Whither CERNLIB (, Paw, Geant3.21) in Debian? RFA / future plans
[Please follow up ONLY to debian-science] [CC'ed to the Fedora CERNLIB maintainer and to Ubuntu MOTU-Science, for their information] I'm cross-posting this to debian-user on the off chance there may be some Debian-using physicists there who don't follow debian-science, which seems to have morphed into a developer list lately. I'm seeking one or more people to take over maintenance of the following FORTRAN- and physics-related source packages. cfortran cernlib paw geant321 mclibs (This message does not apply to my unofficial Geant4 .debs) Since my employment is no longer in the physics field, and family obligations continue to reduce my amount of free time for Debian, I no longer have time or interest to maintain these packages properly. The maintainer switch can be gradual (co-maintainership for apprenticeship for a while is OK by me, but can't be too much of a demand on my time) but needs to be complete well before the release of Squeeze (Lenny+1). Please follow up to [EMAIL PROTECTED] if you want them. Nasty details below. For the reasons below, I think these packages are much too difficult to be maintained by QA or by a team (e.g. debian-science) that is not specifically targeting them. Therefore, if there are no offers, I will request their removal from Debian after the release of Squeeze (assuming they do not get RC bugs filed against them prior to that); until then I will maintain them on a best-effort basis, with a note in NEWS.Debian to the effect that they will go away after Squeeze release. Prospective maintainers should be aware that the packages are particularly challenging for the following reasons: * Most of the code is written in either KR C or FORTRAN IV, with arch-specific #ifdef's around code for cutting-edge machines like PDP 11s. (That is not a joke.) * Bugs in new versions of gfortran often break various bits of code on obscure architectures and make the test suite fail. * The upstream build system is based on imake. * The Debian packaging is dpatch + tarball-in-tarball, with some hacks to make it possible to use Debian patches for building on non-Debian systems. * GUI code is based on Motif ... building the programs against Lesstif seems to mostly work (with some patches) but there may be as-yet undiscovered GUI bugs. * There was a lot of non-free stuff that had to be cut out of the source packages to get the existing tarballs suitable for main. However, this may not cause you much work, since... * Upstream has been dead since 2006, and moribund for several years prior to that. * There is no upstream support for shared libraries, nor for builds on Linux arches other than m68k or i386. Large patches have been hacked into the Debian source packages adding these. * Upstream source is not 64-bit clean. Properly fixing this would require man-years of work. In the meantime, a fellow named Harald Vogt has hacked together some patches that mostly make things work on 64-bit machines... but the patches are so fragile that they break unless programs are statically linked against the CERNLIB libraries on 64-bit. I will wait a week or two to hear from prospective maintainers. If there are any, I will email them with more info at that time. If not, I'll submit official RFAs to the BTS and send a note to debian-science that PAW and friends are slated for doom. In the meantime, prospective maintainers may want to take a look at my CERNLIB on Debian pages at http://people.debian.org/~kmccarty/cernlib/ and also at the source packages (prepare to be horrified). best regards, -- Kevin B. McCarty [EMAIL PROTECTED] WWW: http://www.starplot.org/ WWW: http://people.debian.org/~kmccarty/ GPG: public key ID 4F83C751 signature.asc Description: OpenPGP digital signature
Re: mac grapher for linux
Hi lists, Jordi Gutiérrez Hermoso wrote: I'm forwarding this to the d-science list, where this stuff is often discussed. On 13/05/2008, Jimmy Wu [EMAIL PROTECTED] wrote: Has anyone here ever used the Grapher application that all Macs come with? It is a very nice piece of software with support for 3D and many types of equations, and it's very useful for math/science purposes. Is there something comparable for Debian or Linux in general? I was looking on Google but couldn't find any convincing replacements. Most of the stuff I see is very rudimentary 2D graphers that only allow a very limited amount (like 3) equations. I saw labplot, but I didn't try that because I don't want all the KDE dependencies (which, as an xfce user, means a lot of extra packages). I also gnuplot, which seems to be decent except it doesn't have the nice friendly GUI of mac's grapher. Neither of these are great as substitutes, but I haven't found anything better. So basically, I was wondering if anyone had any ideas before I go invest the time into learning how to use gnuplot? Let me put in a plug for ROOT. It's a (huge) piece of software used mostly by the HEP community, but no reason that others shouldn't be able to use it. There are not yet official Debian packages (except in experimental), but unofficial ones can be obtained from http://mirror.phy.bnl.gov/debian-root/ . The plan is to upload official packages of ROOT version 5.18 in time for them to end up in Lenny. It might not be exactly what you are looking for in terms of user-friendliness, but if you are somewhat familiar with C++ or Python (it has bindings for both) it can be learned pretty quickly and is very powerful. See for instance this page: http://root.cern.ch/root/soeren/ or the big tutorial at http://root.cern.ch/root/Tutorials.html . Just to give you an idea of what can be done, one can type root on the command-line and then run the following code within ROOT's C++ interpreter to get a nice 3D plot of the function f(x,y) = sin x sin y in the range x in (-10, 10), y in (-10, 10). TF2 func(func, sin(x)*sin(y), -10, 10, -10, 10); func.Draw(surf4); There are many, many options for controlling the appearance of the resulting plot. Or if you prefer Python over C++, you can instead (after installing the libroot-python-dev package) start up the Python interpreter and type into it: import ROOT func = ROOT.TF2(func, sin(x)*sin(y), -10, 10, -10, 10) func.Draw(surf4) I think the existing unofficial packages were built against python 2.4, so you'd have to install and run that version of Python explicitly, but the forthcoming official ones will use python 2.5. ROOT also has Ruby bindings but I'm not familiar with that language. If Christian Holm (the package maintainer) is around on this mailing list, he might have more to say about ROOT. best regards, -- Kevin B. McCarty [EMAIL PROTECTED] WWW: http://www.starplot.org/ WWW: http://people.debian.org/~kmccarty/ GPG: public key ID 4F83C751 signature.asc Description: OpenPGP digital signature
Re: [OT] C++ programming: keeping count of data items read from file
H.S. wrote: Michael Marsh wrote: Can you read full lines out into, eg, a stringstream, and parse your doubles out of that? You'd hit an EOF at the end of each line in that case. I'm not sure how you'd get stream out line-at-a-time, though there may be a stream operator that sets the appropriate behavior. Yup, that could be done by getting a line till the end of \n character and then parsing the line. I was just wondering if there was any other way (was trying to avoid parsing). std::getline() ? I think there is a version that outputs the result as a string, from which you can then create an istringstream. You can read numbers from an istringstream the same way as from cin with operators, so the amount of parsing should be minimal (as long as the file is not corrupt). best regards, -- Kevin B. McCarty [EMAIL PROTECTED] WWW: http://www.starplot.org/ WWW: http://people.debian.org/~kmccarty/ GPG: public key ID 4F83C751 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how th generate like oficial site from packages (or CD's)
abelahcene wrote: How the create like an official site from my packages ? Hugo Vanwoerkom wrote: You mean this? http://people.connexer.com/~roberto/howtos/debrepository#setup You should probably note that there is a big warning basically saying these directions are obsolete, look at [1] instead near the top of that page. [1] http://www.debian-administration.org/articles/286 best regards, -- Kevin B. McCarty [EMAIL PROTECTED] WWW: http://www.starplot.org/ WWW: http://people.debian.org/~kmccarty/ GPG: public key ID 4F83C751 signature.asc Description: OpenPGP digital signature
Re: fixing .txt files sent from a MS$ user
Douglas A. Tutty wrote:
Hello all,
My niece sends some of her schoolwork to my wife (e.g. essays) for her
to read. First she sent .doc files which I can't access properly (no, I
do no run OO) although I could get the jist. I then suggested that she
send plain text.
I don't run any locales but have LANG=C.
Her plain-text files are fine, except that apostrophies (singel-quote)
(') are replaced with \222, opening quotes () with \223, and
closing quotes () with \224.
Other than writing a python script, can someone give me a simple command
to fix this so that it is not so distracting? Ideally, I'd put it into
an executable file and pipe the file through it on it's way to lpr.
e.g.
$ cool-writing.txt | antiAJ | lpr -Pepson
Use tr? It's a very simple program meant for exactly this kind of task.
The first argument is the list of characters to replace, and the second
is the list of characters (in order) to replace them with.
tr \222\223\224 \'\\ cool-writing.txt | lpr -Pepson
N.B. the ' and characters need to be escaped from the shell in the
second argument on the command line above.
tr is even in coreutils so you don't have to install anything extra.
best regards,
--
Kevin B. McCarty [EMAIL PROTECTED]
WWW: http://www.starplot.org/
WWW: http://people.debian.org/~kmccarty/
GPG: public key ID 4F83C751
signature.asc
Description: OpenPGP digital signature
Re: Debian website broken ???
Jarek wrote: Hi all! Does anyone know what happend to debian.org ? Most of package lists and package info pages are broken. Like this: http://packages.debian.org/stable/text/ Hm, interesting. The pages for oldstable, testing and unstable all seem to work for me, it is only the pages for stable that are broken. Is it possible this is just a temporary state while p.d.o is in the process of being updated? I think it gets updated every 24 hours but I don't know the exact times at which this happens. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: /etc/ld.so.conf for dpkg-buildpackage issue
Scott Edwards wrote: I'm building a lib under sarge in a debootstrap chroot as a normal user (and fakeroot). When it gets to this part, it's trying to modify the system /etc/ld.so.conf (which is naughty afaict). What's the proper way to prepare this? Thanks - Scott. Hi Scott, It looks like the version of libcommoncpp2 you're trying to compile, 1.5.1, is available as a Debian package in unstable. Have you tried building from that source package under stable? It may work, depending upon whether or not the build-dependencies can be met in Sarge. Even if trying to build under Sarge doesn't work, you could try looking at the diff.gz [1] to see how the Debian maintainer solved this problem. (He _must_ have solved it somehow, otherwise the package would FTBFS on most of the buildds [2].) I'm working from the assumption that your source tree, although Debianized, is not the official Debian source package. If you *are* building from the official Debian source package from unstable and this happens anyway in your Sarge chroot, I have no idea. [1] http://ftp.debian.org/debian/pool/main/libc/libcommoncpp2/libcommoncpp2_1.5.1-4.diff.gz [2] Not on alpha, mips, or mipsel, since IIRC fakeroot doesn't work there and they actually build things using sudo. Hope this helps, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Changelogs on packages.debian.org
I seem to be having issues getting any sort of change logs on packages.debian.org either via aptitude on the server or via web browser on multiple computers networks. This is leading me to believe something is gone awry on the website. Any idea on when this functionality will be restored? I verified the problem, and submitted it as a bug [1] against the www.debian.org pseudopackage [2], so hopefully soon. [1] http://bugs.debian.org/399236 [2] http://bugs.debian.org/www.debian.org regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OpenSSL version 0.9.7e ?!
Hi Nicolas, Nicolas Pillot wrote: I had a strong *shrug* when i noticed that my stable system (originally woody, upgraded to sarge without kernel change) still had ^^^ If you are running Debian-provided kernels, you *really* should upgrade to a kernel from Sarge. The kernels from Woody have not been security-supported for a LONG time, and there have been quite a few serious security issues discovered in the kernel since then. This is potentially a much bigger deal than the OpenSSL issues you are concerned about. OpenSSL version 0.9.7e installed, despite a dayly dist-upgrade. After looking at debian's sarge repository, i saw that the most up to date package is 0.9.7e-3sarge4, which i have (0.9.7 dates back from 2004). My question is, why on earth don't we have a newer version ? I counted about 12 different releases, either 0.9.7- or 0.9.8-based, each including security fixes. I could understand the will not to upgrade to 0.9.8, but i count 7 more recent 0.9.7 versions (up to 0.9.7L version, and the stable debian package build version is -4... Even with a backport of the security fixes, i can't guess how the ssl pacakge i have 0.9.7e-3sarge4 could be the most up-to-date one (security wise). I though that all the security fixes were included into sarge, am i wrong ? If someone could give me some details, i'd be quite happy to learn :-) Debian does not put new upstream releases, even point releases, into a stable distribution. What happens is that only the security fixes are backported into a package in stable. This minimizes the possibility for the stable release to be de-stabilized by new code introduced upstream. So while the version of libssl0.9.7 in Sarge is 0.9.7e-3sarge4, it should nevertheless incorporate all the security fixes present in 0.9.7L. If you look at the top several entries in /usr/share/doc/libssl0.9.7/changelog.Debian.gz, you can see that the following problems have been fixed in the Debian package in Sarge since the upstream release of 0.9.7e: CVE-2006-2940 CVE-2006-2937 CVE-2006-3738 CVE-2006-4343 CVE-2006-2940 CVE-2006-4339 CVE-2005-2969 CAN-2004-0975 If you are aware of other security-related bugs that have been fixed in the latest upstream version of openssl but are not fixed in the Debian package in Sarge, please contact the Debian security team or file a severity grave bug in the Debian BTS! best regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: tremendous size for .xsession-errors
Deephay wrote: I have a problem with my .xsession-errors file, this file will grow tremendous size each day (956 MB today, cleared yesterday, with a 4 GB size). The major source of errors comes from gnash (I am using the amd64 port which adobe flash player won't work), gnash is not matured yet and will produce lots of error messages with certain kind of flash movies which can be easily found on many websites, all of the messages are logged in the .xsession-errors (millions of lines!), mplayer and other applications produce some messages as well. So I want to know that is there a way to turn down the verbose level of the error log, or should I just add an entry in the crontab to clean it every ten minutes? TIA! If you never care about any errors that appear in that file, maybe making it a symlink to /dev/null would work? I don't know a way to reduce the verbosity offhand. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Heads up - mplayer now in Sid!
Hi all, Maybe no one has noticed yet... I thought people might be interested to know that mplayer is now available in Debian Sid: [EMAIL PROTECTED] (sid)[21]:~% apt-cache policy mplayer mplayer: Installed: (none) Candidate: 1.0~rc1~svn20199-1 Version table: 1.0~rc1~svn20199-1 0 500 http://ftp.us.debian.org sid/main Packages (Note that until #395307 is resolved, mozilla-mplayer will not be installable, however.) Many thanks to the maintainers, FTP-masters, mplayer upstream, and all the others who made this possible! Some possibly interesting relevant links: http://wiki.debian.org/MPlayer http://people.debian.org/~mjr/legal/mplayer.html http://lists.debian.org/debian-devel/2006/09/msg00915.html (n.b. I have absolutely nothing to do with mplayer in any way, I just thought this would be interesting news for debian-user) regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 signature.asc Description: OpenPGP digital signature
Re: apt bug 388708
Francesco Pietra wrote: Still not found apt_0.6.45.i386.deb should be apt_0.6.45_i386.deb (there's an underscore separating the version number and architecture, not a period as in RPMs) I found at snapshot.debian.net apt_0.645exp2.tar.gz Maybe snapshot.debian.net wasn't working when you checked it? I went to the front page, entered apt in the package name field and clicked details. That gives a table of package versions. In the second column (date), click on the date corresponding to version 0.6.45 (2006/07/28) and you get the set of apt packages from that date. Anyway, this should be a direct link to the .deb file you need: http://snapshot.debian.net/archive/2006/07/28/debian/pool/main/a/apt/apt_0.6.45_i386.deb Hoping this helps. best regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: modern Gnucash still needs libglib1.2?
Andrew Sackville-West wrote: On Tue, Oct 03, 2006 at 08:34:02AM -0400, Roberto C. Sanchez wrote: On Tue, Oct 03, 2006 at 06:12:42AM +0200, Tshepang Lekhonkhobe wrote: Hi, I just noticed that 'gnucash' depends on 'guile-g-wrap', which in turn depends on 'libglib1.2'. Is there anything available there not yet in 'libglib2.0' or has 'guile-g-wrap' not ported yet? IIRC, there is currently an effort under way to eliminate all the gtk1.2 and related dependencies from gnucash. It appears that the transition is not yet complete. AFAIK, from lurking in gnucash-devel and -user, that conversion was completed with the release of gnucash 2.0. That is gnucash has completed its gtk2 port. They (gnucash) however, do not maintain guile-g-wrap and can't be responsible for the dependencies of their dependencies... heh. Doesn't it cause segfaults if gnucash is linked (directly or indirectly) against two different ABI versions of libglib? Or does glib use symbol versioning to prevent that? -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal X-Windows Setup
Roberto C. Sanchez wrote: On Mon, Oct 02, 2006 at 12:24:10PM -0400, Stephen Yorke wrote: I am looking for a basic X-Windows setup...I have tried several things but I am just not getting it. Does anyone have some sort of package list which will get me a BASIC/MINIMAL X setup. Anytime I do anything with X-WINDOW-SERVER-* I get tons of dependencies and I do not want them. Mainly, I want an X setup which will allow me to view an X session log in with GDM and have Fluxbox as my window manager. Maybe I am going about it all wrong but I know I am going through hell trying to figure it out. You need at least x-window-system-core (if that is still the name of the metapackage). If you choose xdm, that will be better than gdm (which will pull in quite a few GNOME libs. Actually, x-window-system-core and x-window-system (in Sid Etch) are identical meta-packages that do nothing but Depend upon the xorg meta-package. The basic problem is that any sane installation of X should probably include the xbase-clients package, which includes rather useful things like startx and xauth scripts. Unfortunately it also includes a whole lot of junk that pulls in a lot of library dependencies. But until this package is split apart, I don't see any way around this problem. See for instance bug # 332521. I guess one could at least skip a few dependencies (like imake) by not installing the xorg metapackage, and installing only the following minimal list of packages and their dependencies: xserver-xorg xfonts-base xfonts-100dpi xbase-clients xkb-data xterm xdm (As Roberto says, xdm will pull in fewer packages than gdm. Choose xfonts-75dpi and/or xfonts-scalable as well as or instead of xfonts-100dpi if you wish. You can try not installing xbase-clients, but I don't know how well things will work.) ... and then uninstalling all the xserver-xorg-* packages except the ones appropriate to his system. Note, I haven't at all tested this suggestion, so caveat emptor. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
RE: Minimal X-Windows Setup
Stephen Yorke wrote: I have tried the EXACT listing you have here but it still puts on all the xserver-xorg-video-* stuff...I do not know exactly how much overhead these packages add but I would really rather using just the VESA driver and remove the rest...but a question about that...how do I force the usage of just the VESA driver? If you're using aptitude or apt-get at the command line, try putting xserver-xorg-video-vesa at the beginning of the list, like this: aptitude install xserver-xorg-video-vesa xserver-xorg [...] or else just install xserver-xorg-video-vesa FIRST and then all the other stuff in a second run. The trick here is that the -vesa package Provides xserver-xorg-video. So if you install xserver-xorg-video-vesa FIRST, then xserver-xorg's dependency on xserver-xorg-video-all | xserver-xorg-video is satisfied without APT having to pull in xserver-xorg-video-all (a meta-package that, as its name says, pulls in every video driver). Similarly you can specify a particular package that Provides xserver-xorg-input to prevent the xserver-xorg-input-all meta-package from pulling in all the input drivers. Just make sure you know exactly which one you need! And if you change your hardware in the future, remember that you may need to install new -video-something and/or -input-something packages to suit. best regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt bug 388708
Florian Kulzer wrote: Something like sudo dpkg -i /var/cache/apt/archives/apt_0.6.45_i386.deb will install the last working version from your package cache. (Change the i386 part to match your architecture.) If you do not have the 0.6.45 deb file in your package cache, then you can download it from the Debian snapshot site: http://snapshot.debian.net/archive/2006/07/28/debian/pool/main/a/apt/ I see that the buggy version of APT (0.6.46) has propagated into testing. It looks like there are plans to fix #388708 -- see the bug log page. A version 0.6.46.1exp1 has been uploaded to experimental with this bug fixed, and you may want to test that version of the package. I hope this bugfix can make it into Etch! regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal X-Windows Setup
Kevin Mark wrote: Hi Stephen, 'x-window-system-core' 'x-window-system', so install the former. Nope, they're exactly the same in Etch and Sid. Assuming there are no sarge, sarge security, or backports entries in your sources.list, try apt-cache show x-window-system-core | egrep '^(Depend|Suggest|Recomm)' apt-cache show x-window-system | egrep '^(Depend|Suggest|Recomm)' to see for yourself. Unfortunately there seems not to be an xorg-core meta-package. but it maybe the dpkg front-end that you are using that is adding 'suggests'. aptitude normally adds this, so you may see more things ^^ ITYM recommends being downloaded. So checkout the aptitude options to turn this off and it will install less things. best regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: aptitude update and pdiff files
Urs Thuermann wrote: For a couple of month I have seen an annoying change in behavior of aptitude update and apt-get update. Formerly, a couple of package files where downloaded which took a few seconds. Now I normally see that aptitude update downloads hundreds of files named like Get:282 2006-09-11-1318.15.pdiff [13.3kB] Get:283 2006-09-11-1318.15.pdiff [13.3kB] Get:284 2006-09-11-1318.15.pdiff [34.3kB] Get:285 2006-09-11-1318.15.pdiff [34.3kB] Get:286 2006-09-11-1318.15.pdiff [34.3kB] and this takes quite long. From the name I guess these are only differences two package files so that less data have to be transferred. However, the overhead of nearly 300 TCP connections causes this to take *much* longer on some systems than simply downloading the whole package file. Try putting this line in /etc/apt/apt.conf (creating the file if it doesn't already exist) to turn off pdiffs: Acquire::PDiffs false; I guess you can put it in a separate file in /etc/apt/apt.conf.d/ instead if that directory exists on your system. Not sure why this isn't in the apt.conf(5) man page; maybe because it was just added recently. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to list all the non-free packages installed
Stefan Monnier wrote: Every once in a while I need to install a non-free package, so my system does have some non-free packages installed. What command can I use to find out what packages are those (so I can remove them or replace them with free alternatives)? grep-status and grep-available are very powerful tools for answering this sort of question. To print out installed non-free packages: grep-status -FSection -r 'non-free/.*' -a \ -FStatus 'install ok installed' -sPackage -FSection tells it to search the Section field for the regexp (-r) non-free/.*; -a gives a logical and; the string after -FStatus tells it to look only for currently installed packages; and -sPackage tells it to output only the Package field of matching packages. As others have already said, vrms provides a shortcut for this particular query. Similarly for packages from `unstable, or `experimental'? This one can't be done by grep-status et al because information about the distribution from which the package came isn't provided in dpkg's status file. But apt-show-versions can help. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xlibs in etch
Someone anonymous wrote: I recently upgraded from sarge to etch, and I have been able to get almost everything working, but I tried to upgrade to a newer version of vlc, and it says: --- The following packages are BROKEN: vlc wxvlc 0 packages upgraded, 2 newly installed, 0 to remove and 0 not upgraded. Need to get 3237kB of archives. After unpacking 9294kB will be used. The following packages have unmet dependencies: wxvlc: Depends: xlibs ( 4.1.0) but it is not installable vlc: Depends: libdvbpsi2 which is a virtual package. Depends: libflac4 which is a virtual package. Depends: xlibs ( 4.1.0) but it is not installable --- When I try to install xlibs, it says: --- No candidate version found for xlibs --- I uninstalled the old version of vlc as part of my attempt to get the new one to install, and I can't even get the old (sarge) version to reinstall. There is no version of vlc in testing, and judging by the excuses page [1] there may not be for quite some time, due to its many dependencies and (even worse) an old licensing bug (#324978) [2]. Probably when you attempted to upgrade vlc, APT was just trying to re-install the old packages from Sarge. What does apt-cache policy vlc show you? xlibs is an old obsolete meta-package no longer existing in Etch or Sid, and there are packages in Etch/Sid that specifically Conflict against its dependency xlibs-data in order to force it off the system. [1] http://qa.debian.org/developer.php?excuse=vlc [2] http://bugs.debian.org/324978 Your options are basically (a) upgrade to Sid, (b) attempt to download and individually install the wxvlc and vlc debs from Sid on your Etch machine (this may not work due to newer library dependencies in the Sid packages), or (c) try to build the vlc source package from Sid on your Etch machine, and (if successful) install the resulting binary debs. Probably (b), if it works, or (c) if not, is the best idea since you already have everything else working well. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problem with: #include X11/Xlib.h
I wrote: Hmm, then try (re-)installing the latest version of the x-dev package from backports.org (version 6.9.0.dfsg.1-5bpo2). The Xfuncproto.h in there does include the _X_SENTINEL stuff. gustavo halperin wrote: Thank you, you are right. I fixed it now like you say. BTW: So, maybe I must to update all the packages from the Xorg family. For example if I have currently installed the xlibmesa-gl version 4.3.0.dfsg.1-14sarge1 mean that also this package I must upgrade ?? In principle you should be prevented from installing incompatible sets of packages by versioned dependencies. In practice, sometimes not all of them are noticed or deemed important. Apparently in this case no one realized that libx11-dev 6.9 required x-dev 6.9 :-( (This has been fixed in Sid and Etch, by the way, but I seem to recall that the monolithic XOrg backports are unlikely to be updated.) So it's usually safest to upgrade everything related at once. And if your answer is yes, there are any way to see which of this Xorg family package I have installed and not missing no one ?? Try something like this to find all the Xfree86 packages still on your system: COLUMNS=200 dpkg -l | grep '^ii' | grep '4\.3\.0\.dfsg\.1' (This only works because the Xfree86 packages in Sarge have a very distinctive version number.) The following is a little more tricky, and requires you to install the grep-dctrl package, but it's technically nicer: grep-status -FSource xfree86 --and -FStatus 'install ok installed' -s Package,Version (Search the dpkg status file for packages coming from source xfree86 that are currently installed, and print out the package names and versions.) And to find the backports you have installed from XOrg 6.9: grep-status -FSource xorg-x11 --and -FStatus 'install ok installed' -s Package,Version regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problem with: #include X11/Xlib.h
gustavo halperin wrote:
I have a code that include the file /usr/include/X11/Xlib.h, but when I try
to compile I receive the parser error, see below:
In file included from cube.cpp:32:
/usr/include/X11/Xlib.h: In function `char* XSetOMValues(_XOM*, ...)':
/usr/include/X11/Xlib.h:3578: error: parse error before `(' token
/usr/include/X11/Xlib.h: In function `char* XGetOMValues(_XOM*, ...)':
/usr/include/X11/Xlib.h:3583: error: parse error before `(' token
..
How I can resolve this problem,
I assume that you are using the Xlib.h from the Sid or Etch libx11-dev
package, since it has definitions for the functions XSetOMValues() and
XGetOMValues() near the line numbers printed in the compiler error.
(When reporting error messages in the future, please specify whether you
are using Sarge, Etch or Sid.)
It looks to me like the compiler is complaining about the _X_SENTINEL(0)
in this code in Xlib.h:
extern char *XSetOMValues(
XOM /* om */,
...
) _X_SENTINEL(0); // --- Line 3578
extern char *XGetOMValues(
XOM /* om */,
...
) _X_SENTINEL(0); // --- Line 3583
The _X_SENTINEL macro is defined in /usr/include/X11/Xfuncproto.h -- try
installing the latest version of the x11proto-core-dev package that
supplies this file. (Should be version 7.0.4-3 in Sid or Etch.)
Here is another possibility if the above suggestion doesn't work. The
definition of _X_SENTINEL includes the GNU C extension of the
__sentinel__ attribute, but only when __GNUC__ is defined and = 4.
Is it possible that you are using an older compiler that doesn't
recognize __sentinel__, but for some reason your build has defined
__GNUC__ to have a value of 4 or greater anyway?
Let the list know if these suggestions don't help. It might also be
helpful if you posted the problematic file cube.cpp.
regards,
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG: public key ID 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problem with: #include X11/Xlib.h
gustavo halperin wrote: Kevin B. McCarty wrote: The _X_SENTINEL macro is defined in /usr/include/X11/Xfuncproto.h -- try installing the latest version of the x11proto-core-dev package that supplies this file. (Should be version 7.0.4-3 in Sid or Etch.) OK, I'm using Debian Sarge version, but my X is Xorg from www.backports.org/backports.org/ version 6.9. . This unofficial APT repository unfortunately haven't the x11proto-core-dev package, but the file /usr/include/X11/Xfuncproto.h is there but without the _X_SENTINEL definition. Hmm, then try (re-)installing the latest version of the x-dev package from backports.org (version 6.9.0.dfsg.1-5bpo2). The Xfuncproto.h in there does include the _X_SENTINEL stuff. /* _X_SENTINEL BS */ #if defined(__GNUC__) (__GNUC__ = 4) # define _X_SENTINEL(x) __attribute__ ((__sentinel__(x))) # define _X_ATTRIBUTE_PRINTF(x,y) __attribute__((__format__(__printf__,x,y))) #else # define _X_SENTINEL(x) # define _X_ATTRIBUTE_PRINTF(x,y) #endif /* GNUC = 4 */ If you don't want to install the newer x-dev package from backports, then fixing Xlib.h by sticking this at the beginning seems fine to me -- this is the same as the block of code in my version of Xfuncproto.h in Sid. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: fluent like software
belahcene abdelkader wrote: The fluent software is to expensive for our small institute, At least (despite the price) it's available for GNU/Linux (I've installed it on Debian sarge machines for my group, and never had any difficulties), which is more than can be said for a lot of proprietary software! I am looking software which is around it, not necessary having all that properies. thanks for help best regards Maybe try asking on the debian-science list? regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: lyx-common 1.3.4-2 file has got a loop hole
formless void wrote: Has anyone noticed that this file is buggy? I'm using Xandros desktop and have tried to install it onto my system. However, it asks for lyx-qt and lyx-xforms. When I tried to install the lyx-qt and lyx-xforms, they were asking for lyx-common. I have noticed that when lyx-common asked for lyx-qt and lyx-xforms, it did not ask for version numbers which are unusual What program are you trying to install lyx-common with? If you are using Synaptic, aptitude, apt-get, or another APT front end, try installing one of lyx-qt or lyx-xforms, and lyx-common should get pulled in automatically. If you are using dpkg, you will need to tell it to install two of the .deb files at once: dpkg -i lyx-common*.deb lyx-qt*.deb (if you want the Qt interface) dpkg -i lyx-common*.deb lyx-xforms*.deb (xforms interface) It is highly recommended that you use some APT front-end, not try to install things with dpkg. Dpkg is a very low-level tool and not really suited for package management by itself. To answer the only question that you explicitly asked :-) you're correct that circular dependencies are usually considered to be a packaging bug. This one has already been reported at http://bugs.debian.org/341848 although it has not yet been fixed. If you have this kind of problem with other packages in the future, consider filing a similar bug report. P.S. Since you are using Xandros, you should probably go through their help forums first, not Debian's. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt-get install: unmet dependencies
[EMAIL PROTECTED] wrote: # aptitude try to delete fastforward and install sendmail-bin simultaneously (Reading database ... 46737 files and directories currently installed.) Unpacking sendmail-bin (from .../sendmail-bin_8.13.6-1_i386.deb) ... dpkg: error processing /var/cache/apt/archives/sendmail-bin_8.13.6-1_i386.deb (--unpack): trying to overwrite `/usr/sbin/newaliases', which is also in package fastforward abort-install of sendmail-bin From this error, it looks like sendmail-bin and fastforward packages include the same file /usr/sbin/newaliases. The best way to fix the situation: dpkg -i --force-overwrite /var/cache/apt/archives/sendmail-bin*.deb The --force-overwrite option to dpkg tells it to ignore the problem and overwrite fastforward's copy of the duplicate file. This should be enough to get you past the other errors. Generally it's a bug for packages to include the same file without taking special measures. The bug would be in either sendmail-bin or fastforward. I can't find fastforward listed in the Debian package database -- is it an unofficial package? -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Stupid shell script question about read [SOLVED]
I wrote:
Could someone tell me why the following works in zsh but not in
bash/posh/dash?
benjo[3]:~% echo foo bar baz | read a b c
benjo[4]:~% echo $a $b $c
foo bar baz
Thanks everyone for the enlightening answers! So just to summarize, the
problem is that the pipeline is treated as a subshell, and so the
variables $a $b and $c are defined within the subshell but not the
main shell.
These seem like the best solutions to my problem:
* Bash-specific (i.e. not POSIX-compliant) :
David Kirchner wrote:
I'm not sure of the POSIX way to use read in this manner, but I found
this on Google/A9:
http://linuxgazette.net/issue57/tag/1.html
The example he gives, with the () syntax, worked in bash, but not
in Debian or FreeBSD's /bin/sh.
Almut Behrens wrote:
In more recent bashes, the following should work as well
#!/bin/bash
read a b c `echo foo bar baz`
echo $a $b $c
The (here strings) are an extension of the here document syntax,
IOW, the string given after is supplied as stdin to the command.
* POSIX-compliant:
Bill Marcum wrote:
I think the POSIX way would be
echo foo bar baz | { read a b c; echo $a $b $c; }
Not too bad if what you want to do inside the { } braces is pretty short.
Almut Behrens wrote:
#!/bin/sh
eval `echo foo bar baz | (read a b c; echo a='$a';b='$b';c='$c' )`
echo $a $b $c
To get the variable's values from the subshell back to the main shell,
a shell code fragment is written on stdout, captured with backticks,
and then eval'ed in the main shell... (this is the moment when I
usually switch to some other scripting language -- if not before :)
Ugh. Does get the job done though. I guess one has to be a little
careful about escaping special characters in this case? Here's the
safest version I've found so far -- single quotes in the input have to
be special cased with the sed command, and the -r flag to read keeps
it from eating backslashes.
# set some variables to nightmarish values for testing purposes
d='ab\q' # literal value is ab\q
e='$d' # literal value is $d
f='ba'r' # literal value is 'ba'r'
# here's the meat of the code
result=`echo $d $e $f | sed s/'/\'\\\'\'/g | \
( read -r a b c; echo a='$a' ; b='$b' ; c='$c' )`
eval $result
# test that $a $b $c have the right values
echo $a $b $c
Tested on Sarge with zsh, bash, dash and posh :-)
Of course, replace this:
echo $d $e $f
with whatever is producing the output that needs to be put into $a $b $c
Personally, I'd rather constrain my script to work only with bash and
use or () operators than to write something like the above!
(N.B. every method above still needs the -r flag to read if the input
might contain backslashes.)
best regards,
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG: public key ID 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem
Olafur Jens Sigurdsson wrote: Þann 2006-03-01, 12:54:20 (-0500) skrifaði Kevin B. McCarty: Olafur Jens Sigurdsson wrote: Hi again, after trying a few stuff out I think that when I try to print using lp -dojs (or lpr -Pojs, where ojs is the name of the printer queue I set up for the printer) then that program tries to contact the server of the school, and not my local cups server where of couse there is no ojs spool. How do I tell lp (or lpr) to use the localhost cups server and not the one that the school provides? Have you tried changing the setting of ServerName in the file /etc/cups/client.conf (and then running /etc/init.d/cupsys restart) ? /usr/bin/lpr is the version of lpr in the cupsys-bsd package, right? (Run dpkg -S /usr/bin/lpr to check) Kevin you master! Thank you a billion times, now it works like a charme! I had looked throug the cupsd.conf file a few times but by some reason I overlooked the client.conf file :-) But then I have the question why on earth did it choose my school CUPS deamon instead of the localhost to put as a default server? Should I file this as a bug? That is rather strange. I'm surprised that localhost isn't the default. Maybe CUPS checks for print servers broadcasting their availability on the net, and only falls back to localhost if none are found? I'm no CUPS expert, though, so maybe someone who knows more about it could correct me and suggest whether this is really a bug. best regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Stupid shell script question about read
Hi list, Could someone tell me why the following works in zsh but not in bash/posh/dash? benjo[3]:~% echo foo bar baz | read a b c benjo[4]:~% echo $a $b $c foo bar baz If I try the same with bash (or other sh-compatible shells), the variables $a $b and $c are unset. From the bash man page: read [-ers] [-u fd] [-t timeout] [-a aname] [-p prompt] [-n nchars] [-d delim] [name ...] One line is read from the standard input, or from the file descriptor fd supplied as an argument to the -u option, and the first word is assigned to the first name, the second word to the second name, and so on, with leftover words and their interven‐ ing separators assigned to the last name. So read claims to read from the standard input, but it doesn't actually seem to happen when a pipe is involved. Posh and dash behave like bash in this respect, so I guess that this is not a bug, and that what zsh does is actually an extension. So, what is the correct POSIX-compatible way to get read to work as I want? Thanks in advance, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing problem
Olafur Jens Sigurdsson wrote: Þann 2006-02-13, 13:07:46 (+) skrifaði Olafur Jens Sigurdsson: Hi I have a problem printing to a network printer here at school. The spool that the school provides is no good so I am trying to create my own on my computer and talk directly to the printer (then I allso have better control over my printing, I like that more). Hi again, after trying a few stuff out I think that when I try to print using lp -dojs (or lpr -Pojs, where ojs is the name of the printer queue I set up for the printer) then that program tries to contact the server of the school, and not my local cups server where of couse there is no ojs spool. How do I tell lp (or lpr) to use the localhost cups server and not the one that the school provides? Have you tried changing the setting of ServerName in the file /etc/cups/client.conf (and then running /etc/init.d/cupsys restart) ? /usr/bin/lpr is the version of lpr in the cupsys-bsd package, right? (Run dpkg -S /usr/bin/lpr to check) Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bug with the mailing list archive
Cedric BRINER wrote: I'm facing some uncomprehension with aptitude vs apt-get. And I remembered that there was a thread about this on the mailing list archive: http://lists.debian.org/debian-user/2005/02/msg00372.html but when I click on: Index(es): * Date * Thread --- this one I don't find this thread !!! As already pointed out, this seems to be a bug in the mail archiving software. In the meantime, the URL you want is the following one, since the thread starts on the second page of the February 2005 archive: http://lists.debian.org/debian-user/2005/02/thrd2.html#00372 regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sometimes nothing happens when printing to network printers
Robert Rothenberg wrote: I've been having trouble printing to network printers from KDE and Gnome applications lately, using CUPS or LP. No errors show up, and as far as the applications are concerned, the documents printed. But the documents are never actually sent to the printers. There's no problem with the printers. They're mostly HP LaserJets. I've checked with the network admins and the other users in the building. They've had no such problems. Killing cupds and restarting it sometimes works. Rebooting my machine always works. But that's a Windows-style kluge--I'd like to actually fix the problem. Does anyone have any idea what could be wrong, and how to fix it? I had a very similar problem (running Debian Sarge with CUPS) not too long ago. I was able to print to a network printer as root but not as my normal user. But I could still print test pages from the CUPS HTTP interface on localhost:631. I eventually ran lpr under strace and found out that CUPS was looking at a dot file in my home directory, which apparently had incorrect settings or bad syntax or something. Deleting the file fixed the problem. If I remember correctly it was called .lpoptions. Maybe this will help you? Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: C++: Math function round() does not compile - why?
Oliver Elphick wrote: On Tue, 2006-02-14 at 14:11 +0100, Jan C. Nordholz wrote: Hi Oliver, g++-2.95 -c -pipe -DQWS -fexceptions -fno-rtti -Wall -W -g -std=c99 -I/opt/Qtopia/include -o orderimpl.o orderimpl.cpp orderimpl.cpp: In method `void OrderImpl::calculate_line(int)': orderimpl.cpp:84: implicit declaration of function `int round(...)' this sounds like math.h doesn't give you the prototype for round(). Try to run math.h through the preprocessor alone (g++-2.95 -std=c99 -E /usr/include/math.h), and grep for round. I don't get the definition with g++-3.3 alone, but with g++-3.3 -std=c99 I do... YMMV. I've got no g++-2.95 here to continue the tests. with g++-2.95, this gives no prototypes for round(). So it appears that -std=c99 does not work. Is there some way to achieve this for 2.95? (This is now academic, since I have written my own version for this application.) Maybe try defining _ISOC99_SOURCE in your source code before #including math.h? (You might need to put the #define before _any_ #include statement and maybe specifically #include features.h.) For instance: #define _ISOC99_SOURCE #include features.h #include math.h If I've put together the clues correctly, setting this macro is equivalent to the -std=c99 option. See the comment at the top of /usr/include/features.h (after the copyright blurb) for more information. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Inverse and Forward Search : This is not what I expected in TeTeX
ahmet nurlu wrote: I was using MikTeX and YAP previewer for my tex files in Windows. My editor of choice was WinEdt. Recently, I switced to Debian/sarge. I was expecting the same functionality of inverse and forward search in TeTeX as in MikTeX. However what I observed in TeTeX is different in MikTeX. [snip] In my Debian/Sarge system, I use TeTeX(ver:3.0.13) and Xdvi(ver:22.84.9) Hi Ahmet, The version of TeTeX in Sarge is 2.0.2, and the version of xdvi in Sarge is 22.40v. Are you using a backported Debian package (maybe from backports.org), or a locally compiled version from TeTeX upstream? If you are using a Debian package of some sort, I recommend that you send the exact email you sent to debian-user as a bug report against one of the TeTeX packages, maybe tetex-bin. (Set the severity to wishlist or minor since the bug isn't incorrect behavior, it just isn't as helpful as it could be.) This way the TeTeX maintainers will be much more likely to see your message and hopefully fix the problem. If you are using a backport from backports.org or some other unofficial source of .debs, be sure to say so in the bug report. If on the other hand you have a locally compiled version from upstream, probably you should send the bug report directly to the upstream TeTeX authors. Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: install krita. Broken pipe. Do I need a plumber?
Hi Scott, Scott wrote: But seriously folks... What's a broken pipe? A pipe is this keyboard symbol: | When you want to use the output of one command as the input of another command, you pipe it, like this: ls -l | sort A broken pipe happens when the command reading input dies unexpectedly, for instance (supposing I was a very fast typist) if I meanwhile ran killall -9 sort in another xterm before ls was finished. dpkg: error processing /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb (--unpack): trying to overwrite `/usr/share/mimelnk/image/x-raw.desktop', which is also in package digikam dpkg-deb: subprocess paste killed by signal ( *Broken pipe* ) Errors were encountered while processing: /var/cache/apt/archives/krita_1%3a1.4.90.1-1_i386.deb E: Sub-process /usr/bin/dpkg returned an error code (1) In this case, the program that dies is dpkg, and the reason it dies is stated on the second line I quoted above: the already-installed digikam package and the new krita package that you are trying to install contain the same file. This is not supposed to happen, and is therefore a bug in one of the two packages. Since your krita package came from experimental, it's probably the one at fault. Please file a bug against it. In the meantime, to work around the problem, you can run dpkg -i --force-overwrite /var/cache/apt/archives/krita_*.deb to tell dpkg that you want it to overwrite the common file instead of griping about it. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gnome firefox thunderbird
Adam Hardy wrote: I keep seeing messages about Firefox and Thunderbird but it's never about this and I can't find any info about how to control this in Gnome: I want to click on a link in Thunderbird emails, and see the Firefox browser pop up a new browser window in the foreground. Currently the link opens in an existing Firefox window AND in the background, so I have to go looking for it and then work out whether I want to go back to the page the browser window showed before. Could be simple, but could be impossible - anybody know? Simple if you know exactly where to look. In Firefox, open the preferences (Edit-Preferences), select the Advanced tab, and click the + to the left of the Tabbed Browsing heading. This should open up the following set of radio buttons: Open links from other applications in: (o) a new window ( ) a new tab in the most recent window ( ) the most recent tab/window Sounds like you want the first of these options. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where to file a package-dependency bug?
Since Andrew Vaughan already answered your first question, I'll just get this one: The Wanderer wrote: (As long as I'm here anyway, a minor question on completely another topic: since I'm told that apt-get and apt-cache are deprecated, what is the intended successor - in aptitude or similar - to 'apt-cache policy'? The aptitude man page tells me nothing useful. I find 'apt-cache policy' more routinely useful than just about any other apt command, and would be very disturbed to lose its functionality.) As far as I know, apt-get / apt-cache are not deprecated, they just have fewer features than aptitude / synaptic / etc. and probably won't have many new features added. For instance, apt-get doesn't keep track of which packages were installed automatically for dependency reasons, doesn't install Recommends, etc. Apt-get is sort of the reference implementation of APT. On the other hand, aptitude doesn't have the functionality of (for instance) apt-get source or apt-get build-dep. I think you can continue using apt-cache policy without any worry that it will go away soon. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where to file a package-dependency bug?
The Wanderer wrote: In current sid, the x-winow-system-core package depends on both xfonts-75dpi and xfonts-100dpi. The package description for xfonts-75dpi includes the note: == This package and xfonts-100dpi provide the same set of fonts, rendered at different resolutions; only one or the other is necessary, but both may be installed. == xfonts-100dpi contains a mirror of the same sentence. This is either a bug in x-window-system-core (AND dependency rather than OR dependency) or a bug in both xfonts-75dpi and xfonts-100dpi (incorrect description). Under which package should I file the bug report? I'd file it against x-window-system-core (with severity minor or wishlist, since nothing is really broken by the existing situation). The xfonts-*dpi descriptions are true enough, you only need one of the sets of fonts installed. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where are GTK devel packages?
Thomas H. George wrote: I am trying to run fakeroot debian/rules binary to install mplayer. The program aborts with the message, the GUI requires GTK devel packages (which were not found) My setup is Sarge with XWindows started with gdm but I probably do not have the whole Gnome package installed as I normally use icewm. I don't need a lot of bells and whistles as the programs I use: openoffice, xfig, gimp, cups, pysol and mozilla work fine but I would like to add mplayer-plugin to watch TV clips. You can see exactly what packages are required to build mplayer by looking for the line beginning with Build-Depends: (and maybe also a line beginning with Build-Depends-Indep:) in the mplayer-version/debian/control file. I'd guess you at least need to install libgtk2.0-dev. But are you aware that Christian Marillat (a Debian developer) has already created unofficial Debian packages of mplayer? As root, add the following lines to your /etc/apt/sources.list: deb ftp://ftp.nerim.net/debian-marillat/ sarge main deb-src ftp://ftp.nerim.net/debian-marillat/ sid main then run apt-get update ; apt-get install mplayer-386 mozilla-mplayer and you should be all set. (Note: you can install mplayer-586 or mplayer-k6 instead of mplayer-386 to get a version more optimized for your system, depending on your CPU.) regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where to file a package-dependency bug?
The Wanderer wrote: (In the process, I've discovered that reportbug does not permit a bug to be reported on a dependency package, but only on the packages it depends on. This is probably another bug...) I didn't believe this at first but it's true! I think it's planned behavior, though. Looking into the reportbug code, you can force reportbug to submit a bug against the dependency package when the mode is set to advanced or expert. (This can be changed in ~/.reportbugrc) There are other sneakier ways, too, although the fact that they work is probably unintentional. By the way, while checking into this, I found that the reportbug in sid actually crashes when one runs reportbug x-window-system-core. Just filed as #345699. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Any interest in Debian packages of Geant 4?
Hi all, Here is a question for other HEP / nuclear physicists on the lists. Please follow up only to debian-science. Needing to do some Monte Carlo simulations for the last bit of work on my thesis, I was once again reminded how much of a pain compiling Geant4 from source is. As a result I've spent some time over the last few days Debianizing it. The packages are still very rough so I'm not making them public (yet), but I am curious to know whether there is enough interest in Geant 4 on Debian to make it worthwhile to eventually have it in the official archive. The total installed size of the packages is 120 MB, not including the data files [1]. I would be reluctant to put this huge chunk of code in the archive unless there is a demand for it. (Plus *another* 100 MB for James Ferrando's unofficial package of the CLHEP library, although this could be reduced to 50 MB by including only the merged libCLHEP.so and not the component libraries libCLHEP-whatever.so.) The Geant 4 packages could be cut down from 120 to 97 MB by only distributing the static libraries (upstream doesn't version the shared libraries so they aren't very safe to use), but still! For comparison, in Sid, a reasonably complete OpenOffice.org 2 installation from Debian packages is 200 MB, and tetex-bin + tetex-base + tetex-extra is 127 MB -- both of these have a lot more users than Geant 4 would. If there isn't enough desire for Geant 4 in Debian's official archive, I might be able to keep unofficial .debs around in a separate repository when they are more polished. I'll post occasional updates on the progress of this to debian-science. [1] The data files total 473 MB when uncompressed, and are probably non-free. Fortunately they are optional. I would definitely not upload them into Debian, but instead would create a geant4-data-installer package in contrib that downloaded them from CERN's web site - let CERN pay for the bandwidth! Happy Holidays to all, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: illegal access using ssh
Amish Rughoonundon wrote: I was looking at my auth.log file and I saw a bunch of these things: Nov 28 16:22:41 localhost sshd[11363]: Illegal user nobody from 212.0.148.2 I was wondering if there is a way to filter the ip allowed to access the computer and allow only 1 ip (mine) to do so. Thanks a lot, Sure! You could set up a firewall (iptables), but tcpwrappers is probably easier. On the server, put this line into /etc/hosts.allow : sshd: 1.2.3.4 (of course, replace 1.2.3.4 with the IP of the computer you want to permit to log in) and this line into /etc/hosts.deny : sshd: ALL When a new connection attempt is made, the ssh daemon will first check hosts.allow to see if the host is permitted. Hence your client system will be allowed to log in. If the host isn't listed in hosts.allow, sshd next checks hosts.deny to see if the host is forbidden. Since that file has the entry ALL, all other hosts will be rejected. The man page for hosts_access(5) has more information. By the way, you can replace the daemon name sshd with ALL in both config files, to make this behavior occur for all daemons on the server that use tcpwrappers. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Warning: Ending maintenance of my Mozilla/Firefox/Thunderbird backports to Sarge
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Dear all, Since the Debian Security Team is doing a good job of keeping up with security updates for the Mozilla family in Sarge, much better than I can, I am ending maintenance of my Mozilla backports to Sarge located at http://borex.princeton.edu/~kmccarty/ . If you have any of my backports installed, please install the packages from the Sarge security updates instead. Since (due to poor planning on my part) the version numbers of my backports are higher than those of the Sarge security updates, this sidegrade is slightly non-trivial. The procedure I outline below should ensure a safe transition. 1) Log in as (or su to) the root account 2) Ensure that your /etc/apt/sources.list contains a line for Sarge security updates. The required line looks like this: deb http://security.debian.org/ sarge/updates main contrib non-free (Remove contrib and/or non-free from the line if you don't want them.) 3) Run aptitude update 4) If you do not already have the apt-show-versions package installed, install it by running the command aptitude install apt-show-versions 5) Run the command apt-show-versions | grep kmccarty This step lists the packages for which you have my backports installed (one per line). If there is no output, you need go no further. 6) Change directories into a temporary directory. For every package listed in step 5, run the command aptitude download package/sarge (where package is the package name). If that doesn't work, try aptitude download package/stable instead. Whichever of these commands works will download the .deb file to your current directory. 7) Run the command dpkg -i *.deb to install all the downloaded files. 8) To ensure that the packaging tools completely forget about my backports, run this command: aptitude update apt-show-versions -i I have already made my backported packages inaccessible, so you will now be on track to install additional security updates from Debian down the road. My apologies for the inconvenience. regards, - -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFDfNNDfYxAIk+Dx1ERAtyaAKCb6kybafeMLrQVy79ifkiFHfgU8ACdEBU5 41SDiQCnU8Ro5HW8RYo3tjg= =Sx8j -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xserver-xfree86 and xserver-xorg problem
James Vahn wrote: Kevin B. McCarty wrote: For now is right. This is probably not such a good idea in the long run. If you are running etch or sid (which I assume you are, otherwise you wouldn't need snapshot.debian.net to get XFree86) then you won't get any security fixes for XFree86. Better to add the following to sources.list instead: Are you sure? One came through just the other day. Snapshot: 4.3.0.dfsg.1-14sarge1 Security: 4.3.0.dfsg.1-14sarge1 Hmm, looks like snapshot.d.n also includes stable security updates. I wasn't aware of that; my apologies. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Downgrading a system to a particular distribution
Alf wrote: Now my system is no very stable, and some applications crash now and then. Because of this, i'd like to return to a more stable state. Is there a way to uninstall all packages from unstable branch? I'd like to get my system back to testing, say, or stable. Downgrading in general isn't supported by Debian. If you want to try anyway, though, here are some guidelines. I've done this before to take a computer with an old version of Sid down to Sarge. I did this before the XFree86 - Xorg transition happened; things will be more difficult now. - Remove all lines referencing sid or unstable from /etc/apt/sources.list. If you want to downgrade all the way to stable, also remove all lines referencing etch or testing. - Run apt-get update - Install the package apt-show-versions. Run the following command, and it will tell you all the packages installed on your system that are newer than the version in testing (or stable, depending on how far you want to downgrade): apt-show-versions | grep 'newer than version in archive' - To downgrade something that shows up in this list, run apt-get install package/stable. For instance, to downgrade the tar package, you would do apt-get install tar/stable. Note that you may have to replace stable with sarge depending on which you refer to it as in your /etc/apt/sources.list file. If you only want to downgrade to testing, use etch or testing instead, of course. - Downgrade applications first, then libraries. In general, if package A is an application depending on libraries B and C, and library B depends on library C, then you want to downgrade them in the order A,B,C. This is because an older application can use a newer library (so long as the soname has not changed), but a newer application may not be able to use a library older than the version against which it was compiled. - Sometimes you may have to downgrade several packages at once (e.g. apt-get install libgtk2.0-0/stable libgtk2.0-dev/stable) in order to work around versioned dependencies. You'll know this is necessary if APT wants to remove packages or complains about packages having unmet dependencies. - Some packages may exist in sid but not in testing/stable. You can find them with apt-show-versions | grep 'No available version in archive' If you don't want them, try apt-get removeing them to see what happens. If APT wants to remove a bunch of other things, let it -- keep a list of what gets removed, then reinstall it from stable. - If at any point you run into a problem like maintainer scripts failing in non-trivial ways, or segfaults, or something like that, back up all your data and reinstall from scratch -- it's probably easier. Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how to add java support to mozilla in woody
Roberto C. Sanchez wrote: On Wed, Sep 28, 2005 at 07:57:07AM +0800, [EMAIL PROTECTED] wrote: Thanks in advance! Get java-package from Sarge, download a suitable JRE or JVM (from Sun, IBM, Blackdown, etc) and create a .deb. Install it and it will install the correct mozilla plugin. I don't know if that will work - he was asking about Woody. I seem to remember that the Mozilla plugin ABI changed between Woody and Sarge because Mozilla in Woody was compiled with g++ 2.95 and that in Sarge, with g++ 3.3. As a result, old Java plugins didn't work with newer Mozillas, and I bet the same is true the other way around. I don't have any better suggestions, though, so he may as well try it. Maybe he can find a sufficiently old JRE? regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how to add java support to mozilla in woody
[EMAIL PROTECTED] wrote: Thanks in advance! What version of Mozilla are you running? The version that was included with Woody, 1.0, is horribly outdated and has never received security updates. Even the ones available at backports.org are nine months old and many new security flaws in Mozilla have been found and fixed since then. For a desktop system, I suggest you upgrade to Sarge ASAP. I'm not sure it's even possible to find a Java plugin compatible with a Mozilla built on Woody - see my other email in this thread. Once you upgrade to Sarge, it's not difficult to obtain a Java .deb via java-package. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xserver-xfree86 and xserver-xorg problem
James Vahn wrote: cc wrote: 'apt-get install xserver-xfree86 ' will install 'xserver-xorg' instead of xserver-xfree86, whats the matter ? i want to install xfree86, not xorg me too.. so i put this line in sources.list : deb http://snapshot.debian.net/archive pool xfree86 and all is well again. For now. For now is right. This is probably not such a good idea in the long run. If you are running etch or sid (which I assume you are, otherwise you wouldn't need snapshot.debian.net to get XFree86) then you won't get any security fixes for XFree86. Better to add the following to sources.list instead: deb http://security.debian.org/ sarge/updates main contrib non-free This way your XFree86 packages will at least get security fixes for the lifetime of Sarge. There was one just a week ago: http://www.debian.org/security/2005/dsa-816 regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Man-DB is crazy
Bob Proulx wrote: Kevin B. McCarty wrote: Once a day I get the following email message from the man-db cron job: /etc/cron.daily/man-db: mandb: warning: /usr/share/man/man1/x-terminal-emulator.1.gz is a dangling symlink It would be a lot easier to use the -L option and do it in one command. ls -lL /usr/share/man/man1/x-terminal-emulator.1.gz -rw-r--r-- 1 root root 1369 2005-03-10 14:55 /usr/share/man/man1/x-terminal-emulator.1.gz Thanks, that's a useful option to ls! I learn something every day :-) In spite of your evidence to the contrary I believe that the link in question must really be dangling. I just can't believe otherwise because of the series of symlinks. I feel certain there was a mistake in there somewhere. Unfortunately I get the exact same thing you did: benjo[3]:~% ls -lL /usr/share/man/man1/x-terminal-emulator.1.gz -rw-r--r-- 1 root root 1369 2005-03-10 16:55 /usr/share/man/man1/x-terminal-emulator.1.gz (I take it you're using gnome-terminal too.) So I'm still not convinced of man-db's sanity. Although I didn't get such an error message in my email this morning, so maybe the situation has resolved itself somehow (maybe I installed another package that caused man-db to rebuild its database???) I hate when computers act non-deterministic. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Request: Anyone able to test printing with CUPS on *woody*?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi folks, I've come across a strange behavior printing a certain postscript file with the CUPS server in Sarge. (See bug # 329207 for further details.) I would like to request someone who is still running oldstable - Woody - - to see whether or not two files will print correctly on the Woody version of CUPS. Specifically, I need someone who is using the generic Foomatic/Postscript driver on their CUPS server. If you still have a Woody CUPS server with that driver and you are willing to run the test (just printing a couple of one-page PostScript files), please email me privately for further details. Thanks, - -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFDMWSjfYxAIk+Dx1ERAhvHAJ9dsClt0aD2CFAus+iaur1JfjND7ACgqBeV etGfgYL5BovF+hv3p6O7YBs= =mrDT -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Man-DB is crazy
David E. Fox wrote: On Tue, 13 Sep 2005 09:44:15 -0400 Kevin B. McCarty [EMAIL PROTECTED] wrote: benjo[1]:~% ls -l /usr/share/man/man1/x-terminal-emulator.1.gz lrwxrwxrwx 1 root root 42 2005-09-09 10:05 /usr/share/man/man1/x-terminal-emulator.1.gz - /etc/alternatives/x-terminal-emulator.1.gz You might need to follow the links. This happens occasionally over here, and I wondered about it as well. Try looking in /etc/ alternatives, you may find that the file x-terminal-emulator.1.gz points nowhere, or is nonexistant. Or, it points back to a manpage in / usr/share/man/man1 that does not exist. You snipped the part of my email where I did follow the links :-) I eventually got to the OK-looking existing file /usr/share/man/man1/gnome-terminal.1.gz . Thanks, though, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Man-DB is crazy
Hi list, Once a day I get the following email message from the man-db cron job: /etc/cron.daily/man-db: mandb: warning: /usr/share/man/man1/x-terminal-emulator.1.gz is a dangling symlink This in spite of all the evidence: benjo[1]:~% ls -l /usr/share/man/man1/x-terminal-emulator.1.gz lrwxrwxrwx 1 root root 42 2005-09-09 10:05 /usr/share/man/man1/x-terminal-emulator.1.gz - /etc/alternatives/x-terminal-emulator.1.gz benjo[2]:~% ls -l /etc/alternatives/x-terminal-emulator.1.gz lrwxrwxrwx 1 root root 39 2005-08-30 10:38 /etc/alternatives/x-terminal-emulator.1.gz - /usr/share/man/man1/gnome-terminal.1.gz benjo[3]:~% ls -l /usr/share/man/man1/gnome-terminal.1.gz -rw-r--r-- 1 root root 1369 2005-03-10 16:55 /usr/share/man/man1/gnome-terminal.1.gz Anyone have any idea what man-db's problem is and how to fix it? I'm running Debian Sarge, and as far as I remember haven't modified the scripts in /etc/cron.daily. Running /etc/cron.daily/man-db directly (as root) produces no output. Please CC replies to me as I'm not subscribed. Thanks, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debconf error
L.V.Gandhi wrote: While configuring debconf I opted for kde. But when I configure any package now I get the following error. debconf: unable to initialize frontend: Kde debconf: (Can't locate Qt.pm in @INC (@INC contains: /etc/perl /usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.8 /usr/share/perl/5.8 /usr/local/lib/site_perl .) at /usr/share/perl5/Debconf/FrontEnd/Kde/Wizard.pm line 7, line 1.) debconf: falling back to frontend: Dialog From where to get Qt.pm. I am running debian sarge in dell inspiron 600m. Hello, For answers to this sort of question (What package contains file X?) there are two good sources of information. I wish they were more widely known so the frequency of this question was reduced. The first is http://packages.debian.org - go there, scroll to the bottom of the page, and fill out the form Search the contents of packages. Make sure to select your correct distribution and architecture. Going here, for i386 unstable, I find that Qt.pm is in the libqt-perl package. The second is to install the apt-file package, run apt-file update as root, and then (can be as a normal user) run apt-file search filename with filename being the name of the desired file. A third method is to run dpkg -S filename but this only works if the package containing filename is already installed on your computer. Note that none of these methods will be able to find files (usually in /var or /etc) that are created dynamically upon package installation; /etc/ssh/ssh_host_rsa_key.pub is an example of such a file. You may be able to find creators of some of these files by grepping for the filename in /var/lib/dpkg/info where the package installation and removal scripts are stored. I believe there is work on dpkg ongoing that will eventually fix this deficiency. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Troubling security news for sarge users of mozilla, firefox, thunderbird...
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ralph Katz wrote: Troubling security news for sarge users of mozilla, firefox, and thunderbird is unfolding on debian-security: [snip] Alexander Sack has backported thunderbird 1.06: http://www.asoftsite.org/s9y/archives/84-Sarge-Backport-Release-Thunderbird-1.0.6-1-and-enigmail-0.92.0-1-uploaded.html Yet I'm wondering how other sarge users are going to address the security concerns? Backport, install from Mozilla tarball, upgrade to unstable, change to kmail/konqueror? I backported the newest versions of mozilla, firefox, thunderbird and enigmail to Sarge. (Sorry, no galeon or epiphany since I don't use them.) Anyone who wants them and is willing to trust me (shouldn't be too scary since I'm in the NM queue :-) can get the packages here: deb http://borex.princeton.edu/~kmccarty/ sarge main deb-src http://borex.princeton.edu/~kmccarty/ sarge main Please note I can't promise that I will keep up with future new releases in Sid. I would much rather these packages live at backports.org but it doesn't seem to have them available yet. And as usual there is no warranty; if something breaks you get to keep both pieces. Two notes: 1) I jumped the gun and backported Mozilla 1.7.11 even though the latest version available in Sid is still 1.7.10. 2) Regarding Enigmail, only the mozilla-thunderbird-enigmail package will work; the mozilla-enigmail package doesn't yet support the new Mozilla version. regards, - -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFC8jCufYxAIk+Dx1ERAnL3AJ9o03NiDCvcXkuF+/THzC0UZiBQSQCgjoeu le/0L1scOCJUYyDrow2lAvI= =dST1 -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: x-windows-system package
Hi Paolo, Paolo Pantaleo wrote: I just installed the x-windows-system package, and i found some unpleasant things: 1)xdm is installed and no question is done about it during configuration This is by design. x-window-system depends upon all programs that are traditionally part of an X release (including xdm), even binaries that many people don't find useful. You probably only want to install x-window-system-core. 2)the keyboard doesn't work properly as said in a previous message ALT GR + key doesn't work, there seems to be some problems with xkb 3)the module xtt is not included by default in the configuration process, it must be selected by hand, if not selected the X server wont start, telling it cannot find the fixed font. These I'm afraid I can't answer. If these are bugs (or better suggestions i want to make) how can i make my ideas reach the right people (who maintains the X system, or whatever)? You can file a bug (severity wishlist if it's a suggestion) against one of the X-related packages, or you could directly mail debian-x@lists.debian.org which is where all X-related bug reports go anyway. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Help/Question: Can't get xlibs package to install all of it's files
Lance Rushing wrote: I'm trying to get /etc/X11/xkb/symbols/altwin (an other files) to install from the xlibs package. If I: # dpkg --contents xlibs_4.3.0.dfsg.1-14_all.deb I get of list of the contents and it shows the files I am interested in. But if I: # dpkg --install xlibs_4.3.0.dfsg.1-14_all.deb the files don't get installed. I've extracted the files and the controls from this package to check if there is a preinst script or something. But there are none. The only control files are: 'conffiles', 'list', and 'md5sums'. Hi Lance, This is probably because most files under /etc are conffiles, and dpkg will not install a missing conffile by default if you reinstall an already-existing package. (It assumes that if you previously deleted the conffile, you must have done it for a good reason.) The --force-confmiss option to dpkg will probably do what you want: dpkg --force-confmiss --install xlibs_4.3.0.dfsg.1-14_all.deb regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: undefined reference to 'something'
Yinghong Zhou wrote:
I also used rtc() in my program. The linker doesn't report error on this
one. Following is how I declare the functions in my program:
extern C {
void wssmp( int* N, int* IA, int* JA, double* AVALS, double* DIAG,
int* PERM, int* INVP, double* B, int *LDB,
int* NRHS, double* AUX, int* NAUX, int* MRP, int* IPARM,
double* DPARM );
double rtc();
void wsetmaxthrds(int*);
void wsetnobigmal();
}
With functions having this sort of signature, I am guessing that the
library was written in FORTRAN?
Here is how my Makefile looks:
all:
g++ -o wgsmp wgsmp_ex1.cpp -L../lib/libgoto_opteron-64-r0.99-3.so
-L../lib -lwsmp64 -lpthread -Wno-deprecated
The author said it may have something to do with the compiler inserting
trailing underscores, etc.
G77 for reasons unknown likes to add an underscore to the end of the
names of functions defined in FORTRAN. So in your code, you probably
need to declare and use wssmp_() instead of wssmp(), etc., everywhere.
To avoid the need for the ugly underscore, you can
#define wssmp wssmp_
(etc.) at the top of the file where you declare the functions.
regards,
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG: public key ID 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: undefined reference to 'something'
Yinghong Zhou wrote: Thank you very much for your reply! Your are absolute right. The wsmp was written by Fortran. After I added the underscore, I have new undefined reference coming up. Here is a few examples of them: : undefined reference to `ftn_i_jishft' ./lib/libwsmp64.a(torder.o)(.text+0x11f9): In function `fllagbgref5_': : undefined reference to `ftn_i_jishft' ./lib/libwsmp64.a(torder.o)(.data+0x10): In function `.C4_283': : undefined reference to `pgf90_compiled' [snip] Could you please tell me how to solve those errors. They are not functions in wsmp, but called by wsmp. Try adding the FORTRAN library at the end of the linking command. G++ doesn't know to look for it if you don't tell it. So if the wsmp64 library was originally compiled by G77, you would need to add -lg2c at the end of this command: g++ -o wgsmp wgsmp_ex1.cpp -L../lib/libgoto_opteron-64-r0.99-3.so -L../lib -lwsmp64 -lpthread -Wno-deprecated However it looks like your libwsmp64 was instead originally compiled with pgf90, judging by the undefined pgf90_compiled references. I don't know offhand what magic incantations are needed to make G++ link against pgf90 libraries. Section 6.13 of this web page I found via Google looks like it might be helpful: http://www.dcsc.sdu.dk/docs/PGI/pgiws_ug/pgiug_07.htm#Heading97 The table of contents for the page is here: http://www.dcsc.sdu.dk/docs/PGI/pgiws_ug/pgiug_.htm Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: big trouble in upgrading from Debian 3.0 to 3.1
Paolo Alexis Falcone wrote: Don't use dselect for distribution upgrades. Use apt-get directly, e.g. apt-get update apt-get dist-upgrade As others have pointed out already, this isn't recommended by the release notes either. The suggested upgrade path is to use the version of aptitude from Sarge: http://www.debian.org/releases/sarge/i386/release-notes/ch-upgrading.html as aptitude is smarter about dependencies than apt-get. (It also has the nice feature of remembering which packages were installed only in order to fulfill dependencies.) regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Am I hacked?
Michal Sedlak wrote:
I am nearly sure that my server was hacked, but I want to be sure. Can
anybody say me if it is true.
Here is tiger script output. Do you have any ideas how to repair it {no mkfs
funny stuff please}
There are some line interesting. I have one for every critical system command
like {login, su, etc}
--WARN-- [sig004w] None of the following versions of /bin/netstat
(-rwxr-xr-x) matched the /bin/netstat on this machine.
and something like this for some kernel modules
--FAIL-- [lin005f] Installed file `/lib/modules/2.6.8-2-386/modules.symbols'
checksum differs from installed package 'kernel-image-2.6.8-2-386'.
Could you try running chkrootkit and send the results to this list? A
Debian package exists, but you may want to install it manually (install
the package to another machine and copy over the files) if you don't
know whether apt-get et al. have been trojanned.
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG: public key ID 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Oldstable? Abandoned Packages? Alternatives?
Thomas H. George wrote: An academic question: If I wrote a standalone program in C strictly for my own use what would be its lifetime? I have heard it argued that C (and I assume gcc) is here forever as it is the preferred language for writing operating systems. If you wrote a program using only C and the standard C library functions, the binary will work for probably many years (depending on how long it is until any transition from libc6 to libc7), and the source code will probably compile for decades (if appropriate arguments for backwards compatibility are given to the compiler). However, almost any program needs to be linked against some additional libraries for using graphics, etc, unless you want to write _everything_ from scratch. So really the binary (source code) will work only until one of the libraries it needs changes ABI (API, respectively), and the old version of the library is no longer available for Debian and can't be installed. This is probably on the order of a few years. Hope this answers your question, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Maple 9.5 interfaces on Debian box
Jerome BENOIT wrote: I have just installed Maple 9.5(2) on my Sarge box: 1] I have some trouble with the keyboard when I use the classical interface (Motif); Try toggling the NumLock key? I remember that Mathematica (which I believe is also Motif) behaves strangely if NumLock is on. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Multiple installed kernel-image packages?
Hi Paul, Paul Gear wrote: Thanks for the detailed response. Are you saying that once my system is installed (on 2.6.8, as it happens), it will never get an upgrade to 2.6.9 (once it is released) unless i explicitly install it? Does the fact that i asked for kernel-image-2.6-686 have any bearing on the situation? I thought this always pointed to the latest released 2.6 kernel image. Ah, that was a complication that I wasn't sure whether or not to bring up. Yes, kernel-image-2.6-686 is a metapackage [1] that always depends on the latest kernel version, so if you have it installed, running an apt-get dist-upgrade (not just upgrade) will see to it that you have the latest 2.6 kernel-image package installed. I believe (but haven't verified) that if you use aptitude for upgrades instead of apt-get, your old kernel images will then be uninstalled automatically, too. [1] metapackage means an empty package whose only purpose is to depend upon other package(s), making it easier to install them. I'm happy with removing the old ones myself. The only drama with the way you explain it is: how do i know when 2.6.9 is released except by checking for it manually every day/week/whatever? Well, other than having it done automatically by having the kernel-image-2.6-686 metapackage installed, you can for instance glance at the list of new packages in the aptitude UI. However there are often several Debian revisions of each kernel version; so apt-get upgrade WILL upgrade you from Debian release 2.6.8-6 to 2.6.8-7. Needless to say, you CANNOT install two Debian releases of the same kernel version at the same time. Presumably these Debian revisions are only released to fix security problems or other major bugs? For kernels in stable, that's right. For kernels in unstable and testing, they may go through several iterations of adding new out-of-tree drivers and other niceties in order to make them suitable for release in the next version of stable. Right now the plan is to release 2.6.8 with sarge, so a lot of effort is being concentrated on that, with a bit less on 2.6.9 (for people who intend to keep running sid). Presumably work on 2.6.9 will soon be moved to 2.6.10. ... Running grep-available -FProvides -sPackage kernel-image will give you a list of kernel packages known to APT on your architecture. (The grep-available command is in the grep-dctrl package.) How is that different from what apt-cache search --names-only kernel-image shows me? The latter shows every package with kernel-image in the name; the former only shows packages that explicitly claim to provide a kernel-image. There isn't much difference, except that your command shows the latest kernel image metapackages and the one I posted doesn't. As usual in Linux there's more than one way to skin a cat. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Multiple installed kernel-image packages?
Hi Paul, Paul Gear wrote: A quick question: is there a way to get apt to install new kernel-image packages rather than upgrade them, and keep the existing kernel-image package installed as well? Back on Red Hat, i could 'rpm -iv' (install) a new kernel package rather than 'rpm -Uv' (upgrade), and it would update grub's menu.lst and make the new one the default without affecting the currently-installed one. Is there an equivalent to this under Debian? If you are asking whether you can install (for instance) kernel 2.6.8, kernel 2.6.9, and kernel 2.6.9 compiled for SMP all at once, the answer is yes. In fact this is the default behavior. This is because Debian provides kernel versions as different packages, not as different versions of the same package. They have no files in common so, assuming you're using a modern Intel chip, it should be as easy as: # apt-get install kernel-image-2.6.8-1-686 # apt-get install kernel-image-2.6.9-1-686 # apt-get install kernel-image-2.6.9-1-686-smp As a corollary, if you are currently using kernel 2.6.8 and want to upgrade to 2.6.9, you will have to explicitly ask APT to install kernel 2.6.9, because apt-get upgrade will not do the trick. Likewise you will have to explicitly remove any old kernels that you are no longer using. However there are often several Debian revisions of each kernel version; so apt-get upgrade WILL upgrade you from Debian release 2.6.8-6 to 2.6.8-7. Needless to say, you CANNOT install two Debian releases of the same kernel version at the same time. For the sake of completeness, in case you aren't already aware: the -1 in the package names is a result of the kernel ABI in Debian packages changing at some point in the past, and the -686 or -686-smp suffix tells you the subarchitecture of CPU for which the package was compiled. Running grep-available -FProvides -sPackage kernel-image will give you a list of kernel packages known to APT on your architecture. (The grep-available command is in the grep-dctrl package.) regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian and Dell?
Hi Ed, Ed Sutherland wrote: I just purchased a Dell for my home office and am interested in using Debian on a partition. As my only Linux experience comes from a Mac, I have some questions: 1) Will I be able to easily dual-boot Windows or Linux using yaboot, or will I need to go through some BIOS mumbo-jumbo? The equivalent x86 tools are lilo or grub. One thing you need to be aware of: the x86 BIOS is much dumber than OpenFirmware. So if you use lilo, you must re-run the /sbin/lilo program every time the kernel location changes, even just the inode (e.g. by installing or upgrading a kernel image), otherwise you may find yourself unable to boot. The Debian kernel packages do this automatically upon install / upgrade, but you should be aware of it if you compile your own kernel. I think grub gets around this limitation in some way, but I'm not completely sure as I don't use it. Your BIOS may have some silly boot virus protection which you'll have to disable so it doesn't complain when lilo or grub overwrites the master boot record; this should be an option if you press F8 while booting. Don't know about your second question so I'll skip it. 3) Does the i86 side of Debian better support Web graphics and animation formats -- such as shockwave? There is a package in contrib that installs the Macromedia Flash player Web browser plugin. There is Acrobat Reader, for which Christian Marillat provides unofficial packages, which can read some PDF files that xpdf and gv choke on. For videos: there is MPlayer (packages also provided by Marillat), although you're probably aware of it since it's available on powerpc too. There is also RealPlayer (not in Debian, but can be obtained at www.real.com ). As far as I know Shockwave is not supported natively even on x86 Linux, but there is CrossOver, a proprietary ($20?) version of Wine that allows you to run a lot of Windows plugins and programs seamlessly; maybe it supports a Shockwave plugin. Hope this helps, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Odd eggdesktopentries.c messages when Gnome packages installed
Hi list, Since a few days ago I've been getting strange error messages when various Gnome-based applications are installed. During today's install from unstable: Setting up gnome-games (2.8.1-3) ... ** (process:10974): CRITICAL **: file eggdesktopentries.c: line 2223 (egg_desktop_entries_add_group): assertion `egg_desktop_entries_lookup_group (entries, group_name) == NULL' failed [and repeated twice more] Similarly with bug-buddy, gnumeric, etc. This doesn't seem to abort any package installations, so maybe it's only an aesthetic problem. Anyone know what it's a symptom of? thanks, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Odd eggdesktopentries.c messages when Gnome packages installed [solved]
Ron Johnson wrote: Try Googling for egg_desktop_entries_add_group assertion. Ah, thank you! For anyone else having this problem, the cause appears to be a botched mplayer.desktop file in the unofficial mplayer-arch packages; to fix the problem, edit /usr/share/applications/mplayer.desktop so that the contents aren't repeated. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] Can I make a prototype Free and sell the final product?
William Ballard wrote: I'm hacking together a little prototype for somebody, mostly as a curiosity and kindness. It'll be just a prototype. If they like it, I'll flesh it out and charge them money. Regardless, I'd like to put the prototype on SourceForge. (It might help close a sale.) Do I need some kind of dual license? If you are the only copyright holder, you can license any copy in any way you like, regardless of the history of the code's licensing. So you can certainly do what you propose. One thing to consider: once you have released the prototype under an open source license, there is nothing to stop someone else from forking it and adding features to compete with your final commercial product. And if the license in question is the GPL, you will not be able to merge new code from the fork back into your product (without the permission of the forker). Finally, don't sell anyone the _copyright_ on the commercial product, or they might conceivably be able to sue you for continuing to distribute the open source prototype! The case of Eric Weisstein being sued by CRC over his World of Mathematics a.k.a. http://mathworld.wolfram.com comes to mind: http://slashdot.org/yro/01/11/06/2028252.shtml ObDisclaimer: IANAL, and you might get better advice on debian-legal. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debian architectures
Mauro Darida wrote: I cannot quite understand which machines debian architectures are referring to. Of course I know that x-86 are common pcs, sparc are sun workstations, but others are quite cryptic to me. Anyone willing to translate into non-developer language?? Here's a list off the top of my head. Presumably others will add to and correct it. * alpha - Digital (DEC) workstations, usually originally running VMS, although there was a Windows NT port to this for a while. Pretty much legacy-only now. * arm - Includes a lot of embedded systems like routers, some PDAs (e.g. Sharp's Zaurus). * ia64 - Intel's 64-bit Itanium series of workstations. Not doing too well in the marketplace. * m68k - Motorola 680x0 chips: Ancient Apple computers, Amigas, VME crate controllers, embedded systems, ... * powerpc - IBM's PowerPC chips: Modern Apple computers, some IBM workstations (usually originally running AIX, I think?) * s390 - IBM's mainframes: big iron. * mips, mipsel, hppa - These I'm not too clear on. Architectures for which there exist Debian packages, but not officially supported by Debian: * amd64 (formerly x86-64) - AMD's 64-bit Opteron chip series for workstations. Intel has either come out with a clone or is soon to do so. This is rapidly growing in popularity and will almost certainly be supported in the next Debian release after Sarge. * sh (sh3,sh4) - SuperH chips from Hitachi, mostly found in some gaming consoles. Not clear if Debian will ever officially support it due to the rarity and slowness of the machines. There are also hurd-i386 and bsd-i386 which are using kernels other than the Linux kernel, running on Intel 32-bit machines. These are sort of experimental at the moment. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Firefox, Thunderbird etc.
Rogério Brito wrote: A newer Firefox will probably hit testing today. I don't know about Thunderbird, as it seems version 0.9-2 has not yet been uploaded to unstable. The Thunderbird maintainer has unofficial preview packages of 0.9-4 available from his web site: http://www.jwsdot.com/debian/install.html I think he's waiting for people to test them some before making an upload to unstable. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problem with apt-get and circular dependencies
Christian Renz wrote: apt-get wants to install libc6_2.3.2.ds1-18_sparc.deb (to replace libc6-2.3.2-7). However, the installation aborts telling me that You have a cpu which requires kernel 2.4.21 or greater in order to install this version of glibc. I tried to fix it by installing kernel-image-2.4.27-1-sparc32-smp, but to install this, I need a newer version of coreutils and initrd-tools (and others) that update glibc. Which aborts the installation telling me that... and so on. There are messages on debian-sparc giving various solutions. In order of (IMO) decreasing preference: http://lists.debian.org/debian-sparc/2004/11/msg00096.html http://lists.debian.org/debian-sparc/2004/11/msg00093.html http://lists.debian.org/debian-sparc/2004/01/msg00140.html regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: printing from firefox
Rich Wellner wrote: I get an error printing from firefox saying A broken version of the X print server (Xprt) has been detected. Note that printing using this Xprt server may not work properly. Please contact the server vendor for a fixed version. I've updated xprt-common, but that didn't seem to do anything. Are there other packages which I must touch? Hi, Maybe you have the package xprt installed? If so you should remove it and install xprt-xprintorg instead. Apparently the two are supposed to be equivalent, but xprt is pretty broken, so I'm not sure why it is even packaged. I think you can also print to PostScript in Firefox without using Xprint -- in the Print dialog, change the Printer selection box to read PostScript/default, then in Properties you can set options for lpr to use. Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CD malfunction on Sarge testing install Sparc
Walt L. Williams wrote: This may be a little off topic. I just tried to install Sarge testing for Sparc which I downloaded just last on an Ultra2. I was able to boot from the CD but afterwards it (strangely) could not find the CDROM drive which it booted from and therefore would not install. I didn't know who else to report this to so I posted it here. You probably want to email [EMAIL PROTECTED] instead. If they're unable to help, try debian-boot. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
New vim-doc package in unstable is very broken; don't install!
Hi list, I was just bitten by a problem in the preinst of the latest version of the vim-doc package. If you are tracking unstable, put it on hold until version 1:6.3-031+2 becomes available; version 1:6.3-031+1 will delete any symlinks and regular files in the /usr/share/doc directory. You can track the problem at bugs # 280824 and 280825 (be nice to the vim maintainers and don't submit any more duplicates; they are providing a fix as fast as possible). regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New vim-doc package in unstable is very broken; don't install!
On 11/11/2004 06:45 PM, Kevin B. McCarty wrote:
I was just bitten by a problem in the preinst of the latest version of
the vim-doc package. If you are tracking unstable, put it on hold until
version 1:6.3-031+2 becomes available; version 1:6.3-031+1 will delete
any symlinks and regular files in the /usr/share/doc directory.
By the way, if you've already been hit by this, and if it matters to
you, here is an inelegant fix. The following commands will reinstall
all of the packages that do not have a corresponding directory or
symlink in /usr/share/doc. (Not a practical solution for people on
dial-up, I'm afraid; in that case you could examine diff.txt after the
fourth line and recreate the symlinks manually, with some effort
involved in figuring out what links to what.)
cd /tmp
COLUMNS=200 dpkg -l | grep '^ii' | awk '{print $2}' pkgs.txt
ls -1 /usr/share/doc docs.txt
diff pkgs.txt doc.txt | grep '' | awk '{print $2}' diff.txt
xargs apt-get -y --reinstall install diff.txt
note that the flag to ls on the third line is a one, not an ell.
regards,
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG public key ID: 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt-get 'search'?
Tony Uceda Velez wrote: Anyone know what is comparable to Gentoo's 'emerge search' in apt-get? Man pages didn't reflect something similar? Want to search for list of packages that start with a certain string. grep-available -r -F Package '^prefix' The -r tells it to accept regular expressions, and -F Package to look in the Package field. '^prefix' is of course the prefix you want to search for, preceded by a caret. grep-available is in the grep-dctrl package. There is also apt-cache search, which is generally easier to use, although it's harder (maybe impossible?) to fine-tune exactly how it will search. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apt-get -f install question..
Ishwar Rattan wrote: Unpacking replacement libxt6 ... dpkg: error processing /var/cache/apt/archives/libxt6_4.3.0.dfsg.1-8_i386.deb (--unpack): trying to overwrite `/usr/X11R6/lib/X11/app-defaults', which is also in package xnview Hi Ishwar, The package xnview which has the conflicting file doesn't seem to exist anywhere in Debian -- I can't find it at packages.debian.org or with apt-cache. Where did you get it from? In libxt6, the file /usr/X11R6/lib/X11/app-defaults is a symlink to the directory /etc/X11/app-defaults. It isn't clear to me why the xnview package also contains /usr/.../app-defaults; instead it probably should contain an /etc/X11/app-defaults directory and whatever files it needs under there. You should inform the people who provided the xnview package of the problem. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Adding a kernel module - to get iptables to work.
Hi, Dr. David Kirkby wrote: Anyway, this error occured on the default kernel, which I think is 2.2.20. I think iptables runs on the 2.4 kernel and ipchains on the 2.2, so perhaps its not surprising it did not work on 2.2.20. That's right, iptables can only be used with 2.4 kernels and higher. I assume I need to load a module into the kernel, but are not sure if I load it with 'modload', whether I need to rebuild the kernel again, or whether I should remove the iptables package and download the source for iptables. You would use the modprobe command, found in the modutils package (for kernel 2.4) or the module-init-tools package (for kernel 2.6). Alternatively you can have the modules loaded automatically at boot time by putting their names (one per line) in the file /etc/modules (or, for a 2.6 kernel, /etc/modules-2.6). There should be no need to recompile iptables. You may need to recompile your kernel if you did not originally have it configured to build all of the iptables-related modules. But why not use a Debian-provided kernel package, which is certain to contain all of the necessary modules? The modules I use in my firewall (2.4 kernel) are: ip_tables iptable_filter ip_conntrack ip_conntrack_ftp iptable_nat ip_nat_ftp ipt_limit ipt_multiport iptable_mangle ipt_state ipt_REJECT ipt_LOG but you may not need all of these, depending upon what you are doing. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Invisible fonts in Xpdf widgets
Hi all, On my iBook I have for some time had the weird problem that some Xpdf widgets do not show any text. (Running Debian powerpc/unstable.) For instance, the magnification level pop-up menu, the Quit button in the lower right, and the context menu that pops up when the document is right-clicked. This doesn't affect viewing of any PDF document (thank goodness!), only the Xpdf interface. As far as I remember, I haven't made any changes to the Xpdf config files in /etc. No other application seems to be affected. dpkg -l \*font\* \*ttf\* produces the following output (only installed packages shown): ii fontconfig 2.2.3-1generic font configuration library ii gsfonts8.14+v8.11-0.1 Fonts for the Ghostscript interpreter(s) ii gsfonts-x110.17 Make Ghostscript fonts available to X11 ii libfontconfig1 2.2.3-1generic font configuration library (shared l ii libfontconfig1 2.2.3-1generic font configuration library (developm ii mplayer-fonts 3.5-2 Fonts for mplayer ii msttcorefonts 1.1.5 Installer for Microsoft TrueType core fonts ii psfontmgr 0.11.8-0.1 PostScript font manager -- part of Defoma, D ii ttf-freefont 20031008-1.1 Freefont Serif, Sans and Mono Truetype fonts ii x-ttcidfont-co 17 Configure TrueType and CID fonts for X ii xfonts-100dpi 4.3.0.dfsg.1-7 100 dpi fonts for X ii xfonts-75dpi 4.3.0.dfsg.1-7 75 dpi fonts for X ii xfonts-base4.3.0.dfsg.1-7 standard fonts for X ii xfonts-scalabl 4.3.0.dfsg.1-7 scalable fonts for X ii libsdl-ttf2.0- 2.0.6-5ttf library for Simple DirectMedia Layer wit ii libttf21.4pre.2003040 FreeType 1, The FREE TrueType Font Engine, s ii ttf-bitstream- 1.10-3 The Bitstream Vera family of free TrueType f ii ttf-freefont 20031008-1.1 Freefont Serif, Sans and Mono Truetype fonts ii ttf-opensymbol 1.1.2-3The OpenSymbol TrueType font Any suggestions? Thanks in advance! -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: forms.c and forms.h
Paul Akkermans wrote: I want to use the library forms.h and forms.c and buf.h but my Debian system says that it can not find these files. Does anybody know which package I have to download (via apt-get) to get these libraries included in my Debian system? Apt-file gives several possibilities: $ sudo apt-file update $ apt-file search forms.h | grep '/forms.h$' libforms-dev: usr/X11R6/include/X11/forms.h libvdk1-dev: usr/include/vdk/forms.h libvdk2-dev: usr/include/vdk2/vdk/forms.h $ apt-file search buf.h | grep '/buf.h$' genesis: usr/include/genesis/buf.h postgresql-dev: usr/include/postgresql/server/storage/buf.h I'm guessing you want Xforms, in the libforms-dev package, although that doesn't seem to contain a buf.h. It would be helpful if you knew what the library was called, instead of just the filenames of its header files. Using apt-cache show package on each of the above might help you decide. Also, you are unlikely to find a package containing forms.c; any package in Debian would have it compiled into a static or dynamic library (.a or .so file) which you then link against. If you really do want the source code, you can get it with apt-get source package. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mobility radeon driver
Pau Novella Garijo wrote: I have a laptop with a mobility radeon M6 LY video card and I've been looking for a free driver for a week with no results. I've been told that there are free drivers which suports up to M7 cards. I have the exact same card in my iBook and it's working flawlessly, so the following should work if you are running kernel 2.6: * Install the latest version of xserver-xfree86 from unstable. * (Re)configure X and select the ati driver; answer Yes when debconf asks if you want to use the framebuffer. * In your /etc/modules-2.6 file (or in /etc/modules if that file doesn't exist), add these three lines. They MUST be in this order: agpgart uninorth_agp radeon (Maybe the uninorth_agp isn't necessary on non-Apple machines? Don't know.) * In your bootloader config file (e.g. /etc/lilo.conf), add the argument append=video=radeonfb to your default kernel entry, then run the bootloader installer (e.g. lilo) if necessary. * Reboot and start X -- good luck! Run glxgears to see if your video acceleration is working. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: rm difficult filename
John Summerfield wrote: Often, too, you can use the TAB key to advance over problematic characters, a ? to represent one of them and * to represent any number of them. So rm ?rtsp-stream-over-tcp rm *over-tcp Sometimes, but in this case it wouldn't work. Using a * or ? will help with characters that the shell wants to mess with, but - generally has no special meaning to the shell. [1] So rm *over-tcp will be globbed by the shell and passed to rm as rm -rtsp-stream-over-tcp. This doesn't help because the first - is interpreted by rm itself as an indication that the argument is a flag instead of a file. Some people take advantage of this behavior by creating a file called -i in important directories, so rm * will turn into rm -i ... causing rm to ask for confirmation before deleting. [1] Yes, I'm aware of cd - and there are probably other special cases. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cannot install xfree86-common via apt
You wrote: I did upgrade with apt-get upgrade, but i got error that i cant install xfree86-common, i try apt-get -f install and it install the other packages except xfree86-common, the error is: Preparing to replace xfree86-common 4.3.0.dfsg.1-6 (using .../xfree86-common_4.3.0.dfsg.1-7_all.deb) ... Document `debian-x-faq' is not installed, cannot remove. Unpacking replacement xfree86-common ... dpkg: error processing /var/cache/apt/archives/xfree86-common_4.3.0.dfsg.1-7_all.deb (--unpack): trying to overwrite `/usr/lib/X11', which is also in package t1-cyrillic This was mentioned on the debian-x list and seems to be a bug in the t1-cyrillic package. See http://bugs.debian.org/270346 . As a workaround (note: not tested personally), you can probably temporarily uninstall t1-cyrillic, install xfree86-common, and then reinstall t1-cyrillic: # dpkg --purge t1-cyrillic # apt-get install xfree86-common # apt-get install t1-cyrillic This should work since /usr/lib/X11 is shipped as a symlink in xfree86-common and as a directory (the bug) in t1-cyrillic. Good luck, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: file/dir rights management
Martin Henne wrote: On my system I want to keep users from browsing the '/home' directory. Unfortunately, when I do a 'chmod o-r /home', the user can't login via ftp anymore. Hi Martin (I'm BCC-ing you since your email was anti-spam munged), Stefan already addressed your main question, so this is a little off-topic. But if you are concerned about security, you should be much more worried about the fact that you are permitting user logins by FTP than that the users can read some files in /etc. Are you aware that FTP transmits passwords over the 'net without encryption, so anyone snooping on the connection could steal them? I suggest SFTP as an alternative -- it unfortunately isn't as featureful as some FTP clients, but at least it will encrypt passwords. [If you are only using FTP on a closed trusted LAN, of course, I withdraw the comment.] -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk package
Pene Chamuscado del Caballo wrote: From what you have mentioned, it sounds like you are actually trying to compile the editor, and you ran ./configure which gave the mentioned errors? If so, try installing the GTK+ development library: apt-get install libgtk2.0-0 I think you need apt-get install libgtk2.0-dev in order to get the pkg-config file, .so symlinks, etc. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sun Java - revisited
Hi Michael, j2eesdk-1_4_2.bin: error while loading shared libraries: libstd++-libc6.2-2.so.3: Cannot open shared object file: No such file or directory. I do have libstd++6 installed on my system, and apt reports that it is the latest version. I don't see a way to install the missing library, and without it j2eesdk will not install. Guessing that an older version of j2eesdk would install, I tried searching for one on the Sun site, but I couldn't find one. I assume that what you really want is libstdc++, not libstd++ (since I don't think the latter thing exists). With that assumption: $ sudo apt-file update [output snipped] $ apt-file search libstdc++-libc6.2-2.so.3 gcc-2.95-nof: usr/lib/nof/libstdc++-libc6.2-2.so.3 gcc-2.95-nof: usr/lib/nof/pic/libstdc++-libc6.2-2.so.3 libstdc++2.10-dbg: usr/lib/debug/libstdc++-libc6.2-2.so.3 libstdc++2.10-glibc2.2: usr/lib/libstdc++-libc6.2-2.so.3 So probably you want to apt-get install libstdc++2.10-glibc2.2. (The apt-file tool is in a package of the same name, incidentally. You can also look up packages by filename at http://packages.debian.org/ .) regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Trying to build agpgart.o with an rpm from Intel
Hi Eric, And I'm having no luck. Nothing seems to be where the rpm wants it to be. Is there anything I can do to make this rpm work? Why would you need an RPM? The agpgart module should be included with any Debian kernel 2.4 package available in woody, sarge or sid (with the possible exception of the woody boot-floppies kernel). $ dpkg -S agpgart.o kernel-image-2.4.26-1-k7: /lib/modules/2.4.26-1-k7/kernel/drivers/char/agp/agpgart.o kernel-image-2.4.27-1-k7: /lib/modules/2.4.27-1-k7/kernel/drivers/char/agp/agpgart.o What happens if you just run modprobe agpgart as root? Now, I'm not sure whether the XFree86 in woody is new enough to recognize current i810 boards, but maybe you can try sarge or sid. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dpkg / apt equivalent to 'rpm -qf'?
John Hasler writes: Just off the top of my head I see no reason why these files could not be included in the package empty and filled in by the scripts. This would identify the files as belonging to the package and also allow dpkg to remove them, eliminating the need for the postrm to do so. I think the canonical answer is that some programs will behave differently if a config file exists (even if empty) than if it doesn't exist. E.g., /etc/nologin -- you wouldn't want to ship that :-) Then there are some files that it's questionable who they would belong to. For instance, /etc/ld.so.conf needs to be modified by several packages; if it was owned by some package, it would be a Policy violation for any other package to touch it. Then someone would have to write an update-ld.so.conf script, which just seems like overkill. I agree that the vast majority of postinst-created files in /etc don't meet either of these criteria, so the suggestion makes sense there. My understanding is that there is long-term work planned on dpkg to allow registering a list of related files on package installation, even if they aren't actually in the package. -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sarge not printing [WORKAROUND]
On 08/12/2004 05:03 PM, Kevin B. McCarty wrote:
For what it's worth, I started experiencing the same problem beginning
perhaps a couple weeks ago. The weird thing is that my cluster is
running woody except for a few backports! One of them was gs 7.07, so I
downgraded back to the gs in woody, but still no luck. I haven't found
a solution yet.
After a couple hours looking more deeply into this, I found part of the
problem. (This was made slightly easier because I'm running woody +
backports, so the number of possible culprits was relatively small.)
Apparently the semantics of sed changed slightly between 4.0.9-5 and
4.1-1, causing something that uses sed in the long pipeline from lpr -
apsfilter - gs - printer to break. So a workaround is to downgrade
sed to version 4.0.9-5 or earlier, which should result in printing via
apsfilter working once again.
Still researching which package Bug 258042 should be reassigned to --
right now it appears to be one of {sed, a2ps, psutils}.
--
Kevin B. McCarty [EMAIL PROTECTED] Physics Department
WWW: http://www.princeton.edu/~kmccarty/Princeton University
GPG public key ID: 4F83C751 Princeton, NJ 08544
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sarge not printing
For what it's worth, I started experiencing the same problem beginning perhaps a couple weeks ago. The weird thing is that my cluster is running woody except for a few backports! One of them was gs 7.07, so I downgraded back to the gs in woody, but still no luck. I haven't found a solution yet. Someone filed a bug on this problem against gs, although who knows whether that is actually the culprit. The similarly afflicted or the morbidly curious can follow along at http://bugs.debian.org/258042 . regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: unresolved kernel symbols
Hi Tong, I am experiencing unresolved kernel symbol problem when compiling my pctel modem driver. The compilation and installation was ok. When trying to load the module, I get: % insmod -f pctel Using /lib/modules/2.4.25-1-386/misc/pctel.o Warning: Please read the error message output by the kernel: The module you are trying to load (/lib/modules/2.4.25-1-386/misc/pctel.o) is compiled with a gcc version 2 compiler, while the kernel you are running is compiled with a gcc version 3 compiler. This is known to not work. Is the claim of the error message true -- that you compiled the module with gcc-2.95? This may have happened unintentionally if, for instance, /usr/bin/cc or /usr/bin/gcc was a symlink to gcc-2.95 on your system. If so, then the obvious fix is to make cc and gcc be symlinks to gcc-3.3 (this should be the case by default in sarge and sid) and recompile. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cite for print-to-postscript exploit in Mozilla?
On 07/10/2004 12:18 PM, Florian Weimer wrote: 1.7 incorporates some other security fixes, apparently in the area of cross-domain scripting vulnerabilities. So you probably should upgrade anyway. Does anyone know if there is some reason these fixes haven't been backported to woody? regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Cite for print-to-postscript exploit in Mozilla?
Hi, I would like to know where you found the security advisory that you cited in your email to Debian Bugs # 252362 and 247585. Inquiring minds would like to know what sort of exploit can be produced by the print-to-postscript option in Mozilla and Firefox (especially since it is still enabled by default upstream). If serious, it should probably result in the release of Mozilla security updates for woody and backports.org. On 2004-06-03 Rebecca Greenwald [EMAIL PROTECTED] wrote: The direct postscript supports has serious remote exploits which allow malicious pages to execute any shell commands in the content of the current user. Sure, if you want that enable it again. Thanks and regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cite for print-to-postscript exploit in Mozilla?
On 07/09/2004 04:02 PM, Ian Douglas wrote: http://www.imc.org/ietf-822/old-archive1/msg01346.html Is probably what is being refered to... Thanks for the link! (Wow, foreshadowing of virus infections via email attachments...) But is there any way in which Mozilla's print-to-postscript is _less_ safe than using gv to open up a random PostScript file found somewhere on the Internet? Or are the two equally insecure? If the latter, then does it make sense to turn off postscript printing without also removing gv and other PS viewers from Debian? I admit this last question is a bit rhetorical. My point is that, as sysadmin of a physics cluster running Debian/woody on which people frequently look at downloaded PS files anyway, I want to know whether it is really worth my time to upgrade Mozilla [currently running 1.4 from Adrian Bunk's backports], install Xprint from unstable, and go through the apparently non-trivial task of getting it to work well. By the way, is PDF also Turing-complete with the accompanying security issues? regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG public key ID: 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Please test Cernlib PAW on AMD64 and exotic arches (sample test included)
Hi all, I hacked out a port of Cernlib for AMD64, and it should now be available in the AMD64 archive (thanks to Andreas Jochens for building it). If anyone who has an Opteron machine could do some testing on it for me, please let me know the results. I'm particularly interested in whether the FORTRAN interpreter built into PAW works correctly. I would also welcome test reports from any of these other architectures that I don't have access to myself (not being a DD), especially the 64-bit ones: ia64, alpha, hppa, mips, mipsel, arm, s390, powerpc (G5 only) Here is a minimal test you can try even if not familiar with PAW or FORTRAN. Note that in pawtest.f each line begins with 6 spaces except for 10 CONTINUE which begins with 3 spaces. (It's a recursive function to generate the Nth triangular number, for those who are wondering.) $ sudo apt-get install paw $ cat EOF pawtest.f SUBROUTINE PAWTEST(N) SUM=0. DO 10 I=1,N SUM=SUM+I 10 CONTINUE PRINT *,SUM END EOF $ paw [PAW banner output snipped] Workstation type (?=HELP) CR=1 : [hit ENTER once] [More PAW output snipped] [A graphics window will open if you have a $DISPLAY; it can be minimized and ignored.] PAW call pawtest.f(5) 15.0 PAW call pawtest.f77(5) 15. PAW quit $ The important requirement is that both PAW commands above should output the real number 15.0 (with some arbitrary number of zeros after the decimal point). Please email results directly to me so as not to clutter the mailing lists. Be sure to let me know your architecture and sub-arch. On 64-bit sub-arches whose .debs are compiled for 32-bit (sparc64, s390x, powerpc G5), you may first need to touch /etc/disable_64_gcc as root for the test to succeed. Please also let me know how the existence and non-existence of that file affect the test results. thanks and regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Heads up: latest libcurl2 package broken
Hi all, The latest libcurl2 package (version 7.12.0-1) in unstable is broken, see bug # 252348. I strongly recommend staying with 7.11.2-1 or earlier until the bug is fixed. Among other things, it breaks discover and therefore also xserver-xfree86 preinst and postinst scripts. regards, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Permissions on the file /dev/pmu are broken
I just upgraded to the Gnome 2.6 packages in unstable on my iBook laptop. Upon logging in to Gnome 2.6 the first time, I receive an error dialog popup message titled Error whose text is Permissions on the file /dev/pmu are broken. Does anyone know what program pops up this error message (so I can file a bug against it for not giving a better error message), and what the permissions on /dev/pmu are supposed to be? Mine are currently rw---, owned by root.root thanks, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Mozilla Thunderbird questions
Hi all, I switched from pine to Thunderbird last week, and there are two issues that are really frustrating me. 1) How can I get Thunderbird to insert a text file into the body of an email that I'm writing? In pine this is just Ctrl+R, but in Thunderbird the only way I've found is to cut-n-paste from an editor window. Needless to say, this is undesirable for large files. (Selecting Attach from the toolbar or File-Attach from the menu only lets me send the file as an attachment.) 2) How can I force Thunderbird not to wrap specific long lines in an email? If for instance I'm writing a bug report, I want it to wrap paragraphs of text, but not the error messages I include inline. Is the context menu Paste Without Formatting what I'm looking for? Thanks in advance for any hints, -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla Thunderbird questions [partly solved]
Well, here's a partial answer to one of my questions. Posting it here in the hopes it helps others, too. Kevin B. McCarty wrote: 2) How can I force Thunderbird not to wrap specific long lines in an email? If for instance I'm writing a bug report, I want it to wrap paragraphs of text, but not the error messages I include inline. Is the context menu Paste Without Formatting what I'm looking for? Copied from here: http://brianpuccio.net/old/node/view/169 : I found out how to get rid of the on the fly, automagic adjusting of the word wrap in plain text emails. I knew it was called flowed format but I wasn't sure how to turn it off. Yes, I am aware of the advantages it offers, but it also makes me look like an idiot when posting to the mailing lists. So I added these two lines to my prefs.js file: user_pref(mailnews.display.disable_format_flowed_support, true); user_pref(mailnews.send_plaintext_flowed, false); -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
kernel security update seems fixed now
Kevin B. McCarty wrote: Dear all, It seems that at least on ix86, the latest woody security updates for the kernel packages are completely broken. The kernel packages named kernel-image-2.4.18-1-subflavor with version 2.4.18-13 contain only one module. DO NOT upgrade to these packages or your system will be broken when you reboot. It looks like this was fixed yesterday afternoon with the release of -13.1 packages. Major kudos to the security team for getting those out so fast! -- Kevin B. McCarty [EMAIL PROTECTED] Physics Department WWW: http://www.princeton.edu/~kmccarty/Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
