XkbVariant "nodeadkeys" XkbKeymap "xfree86(de)" but I have to press after ~, ', `, ^. Thought it shouldn't be necessary for nodeadkeys. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Do teTeX packages work well?
On Thu, 17 Apr 1997, Karl M. Hegbloom wrote: > Has anyone else tried NTeX at all? I have that installed here; I > chose it because it comes with more documentation than teTeX did. The > installer is Debian compatible; it even provides the right packages > for things that depend on them. > > I'm a LaTeX newbie, so cannot make a very good judgement about which > distribution works better. The main thing I noticed was that NTeX > seemed to have a lot more documentation along with it than teTeX. > I've got Knuth's TeXbook in TeX now; that comes with NTeX, along with > some other books in markup form. I've seen a NTeX version comming with SlackWare one year ago. Hmm, lots of fonts, lots of documentation but useless in my opinion. (I never had the feeling to write like the Klingons and why should I read the documentation for that stuff.) By the way SlackWare switched to TeTeX, too. I think that's a sign if such a distribution does a change. This shouldn't be a flame against NTeX, but I know many users of TeTeX and NO user of NTeX. If you miss any documentation try a CTAN-mirror for instance ftp.dante.de There is nothing important in the world of TeX which you would not find there. Hope this helps Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Netscape/Lynx long startup time - why?
On Wed, 16 Apr 1997, Geoff R Deasey wrote: > If you type ping -c 5 www.microsoft.com > how long does it take before you see the first > line ? > > This almost sounds like a DNS timeout. Here is the output: bridge:~ ! ping -c 5 www.microsoft.com PING www.microsoft.com (207.68.137.59): 56 data bytes 64 bytes from 207.68.137.59: icmp_seq=0 ttl=49 time=214.8 ms 64 bytes from 207.68.137.59: icmp_seq=1 ttl=49 time=213.0 ms 64 bytes from 207.68.137.59: icmp_seq=2 ttl=49 time=218.7 ms 64 bytes from 207.68.137.59: icmp_seq=3 ttl=49 time=207.3 ms 64 bytes from 207.68.137.59: icmp_seq=4 ttl=49 time=207.4 ms --- www.microsoft.com ping statistics --- 5 packets transmitted, 5 packets received, 0% packet loss round-trip min/avg/max = 207.3/212.2/218.7 ms bridge:~ ! Just for comparison the central communication host of our university: bridge:~ ! ping -c 5 mlucom PING mlucom.urz.uni-halle.de (141.48.3.3): 56 data bytes 64 bytes from 141.48.3.3: icmp_seq=0 ttl=251 time=5.6 ms 64 bytes from 141.48.3.3: icmp_seq=1 ttl=251 time=5.5 ms 64 bytes from 141.48.3.3: icmp_seq=2 ttl=251 time=6.0 ms 64 bytes from 141.48.3.3: icmp_seq=3 ttl=251 time=5.8 ms 64 bytes from 141.48.3.3: icmp_seq=4 ttl=251 time=13.6 ms --- mlucom.urz.uni-halle.de ping statistics --- 5 packets transmitted, 5 packets received, 0% packet loss round-trip min/avg/max = 5.5/7.3/13.6 ms bridge:~ ! I don't know what times to expect. But why have lynx to look at www.microsoft.com (or any other host for instance). How to avoid looking at hosts which are situated oversea? Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Emacs keys differences between console and X
On Wed, 16 Apr 1997, Rick wrote: > Activating meta in X is in the XF86Config file. Here is an extract from my /etc/X11/XF86config: # To set the LeftAlt to Meta, RightAlt key to ModeShift, # RightCtl key to Compose, and ScrollLock key to ModeLock: LeftAlt Meta # this line was commented out before #RightAltModeShift #RightCtlCompose #ScrollLock ModeLock I deleted the `#` before "LeftAlt Meta" according to the hint of Rick but it didn't help. The answer to -x is only `beep` end I have to type x anyway. By the way: what do I have to read to learn more about the purpose of `ModeShift`, `Compose` and `ModeLock`? Thanks to Rick, but are there any mor hints Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Do teTeX packages work well?
On Thu, 17 Apr 1997, Mark Phillips wrote: > I notice there are several teTeX packages available. Have many people > tried them? Are there any problems with the packages or are they > stable enough for me to install them and be confident things are > going to work properly? While installing it is a little bit hard to remove the remainder of the older installed TeX packages. (I had to edit some removal script of the old packages but it was no problem in the end.) TeTeX works very good and I'm very happy about the change to TeTeX. > Is the teTeX distribution complete, or is it missing a few things? Nothing is missing. Thanks for the developers to include TeTeX into Debian. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Emacs keys differences between console and X
Hello, I use Emacs at the console (I have an 486/33 :-() and under X. I've done some key bindings in my .emacs (more detailed in an *.el file called from .emacs but this is not the point): (global-set-key [C-left] 'backward-word) (global-set-key [C-right] 'forward-word) (global-set-key [C-prior] 'beginning-of-buffer) (global-set-key [C-next] 'end-of-buffer) (global-set-key [C-backspace] 'backward-kill-word) (global-set-key [C-delete]'kill-word) (global-set-key [M-backspace] 'backward-kill-sentence) (global-set-key [M-delete]'kill-sentence) Well, this works fine under X :-). But under X the -key refuses to work as Meta. In any case I have to press -... . While working at the console my -key works fina as Meta, but the key bindings mentioned above don't work. or any works fine but it seems me that the Control key didn't work together with other special keys. A hint for this behaviour is that says that was pressed (the same when trying seems only the F?-key to be pressed). So the question is: How to enable the as Meta under X and how to make a differnce between for instance and ? Thanks in advance Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
A question about TmView (dvisvga)
Hello, I installed the tmview_96.05-1.deb package which I got from bo/binary/tex some days ago (it disapeared from there now, like some other TeXrelated packages??). At home it works fine at home (360 DPI fonts) but in the university it produces only a very strange vertical pattern. Behind this pattern I can see the shadow of my DVI just like after a curtain. I use 600 DPI fonts. There aren't any error messages. Does anybody have a clue Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: NFS and debian packaging system
On Wed, 16 Apr 1997, Portuesi Simone wrote: > I've to install debian linux in 4 PCs here at the university (now they > have an old slackware). My idea was to have one of them to export users > and their directories plus a directory with apps not in the debian > distribution (a sort of non-local). Up to here no problems arise. > But I'd like also to share man pages, info files and documentation > across the 4 systems but that will break the packaging handling of > Debian (ex. removing a package in one PC will remouve all the > documentation and help on the others), is there a work-around? > In particular is there a way to inform the debian packaging system of > shared files ? I maintain a Debian system with a workstation as NFS server. That means the /usr tree resides on a drive mounted via NFS (ro) and some parts of var resides also there. I linked /var/texmf and /var/catman on a NFS mounted drive (rw). This works fine for me and I'm intended to use this server from several other Linux PCs (but not tried yet). I have to mention that I had serious trouble while installing and if you have problems don't hesitate to ask me special problems. But once installed (using some tricky workarounds) it works fine (but slightly slower). > Also is there a method to install part of a Debian package in a > different directory? (I think the only one is to modify packages by > hand). That's not necessary in any way. The only thing is to link some files in /etc to a common tree but I think this is not necessary and too much work. > Even if it is complex it would be nice if future versions of Debian > would permit such kind of installation or have tools to permit it. OK, tell me your problem. I think you will cope with them with the actual version. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Netscape/Lynx long startup time - why?
On Mon, 14 Apr 1997, Brian C. White wrote: > > Why does it take so long for Netscape or Lynx to start? > > It takes forever because they both read the entire file and run every > single test condition they encounter. > > The latest mime-support (2.12) will work much better with lynx since > it tries to use tests that lynx can recognize internally without forking > an external process. Where is 2.12 available. I can only find 2.11 on my nearest Debian mirror. Lynx takes so terribly long time form me to start that I wonder if I should install a plain ASCII-reader in my Midnight-Commander for html files. Are there possibly other reasons for this behaviour? Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Install base system from frozen (bo)
Hello, I tried to install the install disks from frozen/disks-i368/1997-03-16. I made this disks as described in the docs by dd. After mounting several local disks via the install program I selected the entry to install base system. I have chosen /dev/fd0 as install medium (NICE TO HAVE THE NEW POSSIBILILTIES of mounted disks -- thanks to the developers -- but I had only the disks in my hand at this moment). That caused the following error: Usage: floppy_merge device > You might want ro write down error messages After pressing it complained about many files which aren't accessable. This files are: /target/sbin/unconfigured.sh /target/etc/inet.d /target/etc/inittab /target/etc/kbd ... Switching to another Console I noticed, that there are no directories /target/sbin or /target/etc. What's the matter. I could cope with this problem because I have a running system and are able to update to `bo' without installing from scratch. But I'm afraid that the installation from scratch will fail in any case when a user have to install in this way Greetings Andreas.
By default fvwm2 starts xterm -- why?
Hi, I wonder why fvwm2 starts a xterm when starting. I didn't found any call of xterm while browsing through the startup files in /etc/X11/fvwm2. What should I do if I want to avoid this or what should I do to if I prefer rxvt with a geometry different from the standard size? Thanks for advise Andreas.
Unknown host (more)
Hi once more, I tested the other direction to send mail from an other host to my machine and this was the result: -- Forwarded message -- Date: Wed, 2 Apr 1997 10:17:05 +0200 (CEST) From: MAILER-DAEMON <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Subject: mail failed, returning to sender |- Message log follows: -| no valid recipients were found for this message |- Failed addresses follow: -| <[EMAIL PROTECTED]> ... unknown host |- Message text follows: | Received: from mlucom.urz.uni-halle.de ([141.48.3.3]) by bridge.physik.uni-halle.de with esmtp id m0wCLES-1JC (Debian Smail-3.2 1996-Jul-4 #2); Wed, 2 Apr 1997 10:17:04 +0200 (CEST) Received: from localhost by mlucom.urz.uni-halle.de with SMTP (1.37.109.23/15.6) id AA294219062; Wed, 2 Apr 1997 10:17:43 +0200 Posted-Date: Wed, 2 Apr 1997 10:17:43 +0200 Received-Date: Wed, 2 Apr 1997 10:17:43 +0200 Date: Wed, 2 Apr 1997 10:17:42 +0200 (MESZ) From: Andreas Tille <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Subject: April Message-Id: <[EMAIL PROTECTED]> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Any hints Andreas.
Unknown host
Hi, i installed smail from the unstable tree. I havn't any idea how to configre smail. When I installed Slackware half a year ago I configured nothing and it worked. Now I answered the some questions while installing smail, but any mail to any host (except on my own machine) is returned to sender. Would anyone be so kind to give me a hint, which document to read or which file is responsible for the error which causes the following reply on a Testmail: Date: Wed, 2 Apr 1997 09:41:08 +0200 (CEST) From: <[EMAIL PROTECTED]> To: e2od5 Subject: mail failed, returning to sender Reference: <[EMAIL PROTECTED]> Date: Wed, 2 Apr 1997 09:41:08 +0200 (CEST) From: <[EMAIL PROTECTED]> To: e2od5 Subject: mail failed, returning to sender Reference: <[EMAIL PROTECTED]> |- Message log follows: -| no valid recipients were found for this message |- Failed addresses follow: -| [EMAIL PROTECTED] ... unknown host |- Message text follows: | Received: by bridge.physik.uni-halle.de id m0wCKff-1JC (Debian Smail-3.2 1996-Jul-4 #2); Wed, 2 Apr 1997 09:41:07 +0200 (CEST) Message-Id: <[EMAIL PROTECTED]> Date: Wed, 2 Apr 1997 09:41:07 +0200 (CEST) From: e2od5 (Andreas Tille) To: [EMAIL PROTECTED] Subject: April Thanks for advise Andreas
Can't find unstable/binary/base/modutils_*.deb
There is no modutils_??.deb in /bo/binary-i386/base but modules_2.1.23-1.deb depends on it. Is my ftp server wrong or isn't this file existing? Andreas
Error while installing unstable/binary/base/at_3.0.deb
While doing dpkg -i at_3.0.deb I got the following errormessage. The at_2.9b??.deb in stable works. (Reading database ... 21163 files and directories currently installed.) Preparing to replace at 3.0 (using at_3.0.deb) ... Unpacking replacement at ... /var/lib/dpkg/info/at.postrm: spool}: command not found dpkg: warning - old post-removal script returned error exit status 127 dpkg - trying script from the new package instead ... /var/lib/dpkg/tmp.ci/postrm: spool}: command not found dpkg: error processing at_3.0.deb (--install): subprocess new post-removal script returned error exit status 127 /var/lib/dpkg/tmp.ci/postrm: spool}: command not found dpkg: error while cleaning up: subprocess post-removal script returned error exit status 127 Errors were encountered while processing: at_3.0.deb Is there anybody who observed the same or should I send a bug report (or are there any developers reading this and take it as a bug report and I havn't to send this twice) Andreas.
Re: A lot of errors while installing (dselecting) bo
On Sat, 22 Mar 1997, Rick wrote: > The files are dependant on order. To update properly you should put what > are expected to be the most recent files at the end of the list. > Otherwise you may not be shown newer files that exist in other tree's. > > [stable contrib non-free Debian1.2-updates unstable] Would you recommend that I should try this from scratch or should it be enough to correct this list and continue with dselect. > >findutils_4.1-20.deb: > > unable to create `usr/bin/find': Text file busy > > Should have done a ps -aux to see if find was being used at the time. If > not check for a lingering lock file. I tried successfully another trick. As I wrote I mounted /usr via NFS. I supposed that this would be the problem and tried the following: 1) mv /usr/bin/find /local_usrbin 2) ln -s /local_usrbin/find /usr/bin/find where /local_usrbin is a directory on a local drive (root). This worked for find but not for several other files (for instance mcserv (from mc-package) and ispell). I'm really sure that this is a problem of NFS. I've got I hint from other readers of this list while asking further questions that there is a "File locking"- mechanism of NFS. What about this? How to avoid this or come around this tricky problem? Thanks for the reply. Are there any further hints. Andreas.
Additional remarks to: A lot of errors while installing (dselecting) bo
On Tue, 18 Mar 1997, I wrote: > Does anybody think that it is a problem of the unstable tree. But if this > would be the case anybody would have reported such serious errors before. I tried to verify the problem today by installing stable. --- It was the same ... :-(((. > Please help. I tried very hard, several days and I'm afraid I will go > mad with this stubborn box. I havn't any clue how to cope with this > problems. I seriously need help. Andreas.
A lot of errors while installing (dselecting) bo
Hi, I'm the one who frequently asks questions about installing :-(. I had have many trouble while installing Debian on a NFS-Server (see many many different postings in the past). Now I solved one problem and got two further problems :-(. For the guys who helped me solving my old problem: I mounted only /var/lib/dpkg/methods/ftp/debian via NFS to store only the *.deb files on the server and all the other stuff from /var stay now on my local disk. This works up to getting the next problem. I tried to install /bo (/unstable) because I want to use TeTeX and thought that after quite a long time of testing /unstable will become /stable after some days. Many postings reported that /unstable is quite stable. I used the ftp-method. Is it right, if I selected [bo non-free distrib] or should I use stable (in front of this list) too? When installing the selected files I got many error-messages. Some of them are: 1) at_3.0.deb: /etc/crontab -> no such file or directory ... that was right, because at was the first package and cron wasn't installed. 2) The next stands as an example for an error, which occured at many packages: findutils_4.1-20.deb: unable to create `usr/bin/find': textfile busy Please don't ask whether other users worked in my box: I was the only one logged in and my only job was to install Debian. 3) As a consequence I got after failing to install such packages /usr/sbin/install-info: no such file or directory ... but this file exists 4) OK, some packages were installed but after a number of failing packages dpkg stops and I was asked if I would wish to delete ... Yes couldn't figure out type of at_3.0.deb Couldn't get info for: unstable/binary-i386/admin/at_3.0 I got this message for EACH package I wanted to install (and got by dpkg-ftp in /var/lib/dpkg/methods/ftp/debian). For each stupid package regardless if it was successfully, unsuccessfully installed or skipped because dpkg was broken after a number of errors. And to recognize that it couldn't figure out/get info dselect needs TERRIBLY LONG TIME. Its a questions of MINUTES (I stopped the time for syslinux_1.30-2.deb: it was 2 Minutes and 25 Seconds!!). It seems me that the time it tooks dselect to figure out/get info is longer than getting the file by ftp. So don't tell me please anything about the performance of my NFS-Server. Does anybody think that it is a problem of the unstable tree. But if this would be the case anybody would have reported such serious errors before. Please help. I tried very hard, several days and I'm afraid I will go mad with this stubborn box. I havn't any clue how to cope with this problems. Any help appreciated Andreas.
Re: dselect error (more)
On Fri, 7 Mar 1997, Bob Clark wrote: > Two things. If I remember correctly (from your previuos posts) you have > your packages on an nfs-mounted partition; and a troublesome one on an > HP machine at that. There could be many things not related to deselect > or Linux that may be affecting your system. Second, I imagine that > dselect is doing other things besides announcing that it is "leaving" > certain packages during this time. I think 5-10 seconds might be quite > normal given your configuration. > > How fast is your connection to the HP? I would say "not bad" because I don't *feel* any speed loss while doing some things. > What is the load on the HP? Usually I'm the only one. It's a shame to say that but it isn't used (see the next question for the reason). > How well has nfs been implemented on the HP? If there is any chance to do it in different ways I'm quite sure that the administrator had choosen the bad way. This machine is in a terrible state and it seems me that to become a Server for several Linux boxes would be the a rebirth of it. Fortunately I got the chance to do the job of installing the Linux distribution myself. If the administrator of the HP machine would do that it would never work. > How many packages does dselect have to look at? About 30. But it tooks 5-10 seconds for EACH. The time needed seems to be proportional to the size of the package (emacs needs longer than at for instance). > What kind of load is on both these machines? On both I was the only user at this time logged in and I hasn't done any other thing than dselect. > How much memory do you have? 20MB at the 486/66 > Do you swap to a swap partition? I think this wouldn't be necessary in this condition. > Is it local or nfs-mounted? In case it would be done on a local disk, of course. > These are just a few things that might have an impact on your system > performance. I think most of these things shouldn't matter according to this answers. Correct me if I'm wrong. What about the feeling of you. Takes it so much time on your machine, too? Andreas.
Re: Non-existing packages in dselect
On Fri, 7 Mar 1997, Heikki Vatiainen wrote: > Yes sir, that's it. I did associate the starnge behaviour with dselect even > though it's dpkg-ftp working under the hood. I forgot to check the bugs this > time but I'm not sure if I would have found it under dpkg-ftp. > > Thanks for clearing this to me. > > > You're probably using dpkg-ftp and have hit bug 6210. Could someone give me a pointer to an explaination of this bug. May be I also run in trouble while using dpkg-ftp caused by this bug. Andreas.
Re: dselect error (more)
On Thu, 6 Mar 1997, Bob Clark wrote: > It is "leaving" packages that were *not* installed. Things like pakages > that were present before this instance of deselect. This is what I read on my screen, but to leave a package which wasn't installed should take 5-10 seconds?? That's the time it takes me to leave the room but I thought my Debian box should be a little bit faster when doing quiet easy things. What means "leaving". If "leaving" means to copy each package to each partition on my box the time it tooks would fit my expectation, but I think leaving is something other :-). I thought it may be a hint on any error, if a process which should be done quiet fast takes a much longer time. Correct me if I'm wrong Andreas.
Re: dselect error (more)
Under the subject: dselect error I posted some odd things. Now a further question: After the emergency stop caused by errors dselect asks, if I want to delete the installed stuff. I said yes and now I get messages of the type: leaving: stable/.../package.deb for each package. But it takes a terribly long time (about 5-10 seconds per package) though nothing was installed. What is dselect doing in this time???
Re: dselect error (Was: Access to NFS-Server)
On Thu, 6 Mar 1997, Ronald van Loon wrote:
> Try adding the -P option to the df command in
> /usr/lib/dpkg/methods/ftp/install:
>
> line 230:
>
> my $avsp = `df -P -k $::dldir| awk '/\\// { print \$4}'`;
>
> line 236:
>
> $avsp = `df -P -k $::dldir| awk '/\\// { print \$4}'`;
This helps to *get* all files I want to install :-), but the problem
while installing remains :-((.
OK, the problem of free disk space was solved and I suggest to
insert the '-P' option in the install script.
But what about the other problem? I had the hope that it would disapear
but after
Do you want to install the files fetched [y]:
I got:
Installing files...
dpkg: error processing debian/stable/binary-i386/admin/acct_6.2-2.deb
(--install):
failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists
dpkg: error processing debian/stable/binary-i386/admin/adjtimex_1.2-5.deb
(--install):
failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists
dpkg: error processing debian/stable/binary-i386/admin/at_2.9b-1.deb
(--install):
failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists
dpkg: error processing debian/stable/binary-i386/admin/cpio_2.4.2-8.deb
(--install):
failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists
(and so on -- now I have some more files than before got by ftp)
dpkg: too many errors, stopping
Errors were encountered while processing:
debian/stable/binary-i386/admin/acct_6.2-2.deb
debian/stable/binary-i386/admin/adjtimex_1.2-5.deb
debian/stable/binary-i386/admin/at_2.9b-1.deb
debian/stable/binary-i386/admin/cpio_2.4.2-8.deb
Processing was halted because there were too many errors.
DPKG ERROR
I can't imagine any reason for this behaviour. What the hell is the
task of the files in tmp.ci? An `ls -l' says:
drwxr-xr-x root root tmp.ci
I had the supposition that dselect
uses some quite different rights from root but this makes no sense at all.
This supposition is provided by the following fact. Former I mounted only
/var/lib on the HP-server. Now I tar-ed the complete /var-tree, mounted
/var on the HP-server and un-tar-ed it. I've got some mysterious error
messages from tar when unpacking the *.deb files:
tar: cannot chown file var/debian/stable/binary-i386/.deb to uid 47653 gid
64696
tar: cannot chown file var/debian/stable/binary-i386/.deb to uid 2282 gid
64976
The "possibility" of the second uid/gid combination was much higher (1:5)
but it seems to be randomly if a *.deb package failed to chown to one or
the other user. What about this uid/gid numbers. May be the root of my
problem is here.
Any more ideas to fix this bug out there? Without any clue
Andreas.
dselect error (Was: Access to NFS-Server)
Hi once more after some conversation with Jeff Gunter <[EMAIL PROTECTED]> from which I've got some new ideas I post some parts of this discussion to the list and add some new aspects of my problem. (The new aspects are the reason why not posting only to Jeff.) JG>Sounds like dselect is df-ing the /var to figure out if there is file space. JG>That's a bug in my opinion. It should probably also check to see if you JG>have a remote mounted /var/lib or /var/lib/dpkg or /var/lib/dpkg/methods/ or JG> /var/lib/dpkg/methods/ftp etc. I tried to verify this idea by moving the complete /var-tree to the HP-machine. I thought it wouldn't be a good idea because of the low speed of NFS-access, but now I tested this. The result was the same: Approximate total space required: 28526k Available space in debian: 40%k Space required is greater than available space, you will need to select which items to get. But a df says: ... Available ... Mounted on /dev/hda25288 / hp.workstation://usr1038027 /usr hp.workstation://var1038027 /var It seems to be enough space to store under /var where the packages are stored while installing. JG> packages are compressed and I *think* they get uncompressed BEFORE JG> installation and so require more file space. JG> JG> What happens if you only install one very small package? Does it put JG> things on the HP ok? Assuming you are installing via ftp ... you could JG> look on the HP and see if the files got put there ok Yes, the packages come in /var/debian/stable/binary-i386/admin. Now I observed a further strange behaviour: I selected only the packages in the beginning of the dselect-list which were preselected as default. Some packages I deselected because I thought I wouldn't need them in the first step. For instance I deselected taper-x.x.deb. BUT taper-x.x.deb EXISTS in this path. I think this isn't a clever behaviour. (I checked my list twice to be sure taper is deselected!) OK, the *.deb-Packages are stored on the workstation which should become the NFS-Server for my Debian box. May be that it is usual when installing only the packages in the admin/ directory were stored. While installing them I ended up with error messages. JG> What are the error messages? Here es the output: Unpacking acct (from .../admin/acct_6.2-2.deb) ... Stopped process accounting. dpkg: error processing debian/stable/binary-i386/admin/acct_6.2-2.deb (--install): package control info contained directory `/var/lib/dpkg/tmp.ci/control' dpkg: error while cleaning up: failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists dpkg: error processing debian/stable/binary-i386/admin/adjtimex_1.2-5.deb (--install): failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists dpkg: error processing debian/stable/binary-i386/admin/at_2.9b-1.deb (--install): failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists Errors were encountered while processing: debian/stable/binary-i386/admin/acct_6.2-2.deb (--install): debian/stable/binary-i386/admin/adjtimex_1.2-5.deb (--install): debian/stable/binary-i386/admin/at_2.9b-1.deb (--install): DPKG ERROR I can't imagine any reason for this behaviour. What the hell is the task of the files in tmp.ci? An `ls -l' says: drwxr-xr-x root root tmp.ci But this directory does *not* contain a file `control' in the moment the error messages mentioned above apeared on the screen (may be it is deleted but any rm-call failed but which???). Questions, questions ... JG> This sounds like a good bug, but I wouldn't be surprised if nobody had JG> stumbled across it yet. I also think this but I want to give a description which makes sure that this bug can be reproduced. It me seems so strange that I'm afraid it would be ignored. I want to make sure to give the true reason for this bug. JG> I would check the permissions. Try creating directories and such on the JG> /var/lib filesystem. Yes, I've done this and all worked fine. I had the supposition that dselect uses some quite different rights from root but this makes no sense at all. This supposition is provided by the following fact. Former I mounted only /var/lib on the HP-server. Now I tar-ed the complete /var-tree, mounted /var on the HP-server and un-tar-ed it. I've got some mysterious error messages from tar when unpacking the *.deb files: tar: cannot chown file var/debian/stable/binary-i386/.deb to uid 47653 gid 64696 tar: cannot chown file var/debian/stable/binary-i386/.deb to uid 2282 gid 64976 The "possibility" of the second uid/gid combination was much higher (1:5) but it seems to be randomly if a *.deb package failed to chown to one or the other user. What about this uid/gid numbers. May be the root of my problem is here. Any more ideas to fix this bug out there? Without any clue Andreas.
Re: Access to NFS-server
On Wed, 5 Mar 1997, Jeff Gunter wrote: > On Mar 5, 6:02am, Bob Clark wrote: > > > > cd > > rsh my.pc "cd /usr; tar cf - ." | tar xf - > > This may work to copy the /usr-tree after installing the base disks. But it won't solve the upcoming problem (see below). > > root=hostname[:hostname]... >Give root access only to the root (uid 0) users from a >specified hostname. The default is for no hosts to be granted >root access. > > While this does open up the possibility for someone to spoof your pc's ip >address and mount up the filesystem, it's safer for the HP because somebody who > breaks into the pc can't immediately hop onto the HP as root. OK, this works when copying the /usr-tree. I moved /usr and /var/lib to the workstation. I've done the necessary entries in /etc/fstab and rebooted, so I come up with the small diskspace of my own computer and the huge diskspace made available in /usr and /var/lib via NFS on the HP-workstation. Now I tried dselect and here we are -- me and my very old problem (my problem is about 9 weeks old): After selecting the wanted packages (method ftp) dselect complained: Approximate total space required: 28526k Available space in debian: 40%k Space required is greater than available space, you will need to select which items to get. I wanted to install only a few packages at first (that's why only 28M) and if I call `df' I get: ... Available ... Mounted on /dev/hda25288 / /dev/hda3 74720 /var hp.workstation://usr1038027 /usr hp.workstation://var/lib1038027 /var/lib It seems to be enough space to store under /var/lib where the packages are copied to and under /usr where the packages are installed in. The rest of dpkg's job (which was called by dselect) is a mess of error messages. Is there anybody out there who have any idea why dpkg is so stubborn? I think that the failure is caused by the special configuration with the NFS-server. Should I send a bug report concerning dpkg or should I look for the reason on my own boxes concerning permissions and so on? I'm really helpless (and tried several times since more than two month). Any further help is very appreciated Andreas.
Access to NFS-server
Hallo, because of a lack of diskspace I want to use another computer (HP-Workstation) as NFS-server. So I installed the base disks of debian and wanted to move /usr to the NFS server by tar -cf usr.tar /usr and tar -xf usr.tar in the approriate directory of the server mounted via NFS. But I got: tar: Cannot change mode of file <...> to <644/755>: Operation not permitted for each file which is untared. The files exists after this mess of errormessages and have the right rights but the owner is root (the root of the server) and so rm -r usr ends with rm: <...> : Permisssion denied. OK, the reason is that the information in /etc/exports of the NFS-Server is not correct. I played around with several combinations of /usr/local/linux -access= /usr/local/linux -access=,root= /usr/local/linux -rw= ... Please don't ask me what ever I tried but nothing helped. Yes I did not forget to call exportfs -a after every change in /etc/exports. Is anybody out there who knows the right way to setup the server correctly Any help appreciatet Andreas.
Re: A NFS problem
On Thu, 27 Feb 1997, Martin Stromberg wrote: > Perhaps the server only can export the directories in the export file; > try mounting with "mount -t nfs 141.48.54.2:/usr/local/linux /mnt". > I really don't know a lot about nfs mounting, but trying won't hurt. I forgot to mention that I also tried this. I also tried to mount another Linux box which worked everey time before but failed in the circumstances I described. > If you have another linux box running you could make a statically (or try > with the dynamically) linked ping and copy it to a floppy and then mount > that floppy on /dev/fd0 on the box you are trying to get connected. I will try that. > But I do wonder if you have nfs compiled into the kernel or as module > or not at all? Yes, Susan G. Kleinmann compiled it into the kernel. The error message is completely different if you havn't NFS or NFS has to be loaded as module but can't be found. > Hopefully helpful, Hope that ping gives further hints (I will report tomorrow). Hopefully there are other ideas. Andreas.
Re: The Future of Debian's TeX system: tetex
On Tue, 25 Feb 1997, Christian Schwarz wrote: > The new tetex packages will be included in Debian 1.3. There are now part > of the unstable distribution and will replace the old packages in about a > week. If you want to help you could download these new packages and test > them. In my opinion that's quite a good decision. I think it's a good idea to have something like a standard TeX distribution. It the hint on Debian 1.3 something like a "preannounce" of an upcomming new release in the next weeks? Greetings Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: debian-newbie list
On Fri, 21 Feb 1997, Susan G. Kleinmann wrote: > Of course, we could break it up further. For example, we could have > --an FAQ on installation and booting, and > --an FAQ on using the Debian archives. > --an FAQ on other Debian utilities or Debian practices. > --an FAQ on printing and text utilities > --an FAQ on TCP/IP problems, including PPP. This would be a good idea in my opinion. > An argument could be made that it might be a good idea to have user > lists that were broken up along the same lines as the FAQs. This is possibly necessary, because there is so much traffic on this list. Subscribing this list my dayly mail incresed from about 20 to more than hundred letters. I'm afraid getting so much per day I sometimes realize some important mails while pressing 'd' (delete this letter) while scanning only the subjects. Thanks for doing the job to write the FAQs Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Any help appreciated
Hello, this is my 4-th try to solve my problem. If somebody is interested in my concrete problem I can post my further letters under the subjects "Installation failure", "Once more: Installation failure" and "Instalation problems, anyway" with private mail. I don't want to bother the readers of this list with this question which seems to be not so frequently that somebody is able to answer on spot. I hope it would be helpfull if someone would be so kind to build a resque disk for me which contains a kernel including nfs and the network driver for my card (smc8000, which needs wd.c). I would be very grateful if somebody would do this job which could possibly solve my problem, because I'm not able to do that. I don't need any CD-ROM driver or SCSI devices. If somebody sees an other solution please tell me. I very urgently need help. Thanks for any help which brings me one step nearer a running Debian system Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Once more: Installation failure
Hallo, I posted this message last week. Because I didn't got any answer I try to give more details in the end of my posting. I have installed Debian 1.2 successfully at home :-). Now I wanted to install it also at the university. I tried to do this by the "ftp"-method but failed. The problem seemed to be the following: I have a partition mounted on /var of about 80MB. To download the packages temporaryly they were stored somewhere in /var/lib/anywhere. To get enough space I mounted /var/lib on another computer (per NFS) with more than 1GB free space. But dselect (and freinds) seems to ignore this space and stops by saying "Not enough memory" (This are not the correct words -- ask me if I should try it again to get the correct answer.). So I downloaded all files to /var/lib/debian and tried the method "mounted disk". But I also failed and I am wondering what this errormessage would mean. When selecting "Install packages" I get the following: dpkg: error processing /path_to_package/package.deb failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists After a number of this messages dpkg stops caused by to many errors. What the hell should be the reason for such a message. I tried to set all files in /var/lib/dpkg/tmp.ci and the directory itself by chmod 777 * but nothing helps. May be I missed the following special environment: In the university we have some PC's with very small diskspace ( < 200MB ) all working in a Novell network. But it's to less for Linux :-). Because we have an HP workstation with relatively less trafic I thought we can use it as Linux fileserver per NFS. The administrator of this workstation (who isn't using Linux himselves) created a user `linux' on this workstation with the homedirectory /usr/local/linux. He added in /etc/exports: /usr/local/linux -access=[my.own.machine],root=[my.own.machine] so that I have access to this directroy being root on my Linux PC. I created /usr/local/linux/usr and /usr/local/linux/var/lib and added in my /etc/fstab two lines to mount /usr and /var/lib, respectively. Than I tared the /usr-tree, installed after installing the Debian install disks and untared it on /usr/local/linux/usr. I need to use /usr/local/linuc/var/lib to store the amount of *.deb files. I hope this additional information gives a hint to solve my problem. I'm in serious trouble. Please help. Thanks for all help. Andreas. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Installation failure
Hello, I have installed Debian 1.2 successfully at home :-). Now I wanted to install it also at the university. I tried to do this by the "ftp"-method but failed. The problem seemed to be the following: I have a partition mounted on /var of about 80MB. To download the packages temporaryly they were stored somewhere in /var/lib/anywhere. To get enough space I mounted /var/lib on another computer (per NFS) with more than 1GB free space. But dselect (and freinds) seems to ignore this space and stops by saying "Not enough memory" (This are not the correct words -- ask me if I should try it again to get the correct answer.). So I downloaded all files to /var/lib/debian and tried the method "mounted disk". But I also failed and I am wondering what this errormessage would mean. When selecting "Install packages" I get the following: dpkg: error processing /path_to_package/package.deb failed to rmdir/unlink `/var/lib/dpkg/tmp.ci': File exists After a number of this messages dpkg stops caused by to many errors. What the hell should be the reason for such a message. I tried to set all files in /var/lib/dpkg/tmp.ci and the directory itself by chmod 777 * but nothing helps. What's wrong here. Is it me? Thanks for all help. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
