Re: [gentoo-user] Doubt about qt and kde flags
A. R. schreef: > On 5/13/05, Allan Spagnol Comar <[EMAIL PROTECTED]> wrote: > >>Hi All, I am having a little doubt about qt and kde user flags, I am >>using Gnome in my system but I like some kde programs, so I have gtk >>and gnome flag, should I use kde and qt flags to or should I put the >>-kde and -qt ? >> >>-- > Hi, > > It is my understanding that even if you disable those flags > (-kde -qt) and you try to emerge a package that depends > on kde and qt libraries, those will be compiled as well. > > As an example, I am running xfce4 (gtk2) and my use flags have > -kde -qt, if I try to emerge k3b, it shows that it wants to emerge: > (just a snippet) > > x11-libs/qt > kde-base/kde-env > kde-base/kde-libs > > The flags only affect what a given package is *able* to support. > > In this example, if I want to be able to use k3b, some > kde and qt libraries need to be compiled so it can run, > and that is completely different than the software > supported by k3b that can be customized through altering > the use flags. > > HTH, > > -AR > This is true, but not exactly clear. Continuing with the K3b example (because it's a good one for this purpose), I have often compiled K3b without the -kde and -arts USE flags, because by preference I use GNOME or Openbox or IceWM, not KDE. Because K3b relies on kdelibs and qt before it can be compiled at all, those will be compiled when I emerge K3b, no matter what my USE flags say. They are required dependencies. However K3bsetup2 is an optional dependency. It is required if you want to integrate K3b into KDE (+kde), and sound output (the "rasberry" sound when a burn fails, the fanfare when it succeeds, etc) is an optional dependency that will be compiled with the +arts flag, also for integration with KDE. So if I am running K3b under GNOME, and thus have emerged it -kde and -arts, I won't have K3bsetup2 or the sound business, but then again, I'm running it under GNOME, so I don't actually care that much (I can do the setup manually if I must, and it's not worth it to have aRTs just to hear a fanfare when my burn succeeds). That's what an optional dependency is. If I don't have KDE installed, though, and try to emerge K3b +kde and +arts, K3b will not only emerge the required kdelibs, but also kdebase, because k3bsetup2 (the optional dependency added by +kde) requires that as well. So basically I'll get KDE, because I requested that K3b be integrated with KDE-- and that obviously requires that a viable KDE installation be present, so it will be created if it does not exist beforehand. So if you want your programs to have all possible integration into the KDE environment (perhaps you want to give it a spin, even I think that 3.4.0 is pretty nice), then use +kde and +arts. +qt you probably want if you want to use QT based programs (some of which are not related to KDE, for instance DVD:Rip, iirc, and probably Quanta). But if you know that you just want some KDE based programs, but not KDE itself, you can try using -kde and -arts (it should always be -both, because aRTs pulls in KDE if present). However, you should be aware that there are not all that many KDE-based programs that don't require KDE to be present to run, but only the libs. That may have changed with the split ebuilds, but up to now the only ones I know of are K3b and Krusader (mostly because they're the only ones I use when not using KDE). It's not like you can so much install Konqueror without KDE, for example. But again, that may have changed. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: encrypted dvds no longer correctly decrypted under gentoo
Antoine schreef: >>I thought it might be a problem with the ugly state of my portage but I >>get a slightly different problem on a recently installed laptop running >>gentoo ~x86. On the laptop a dumpstream gets most of the way through and >>then starts transferring bytes about 8 at a time - with about 30 seconds >>between reads - clearly not an option for the last 1 gig! >>Seeing the problems I tried (on machine 1) remerging and the merging >>~x86 versions of libdvdread and libdvdcss but no change. >>Any help or even similar stories would be most appreciated. > > > Is it possible that these problems come from /usr/src/linux pointing to > the ck sources instead of the 2.6.8 gentoo std that I am now running? > Should I try to recompile? > Cheers > Antoine I don't so much see how that could be involved. But you say these are rented DVDs? Are they new films (such as "Kingdom of Heaven", rather than older films such as "Fargo")? The causes I see as more likely than the kernel link are: 1) dirt (rented DVDs pass through a lot of hands) 2) new encryption that libdvdcss cannot handle. After all, that regstered encoding copy-protection business that the media providers have been pushing/implementing (I don't follow RIAA/MPAA news that closely) is certainly more likely to be on DVDs available for rental, since "they" know that it's a source to rip DVDs. So that's where I'd put it if I thought it was important to make it (meaning on DVDs destined for rental outlets). Plus, you *played* it under PowerDVD, but you didn't try to *rip* it under Windows-- I suspect that it wouldn't work there either, and that PowerDVD and other Windows players have the facility to bypass the copy protection for reading, but that the DVD itself is "copy-proofed" against both Windows and Linux tools. Clearly some of them are-- I just read (not a week or two ago) about a French claimant who won (!!) a case claiming that DVD copy protection violated his privacy rights (he was trying to copy a DVD to a video tape-- the story is at http://www.foxnews.com/story/0,2933,154663,00.html ). So clearly the technology is in use. Perhaps it's bitten you in the figurative butt. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] AMD64: "vanilla-sources" and the risk of using them
Michael Haan schreef: > I *think* I know what they are, what risk do I run by using them? > If you think there's a risk, then I don't think you know what they are :-) . The "vanilla" sources are the same sources you would get on kernel.org. No extra patches (as you would find in gentoo-dev-sources) or optimizations/patches (as you would find in ck-sources) and no "bleeding edge" patches (as you would find in mm-sources). The kernel doesn't get any more risk-free than vanilla-sources, because if those sources are broken then Linux is broken. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] next step X
Mark Knecht schreef: > On 5/14/05, A. Khattri <[EMAIL PROTECTED]> wrote: > >>On Sat, 14 May 2005, Mark Knecht wrote: > > >>>The first problem is that Mozilla is not in this system and it's not >>>in the world file. Why is emerge trying to bring it in? >> >>Add the -t flag to print the dependency tree. >> > > > Thanks. I'm surprised that the Gnome ebuid depends on Mozilla. I have > Firefox installed. I would have hoped that would be enough. The gnome metapackage includes epiphany, which does depend on Mozilla. Evolution (also included) might depend on Mozilla as well. > > I'm looking at installing Gnome-light instead but I'm not clear if the > emerge -C gnome step will uninstall everything and cause me to have to > completely rebuild all parts of Gnome. No, it won't uninstall everything. In fact, it won't uninstall anything except the meta-build itself. It's kind of annoying, as I just did this myself, and still don't know how to clean my system of more GNOME cruft than the obvious (Epiphany, Mozilla, Evolution and Evolution-data-server). But in any case, I think that what uninstalling the metapackage does is frees the underlying packages, so you *can* uninstall them, without having your next emerge -u world try to drag Mozilla, Epiphany, et al back in (because the metapackage depends on them). Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] next step X
cfk schreef: > Gentlemen: > > Here's what I have found over the last day or so in trying to get to X > functionality. This is with a computer with more then one distribution, and > the others all have X functionality. The computer is an Intel 810 motherboard > with the i810 integrated graphics device. > > I can see there is no /dev/mouse0 or /dev/agpgart on the Gentoo partition as > there is on the Fedora partition, so part of the installation is incomplete. > > I copied the known functional /etc/X11/xorg.conf from the known functional > Fedora X installation. > > When I invoke, startx, I get two errors. One has to do with the > missing /dev/agpgart and the other has to do with the missing /dev/mouse0. > > I did try to follow the "X Server Configuration Howto" at gentoo.org with > 'emerge xorg-x11', 'env-update', 'xorgconfig' and others, but I obviously > missed a step and my naivety is showing a little bit. > > What installation step have I missed that precluded Gentoo from creating > these two device nodes (and perhaps some others) that keep me from getting to > the next step, an X windows screen (forget about KDE for now, I would be > happy just to get a pointer moving around on the screen). > > Charles Krinke I think (read, "suspect") that you have missed one step, and mistaken one step. I suspect that your mouse is fine, but you're looking for the wrong device. Is your Fedora installation based on a 2.4-series kernel? If you change your device in xorg.conf to /dev/input/mice (the default for the 2.6-series kernel you now most likely have), I bet it will work. As for the missing agpgart, that could well be because your motherboard needs a specific driver (for instance, my VIA KM266A board needs both /dev/agpgart in the kernel, and also via-agp either statically or as a module, or else I won't have AGP support). So I would check your kernel config and make sure that any chipset support for the AGP bus that you may need for your specific mobo has been compiled either statically, or as a module. If a module, don't forget to add it to /etc/modules.autoload.d/kernel-2.6 . Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Problems changing email clients
Hey, all-- So many people on this list have mentioned neat features of Sylpheed-Claws that I wanted to check it out. My problem is that I'm currently using Thunderbird, and I want to share my stored mail with Sylpheed-Claws. In the event Sylpheed-Claws doesn't suit me, I certainly don't want my mail winding up split all over the place (the "new" mail in the MH directory, and the "old" mail in the directory Thunderbird points to --which is not in ~/.thunderbird, btw). Now, I could always leave the mail on the POP3 server just for that, but I also want to know how Sylpheed will act upon my current settings (labels, filters, etc). And I just don't want to have to manage two mail directories, even temporarily. As I understand it, I could share the mail directory with the sylpheed-claws-mailmbox plugin, but I can't get it to work under either the GTK version (1.0.4) or the GTK2 version (1.9.9). It's a twisted tale of installs, uninstalls and reinstalls (the bulk of which I will spare you), but the upshot seems to be that under 1.0.4 no plugin would load (even the builtins), giving me 'invalid ELF header' errors, and under 1.9.9, the only mbox plugin is for 1.0.4 (GTK1, in other words, but the builtin plugins load fine). As much as I dislike GTK1 apps, I would drop back down to 1.0.4 (now updated to 1.0.4a, I see in just the couple of hours I've been playing with this; just my luck) if it would work, but I presently don't have any such assurance. So here are my questions: 1. Is this even possible in the first place? What I want is to point sylpheed to my current mail directory, which is the same one I've used since before 1) I moved to Holland 5 years ago; 2) I switched from Windows to Linux; 3) Thunderbird even existed (way back when, I was using MozillaMail, and before that Netscape Mail). In other words, I've ported this same mail directory and its contents across an ocean, from ISP to ISP, and from OS to OS-- I really don't want to screw it up now just because I changed mail clients. If I did, I'd be using Opera :-) or (heaven help us) KMail. If I can't change back to my current mail client without endangering my data (which, as you might guess, I somewhat value) via 2 conversion operations (to Sylpheed-Claws and then back again if I don't like it), then the deal is off already. 2. If this is possible, under which version of Sylpheed-Claws is it possible? My guess is that the new GTK2 version is so new that the plugin has not yet been updated to match-- but can I expect this to occur in a reasonably timely fashion (in other words, is the speed of Claws development fast, slow, or non-existent)? If I should drop back to the GTK version, how can I do so so that the plugins will load when I try to activate them? Uninstalling everything and then installing the client and the plugins in the correct order did not seem to work, and I really didn't feel like recompiling libelf for "no reason". So if someone can confirm that either an updated plugin is likely within a few days, or that recompiling libelf would "fix" the GTK plugins, I'd wait or give it a shot. It really looks like a nice program (it did when I looked at it last year, too), and I would regret not having the opportunity to check it out more thoroughly, but I am admittedly (extremely) anal about my mail (which is actually pretty funny, all things considered, but it's my right to be as weird as I want about what I want to be weird about, and I'm exercising it :-) ). TIA for any help/advice, Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Rogue devices at boot
Hey ho again-- I don't *think* this is a major issue, as everything works, but since my system is reasonably stable atm, I'm working on my 'orange flag' items (disturbing things that are not an emergency). During boot, when devices are being set up, services loaded, and drives mounted, I get a whole lotta the following: nbd110: Request when not-ready end_request: I/O error, dev nbd110, sector 0 nbd111: Request when not-ready end_request: I/O error, dev nbd111, sector 0 nbd112: Request when not-ready end_request: I/O error, dev nbd112, sector 0 nbd113: Request when not-ready end_request: I/O error, dev nbd113, sector 0 nbd114: Request when not-ready end_request: I/O error, dev nbd114, sector 0 nbd115: Request when not-ready end_request: I/O error, dev nbd115, sector 0 nbd116: Request when not-ready end_request: I/O error, dev nbd116, sector 0 nbd117: Request when not-ready end_request: I/O error, dev nbd117, sector 0 nbd118: Request when not-ready end_request: I/O error, dev nbd118, sector 0 nbd119: Request when not-ready end_request: I/O error, dev nbd119, sector 0 nbd120: Request when not-ready end_request: I/O error, dev nbd120, sector 0 nbd121: Request when not-ready end_request: I/O error, dev nbd121, sector 0 nbd122: Request when not-ready end_request: I/O error, dev nbd122, sector 0 nbd123: Request when not-ready end_request: I/O error, dev nbd123, sector 0 nbd124: Request when not-ready end_request: I/O error, dev nbd124, sector 0 nbd125: Request when not-ready end_request: I/O error, dev nbd125, sector 0 nbd126: Request when not-ready end_request: I/O error, dev nbd126, sector 0 nbd127: Request when not-ready end_request: I/O error, dev nbd127, sector 0 device-mapper: dm-linear: Device lookup failed device-mapper: error adding target to table This is just the end of the list; the listed nbd devices start from 0. I've checked Google, and now I know that nbd stands for 'network block device". I also know that I have enabled network block devices in my kernel (currently gentoo-sources 2.6.11-r6), because I thought it might be useful when setting up Samba to share to and receive shares from my bf's Windows computer. But even though I have not yet configured Samba (it is, however, installed and running; it just doesn't work because I haven't configured it), and even if nbd has nothing to do for or against it, I don't see that I should be getting 127 unusable devices. Why has udev not removed them, for example (no, I'm not using the tarball; I checked /etc/conf.d/rc)? Can I (as root) remove them without issue? I've just downloaded a new kernel; if I disable ndb, will that get rid of them? Should I actually disable nbd in the kernel (or is it useful for something like Samba)? Or do I just need to configure Samba (client and server) so that *something* knows what to do with these devices, and that will do it? I don't have a clue (obviously), so any suggestions apreciated. In case it's relevant, I will also mention that this installation is 'converted' to real Gentoo from an installation of "that Gentoo-based OS with an installer" (I was just out of patience, but whether it was worth it is offically questionable), so if it's possible that this device creation/persistence is caused by a holdover system utility from that OS which I'm unaware of and which was not removed during the conversion process, please let me know-- that's another aspect of the "orange flag" cleanup that I'd like to handle as well. Thanks, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] different USE flags for apache on two machines?
Craig Duncan schreef: > Stroller wrote: > > >>On May 16, 2005, at 4:41 am, Craig Duncan wrote: >> >> >>>I have two systems (x86 laptop and a x86 1u server) both of which have >>>been updated and now link to the 2005.0 profile >>>(/usr/portage/profiles/default-linux/x86/2005.0). When I emerge apache >>>on these systems the USE flags reported are different and I am unsure >>>what I did differently when updating each system - any ideas? >> >> >>You set your USE flags in /etc/make.conf and (optionally) >>/etc/portage/package.use >>Compare these files on the two systems. >> >>Stroller. >> > > Yes, I understand that, but these are the default flags for the ebuild. > I comment out all my own use flags and on my laptop I see the following > defaults > > net-www/apache-2.0.54-r4 +apache2 -debug +doc +ldap -mpm-leader > -mpm-peruser +mpm-prefork -mpm-threadpool -mpm-worker -no-suexec +ssl > -static-modules -threads > > On the server, I do the same, yet the default USE flags are as follows > > net-www/apache-2.0.54-r4 +berkdb -debug -doc +gdbm +ipv6 -ldap +ssl -static > -threads > > Where are the mpm-* flags, why are the defaults on each system different? > > > Craig > > Your USE flags in make.conf and the ebuild are not the only place default USE flags are read from. They are also read from the profile directory. Has one of the computers synced recently, and the other not? Emerge sync also updates profile information, I believe. In any case, I can only think that the /usr/portage/profiles//use.defaults file is not the same between the two computers-- but that may be exactly as it should be. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] adding/removing services
Cheryl Homiak schreef: > At the end of emerges I'm getting something about a couple of programs > saying it isn't adding the to services because some other program, a > program that is no longer even merged, is providing that function. I did > actually delete the old and add the new in rc-update and the new program > is running fine, but is there some way to take care of this. > Thanks. > Hi, Cheryl-- Usually the solution to this is to go to /etc/init.d and remove the script for the offending unmerged program. I'm sure there's some reason that such scripts are not removed by the unmerge, but I don't know what it is. YMMV, but hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] links in Thunderbird
Pingveno schreef:
> Craig Duncan wrote:
>
>> Arran Fraser wrote:
>>
>>
>>
>>> I recently did my first-in-a-long-while "emerge world". Now, when I
>>> click a link in Thunderbird nothing happens. I'd like the link to be
>>> opened in Firefox (of course). I'm using KDE.
>>>
>>> Any ideas?
>>>
>>>
>>> --
>>> Arran
>>>
>>>
>>>
>>
>> Try something like this...
>>
>> $EDITOR ~/.thunderbird//prefs.js and add the following
>> lines
>>
>> And add these lines
>>
>> user_pref("network.protocol-handler.app.ftp", "/opt/firefox/firefox");
>> user_pref("network.protocol-handler.app.http", "/opt/firefox/firefox");
>> user_pref("network.protocol-handler.app.https", "/opt/firefox/firefox");
>>
>> Restart thunderbird and links in email should now open in firefox.
>>
>> Craig
>>
>>
>>
> I'm have approximately the same problem, with Thunderbird instead
> wanting to open links in mozilla-launcher. The problem is,
> mozilla-launcher has the audacity to fail to open *anything*.
> mozilla-laucher sits in /usr/libexec, giving the error "unknown browser"
> when I attempt to launch it.
>
> I tried editing prefs.js, but Thunderbird reverted prefs.js to its
> previous state when I next started it up.
>
> -Pingveno
>
This is odd-- I have been so happy this week because after a good year
of waiting, T-bird and Firefox finally interoperate seamlessly, with no
need for me to write scripts, edit prefs.js or anything of that nature.
What I wonder is:
1) do you have Thunderbird and Firefox set as the default email
client/browser in your desktop environment (if GNOME or KDE)?
2) what does mozilla-launcher itself say (what is the unknown browser
it's looking for)?
Mozilla-launcher is just a script, so open it up in a text editor and
look at it.
Hope this helps,
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: depclean
David Morgan schreef: > On 04:04 Thu 19 May , Ciaran McCreesh wrote: > >>On Thu, 19 May 2005 01:51:23 + (UTC) James >><[EMAIL PROTECTED]> wrote: >>| OK, but this begs another question. >> >>No, dammit! It does *not* beg another question. >> >>Sorry. Pet hate there. >> >>| In general when you do not know what package contains a tool you need, >>| how do you discover the package name containing the tool? >> >>Google it :) I heard a rumour that Marius was working on a program for >>centrally tracking 'what everything installs', but I don't think it's >>beyond the ideas phase yet. It's somewhat tricky to do, what with our >>packages not having static content like other distros do. >> > > > I usually find www.rpmseek.com is a good place to look (you need to do > an extended search and choose 'package contains file'). > > If the tool is already installed, you can do equery b toolname, and > that'll tell you what package provided it (you'll need to emerge > gentoolkit first though). I realise that this probably isn't what you > were asking, but thought you might be interested anyway. > Another good place to look (after Google, which will often hit in rpmseek or rpmfind as well as my suggested search site) is the Debian package database at http://www.debian.org/distrib/packages . There you can search through the contents of packages, which is useful since you already know that the package name is different from the name of the file you're looking for. The package names under Debian are not always the same (usually close enough, though), but it does give you a starting point at www.gentoo-portage.com , which can sometimes be helpful, or at packages.gentoo.org, where you can go to the package's homepage and see what it contains. Or of course, once you've got some prospective package names, you could just fetch the packages in question (the Debian package search results in both findutils and slocate to get updatedb, for example) and look to see if they're appropriate, or read the ebuild for the packages in question to see what's what. Admittedly, some kind of package content search function somewhere would be nice, but I am not sure it's possible (or somebody would probably have done it already). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] nxserver-freenx setup
Grant schreef: > but could you tell me how to see what is in my path and how to add to > it? > To see what your current PATH is: echo $PATH To add to the PATH for the current session: export PATH="/additional/path:$PATH" (I think; make sure that's right-- ok, it is right, but maybe without the quotes according to man export). What I usually do (because I often can't quite remember the correct syntax of an "append to export" comand) is: echo $PATH or export (without parameters), then copy the current PATH to a text file, so if I mess it up I can export PATH="(the original PATH)". Of course I could just read man export like a reasonable person, but who has time for that ;-) ? Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Changing user's Java VM
Hey, ho-- So I used java-config to switch the user's (my) Java VM from blackdown to Sun (blackdown is only 1.4.2 but Azureus prefers 1.5). I don't want to change the whole system VM (because I'm a big chicken). But anyway, when I switch VMs with java-config, I get this message: $ java-config -s sun-jre-bin-1.5.0.03 Env files in /home/motub/.gentoo updated. Source these in your shell's profile. Am I the only one who doesn't know quite what to do at this point? I did figure it out, but I don't find this message at all clear, since the only 'source' I've done up to this point is 'source /etc/profile', which is of course not available to me as a user. So fine, I read the message again, and so tried $ source ~/.gentoo which (unsurprisingly, when you think about it): -bash: source: /home/motub/.gentoo: is a directory Fine, so then I do $ source ~/.gentoo/* and then I was able to get $ java-config -f sun-jre-bin-1.5.0.03 (And now Azureus opens the Configuration tab without crashing, yay!) But I know I was just lucky. I was making this up/guessing wildly as I went along. That command worked, but I don't even know if that's "right" (the asterisk). I suppose the source command sensibly only sourced the "source-able" files in that folder, but I don't know which ones they were, what needed to change, or if everything that needed to change in fact did. An export shows that PATH and JAVA_HOME variables are right (they point to Sun Java), but the JAVAC and JDK_HOME variables still point to blackdown. Is this a problem? I don't know (and I don't like that :-) ). Have I done this correctly? Should I have known how to do this (source files for a user, rather than root/system-wide) after installation (for all my time with Gentoo, I haven't really learned all that many system commands that fall outside the scope of installation routines)? Was there a way to 'intuit' this from the message? This whole ~/.gentoo folder seems new to me. Or should I submit a bug against java-config to request that this message be a bit more specific in its instruction? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge gnome-light
Gustavo Varela schreef: > When I do emerge gnome-light this error appears, any idea of a solution??? > > > > make[4]: Leaving directory > > `/var/tmp/portage/howl-0.9.6-r2/work/howl-0.9.6/src/lib/howl' > > make[3]: *** [all-recursive] Error 1 > > make[3]: Leaving directory > > `/var/tmp/portage/howl-0.9.6-r2/work/howl-0.9.6/src/lib/howl' > > make[2]: *** [all-recursive] Error 1 > > make[2]: Leaving directory > > `/var/tmp/portage/howl-0.9.6-r2/work/howl-0.9.6/src/lib' > > make[1]: *** [all-recursive] Error 1 > > make[1]: Leaving directory > > `/var/tmp/portage/howl-0.9.6-r2/work/howl-0.9.6/src' > > make: *** [all-recursive] Error 1 > > > > !!! ERROR: net-misc/howl-0.9.6-r2 failed. > Remove "howl" from your USE flags (or add "-howl" to the gnome-base/gnome-vfs entry in /etc/portage/package.use)? It's a hack (I have no idea whether you need howl and its dependencies or not), but at least it hopefully allows you to get gnome-light installed, and then you can figure out why it's broke if you do in fact need it. Obviously, if you don't need it, then you're good to go. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] wine
Brett I. Holcomb schreef: > Who's doing the porting and where do we get them. Some time ago I > wanted to get some of the games and all the links pointed to Loki who > was no more. > Loki may be no more, but the Loki Installer is alive and well, and quite able to save one's bacon, using either native games that just have Linux binaries and take the data files from the Windows disks, or Windows games that must be run under Wine or Cedega, but whose installers don't work (properly) under either. See Loki Installers for Linux Gamers at http://liflg.org and to buy what native stuff is available, try TuxGames at http://www.tuxgames.com/ . I find their prices a bit outrageous, but you are supporting the community if you buy from them, so that may make up for it. And on a side note-- Half-Life 2 being ported? I don't think so... I'm sure I heard that HL2, like HL, will *never* be ported. But Cedega tied themselves into a pretzel to get it running (and Steam as well, although I heard that a recent Steam patch broke the whole shebang again), and I know that the Wine developers are beavering away madly to get DirectX 9 support into the package (mostly for WoW, but also for HL2, Myst 4 and Pirates!, those being the games that I've heard mentioned being tested against the incomplete patchset). Not that I care, of course... I have an ATI card using the fglrx drivers, so even half the things which *are* supposed to work under Wine and/or Cedega do not (although native stuff generally runs fine, including Doom 3. Too bad I have no interest at all in playing Doom 3, having never liked the Doom series). So I try not to get all excited about things that will work in the near future, as nice as it would be to finish Myst 4 (and Uru). Until ATI gets their driver act together (or I find the money to buy nVidia), I'm kinda SOL in any case. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] wine
[EMAIL PROTECTED] schreef: > Thank you for the explanation. That clears it up. I don't play the Doom > series either - I have Soldier of Fortune and Half-LIfe along with some > others. And I have an Nvidia card . > Yes, well, I hate you :) . But it is important to mention on behalf of any other idio-- umm, unfortunate ATI users-- like myself, that our regrettable choice of video card also has an effect on the situation. But as to the original question, here are some important resources: The Wine application database at http://appdb.winehq.org This resource is not well-maintained atm, but it's trying to come back, so if you are a Wine user, please register, log in, and update application information, or even apply to maintain an application so that the developers can easily see if an app regressed after a monthly release. Wine now also has a Wiki at http://wiki.winehq.org/FrontPage, and the "user wiki" is at http://wine-wiki.org/ . Naturally, this also needs user support to survive, so any Wine users might consider contributing. The Transgaming games database at http://transgaming.org/gamesdb/ . These resources require you to be a TG subscriber to contribute, but anyone can read the data already provided. It's also not that well-maintained (a lot of listed games don't have any info because no one has tested them yet), but it is at least a place to start. TG also has a Wiki, and if the game in question has an entry there, it is linked in the gamesdb, so that's nice. Codeweavers (Crossover Office) also has an application database at http://www.codeweavers.com/compatibility/ . But I must say, in my experience with the CX demo, they hardly need one-- only one "unsupported" app that I threw at it did not run (CloneCD; but it installed fine)-- and that just doesn't run under Linux, period, afaict. I've found Crossover to be really great, pretty easy to use, looks like a real program, and if it ran games instead of only applications that I have no use for (such as Office and Access and the like), I would buy it in a minute. In fact, I'm almost sorry that I have no use for it, because it's the kind of thing I *want* to buy, just to show my support for how good it is. So for the applications that the OP wants to run (namely Office), I would recommend it, but if the OP doesn't want to spend any money, I would recommend Wine with WineTools: http://www.von-thadden.de/Joachim/WineTools/ which should make it easier to get Office installed and running if you really must have it. However, WineTools vastly prefers the "last stable for Office" Wine version, which is considered by the developer to be 20041019 (which is still available in Portage). However, the developer does welcome information on any further applications that users successfully install and run with WineTools, so they can be included in the supported list, which can be found at http://www.von-thadden.de/Joachim/WineTools/wt212jo.html As you might have guessed, I'm a pretty big Wine-head, and follow developments fairly closely. As such, on the whole I would say that no version or variant of Wine is (at this time, but that may be changing, as soon as within the next quarter) sufficient for all cases, unless you have very specific (and limited) needs. If you don't play games, for example (or only simple ones not requiring DX support), then Crossover (or Wine and WineTools) is probably sufficient. If you play games, but only older ones (I'm re-playing Deus Ex and Septerra Core atm), then Wine alone may be sufficient, but you may require more than one version (an installed binary and a compiled CVS, for example), as things have been known to break (regress) from month to month. On the other hand, Wine can really surprise you sometimes with what runs "out of the box"-- on one of my "live on the edge" days, I installed In Memoriam (Missing: Since January in Europe), which is only a year and a half old, under either Wine 20050111 or 20050211. I didn't really expect it to work, especially since the game requires an Internet connection, both because you get a FreeWeb account (which enables a password to be sent to you so you can even play the game), and because it also contains a lot of embedded links in the puzzle graphics to get clues (it's a "pre-murder" mystery adventure game, and I was having Adventure Game Withdrawal Syndrome, which is why I was trying this at all). To my complete and total shock, everything worked fine. I think I may have had to install the Shockwave Flash Player by hand under Wine (but had I been playing this under Windows, I would have had to do that as well, so I don't count that as "tweaking"). But it worked. I told the game to use my GMail account, and it sent the password there (meaning it got the information correctly, because Wine was able to use my DSL-via-the LAN connection properly to send it). Even more amazing, I didn't have to install IE or any such thing-- I clicked on a link in the game, and the game o
Re: [gentoo-user] sylpheed-claws and GTK2+ ?
raptor schreef: > i see there is a port of sylpheed-claws to gtk2+, but dont see an ebuild for > it ? > Does anyone used this version, is it already functional ? do u have some > ebuild ? > Can it coexist with the older version ? > > thanx alot in advance Yes there is an ebuild (1.99), it's just masked (hard-masked and ~x86 masked). It is functional in the basics, but all of the available plugins have not been updated to work with this version (and those that are have also been masked). The GTK and GTK2 versions can apparently be installed side-by-side (at least, the first time I recently installed the GTK2 version, the GTK version installed as well, but I don't think it was as a dependency since I was later able to install just the GTK2 version by itself), but it looks like one overwrites the other (afaik, the binary names are the same), so I wouldn't suggest installing both of them. At this time, the GTK2 version is not useful to me without either a plugin that is not yet available, or changing my mail system from POP3 to IMAP, and I hate GTK1 (it's just too ugly for me and I try to keep it to the bare minimum), so that's all I know-- I have Sylpheed-Claws GTK2 installed, but haven't used it. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] wine
[EMAIL PROTECTED] schreef: > Some of us are just lucky . I gave up ATI after the Mach 64! > Unfortunately, we all can't just go out and upgrade - I know that > for sure! > > Thank you for the excellent summary - this will be filed away and > when I find time I'll go visit these sites. I did use Wine and Winex > considerably about a year ago - I was doing CVS from TG (wrote a > Gentoo ebuild) and giving them feedback Obviously before they made everybody (Gentoo, Debian, etc) stop that horrible "easy install of CVS version" business. Which I could see their point about (even in some respects that I could agree with, alongside the many respects in which I didn't agree at all), but it certainly didn't do anything to help their reputation, which is still pretty bad, even among subscribers, and quite justifiably so, imo-- I certainly don't like them, and I am willing to go quite a few extra miles to avoid using their product). > and I did a bunch with Wine CVS on Gentoo - ebuild there, too and bugs > to Wine bugs . I used the wine apps db but as you said it was pretty > much dead; used Frank's Corner, too. I had Half-Life and Soldier of > Fortune running Isn't SoF native? http://www.linuxjournal.com/article/4381 ... Yes, it's available at TuxGames. > and was working on SOF II and then due to time (Work, family, and > solving a problem on my main Linux box that has been haunting me for a > couple of years) I had to stop working on it so I've been out of it for > a while. For me Wine is about running everything I can under Linux so I > can decrease my use of Windows - I will still have to have Windows for > the InstallShield programming I do and for MS Flight Simulator since I > beta test that but I'd love to have everything else on Linux. I use > OpenOffice so running office on Linux is not a consideration. Same with IE > - I avoid it on > Windows so I sure don't need it on LInux .. Well, of course IE is a special case. I would never use it as a web browser under Linux (I didn't even use it under Windows unless I had to for those few sites that really only would load in IE), but 1) Some programs (like Office, but Office is far from the only one) require the IE backend before they will install or run; 2) If you do web development, you need IE to test your site or page. After all, it's still a majority of the market, so you'd have to know how your page renders in it. But for development like the kind you're doing, it almost sounds like it might be worth it to have VMWare (not sure if that would help with MS Flight Simulator, though). > > After trying TG and Wine I decided I would use Wine first above all > else and not use TG. I'd see if I could get the apps working in > Wine and submit bugs as I found them. Now there's Cedega but I > really don't want to pay for anything and I'd like to support Wine > first.. I don't mind paying if the product is worth it, but there are just too many things I don't like about Cedega-- can't try before you buy (and they don't even accept PayPal, so I actually can't buy, as I don't have a credit card); don't like how they decide what to support; don't like how if something "not important" (i.e., not "super-game of the quarter") breaks, you can just go jump; not too happy with their statement that if "thus-and-so" doesn't work with ATI cards, you can pretty much go jump, as I suspect that at least some of the problems could be worked around if TG really communicated with ATI (which "go jump" does not imply is occurring); they're not really the best FOSS supporter for me to twist myself into a pretzel to pay them $15 USD (converted to Euros to pay to a friend with a credit card) in order to buy their product without even a firm assurance that I will get what I paid for (given that I have an ATI card). Codeweaver's Crossover Office costs money too, but they're a great FOSS supporter (plenty of their developers develop Wine), they support Wine financially, they have a demo, and their product is quite good. So it's not just a matter of money, but I prefer to support Wine, because I *can* support Wine. I can lurk on the dev list and see what's in the pipeline. I can add to the appdb and the Wiki. I can give and receive the benefits of FOSS, which are worth more to me than being able to play Half-Life 2 (despite the fact that my boyfriend bought it on the day it came out-- before the boxes were even unpacked to the shelves-- and it's driving me slightly bats not to play it for myself. Nothing said about Myst 4 or The Sims 2, which are both dearer to my personal heart). But, like you, I will suck it up. I will not install Windows (I *hate* dual-booting). I will wait. The nice thing about Wine is that (regressions aside), it really gets noticeably better quite fast-- and because it's a "generalist" rather than a "specialist", when things get better, they get better across the board (a big swath of DX9 games will suddenly work, not just KOTOR 2 or HL 2--
[gentoo-user] [OT]Any multi-gnome-terminal users here (GUI-related question)?
Hi, Just wondering if anyone might know what's the deal here, as the docs on the MGT site don't seem to be helping, nor the MGT ML archives. I'm running MGT under KDE, because GNOME is broken (again, don't ask, no idea). I love MGT, but it is an (as always, hideously unnatractive) GTK 1 application. I have GTK-QT installed, and GTK apps are supposed to be using it... The Gimp does (although I'm not sure if my Gimp is GTK 1 or 2), but MGT just won't. I'm using Mosfet's Liquid as my KDE theme, and I've installed both GLiquid and XLiquid for GTK 1 from kde-look.org, I've changed the theme in both gtk-chtheme (which is GTK2), switch2 (which is GTK1), and even in the GNOME theme manager (which I managed to run under KDE), but nothing seems to affect the style of MGT (the colors are fine, weirdly enough). None of the command-line options displayed by --help seem to be relevant. I looked at the screenshots on multignometerm.sourceforge.net, and apparently MGT does respond to theming-- at least the screenshots display GTK styles that I'm familiar with, but not on my PC. Does this by chance ring a bell with anyone as to what the problem might be? Since the colors are OK, I can live with it, but it's freaking me out mildly. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] kernel building tools
Mike Owen schreef: > On 5/25/05, James <[EMAIL PROTECTED]> wrote: > >>Hello, >> >>Is it OK to use 'make oldmenuconfig' to ensure that the options I had >>selected in a 2.6.x kernel also are selected for the newer 2.6. kernel? >>Isn't 'make oldmenuconfig' deprecated for 2.6 or does it still work? >> >>Also I perviously used xconfig (make xconfig) in lieu of make menuconfig, >>but I cannot seem to find anything other than menuconfig. Surely >>there is a nicer gui to use to build kernels and track options selected >>in various kernel builds than the ole standby 'make menuconfig'. >> >> >>ideas? >> >> >>James >> > > > I typically just use "make oldconfig", as the number of changes from > one version to the next aren't that great normally. Doing a make > oldconfig will prompt you for each new feature, so once you have your > baseline kernel set, make oldconfig is real quick. When moving between > different -rX versions, it often won't prompt at all. > > Mike > I've actually been upgrading within a series (gentoo-sources, -r6 to -r8), and across series (to ck-sources and mm-sources)... and been paying attention closely to this exact issue when I did so. And what I noticed is that when I switched the symlink and ran make menuconfig, this was the message: mutable linux # make menuconfig (first time config of newly installed mm-sources, booted into gentoo-2.6.11-r8) HOSTCC scripts/basic/fixdep HOSTCC scripts/basic/split-include HOSTCC scripts/basic/docproc SHIPPED scripts/kconfig/zconf.tab.h HOSTCC scripts/kconfig/conf.o HOSTCC scripts/kconfig/kxgettext.o HOSTCC scripts/kconfig/mconf.o SHIPPED scripts/kconfig/zconf.tab.c SHIPPED scripts/kconfig/lex.zconf.c HOSTCC scripts/kconfig/zconf.tab.o HOSTLD scripts/kconfig/mconf HOSTCC scripts/lxdialog/checklist.o HOSTCC scripts/lxdialog/inputbox.o HOSTCC scripts/lxdialog/lxdialog.o HOSTCC scripts/lxdialog/menubox.o HOSTCC scripts/lxdialog/msgbox.o HOSTCC scripts/lxdialog/textbox.o HOSTCC scripts/lxdialog/util.o HOSTCC scripts/lxdialog/yesno.o HOSTLD scripts/lxdialog/lxdialog scripts/kconfig/mconf arch/i386/Kconfig # *# using defaults found in /boot/config-2.6.11-gentoo-r8* So in fact, makeoldconfig appears to be a bit obsolete, as the new kernel used my old kernel settings as the default. This may be because I had installed my kernel with make install, so there *was* a config to be found in /boot, but it did save a lot of work. For changing series, I definitely prefer to look at the kernel config anyway, to see what options the series' patches have added, but for upgrading within the same series, if nothing has really changed, it's nice to know you can just make menuconfig, save immediately (because you don't have to do anything, but you can also confirm that the settings really are the same if you like), and head right into make modules_install, etc. I find that actually faster than selecting X number of new (and usually irrelevant to me) options with make oldconfig anyway (n, n, n, etc). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] kernel building tools
Neil Bothwick schreef: > On Thu, 26 May 2005 10:24:52 +0200, Holly Bostick wrote: > > >>For changing series, I definitely prefer to look at the kernel config >>anyway, to see what options the series' patches have added, but for >>upgrading within the same series, if nothing has really changed, it's >>nice to know you can just make menuconfig, save immediately (because you >>don't have to do anything, but you can also confirm that the settings >>really are the same if you like), and head right into make >>modules_install, etc. I find that actually faster than selecting X >>number of new (and usually irrelevant to me) options with make oldconfig >>anyway (n, n, n, etc). > > > The disadvantage of this approach is that you don't get to see what new > options have been added. > > Within the series, I probably don't care (i.e., moving from gentoo-sources -r6 to -r8, which I only even did because -r6 was removed from Portage). For changing series (moving from gentoo-sources to mm-sources), I do care, but that's why I manually go through the menuconfig in that case. $DEITY knows I've compiled enough kernels to recognize most new stuff, and if I'm not sure, I read the option's Help, as I've always found it invaluable. But my hardware is all pretty old-- KM266A mobo/Athlon XP 2200+ (so no hyperthreading and no amd_64), no SATA, no PCI-e, only usb devices a mouse and gamepad, no mp3 player/iPod, no usb keys (no usb 2.0, even), no external hdds or devices, no bluetooth or firewire, no mobile phone, external sound card is a C-Media 8738 (second only to my onboard VIA 8233 for bog-standardness), no TV-in or out (unless I try out the ati-gatos drivers for TV-in, but I don't feel like being bothered yet), no wireless, heck, no scanner or printer connected to this box (yes, I'm a dinosaur, but I get along)-- so most new options really don't make the first bit of difference to me, as the kernel options I need have been stable for eons. The major ones that do change (video card, the eternal automount-patch war between supermount, submount and hal/dbus/ivman), are external, and for kernels like ck-sources, the new stuff (schedulers) are generally defaulted on anyway, which is fine by me. So I mostly just check for sightseeing purposes (to see what's available/stablilizing for future upgrade planning), and on the off chance I want to try something out, as well as to confirm that the options that I must have turned off (DRM, Registers) are in fact off. Afaics, I don't need to chase the future with the kernel; I just keep it more-or-less up to date for increased stability and patch issues, and looking for speed increases. But obviously, others may have different needs. Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Unmerging "extra" python
Yes, I've done something dopey. Checked the forums but didn't see anything helpful before the fact, as it were. The short version is that an program I emerged from overlay wanted Python 2.4 (I have 2.3.5 already). The Python dep was installed in a new slot, but the program wouldn't compile anyway, so in a belated burst of sanity, I removed the python USE flag, after which it compiled fine. Great, but now I've got this extra Python on my system that nothing uses. I did do a python-updater-- also stupid--but the one program that seemed broken on this basis removed the "empty bytecode" in the 2.4 folder when I re-emerged it, which fixed that program (don't remember what it was atm). The program that remains broken is Pysol. I've re-emerged it, but it just won't start, claiming that it can't find pysol_24.pyc. OK, I understand that (only pysol_15 through pysol_23.pyc exist). I'm figuring that Pysol either is not ready for 2.4 or is confused by the new slot. Even if this was fixable, I don't want to upgrade Python to 2.4 at this time, nothing needs Python 2.4 (since the program that wanted it is now compiled without Python support, which I hope will be OK), and its presence seems to be breaking things. But if I try to emerge -Cav =dev-lang/python-2.4.1 (presumably leaving 2.3.5 intact), emerge throws me a message so alarming that I CTRL-C before the unmerge can begin: !!! Trying to unmerge package(s) in system profile. 'dev-lang/python' !!! This could be damaging to your system. >>> Waiting 10 seconds before starting... >>> (Control-C to abort)... Press Ctrl-C to Stop in: 10 9 8 7 6 Now of course I know that Portage depends on Python, and I certainly don't want to mess up Portage, so how do I get rid of this version of Python (or how do I recover if there's no way to get rid of it without breaking anything)? Any help appreciated, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] sylpheed-claws and GTK2+ ?
Neil Bothwick schreef: > On Tue, 24 May 2005 15:41:16 +0200, Holly Bostick wrote: > > >>At this time, the GTK2 version is not useful to me without either a >>plugin that is not yet available, or changing my mail system from POP3 >>to IMAP, and I hate GTK1 (it's just too ugly for me and I try to keep it >>to the bare minimum), so that's all I know-- I have Sylpheed-Claws GTK2 >>installed, but haven't used it. > > > What plugin do you need? I'm using 1.9.11 (by renaming the 1.9.9 ebuild) > with POP3 with no problems. > > I believe I need mailboxmdir-- I want to share my current mail folder from my current Thunderbird with Sylpheed-Claws, until I determine if I want to switch. According to package.mask, the version of this plugin that works with 1.9.9 would be masked (and was masked during my original borked install), but now does not appear in Portage at all, masked or otherwise. This seemed to happen about the time the sylpheed-gtk2 package was upgraded (like the day after I had installed it), so I assumed that the plugin broke, and thus was removed. But since I don't know what I'm doing in this regard, I could be wrong. I did ask the list how to deal with this, and it was suggested to just switch to IMAP, but I know even less about that :) . Naturally, this whole thing is not urgent, as Thunderbird is working fine for what I ask of it, but Sylpheed-Claws seems as if it might have features that I definitely have noticed that Thunderbird lacks. But I don't so much want to convert my current mailbox folder or have to manage a copy of same, without knowing if the program suits me. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Unmerging "extra" python
Jason Stubbs schreef: > On Thursday 26 May 2005 23:36, Neil Bothwick wrote: > >>On Thu, 26 May 2005 16:24:28 +0200, Holly Bostick wrote: >> >>>Now of course I know that Portage depends on Python, and I certainly >>>don't want to mess up Portage, so how do I get rid of this version of >>>Python (or how do I recover if there's no way to get rid of it without >>>breaking anything)? >> >>Portage depends on Python 2.3, unmerging that and leaving 2.4 could break >>things, the other way round shouldn't. > > > Portage works with >=python-2.2. Unmerging either 2.3 or 2.4 should not cause > any troubles with portage at all as the libraries and scripts live outside > the python directories. OK, that's reassuring-- but what still concerns me a bit is that when I request to unmerge a specific version of Python, Portage: 1) does not confirm (before getting hysterical) the version I am unmerging; 2) gets all hysterical as if I am uninstalling a mission-critical application -- which it "knows" I am not, because the original ebuild was not only installed to a new slot, but also gave me a statement that my original Python was being retained for compatibility, etc, etc. Of course I know that that's just an echo of something the ebuild writer typed, but still, it does seem as if Portage has the ability to recognize that a valid version of Python will remain installed after my operation is complete, and the alarm with which it reacts is confusing insofar as I cannot be certain that the correct version is about to be uninstalled, and there is no way for me to 'remove' or verify which Python is being considered a critical system application (my virtuals seem to only say "Python", not a particular version thereof, and if any other such system files specify a particular version of Python, I don't know what file that might be. Naturally, a qpkg search gives me the same list of reverse dependencies for both versions). > > >>However, it is worth playing safe, so use quickpkg to build a binary >>package of any potentially important package before unmerging it. Even if >>you break portage, you can still install a binary package with >> >>tar xf /path/to/package.tbz2 -C / > > > This is always good advice. Yes, and I'll be taking that, too. So I should quickpkg 2.3? and Portage? Fair enough. Thanks, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Unmerging "extra" python
Jason Stubbs schreef: > On Friday 27 May 2005 01:07, Holly Bostick wrote: > >>Jason Stubbs schreef: >> >>>On Thursday 26 May 2005 23:36, Neil Bothwick wrote: >>> >>>>On Thu, 26 May 2005 16:24:28 +0200, Holly Bostick wrote: >>>> >>>>>Now of course I know that Portage depends on Python, and I certainly >>>>>don't want to mess up Portage, so how do I get rid of this version of >>>>>Python (or how do I recover if there's no way to get rid of it without >>>>>breaking anything)? >>>> >>>>Portage depends on Python 2.3, unmerging that and leaving 2.4 could break >>>>things, the other way round shouldn't. >>> >>>Portage works with >=python-2.2. Unmerging either 2.3 or 2.4 should not >>>cause any troubles with portage at all as the libraries and scripts live >>>outside the python directories. >> >>OK, that's reassuring-- but what still concerns me a bit is that when I >>request to unmerge a specific version of Python, Portage: >> >>1) does not confirm (before getting hysterical) the version I am unmerging; > > > --pretend or --ask as per usual. For a system package, this does not seem to work: mutable root # emerge -Cav =dev-lang/python-2.4.1 cfg-update 1.7.1 : Building checksum index... (takes a few seconds) done! >>> These are the packages that I would unmerge: !!! Trying to unmerge package(s) in system profile. 'dev-lang/python' !!! This could be damaging to your system. >>> Waiting 10 seconds before starting... >>> (Control-C to abort)... Press Ctrl-C to Stop in: 10 9 8 7 6 Exiting on signal 2 As you see, I did use -a, and maybe it was going to ask me in 4 more seconds (as I do get the correct "These are the packages I *would* unmerge"), but the very fact that there is an immediate countdown (which does not happen when you are "asking" first) is scary enough that I would never let the countdown finish just to see if Portage really *was* going to wait for me to confirm before proceeding. At least not without doing the quickpackages first. It's just unexpectedly contradictory behaviour. > > >>2) gets all hysterical as if I am uninstalling a mission-critical >>application -- which it "knows" I am not, because the original ebuild >>was not only installed to a new slot, but also gave me a statement that >>my original Python was being retained for compatibility, etc, etc. > > > $ grep python /usr/portage/profiles/base/packages > *dev-lang/python > > Portage checks the package you are about to unmerge against the list of > "system" packages and then warns if it matches. Slots aren't taken into > account at all but probably should be. That would likely reduce the "mixed-message factor" I seem to be sensing here. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Unmerging "extra" python
Jason Stubbs schreef: > On Friday 27 May 2005 09:36, Holly Bostick wrote: > >>>>>These are the packages that I would unmerge: >> >>!!! Trying to unmerge package(s) in system profile. 'dev-lang/python' >>!!! This could be damaging to your system. >> >> >>>>>Waiting 10 seconds before starting... >>>>>(Control-C to abort)... >> >>Press Ctrl-C to Stop in: 10 9 8 7 6 >> >>Exiting on signal 2 > > > This is a bug. It will give the normal --pretend output after the countdown > is > finished, but there shouldn't be a countdown at all when doing --pretend. > > >>>Portage checks the package you are about to unmerge against the list of >>>"system" packages and then warns if it matches. Slots aren't taken into >>>account at all but probably should be. >> >>That would likely reduce the "mixed-message factor" I seem to be sensing >>here. > > > Okay. So no countdown during --pretend as well as using slot checks. Care to > open a bug so I don't forget please? :) > > Regards, > Jason Stubbs Just one? Shouldn't it be two (one for the countdown and one for the slots)? In the name of neatness (since I wouldn't know how to reasonably label a multi-headed bug like that anyway?). OK, on my way (dancing) to b.g.o (I found a bug!! Yay, me!! Helping the community, hoorah!) BRB with the bug numbers. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Unmerging "extra" python
Jason Stubbs schreef: > Okay. So no countdown during --pretend as well as using slot checks. Care to > open a bug so I don't forget please? :) > > Regards, > Jason Stubbs http://bugs.gentoo.org/show_bug.cgi?id=94131 http://bugs.gentoo.org/show_bug.cgi?id=94132 Glad to be of service :-) . Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Odd behaviour from azureus-bin
This is probably an upstream issue (if it is an issue), but I'm wondering if anyone else is seeing this. I have azureus-bin 2.3.0.0 installed from the ebuild, and I run it under KDE 3.4.0. I have blackdown set as the system Java VM, but I use java-config to change the user VM to Sun before running azureus, as azureus vastly prefers 1.5.0 to 1.4.2 (but 1.5.0 is not appropriate for most other Java apps, so I don't want to change the system VM). Now, here's the weird part. If I run azureus from the KDE menu entry, the program runs, but all kinds of crashes occur, mostly related to the system tray: 1) If Firefox or Thunderbird (usually both) are docked into the system tray using alltray, one or both of them will crash silently at some point within an hour or two. Thunderbird can usually be restarted (crashes cleanly); Firefox has to be killed from KSysGuard before restarting (crashes dirty). 2) Azureus itself (which is also docked to the system tray), may also crash silently for no apparent reason, on those occasions I remembered to close Firefox and Thunderbird before starting it. When restarting, it may or may not mention that it crashed (usually you get a little box saying that Azureus did not close cleanly and you should check the log to find out more). 3) Attempting to open the Configuration tab (not the Wizard, which works fine but is limited in scope, but the full-bore tab), without fail causes a silent crash to desktop (so I can't even tell Azureus not to dock to the system tray). But if I run Azureus from a terminal, none of this happens, and I can open the Configuration tab. It's slowish, but it does open, and the program does not crash. I cannot figure out why this should be or where this bug might originate from. The original KDE menu entry was 'azureus --ui=swt', but I even changed that to simply azureus, and turned off startup-notification, and the behaviour didn't change. Next I'll try checking "run in a terminal", but I can't test that until this session finishes (as this house of cards is fragile enough that I'd like to get the result while I can). I also recognize that the 3.4.0 Kicker panel and system tray are in and of themselves fairly buggy (at least on my system they crash regularly for no apparent reason), but that doesn't explain why a completely separate program should exhibit more instability when being run from the K-menu than when being run from a terminal, using the exact same command (except that when I run it from a terminal I append '&' . Surely that can't be it). I know we're waiting for the ebuild for 2.3.0.2, and I know that's a bugfix release, but if the site had a Changelog, I didn't notice it. And of course we're waiting for KDE 3.4.1, also a bugfix release. So maybe one or the other will solve this, but at the moment I can't even speculate what update I need to be anticipating. Anyway, does anybody have any idea what might be causing this weirdness, and have a suggestion as to what I can do, or to whom I should report, if reporting seems in order? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Multipurpose compression prog with GUI
Ryan Viljoen schreef: > Hi > > I am looking for a compression program that covers creating and > extracting archives including .zip , .tar , .tar.gz , .tar.bz2 etc. I > am using fluxbox with rox. > > Thanks > Rav > I've gotta say, I'm very fond of file-roller (the GNOME archive utility). I know that it comes with the baggage of several, possibly unwanted, GNOME-related dependencies, but the 2.10 version opens *every kind of archive*-- including p7zip (assuming you install p7zip for Linux, which is available via Portage, or the relevant backend for the archive type, such as unace or rar)... which arK (the KDE archive utility) does not (maybe p7zip support will be in 3.4.1 or 4.0, but why wait?). Alternatively, WinRAR, which also opens pretty much everything (archives, *.iso files, *.cab files, you name it), including p7zip, runs quite well under Wine, and of course avoids the whole issue of installing additional GNOME libraries. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: kernel building tools
James schreef: > Neil Bothwick digimed.co.uk> writes: > > > >>make menuconfig >>make && make modules_install && make install > > >>is easier and accomplishes the same, and also updates the vmlinuz and >>vmlinuz.old symlinks in /boot, removing the need to alter your grub >>config. > > > > OK, I but often I like to keep the older kernel versions and grup entries > around for a while to test the differences in the various kernels. Won't > this approach overwrite entries in grub or does it just make another > entry? Some of my grub.conf files get pretty ugly... Neither. Make install does not write to grub.conf-- you have to do that yourself. If you were using LiLO, that would be a different story. However, what Neil is on about (again :-) ), is that a "properly" set up grub.conf would ideally point to vmlinuz for the current kernel, and vmlinuz.old for the previous kernel. If you don't use specific titles that include a kernel version number, but instead generic titles such as "Gentoo_current" (pointing to vmlinuz) and "Gentoo_previous" (pointing to vmlinuz.old), then you will not have to do any editing of grub.conf after installing a new kernel with make install, because the currently running kernel will be symlinked to vmlinuz.old, and the new kernel will be installed and symlinked to vmlinuz. Any older kernels yet will not be symlinked, but can be accessed directly (vmlinuz-2.6.9-gentoo-r8, for example), as their System.map and .config will also be named with the complete version number. This does mean you have to manually remember which kernel is newly installed and which one is "previous", but that's the price you pay. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] portage v. yum with regards to java
Ciaran McCreesh schreef: > On Sat, 28 May 2005 12:43:49 -0500 "Boyd Stephen Smith Jr." > <[EMAIL PROTECTED]> wrote: > | Besides the /opt vs. /usr problem, I don't like how I have to > | delete /mnt/floppy and /mnt/cdrom everytime baselayout (?) is updated, > | since the "correct" place (and the one I use) for these mount points > | is /media/floppy and /media/cdrom. > > /media considered silly. > I see your "silly", and raise you "confusing", and "annoying", but what is one supposed to do about dbus/hal in that case, since no matter how silly, confusing, and annoying, /media/foo is the mount points its going to make... automatically, even? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Printer setup tool
[EMAIL PROTECTED] schreef: > > None of the other logs have anyhting that jumps out at me, but this would > imply that cupsd is dying because my machine is naming itself "tux" and > cups cannot determine that means Localhost here. Any gurus know how I > should go about fixing this? > > Creighton > Sounds like you need to edit /etc/hosts to tell the system that tux is a synonym for localhost: 127.0.0.1 localhost.localdomain localhost tux ought to do it. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how to delete messages from server in Kmail
askar ... schreef: > I see. Very sad. > > Do other mail clients like sylpheed-claws or thunderbird support this feature? > > askar > Sorry, askar, just got around to checking this. YES, Thunderbird does support this feature: (translated from Dutch, may not be exact, but should be close enough) Edit=>Account Preferences=>Server Preferences: Check : "Leave messages on server" and the then-activated sub-checkbox: "Until I delete or move them from the Inbox" Don't know the effect if you use message filters, but probably not good (since those would move the messages from the Inbox). But still closer than you are, and better than setting up an additional PC to enable IMAP, of all things. Sylpheed-Claws (GTK2) does not seem to support this feature "out-of-the-box", but a plugin may-- I'm not too familiar with Sylpheed-Claws yet. Hope this helps, Holly > On 5/29/05, Dmitri Vassilenko <[EMAIL PROTECTED]> wrote: > >>On Saturday May 28 2005 23:26, askar ... wrote: >> >>>Does anybody know how to delete messages from server when the messages >>>deleted in Kmail ? -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] kde 3.4.1?
Volker Armin Hemmann schreef: > On Monday 30 May 2005 12:11, Jan Meier wrote: > >>>Anyone have an idea when this show goes on the road? >> >>There is actually no release announcement at kde.org. >>Strange that there is a ebuild for that. >> >>Greets >>Jan > > > well, it is usuall, that the ebuilds are ready, before a kde release is out - > hardmasked. > > Maybe he unmasked completly kde3.4.x, or some dev unmasked 3.4.1 by accident, > but I would not care too much about it. 3.4.1 should come soon ;) According to /usr/portage/profiles/package.mask, the tarballs aren't available yet (so I didn't bother unmasking). I thought maybe the ebuilds were nonetheless in Portage because the devs have some secret source for preliminary testing or something. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Unmerging "extra" python
Ow Mun Heng schreef: > On Thu, 2005-05-26 at 18:07 +0200, Holly Bostick wrote: > >>Jason Stubbs schreef: >> >>>On Thursday 26 May 2005 23:36, Neil Bothwick wrote: >>> >>> >>>>On Thu, 26 May 2005 16:24:28 +0200, Holly Bostick wrote: >>>> >>>> >>>>>Now of course I know that Portage depends on Python, and I certainly >>>>>don't want to mess up Portage, so how do I get rid of this version of >>>>>Python (or how do I recover if there's no way to get rid of it without >>>>>breaking anything)? > > >>Yes, and I'll be taking that, too. So I should quickpkg 2.3? and >>Portage? Fair enough. > > > > Just for your info, I went down this path like 2 weeks ago when I wanted > something new (i can remember what was it but I reckon it was either > gdekslets or some other app) and I just unmasked 2.4 and 2.3 stayed > back. > > Of course I was surprised to still see 2.3 there so I promptly unmerge > it. Did I see issues? You Bet! Epylog didn't work and so didn't > gdekslets. > > The solution was to re-emerge both of these (revdep-rebuild didn't help > at all) and then everything works. > > That's my experience. > Thanks for the tip-- in the case that I want to emerge the newest gDesktlets in order to try that spiffy new RSS tickerbar and get off of KDE finally (I really like having a ticker and specifically a scrolling bar, and Knewsticker doesn't so much like running under other DEs), and so wind up with the same situation again, I'll keep it in mind. In any case, I did quickpkg python-2.3.5 and portage, but it was not necessary to use the backups; the unmerge of Python 2.4.1 went fine and all I had to do was re-emerge pysol and java-config. No other applications were disturbed at all. Thanks, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] firefox sage
maxim wexler schreef: > Hello everyone, > > on my former "main" box, K6-II, 500MHz I've managed to > get to startx and, since grokking dialup, am > web-surfing. Here's the issue: when Firefox opened up > the first time it invited me to upgrade which I did by > clicking the link provided. But now when I navigate to > Firefox Extensions and click on Sage install, nothing > happens. > > Is there an ebuild I need? On a related note: as I > discover other programs I like, what is the procedure > for installing them if they are not found among the > distfiles or gentoo pkges? Do you just unpack and run > the install scripts a la slack-ware? > > -mw > Hi Maxim, Really, there are very few programs that you (as a general user) might like that are not found within Portage. Certainly Firefox is-- as a precompiled binary as well as a source tarball. So frankly I don't know what has happened when you upgraded some otherwise-installed Firefox via some internal Firefox script to upgrade to some unknown version. Meaning, I don't know what files and folders you actually have, rather than the defaults installed by Portage. Certainly I've had similar problems when my Portage-installed Azureus client updates itself via the update utility (rather than through Portage). It's just not a good idea. That said, the first thing that occurs to me is that the Sage extension (and possibly all of your extensions) may only be installable by root, because the file where this "global" extension is kept is only writeable by root. Most extensions install to ~/.mozilla/firefox/extensions, but some require being in the application's directory rather than the user directory, due to needing access to internal program functions. In the case of a non-standard install, it's quite possible that all extensions need to be installed by root. What you describe is the behaviour I've sometimes seen when an extension must be installed by root and cannot be installed by a user (depends on extension; some of them try and fail to install, some won't even try). It's also possible that the download has been blocked; the bar that appears on top of a page when Firefox blocks a download from a non-authorized site is easy to miss. If this is the case, it's also easy to fix; the bar tells you where to click and then displays the Reject/Allow dialog, so you can easily allow the site through (if you trust it). By default, the only site allowed is the new https site (Mozilla Update), so if you aren't using that, you probably should be. However, I know of several other "semi-official" sites, such as Extension Room, which used to be official but are now "deprecated" by the new site, yet are still useful on occasion. Such sites will be blocked for downloading by the browser until you authorize them, but I see no issue with authorizing them if needed or wanted. So it's probably one of those two things. But if I was you, I would emerge sync and install Firefox through Portage before going any further anyway. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Printer setup tool
[EMAIL PROTECTED] schreef: > It no longer tells me that it cannot find "tux" but it still says that > port 631 is aready in use and dies. > > Creighton > I would first open the CUPS administration web interface (http://localhost:631 in your web browser) and see if the printer had some stuck jobs in the queue, and stop them if so. If that didn't work, I might even go so far as to remove the installed printer, stop CUPS, restart CUPS, reopen the web interface and reinstall the printer. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Printer setup tool
[EMAIL PROTECTED] schreef: >>>[EMAIL PROTECTED] schreef: >>Holly Bostick wrote back: >>>It no longer tells me that it cannot find "tux" but it still says that >>>port 631 is aready in use and dies. >>> >> >>I would first open the CUPS administration web interface >>(http://localhost:631 in your web browser) and see if the printer had >>some stuck jobs in the queue, and stop them if so. If that didn't work, >>I might even go so far as to remove the installed printer, stop CUPS, >>restart CUPS, reopen the web interface and reinstall the printer. > > I can't, cupsd dies quickly. > > > Well, that's a problem. At what point does it die (what are you doing when it dies), and what does it say with its dying breath (error message)? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] firefox sage
Christoph Gysin schreef: > maxim wexler wrote: > >>Yes, thanks, emerge --sync is awesome! But must it run >>so long? I started more than an hour ago and it's >>still churning away. I noticed it started another >>server(went from Xeon to P4) and kept on going. Does >>it know enough to stop? Is it repeating itself? At >>some point I'm going to need to use the phone; what'll >>I do then? > > > You can safely abort emerge sync with Ctrl-C while it is syncing with the > remote > server. It will continue updating your tree the next time it gets invoked. > After the sync, while updating the portage cache (no network needed for this) > you could suspend it from your shell witch Ctrl-Z. Resume it with 'fg' > > Christoph But isn't emerge --sync a different command from the more traditional emerge sync ? And could that be causing the problem? If not, my questions would be 1) What is the speed of your Internet connection? 2) What is your geographical location? 3) What mirrors are you setting in /etc/make.conf? On my PC (AthlonXP 2200+, ADSL via LAN, located in the Netherlands and using the Europe mirrors to sync via SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"), an emerge sync takes about... just a minute, I'll do one now and tell you in a minute 10 minutes: Tue 05/31/05 15:05 /usr/local [EMAIL PROTECTED] -> emerge sync cfg-update 1.7.1 : Building checksum index... (takes a few seconds) done! >>> starting rsync with rsync://140.105.134.102/gentoo-portage... >>> checking server timestamp ... - .*. / University of Trieste \ /V\ / Department of Physics \ (/ \) / Trieste, Italy \ ( ) ^^_^^ Welcome to rsync1.it.gentoo.org - Intel Pentium IV 3.06Ghz 2GB Ram, IBM xSeries 205 max simultaneous connections : 10 max daily connections: 5 per IP address - For any communication about this mirror please contact [EMAIL PROTECTED] - receiving file list ... 1 file to consider timestamp.chk 32 100%0.00kB/s0:00:00 Number of files: 1 Number of files transferred: 1 Total file size: 32 bytes Total transferred file size: 32 bytes Literal data: 32 bytes Matched data: 0 bytes File list size: 32 Total bytes written: 222 Total bytes read: 948 wrote 222 bytes read 948 bytes 780.00 bytes/sec total size is 32 speedup is 0.03 - .*. / University of Trieste \ /V\ / Department of Physics \ (/ \) / Trieste, Italy \ ( ) ^^_^^ Welcome to rsync1.it.gentoo.org - Intel Pentium IV 3.06Ghz 2GB Ram, IBM xSeries 205 max simultaneous connections : 10 max daily connections: 5 per IP address - For any communication about this mirror please contact [EMAIL PROTECTED] - receiving file list ... 117192 files to consider ./ app-accessibility/gnopernicus/ app-admin/ app-admin/logrotate/ app-admin/modlogan/files/ app-admin/superadduser/files/ app-admin/sus/ deleting app-cdr/cdrtools/cdrtools-2.01-r2.ebuild deleting app-arch/dpkg/files/digest-dpkg-1.10.26 deleting app-arch/dpkg/dpkg-1.10.26.ebuild Number of files: 117192 Number of files transferred: 1671 Total file size: 92305590 bytes Total transferred file size: 3745266 bytes Literal data: 3745266 bytes Matched data: 0 bytes File list size: 2594553 Total bytes written: 33600 Total bytes read: 4157651 wrote 33600 bytes read 4157651 bytes 19270.12 bytes/sec total size is 92305590 speedup is 22.02 >>> Updating Portage cache: 100% Tue 05/31/05 15:14 /usr/local [EMAIL PROTECTED] -> I know I have a pretty fast connection (8000/1024), but even on dialup, I can't see an hour or more being in any way normal, unless: 1) you have connection problems (firewall?) 2) you are conneting to a server with connection problems (server is naturally slow, node between you and server experiencing problems) 3) 1 + 2 combined 4) you're syncing the entire mirror (as if you were a mirror yourself), rather than just the tree. Please correct me if I'm wrong, but I t
[gentoo-user] It's baaaaack...! (Python 2.4.1)
Hi all, I seem to have painted myself into a corner, and hope that someone can see a way out before I repeat my previous mistakes. The short version is that, after my previous thread about removing an extra version of Python (2.4.1 when I already had 2.3.5 installed), Portage now wants to re-emerge Python 2.4.1, and I can't seem to prevent it. At this time, I do not want to emerge Python 2.4.1 to a new slot, as it breaks Pysol seemingly unrepairably (java-config also broke, but was repaired by a re-emerge, Pysol was not repaired by a re-emerge). Yes, I know Pysol is not "important" per se, but I don't want to break a package I use, and the breakage indicates that other things that I value more highly might get broken (I just don't know what they are). The long version of what I was trying to do and the current situation: This all started because I wanted a reliable system for automounting removeable media (CD/DVD). I love supermount, but I'm trying to live in the present. Submount works, but is not reliable (stops working after some period of time, dismounts automatically but doesn't mount automatically), and my previous experience with hal/dbus/ivman is not good. However, there is a new version of ivman which sounds really good, but it's only available from the gentopia overlay. Fine. Installing that upgraded hal, dbus, the new ivman, and installed Python 2.4.1. Except that dbus wouldn't compile with the python USE flag set, so I removed it, which allowed it to compile, but meant (afaics) that I no longer needed Python 2.4.1. So, I got rid of it, as you know from my "Unmerging 'extra' Python" thread. The whole hal/dbus/ivman scheme isn't working anyway atm because (to the best of my understanding): 1) I have not yet configured, compiled and installed a kernel with the correct version of the inotify patch that either hal, dbus or gamin needs (I've emerged mm-sources 2.6.12-something but have not yet compiled it); 2) GNOME is broken atm (which hopefully I can fix when I get the kernel upgraded and the gentopia backend working); 3) KDE doesn't like the upgraded versions of hal and dbus, and so the media kio slaves don't work on my current setup. #3 is the lynchpin here, as an emerge sync has just made KDE 3.4.1 available. This wants to downgrade dbus and hal, which is fine, and wants to install Python 2.4.1, which is not. I then removed the gentopia overlay, and ran the potential uaDtv world again; now Portage wants to downgrade hal, dbus, ivman, and gamin (also fine), but still wants to emerge Python 2.4.1. Here's the relevant output: [nomerge ] net-www/mplayerplug-in-2.80 +gtk2 [nomerge ] net-libs/gecko-sdk-1.7.5 +crypt -debug +gnome +java +ldap -moznomail -moznoxft -postgres +ssl -xinerama +xprint [nomerge ] app-crypt/gnupg-1.4.1 +X +bzip2 +caps +curl -debug +ecc +idea +ldap +nls +readline (-selinux) -smartcard +zlib [nomerge ]mail-mta/ssmtp-2.61 -ipv6* -mailwrapper -md5sum +ssl [ebuild U ] net-mail/mailbase-0.00-r9 [0.00-r8] -pam 0 kB [ebuild U ] kde-base/ksvg-3.4.1 [3.4.0] -arts* -debug +kdeenablefinal +kdexdeltas -xinerama 0 kB [nomerge ] x11-misc/xscreensaver-4.20 -debug +gnome +gtk +jpeg -kerberos -krb4 +motif +offensive +opengl -pam -xinerama [nomerge ] gnome-extra/yelp-2.6.5 -debug [nomerge ] gnome-base/libgnome-2.8.1 -debug -doc [nomerge ]gnome-base/gnome-vfs-2.8.4-r1 -debug -doc +gnutls +hal -howl -ipv6 +samba +ssl [ebuild UD] sys-apps/hal-0.4.8 [0.5.2] -debug -doc -livecd -pcmcia 1,321 kB [ebuild UD] sys-apps/dbus-0.23.4 [0.33-r1] +X -debug +gtk +mono -python +qt +xml2 1,258 kB [ebuild U ] kde-base/kdeartwork-styles-3.4.1 [3.4.0] -arts* -debug +kdeenablefinal +kdexdeltas -xinerama 0 kB [ebuild U ] kde-base/kdelibs-3.4.1 [3.4.0-r2] +alsa -arts* +cups -debug -doc +jpeg2k +kdeenablefinal -kerberos -openexr +spell +ssl +tiff -xinerama -zeroconf 16,458 kB [ebuild UD] app-admin/gamin-0.0.26-r6 [0.1.0-r1] -debug -doc 472 kB [nomerge ] sys-libs/libstdc++-v3-3.3.4 -debug +nls [nomerge ] media-libs/freetype-2.1.9-r1 -bindist -debug -doc +zlib [nomerge ]sys-libs/glibc-2.3.5 -build -debug -erandom -hardened (-multilib) +nls -nomalloccheck +nptl -nptlonly +pic +userlocales [nomerge ] sys-devel/gcc-3.4.3.20050110-r2 (-altivec) -bootstrap -boundschecking -build -debug +fortran -gcj +gtk -hardened -ip28 (-multilib) +multislot (-n32) (-n64) +nls -nocxx -nopie -nossp -objc -static [nomerge ] sys-devel/gcc-config-1.3.10-r2 [nomerge ] sys-apps/portage-2.0.51.19 -build -debug (-selinux) [ebuild NS ]dev-lang/python-2.4.1 +X -berkdb -bootstrap -build -debug -doc +gdbm -ipv6 +ncurses -nocxx +readline +ssl +tcltk -ucs2 0 kB I just don't see where Python 2.4.1 is coming from (since everything in the -D tree for it was perfectly fine with Python 2.3.5), and the new stuff that originally installed it is a) being downgraded to a version
Re: [gentoo-user] It's baaaaack...! (Python 2.4.1)
Neil Bothwick schreef: > On Tue, 31 May 2005 17:06:55 +0200, Holly Bostick wrote: > > >>The short version is that, after my previous thread about removing an >>extra version of Python (2.4.1 when I already had 2.3.5 installed), >>Portage now wants to re-emerge Python 2.4.1, and I can't seem to prevent >>it. > > > You would have unmasked Python 2.4+ to merge it, have you masked it > again? If not an emerge --deep would try to merge the latest unmasked > version, which is 2.4.1 on your system. > > Thank you, Neil, and everyone who reminded me that I had forgotten all about package.unmask. Remasked Python and it's fine. At least I did say I was hysterical in the first place; as you can see from the 'twisted tale', there's just too many irons in the fire (to do just this one stupid task) for me to keep track of properly... and I consider my current install "reasonably stable" ...! (which basically means "stable enough to attempt setups like hal/dbus/ivman from overlay"). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] KDE 3.4.1 released
Andreas Karlsson schreef: > Hi, > > KDE 3.4.1 was released today: > > http://www.kde.org/announcements/announce-3.4.1.php > > I am compiling it right now. Just a little question regarding aRTS. I > compiled > into 3.4.0 but now I have decided to dump it. Could there be any problem > compiling with USE="-alsa" for 3.4.1 (if there is, I would probably have to > recompile again, as I use that right now :) )? > > Best regards, > Andreas Karlsson Do you mean -alsa, or -arts? If the latter (which makes more sense, given your first sentence), I hope not, as I am about to do the exact same thing. If you are using aRTS, then -alsa will compile arts without ALSA support (which means, afaik, that you'd better be using OSS, or you will have no sound, as artsd won't be able to access the ALSA devices, although OSS emulation *might* still work; I'm not sure). If you are using alsa, then -arts will compile KDE without support for the artsd daemon, which means no system sounds and the like, but multimedia programs that can use ALSA or ALSA's OSS emulation should be fine. At least, I hope so, or I'll be two days compiling arts support out and back in myself :) . Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Good diagram app
[EMAIL PROTECTED] schreef: > > > On Tue, 31 May 2005, Bruno Lustosa wrote: > >> Hello. >> >> I am in need of an application like Visio. I tried messing with dia, >> but it's still far from what I need, for it lacks lots of useful >> features. >> What program do the list recommend? > > > Take a look at Inkscape. Many very nice features, some overlap with dia, > some quite different. > > www.inkscape.org KOffice has just expanded with a new diagram tool called kivio . It looked pretty good, almost made me wish I had some use for a diagram tool. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] firefox sage
maxim wexler schreef: >>But isn't >> >>emerge --sync >> >>a different command from the more traditional >> >>emerge sync >> >>? > > > In An Intro to Portage it's emerge --sync and if that > fails, emerge-websync but that failed too. > > >>1) What is the speed of your Internet connection? > > > Slightly faster than I can type ;( > > >>2) What is your geographical location? > > > SW Canada > > >>3) What mirrors are you setting in /etc/make.conf? >> > > > ? All I got is > PORTDIR=/usr/portage No, what I was asking is what this line says: SYNC="rsync:///gentoo-portage" The idea being that perhaps you were attempting to sync with a mirror that is far away from you, and that server perhaps is overloaded, or there is a transmission problem between you and it, due to the distance. For instance, I'm in Western Europe, so setting my rsync server to a mirror in Asia or South America is just asking for more problems than I need, because in order for me to communicate with a server so far away, my packets have to travel via Internet nodes/servers which could go down through no fault of mine or the server I'm trying to connect to. In such a case, someone in the area (Asia or South America) might themselves be able to connect to the server in question, but I would not be (because they don't have to travel through the node that I do to connect). So I was just trying to confirm that this was not (or was) a possible source of your problem. > > >>10 minutes: > > > After 4 hrs it quit. A list of /usr/portage revealed > no changes. But this morning after entering emerge > --sync I note the tree *is* > being updated. >>Number of files: 117192 >>Number of files transferred: 1671 >>Total file size: 92305590 bytes >>Total transferred file size: 3745266 bytes >>Literal data: 3745266 bytes >>Matched data: 0 bytes >>File list size: 2594553 >>Total bytes written: 33600 >>Total bytes read: 4157651 >> >>wrote 33600 bytes read 4157651 bytes 19270.12 >>bytes/sec >>total size is 92305590 speedup is 22.02 > > > This is similar to what I see followed by millions of > lines like this: > > /(something)/(something)/(something) > some-number 100% 0.00 kB/s 0:00:00 > > > Yesterday it quit with the message, error in rsync > port datastream(code 12) at io.c(201,189). > That indicates, afaik, a (usually temporary) server problem, and happened to me a couple of days ago. Since today we have all noticed that KDE 3.4.1 was released, and that's a lot of packages, it's quite possible that you tried to contact the server at the time it was trying to update itself, which is, not surprisingly, likely to fail. How long did your sync today take, and does an emerge -up world or emerge -ua world seem to indicate that it was successful (that packages you have installed have updates available)? Don't worry, the -p switch stands for --pretend, and the -a switch stands for --ask, so nothing will be updated without your approval by running either of the commands. We just want to know if the sync... synced. Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] [OT] Can't load GNOME as user
Hi, I can't stand it anymore, so I thought I'd fish here for ideas. The long and the short of it is that I cannot load the GNOME desktop as a user (works fine as root). The splash screen comes up, but only 2 icons show in the progress bar before the splash disappears-- Sessions, and Window manager, iirc. Already not good (no Nautilus, no Metacity, no gnome-panel...). Then the panel tries to come up, the panel backgrounds (currently set as the default top and bottom panels) display (empty), then disappear. This happens about 5 or 6 times (I suspect related to how many panel applets are attempting to load). No desktop appears, no panels load, right-click on the desktop produces no menu (since Nautilus isn't running, presumably) and since I don't know the GNOME shortcut key to open a Run Box, I pretty much have to Ctrl+Alt+Backspace out to GDM and load another WM. This happens with both the regular GNOME entry and the Failsafe GNOME entry. I have deleted ~/.gconf, ~/.gconfd, ~/.gnome, ~/.gnome2, ~/.gnome2_private, and ~/.gnome_private and allowed my login attempt to regenerate them (supposedly), but this had no effect. Since GNOME loads fine when root logs in, GNOME itself is presumably not broken, but rather the user is. What's left to delete and regenerate? Does anybody know what I might do to fix this (short of creating a new user, which I'm not going to do :) )? Thanks for any help, Holly, hanging out in IceWM while KDE 3.4.1 compiles. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] KDE 3.4.1 released
Antonino Sabetta schreef: > Andreas Karlsson wrote: > >> On Tuesday 31 May 2005 17.45, Antonino Sabetta wrote: >> >>> Andreas, >>> I do not see the relation between dumping Artsd and the -alsa switch. >>> Since >>> I'm having big troubles trying to make things work with >>> alsa+dmix+arts, can >>> you explain how do you intend to configure your sound system? >>> Thanks, >>> Antonino >> >> >> >> Yeah, now I think I was smoking crack or something. What I meant was >> ofcourse dumping arts alltogether, not alsa. *mumble* > > > My question is still valid :) > How are you going to mix sounds without arts? Are you going to use some > other > sound daemon? Or dmix? > Thank you, > Antonino I can't speak for Andreas, of course, but for myself it's not an issue because 1) I don't generally need to mix sounds (I'm not one to watch a movie in mPlayer with kopete or whatever running in the background); 2) I don't generally use system sounds anyway because my PC is next to my boyfriend's. He uses speakers and I use headphones to keep the sound pollution in our computer room to a minimum. I usually don't bother to wear the 'phones unless I specifically need them for a game, movie, or music, so there's not much point in enabling system or application sounds; 3) I invested some 30 Euros in a soundcard that does hardware mixing (could have spent less, actually, but I liked this card and I really didn't want a Soundblaster Live! variant). Vast improvement over my onboard sound chip. So I would answer that you might consider eliminating the need for sound mixing, or investing in hardware that actually supports features you want to use. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] grub still broke
Neil Bothwick schreef: > On Tue, 31 May 2005 14:51:49 -0700 (PDT), maxim wexler wrote: > > >>Whoa! IIRC splashimage is optional. Can somebody weigh >>in here? > > > Yes, it is optional, and worth disabling if you have problems. GRUB will > bail out with no error message if you set an incorrect path for > splashimage. > > You're right, of course, Neil, but from the original post I believe we are not talking about the bootsplash/fbsplash image but the grub menu graphics: ZeeGeek schreef: > On 6/1/05, maxim wexler <[EMAIL PROTECTED]> wrote: > >>That's better. But it still doesn't work. Staring at a >>console right now telling me to please wait. >> > > > that's right, isn't it? it's working if you see the console giving you > the list. if what you are wishing for is that gentoo background > choosing menu, then you'll need splashimage set in your grub.conf > which I didn't see in your first post. > "The gentoo background choosing menu" is what you get in the first part of grub.conf: default=0 timeout=10 splashimage=(hd0,1)/grub/gentoo.xpm.gz not the silent or verbose splash defined by title Gentoo_current (2.6.11-gentoo-r8) root (hd0,1) kernel /vmlinuz ro root=/dev/hda5 video=vesafb-tng:ywrap,mtrr,pmipal,[EMAIL PROTECTED] splash=silent,theme:emergence quiet CONSOLE=/dev/vc1 initrd /fbsplash-emergence Now it's true that both are optional (you can just have a text menu with colors and not a graphic background for the menu, and of course you can just watch text scroll by on a black background during boot), but it seems to me that the operative question here is where does this "Please wait..." message come in? I don't pay enough attention to GRUB's booting process to be sure, but isn't that displayed before the GRUB menu comes up? Or is it just after boot? And which OS are you choosing from the menu again, maxim (assuming you get to a menu)? Or does this affect all OSes in your menu? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Can't load GNOME as user
Chris Woods schreef: > Holly Bostick wrote: > >>Hi, >> >>I can't stand it anymore, so I thought I'd fish here for ideas. >> >>The long and the short of it is that I cannot load the GNOME desktop as >>a user (works fine as root). > > > [...] > > I had this problem pretty persistently for a long time. I believe what finally > fixed it was killing all gnome-session processes, with no X at all running, > and then delete all gnome-related files and directories in $HOME - .gtk* > .gnome* .metacity* .nautilus* Desktop* .gconf* etc - and then restart X by > your usual means, and try logging in to gnome from there. > > HTH, > Chris > All right, I'm making progress, of a sort. I am currently in GNOME, but don't know if I'll make it through a logout/in. Here's what I did: 1) created a new user (I know I said I wouldn't, but it looks like Edward was right) and logged into GNOME as her; 2) changed a couple of things (more on this later) such as the screen resolution and whatnot and logged out 3) logged in as my regular user under another DE, opened a file manager as root and adjusted the permissions of the 'test' user's home folder so I could get into it; 4) copied the test user's GNOME files to my regular user, overwriting whatever was there (easy enough to identify as the user had no files to speak of but what was installed by default). When I logged out and back into GNOME, it started--- with a lot of errors, but it started (mostly; Nautilus won't start, and stuff like the volume control and the show desktop button had to be removed from gconf). But like I said, it might not survive a logout, so I thought I'd post this all now. Here's the thing, though: it looks like at least some part of this was caused by my unorthodox install (which is another reason I wouldn't recommend this method or perform it ever again). I know we don't like to mention 'that Gentoo-based distro' on this list, but I could use some help in cleaning up its mess if anybody might have a clue what happened. Here's the symptoms: When I created the new user and logged into GNOME, the desktop resolution was Vida's 1024x768 instead of my normal 1280x1024; The default wallpaper was the Vida wallpaper Some gDesklets something was listed to start in the "voluntary startup" section of the Session Manager (which I removed). So obviously, I have retained some Vida default config file somewhere (/etc/skel?), and you can be sure I don't want it. Even if it was appropriate (which it ain't), it's presumably a GNOME 2.10 config, and I downgraded to 2.8.3 when GNOME first broke, as I thought that might help. Now I know what the problem probably was-- when converting to Gentoo I clearly missed something out. I hate to ask, and I'll hang my head in shame while the list derides me (although the trouble and embarassment is really punishment enough), but if anyone happens to know enough about what stupid Vida does to tell me what I need to rm -rf --and what, if anything, I need to emerge to replace it, I would be grateful. Would a complete uninstall of all GNOME packages do it? Naturally, if there's a better way, I'd love to hear it. Thanks, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] grub still broke
maxim wexler schreef: >>And which OS are you choosing from the menu again, >>maxim (assuming you >>get to a menu)? Or does this affect all OSes in your >>menu? > > > no choice. After grub-install I get the > > Grub loading stage1.5 > Grub loading, please wait... > > message(white text,black bg). To get back to > Macroshaft I boot into a Win98 CD and run fdisk /mbr > > Ok, now I've got it. The menu doesn't load at all. But your previous post as to formatting the /boot partition made me think of something I had problems like that some time ago, back when I first installed my first Gentoo. Basically what had happened was I got weird and unattributable errors due to my filesystem not being correctly formatted. It was supposed to be formatted, and files were installed to it and everything, but filesizes were being reported differently by different tools and things just didn't work properly. What I wound up doing was using qtparted to delete the filesystem and reformat it. Once the filesystem on the disk was the same as the filesystem that the disk thought it had, everything worked fine. Now, I seem to recall having heard that it is possible to delete and reformat a filesystem without deleting the partition (or damaging the files thereon), but I didn't know enough at the time to do that, so I just deleted the entire partition and recreated it. Since this is /boot, it won't be a tragedy to delete the partition, recreate, format it as ext2 from the start and reinstall grub. But maybe there's a way that you can just reformat the existing partition (again) as ext2, so that it "takes". You might still have to reinstall grub anyway, however at this point that seems like the least of your worries :-) . Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] CFS Howto
Shawn Singh schreef: > Hey all, > > I was attempting to encrypt my home directory using the CFS Howto as > my guide, but I am not getting far at all... > > When trying to emerge cfs, I'm getting the following message: > > convert usr # emerge app-crypt/cfs-1.4.1.14 > Calculating dependencies > > !!! Problem in app-crypt/cfs-1.4.1.14 dependencies. > !!! "Specific key requires an operator (app-crypt/cfs-1.4.1.14) (try > adding an '=')" exceptions > > So, I changed my command to: > emerge -pv =app-crypt/cfs-1.4.1.14 > > and I'm getting the following message: > These are the packages that I would merge, in order: > > Calculating dependencies > !!! All ebuilds that could satisfy "=app-crypt/cfs-1.4.1.14" have been masked. > !!! One of the following masked packages is required to complete your request: > - app-crypt/cfs-1.4.1.14 (masked by: ~x86 keyword) > > For more information, see MASKED PACKAGES section in the emerge man page or > section 2.2 "Software Availability" in the Gentoo Handbook. > > The computer that I'm trying to install this on has a Pentium 4 > processor. Support for NFS has been compiled into the Kernel > (satisfying part of step 3). > > Any suggestions? > > Thank you, > > Shawn > I assume you want to continue running an x86 (stable) system for the most part. If so, then add app-crypt/cfs ~x86 to /etc/portage/package.keywords (create this file if it does not exist). If you'd like to go totally ~x86 (unstable), you can just add ACCEPT_KEYWORDS="~x86" to /etc/make.conf When you've done either one, just emerge cfs normally (emerge cfs), and you should get the ~x86 version. Hope this helps. Holly H -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: When a dependancy built by hand
Harry Putnam schreef: > Jason Stubbs <[EMAIL PROTECTED]> writes: > > >>/etc/portage/profile/package.provided >> > > > Where to look to learn about the required syntax? > man portage :) Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Perl Upgrade Problem
Richard Watson schreef: > Hi - I'm trying to upgrade portage. My existing version of perl comes up as > a block. When I tried to remove the package I received a warning about > damaging the system. Can anyone tell me if it's OK to proceed. > > Output below. > > Thanks a lot, Alan > > ==> > # emerge --pretend perl > > These are the packages that I would merge, in order: > > Calculating dependencies ...done! > [blocks B ] [ebuild U ] sys-devel/libperl-5.8.6 [5.8.5-r1] > [ebuild U ] dev-lang/perl-5.8.6-r4 [5.8.5] > > pisces alan # emerge -C perl > > > !!! Trying to unmerge package(s) in system profile. 'dev-lang/perl' > !!! This could be damaging to your system. > > Waiting 10 seconds before starting... (Control-C to abort)... > > Press Ctrl-C to Stop in: 10 9 8 7 6 > > Exiting on signal 2 > # > > Well, I thought Portage was written in Python, not Perl but no matter. I believe that you are mistaken about what is blocking the Perl upgrade (no fault of your own, block messages are hard to read). I think that the blocking (currently installed) package is perl-core/File-Spec-0.87 not Perl 5.8.5 whatever. So that is what I would unmerge, not Perl, then try the emerge again. It sometimes happens that a *module* can block the upgrade of a language because the module is installed with the upgrade or something along those lines. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Can't load GNOME as user [SOLVED!]
Holly Bostick schreef: > Hi, > > I can't stand it anymore, so I thought I'd fish here for ideas. > > The long and the short of it is that I cannot load the GNOME desktop as > a user (works fine as root). > After my partial success, I did the following 1) a revdep-rebuild 2) upgraded back up to GNOME 2.10 (re-uncommented my keywords and did an emerge -uaDtv world) and it's all working. Thanks for all the help. I really missed GNOME a lot; the simplicity of it is like a breath of fresh air after dealing with KDE for so long. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Perl Upgrade Problem
Richard Watson schreef: > I believe that you are mistaken about what is blocking the Perl upgrade > (no fault of your own, block messages are hard to read). > > I think that the blocking (currently installed) package is > perl-core/File-Spec-0.87 > not Perl 5.8.5 whatever. > So that is what I would unmerge, not Perl, then try the emerge again. > > It sometimes happens that a *module* can block the upgrade of a language > because the module is installed with the upgrade or something along > those lines. > > Hope this helps, > Holly > > Thanks Holly. I'm obviously doing something wrong ... I can't seem to get > emerge to remove the offending package. Below is the output I'm getting. I'm > probably making an obvious mistake ... Regards Alan > > # qpkg -i File-Spec-0.87 > perl-core/File-Spec-0.87 > Handling files and directories portably [ http://www.cpan.org/modules/by > -authors/id/K/KW/KWILLIAMS/.readme ] > > # emerge -C File-Spec-0.87 > --- Couldn't find File-Spec-0.87 to unmerge. > unmerge: No packages selected for removal. > > I notice that the output mentions www.cpan.org. Did you by chance install this module via CPAN? Portage doesn't manage those, so naturally it would not be found by Portage. If so, remove it via CPAN, because Portage clearly *wants* to manage this module, and doesn't want to upgrade Perl with an unmanaged (critical?) module on the loose. Alternatively, I notice this in the Changelog for File-Spec: > 25 May 2005; Michael Cummings <[EMAIL PROTECTED]> +metadata.xml, > +File-Spec-0.84-r1.ebuild, +File-Spec-0.87.ebuild, +File-Spec-3.01.ebuild, > +File-Spec-3.06.ebuild: > dev-perl => perl-core migration > Which means that if you did install the module via portage, you probably installed it as dev-perl/File-Spec-0.87 which no longer exists(?) in Portage; the module is now in a new category: perl-core/File-Spec-0.87 In this case, what I would do is attempt to unmerge File-Spec with emerge -Cav =dev-perl/File-Spec-0.87 This might work, but also might fail, because the original ebuild has been removed from Portage. In that case, I would install perl-core/File-Spec-0.87 over the previous version (assuming you did install it with Portage in the first place), and if it still blocks the emerge, then uninstall it via Portage, which would now be able to recognize that it was installed in the first place. Do you have no recollection of this module and how you might have installed it? Oh, look, this is in the changelog too: > 17 Apr 2005; Michael Cummings <[EMAIL PROTECTED]> > +File-Spec-3.01.ebuild, +File-Spec-3.06.ebuild: > Adding both version 3.01 and 3.06 to the tree. 3.01 gives us cross > compatibility with the version of File-Spec that is shipped with the perl > 5.8.6 core package. 3.06 is just the current stable upstream :) Please see > metadata for explanation of extreme version jump. > So at least I was right about that much-- this module ships with 5.8.6, so if you have it previously installed as a separate module it will conflict and block the upgrade *to* Perl 5.8.6. That brings me to a third thought: It's remotely possible that Portage would be able to upgrade the module as part of the upgrade to Perl, but cannot because: 1) there is no upgrade to the installed module (the module has changed category, and you can't upgrade across categories) 2) the upgrades are ~arch (are you running ppc64 or ia64, where 0.87 is the last stable?), which you may have keyword masked. That would be a weird situation, but I can see how it would result in a Block as the only reasonable error message that Portage could produce (since the way to resolve a block is to unmerge the offending package, which is also how to resolve a category migration). So, in summary: 1) if you installed the module via CPAN, uninstall it via CPAN; 2) if you installed the module via Portage within the last couple of weeks (before the category change) try uninstalling it specifically using Atoms (=dev-perl/File-Spec-0.87), as that is what Portage currently knows the package as; 3) if 2) fails and you did not install the module via CPAN, try installing the new-category package over the old-category package; 3a) see if the module still blocks the Perl upgrade (specifically [B]locks, rather than gives a message that all upgradeable packages are masked, which in this case would be an improvement in the situation, as you could just unmask the keyword-masked package and go on from there); 3b) if Perl is still blocked, then uninstall the whole File-Spec shebang, as emerge -C File-Spec should now work. One way or another, at the end of whichever trail you take, you should be able to upgrade Perl. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Printer setup tool
[EMAIL PROTECTED] schreef: >>>I can't, cupsd dies quickly. >>> >>> >>> >> >>Well, that's a problem. >> >>At what point does it die (what are you doing when it dies), and what >>does it say with its dying breath (error message)? >> >>Holly >>-- >>gentoo-user@gentoo.org mailing list >> > > bash-2.05b# cupsd > cupsd: Child exited with status 98! > > I happens after about 5 seconds. What I am doing doesn't sem to matter. > > And /var/log/cups/error.log says...? This would seem to be a problem setting up CUPS as a whole rather than setting up your printer specifically. Your printer naturally can't work if CUPS doesn't, if you've set it to use CUPS. So you'd first want to find out why CUPS fails, and then setting up your printer should be no problem. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] svgalib: Cannot open /dev/svga Is svgalib_helper module loaded?
Michael Sullivan schreef: > I found a tutorial for svgalib this morning. I've always been > fascinated by graphics programming, but never found a tutorial that was > simple enough for me to follow in the beginning. I liked this tutorial. > It provided source code for a simple c program that just plots a pixel > to the screen. I copied the code into a file and compiled it. It all > compiled fine. When I tried to run the program, I got this: > > svgalib: Cannot open /dev/svga > Is svgalib_helper module loaded? > > > Is this something I can fix, or yet another situation in my quest for > graphics programming knowledge where I'm just out of luck? > Ummm... the error message asked a pretty simple-seeming question: "Is svgalib_helper module loaded?" So as far as that goes, it does seem to be something you can check, and --if the answer turns out to be "no"-- to fix, by loading the module (it's a kernel module created during the install of svgalib, and if you haven't rebooted since you installed svgalib, it won't be loaded). If the module fails to load, what is the error after modprobe svgalib_helper? HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] svgalib: Cannot open /dev/svga Is svgalib_helper module loaded?
Michael Sullivan schreef: > On Fri, 2005-06-03 at 16:44 +0200, Holly Bostick wrote: > >>Michael Sullivan schreef: >> >>>I found a tutorial for svgalib this morning. I've always been >>>fascinated by graphics programming, but never found a tutorial that was >>>simple enough for me to follow in the beginning. I liked this tutorial. >>>It provided source code for a simple c program that just plots a pixel >>>to the screen. I copied the code into a file and compiled it. It all >>>compiled fine. When I tried to run the program, I got this: >>> >>>svgalib: Cannot open /dev/svga >>>Is svgalib_helper module loaded? >>> >>> >>>Is this something I can fix, or yet another situation in my quest for >>>graphics programming knowledge where I'm just out of luck? >>> >> >>Ummm... the error message asked a pretty simple-seeming question: >> >>"Is svgalib_helper module loaded?" >> >>So as far as that goes, it does seem to be something you can check, and >>--if the answer turns out to be "no"-- to fix, by loading the module >>(it's a kernel module created during the install of svgalib, and if you >>haven't rebooted since you installed svgalib, it won't be loaded). >> >>If the module fails to load, what is the error after modprobe >>svgalib_helper? >> >>HTH, >>Holly > > I modprobed svglib_helper and put it into > my /etc/modules.autoload.d/kernel-2.6 file. When I tried to run the > tutorial program again I got this: > > Not running in a graphics capable console, > and unable to find one. > Using SAVAGE driver, 12288KB. Chipset: ProSavage > Not running in a graphics capable console, > and unable to find one. > svgalib 1.9.19 > svgalib: Failed to initialize mouse. > Not running in a graphics capable console, > and unable to find one. > > and put me back at a terminal prompt. How do I fix this one? > Let's see "not running in a graphics-capable console" would imply that neither X nor the framebuffer (I'm kinda leaning towards the framebuffer) is running, and/or that svgalib requires some form of 3D hardware acceleration support which is not installed. I don't know which is the case (I don't know your setup-- i.e., how you're running this program-- or what requirements svgalib has), but presumably your interest in this subject (or the tutorial you're following) should give some kind of a clue as to which area of your graphics subsystem needs to be looked at. As to the mouse thing, well, a Google search for svgalib_helper links to this page, "SM 5 BSZ - Installing svgalib-1.9.19 under different distributions" at http://www.sm5bsz.com/linuxdsp/install/distrib2.htm . Now, naturally, Gentoo is not included, but it does say this for FC3: "To get the mouse going, add a line mdev /dev/input/mice in the /etc/vga/libvga.config file." So I would check to see if there is a libvga.config file on the system, and see if that line is present, and add it if not-- it just might work, although it's also possible that this is related to your original issue with the lack of a graphics-related console insofar as the 'original' basic graphics setup would normally include the mouse, so if the graphics were fixed, the mouse issue might be as well. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] symbolic link to var tmp
Ognjen Bezanov schreef: > I am trying to upgrade my gentoo box but the partition where /var/tmp > resides is too full. > > So I tried making a symbolic link to another disk which had a lot of > space, but then gentoo refused to compile, giving me permission errors. > > So im asking, how can you use /var/tmp on another disk. I dont want to > dedicate the whole disk and mount it to /var, so using links (or > possibly giving portage a differnt path) would be better. > > thanks > You could also try cleaning /var/tmp out: /var/tmp/portage can be emptied in its entirety, but most of the folders there should not be taking up much space, as successfully completed emerges will leave only a couple of bytes or kilobytes inside the program's folder. However, if you have failed emerges, especially for large compiles like openoffice (not the bin), or xorg or something, the folder for that emerge will still be in /var/tmp/portage/program_name/temp/work, because the temporary work files are not deleted if the emerge fails. That can take up a whole lot of space (the emerge for OOo uses some 3GB temp space before completeion). So you might consider deleting any program folders in /var/tmp/portage for programs you know failed to compile and see if that helps. Or, of course you could change your PORTAGE_TMPDIR setting in /etc/make.conf and then change it back when you were done making whatever new arrangements you wanted for the default /var/tmp. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] meta-question: how can I receive the messages that I send to the ML?
Antonino Sabetta schreef: > Hi all, > I apologize for the OT, but probably you know the answer to my question > and will be so kind as to tell me whether is there a way to receive my > posts > just as I receive any other message from this mailing list. > I'm annoyed of reading threads with missing pieces, especially if the > missing pieces are mine :) > Thanks a lot, > Antonino Since this is known to be a Gmail "feature", I would have to say that the answer to your question is "read the list via a mail client or newsreader rather than via GMail". Sorry :-) . I admit that I find GMail of severely limited usefulness, and this is not one of the uses I find it appropriate for, so I may be prejudiced. Alternatively, you could set up POP3 forwarding or notification... that would send (a copy of) even your own message to the forwarding address, wouldn't it? Or do they filter those in the forwarding as well? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Random crash
� schreef:
>
>
> I think I figured it out. Here is my output:
> bash-2.05b$ firefox
> \No running windows found
> /home/omega21/.gtk_qt_engine_rc:62: error: unexpected
> character `{', expected character `}'
> /usr/libexec/mozilla-launcher: line 547: 16841
> Segmentation fault "$mozbin" "$@"
> firefox-bin exited with non-zero status (139)
>
>
> I now realize that Gaim, Firefox & Thunderbird are all
> GTK, and Im using qt-gtk-engine. That must be the
> issue.
>
> Can I get these things to work though with QT-GTK,
> because GTK without a theme -- well, no one deserves
> to see it. :)
>
> Ian
>
>
"GTK without a theme"? GTK-QT *isn't* a theme. It's an engine that
(supposedly) allows GTK (1 and 2) applications to use a KDE theme when
running KDE, so that your GTK and QT applications look the same.
Now, this works fine for me with Thunderbird, Firefox doesn't seem to
accept most theme variables, even under GNOME (it won't pick up desktop
colors, for example), and if Gaim is GTK 1, well... GTK-QT in my
experience works much better with GTK 2 applications rather than GTK 1.
First of all, have you set GTK applications to use the KDE color scheme
in the KDE Configuration Center (Appearance and Themes=>Colors=>use my
KDE color scheme in non KDE applications").
Second of all, if you installed the GTK-QT engine via Portage, you
should also have a "GTK Themes and Styles" entry in the Configuration
menu or in the main menu somewhere (look in Utilities, or Extra). Make
sure that the settings are correct.
But the main problem that I found is that GTK 1 programs especially do
not support all KDE themes. I was using Liquid and no GTK 1 application
wanted to recognize it (though I finally got them to use the right color
scheme). What I finally did was go to kde-look org and find a GTK theme
that I could live with and that was replicated successfully for GTK, GTK
2, and KDE (there are several such themes on the site in the GTK+ 1
section, but many of the ones I tried didn't work properly), and use
that instead.
The theme I chose was QTCurve, and once installed (had to compile it
manually, but that was a simple and standard compile and the
instructions were perfectly adequate to the task), I then set GTK 1 apps
to use it via switch --not switch 2; I specifically installed the 1.0
version in a new slot to have access to GTK 1 theme controls. "switch"
by the way, is the binary name for the package installed by emerge
gtk-theme-switch. I then set GTK 2 apps to use it via the GNOME Control
Panel (if available), switch2 (the binary installed by the 2.0 versions
of gtk-theme-switch), or gtk-chtheme, another GTK (2) theme-setting
utility for those without GNOME. KDE was just set to use the theme
normally via the Configuration Center.
And now all my apps, be they KDE, GTK 1 or GTK 2, look alike to the best
of their ability (since Firefox does not pick up desktop theme colors
using most firefox themes, although istr that there are a couple of ff
themes that do use desktop color schemes. I'm just apparently not using
one).
So my advice to you would be threefold:
1) try installing (gtk) theme engines via Portage, or even manually from
art.gnome.org or gnome-look.org. It's possible that you don't have the
backend for the theme you're trying to use. Be warned ,though-- many of
the gtk-engine packages will install if you have GNOME 2.x installed,
but then if you try to upgrade GNOME (from 2.8.x to 2.10, for example),
the gnome-themes(-extras) package will be blocked by them. However, the
gnome-themes packages don't seem to actually include many of the GTK
engines, though they (obviously) claim to, which is a problem since many
GTK themes are based on some of these engines, like smooth.
2) try a different KDE theme via GTK-QT. I had a lot of problems with
Liquid, but Plastic is much simpler and might work fine.
3) give up on GTK-QT and just use replicated themes. Geramik is in
Portage, replicating Keramik (and Thin Keramik) controls. On
kde-look.org you can find Plastiq (replicating Plastic controls),
QTCurve (modified Bluecurve controls), two versions of Liquid (which
unfortunately don't seem to respect color schemes, but maybe I didn't
follow instructions), and of course, lots of Aqua, among others.
And you naturally have the option to hack a cross-environment theme
yourself, or just forget about this standardized look business and just
use different themes for GTK and KDE apps. It's not so bad as long as
you have the color scheme set to be used across the board.
Hope this helps,
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] grub still broke
maxim wexler schreef: >>Remove the "root (hd0,1)" line. That should (I >>hope) let you boot >>gentoo from the floppy. > > > > Arrrgh! Now when I choose "Gentoo" from the menu: > > Booting 'Gentoo' > kernel /vmlinuz root=/dev/hda4 > > Error 15: File not found > > Press any key to continue... > You might want to have a look at the docs page... meaning, Gentoo Linux Documentation -- Gentoo Grub Error Collection at http://www.gentoo.org/doc/en/grub-error-guide.xml It looks like it might be quite helpful :-) Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Random crash
Ian K schreef: > Holly Bostick wrote: > >>2) try a different KDE theme via GTK-QT. I had a lot of problems with >>Liquid, but Plastic is much simpler and might work fine. >> >> > > This is what worked. I am now using krisp, as opposed to Metal4kde. > Glad to hear it. Krisp is pretty nice (I liked it too). > > >>Hope this helps, >> >> > > It did. Thank you for your time. You're quite welcome. Glad you got it sorted. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] KDE 3.4.1 released
Michael W. Holdeman schreef: > On Sunday 05 June 2005 10:39 pm, Robert G. Hays wrote: > >>[digest-mode reply] >> >><... my PC is next to my boyfriend's. ...> >> >> >> >>Holly, do you have any idea how many hearts (just about including mine, >>at this point!) you just broke with that statement? >> >> > > Ya here too, (but my wife is ok with it!) :) > :-D I would feel bad for you guys, but you lost in a fair fight, even if you don't know it. I met Jorden via a user-to-user hardware help forum (!) on the Internet (!!) and we formed a relationship while 3,643 miles (5,864 km) apart (!!!). So while you two get extra points for compliments (thank you :-D ), he wins on speed (quick to appreciate my general coolness :rolleyes: ), boldness (in daring to make his play), flat-out courage (he staked his life savings to sponsor me as promised when I moved house to the Netherlands from NYC), and stick-to-it-iveness (he hasn't thrown me out yet :-) ). However, despite my widely-acknowledged near-perfection :lol: , I am more than willing to play upon your fragile emotional states after this crushing blow :-) to pitch for work if anybody has got some. Work for money would be (extremely) nice, but I'll take credits on a project that would look good on a CV. Read on if you want to know more; otherwise, feel free to delete this completely OT mail right now :-) . -- PITCH FOLLOWS--- By avocation, I'm a writer. This you may have guessed. For 'fun', I write screenplays, but my 'working' day is spent on how-tos and mini-manuals on forums and mailing lists like this one, as well as my own side projects, most of which are ultimately designed to help me write better how-tos and mini-manuals (though a couple of them are designed to help me write better screenplays). I have been published (by a now-defunct magazine, but I still have the original issues in which my work appeared), and I can demonstrably write to spec and on deadline. For examples of my style, one can of course look in the archives here, or Google for 'motub' (the nick is the brand :-) ), but the best places to look would be LQF, the forums of the Player's Resource Consortium at nwnprc.netforums.co.uk/ , and a hosted piece at www.shell-shocked.org/article.php?id=230 . Unfortunately, the Linux Format forums (where I spent a lot of time before I forced myself to cut back) recently upgraded (which was not by any means the unfortunate part :-D ), and all of the previous post data was lost. So all the 'good stuff' I had there is no longer publically available. By trade, I do Customer Service. I worked for over 10 years in CS in various fields: the CBS television network, a film distribution company, a temporary employment agency (by the name of Personnel Express; if you happen to live in NYC and want to call them, or if you work out of the agency and drop by, tell Stephanie, Paula, Nadine, and Randi--and Gwen if she's still there-- hi for me). I've also done 'undocumented' work in real-estate sales, retail sales, cable television production, telemarketing, and worked as a paralegal in my 18 years of work experience. Most of my CS work was also finance-related-- calculating how much people were getting paid, or how much they were paying us, and explaining to them why the amount was what it was (and making them like it, or at least accept it gracefully). This is probably how I developed the the extreme patience that people on this and other lists have noticed and commented on. Unfortunately, good CS requires exceptional communication skills, and while I am generally considered to speak very good Dutch (I read better than I speak, but probably speak better than I write) for someone who's been here only (now) 5 years, it's just not good enough to do what I do, in the view of employers. In fact, it's not good enough in my own view, either-- because apparently I not only "do" Customer Service, but I "am" Customer Service (in the same sense that I "am" a writer). I find that I treat helping people on forums and mailing lists as if it was my work, and hold my performance to the standards of gainful employment in the Customer Service field. Which says a lot about what makes me tick vocationally, but can be seen as a bit excessive, for no money :-) . That's both a strength and a weakness of mine; I can get somewhat too focused when I have a project on the table (and my definition of what constitutes "a project" can be a bit loose). I'm also very stubbornly independent of mind; I tend to unintentionally but inescapably perform required tasks (correctly, but) by way of my own methodology, which is not appropriate to the stricter traditional corporate atmospheres. I've been told by my supervisor in such a corporation that I'm "too creative", which I must concede was an accurate summation. Another boss told me that I "had balls", which statement I took as a compliment (despite the fact that she was firing me at the time she said it), and which lives on as my v
Re: [gentoo-user] Entrance (enlightenment) problem
Juan Ignacio Sánchez Lara schreef: > Hello, > > I've followed the instructions at > http://gentoo-wiki.com/HOWTO_emerge_e17 to install Enlightenment and > Entrance, but I can't open a desktop session. xdm opens Entrance, but > when I input my user and pass it goes black for a moment (as if it > were changing the resolution) and goes back to entrance login without > prompting a word. > > What can go wrong? > > Thanks I had a similar problem with Entrance (although I fixed the issue, I've since stopped using it and gone back to GDM). Several questions: Are you selecting a specific WM from the drop-down menu, or are you just using "Default"? I am not certain that Entrance has/uses a "Default" setting, wherever it may come from. You'd think from /etc/rc.conf, but heck, GDM doesn't use that to set its default (it asks you, so it has its own config for this somewhere), so maybe Entrance doesn't either (although it never asked me about setting anything as default). Do you have any other WMs installed and listed, and have you tried logging into them instead? Do any of them work, and which ones do not? If Enlightenment itself is the one that doesn't work, did you previously have e16 installed? I did, and even though e17 very kindly moved the previous enlightenment.desktop item in /etc/X11/Sessions to e16.desktop, neither the e16 entry appeared in the list, nor did choosing Enlightenment start e17. This turned out to be because the exec line in the Enlightenment.desktop entry was still pointing to /usr/bin/enlightenment, whereas the binary for e17 is named something like enlightenment-0.17 (to distinguish it from the e16 binary). So I changed the exec line in the script, and then it was fine. However I uninstalled both e16 and e17 (I may reinstall e16, which I liked much better), and switched to fvwm anyway. But e17 sure is pretty, can't argue with that. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] plain-vanilla ethernet; do brands matter?
THUFIR HAWAT schreef: > On 6/8/05, Zac Medico <[EMAIL PROTECTED]> wrote: > ... > >>If your >>local store doesn't know maybe they will let you test >>with a livecd before you buy. >> >>Zac > > ... > > fantastic advice with regards to the live cd > > > -Thufir > Sorry, I can't resist... That's one of the uses that a LiveCD is meant for, but the constraints placed around users' ability to think "outside the box" and for themselves completely blinded you to it. This highlights the synergy between software monopoly and hardware manufacture/distribution preventing the buyer from making an informed decision-- you can't test hardware in a store selling Windows PCs, without paying, taking the unit home and hoping it works. Of course, you *really* don't need to test it anyway, because, under Windows, the hardware supposedly "just works" (even when it doesn't, and when it doesn't it's usually supposedly your fault). This 'buying on spec' may be fine for a $25 ethernet card, but not for a $500 video card, and in fact, isn't *really* fine at all. But we've been rigorously trained to accept it as "the way it is". You wouldn't buy a car from someone who refused to let you take it for a test drive, would you? But the "acceptable standards" for the purchase of computing hardware and software are completely different from almost every other consumer product currently available. But now you are really free to get the information you need to make a proper decision for yourself. You are free to demand that the distributor accommodate your needs (if they won't let you test with the Live CD, take your money somewhere else). It takes some getting used to, but man, it's nice being outside that box (more rightfully a cage). It rains sometimes, but then again, quite often you get rainbows after. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] esd or arts, sound daemon howto
Christoph Eckert schreef: > > Even if you're not running KDE you can install arts standalone > and start it via any login script. Dunno where esound can get > started. rc-update in other words, rc-update add esound default . Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world error
fire-eyes schreef: > When performing an emerge -pv world on an x86 server, I got this today: > > These are the packages that I would merge, in order: > > Calculating world dependencies > !!! All ebuilds that could satisfy ">=sys-libs/pam-0.78" have been masked. > !!! One of the following masked packages is required to complete your > request: > - sys-libs/pam-0.78-r2 (masked by: ~x86 keyword) > - sys-libs/pam-0.78-r1 (masked by: ~x86 keyword) > - sys-libs/pam-0.78 (masked by: ~x86 keyword) > > For more information, see MASKED PACKAGES section in the emerge man page or > section 2.2 "Software Availability" in the Gentoo Handbook. > !!!(dependency required by "net-ftp/proftpd-1.2.10-r5" [ebuild]) > > > !!! Problem with ebuild net-ftp/proftpd-1.2.10-r5 > !!! Possibly a DEPEND/*DEPEND problem. > > !!! Depgraph creation failed. > > I am not quite sure what I should do. Unmask something, for use on a server? > I don't think so... > > Does anyone have any ideas, perhaps I made an error in the recent past that > caused this? Or, maybe I'm not alone? > This isn't an "error"-- certainly nothing you've done. It's just information inviting you to specify how you want to configure your machine. The version of proftpd you are trying to install depends on a version of PAM that is not available to install (because it is masked by ~arch). Proftpd 1.2.10-r5 is also ~arch, according to eix. So you've unmasked the program, but not its dependency, which doesn't make a whole lot of sense :-) . Either unmask pam, remove pam from the use flags for this package (echo 'net-ftp/proftpd -pam >>/etc/portage/package.use) so that the package no longer depends on PAM at all, or remask proftpd and install the x86 version, which will presumably depend on a stable version of PAM. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world error
Mark Shields schreef: > I'd highly recommend you don't add ACCEPT_KEYWORDS="~x86" to your make > file, as this will cause your system to emerge with all ~x86 > (testing/unstable) packages upon next emerge. Add this to your > /etc/portage/package.keywords: >=sys-libs/pam-0.78 then emerge -pv > again. > I agree that adding ACCEPT_KEYWORDS="~x86" is a bad idea if you want to run a (mostly) stable system, but *unmasking PAM is not the only solution!!!* PAM is an *optional* dependency; you can compile the program "-pam" and still use the unstable proftpd without having to also use an unstable PAM. I admit that I don't know how proftpd will react if the system as a whole is not PAM-free (like mine), but it itself is but since it is optional, it ought to work OK anyway (if PAM was a required dependency for the package, there wouldn't be a USE flag for it; it would just be installed). Or one could just use a stable version of proftpd which would not require an update to PAM, but instead use the version of PAM currently installed. Under Linux, if there's not at least two ways to do something, there's something wrong but under Gentoo, if there's not at least three ways to do something, you've (probably) missed (at least) one. :) Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] plain-vanilla ethernet; do brands matter?
Antoine schreef: >>My one piece of advice is to avoid RealTek chipsets (usually RTL-8139). >>These are the winmodems of the network world. Most of their processing >>is done by software, not in hardware. > > > Been solid as a rock for me for ages now (is that 8139too?)... I > occasionally shift dvd images around but nothing more serious than that > though... > Cheers > Antoine Same here; my cheap-as-chips and old-as-methuselah 8139 works fine as well-- in fact, it's the one thing I can count on working right if I go distro-hopping, no matter where I wind up. However, you make a good point, Antoine... the driver that has always worked perfectly for me is 8139too (I have a 8139 A-series). The newer driver for the 8139 C+ (8139CP) gave me nothing but trouble, despite the fact it's supposed to be backwards-compatible. Didn't seem so to me. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] CDRW trouble: can't read superblock
Grant schreef: The discs that won't mount do play in a > CD player just fine. Any ideas? Are these then regular Redbook audio CDs you've burned? Those don't need to be mounted. Just put them into the drive, open your media player of choice and play them as you would a 'retail' audio CD. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] updated gnome packages to 2.10, missing menus
Mats Lidell schreef: > Kurt Guenther wrote: > > I had started a quest to remove evolution, by removing a few evolution > packages, that might have caused the problem with the > not-complete-gnome-to-2.10 update. But as I said the update probably > removed all clues since it reinstalled evolution for me. (So I still > have this path to go down if I dare try it again. ;.-) This time I will > try to do it while not updating gnome at the same time. I did this as well, and also wound up with some mis-matched packages (I was having other problems, so I tried to downgrade to 2.8.3). But anyway, the way I "switched over" from gnome (which includes evo and mozilla) to gnome-light (which doesn't) was (without warranty that this is the 'correct' way, or the 'best' way, just the only way I could manage it fairly reasonably): 1) unmerge gnome. This will not remove anything but the metapackage that demands evo, evo-data-server, and mozilla. So they become orphaned dependencies of a package that you have uninstalled. 2) unmerge evo, evo-data-server, epiphany, and mozilla (if desired. I use Firefox, and most things like liferea and others that formerly hooked only into moz have now been updated to hook into firefox instead/as well, so I really don't need Moz hanging around with its bulky self). 3) sync and emerge gnome-light. Possibly -uD, just to make sure you catch anything that may not have been caught in the original GNOME install, and to confirm that everything that needs updating is in fact up to date with the correct version. But if there are any dependencies that you don't already have, they should be installed just like with any emerge. Seems unlikely, though, all things considered. Afaict, this isn't a "clean" switchover; you'll probably have some leftover gnome meta stuff that would not have been installed if you had originally installed gnome-light; Totem, gnome-volume-manager and possibly any gstreamer backend that may have been installed come to mind. But those are all things you might well use anyway, so you can leave them, or dump them, as you please. At least you've got rid of the big stuff, and you have time to compare the gnome and gnome-light ebuilds at your leisure to see if there's any other cruft you don't want (or run emerge -p depclean, or dep, if that's still around). Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] meta-question: how can I receive the messages that I send to the ML?
Antonino Sabetta schreef: >>I'm popping Gmail too, with KMail, but do see my own posts. > > > My fault, sorry :( > I set up a filter in Thunderbird to move every message sent by me in > the sent folder (so that when I send a msg with gmail via web, I also > have a copy of it in my sent folder locally). This obviously is not > what I wanted for the messages I send to mailing lists :) > Thanks, > Antonino > You may already know this, but you can filter the filter further :-) (I do this, and had to kinda figure it out myself, so I figure I'll share it). The short explanation of this example is that sometimes people mail me off-list, but the subject line is the same as the list thread. That's perfectly fine, but I like mail meant for me personally to be in a separate folder than the rest of the list mail so I don't mistake private conversations for public ones. Obviously I couldn't filter on subject or address to accomplish this. But messages from the list have a special header: List-Id, whereas messages from an individual don't have that header, as it's added by the list server. So go to Message Filters in Thunderbird, and create a new filter (don't forget to name it; I always do). Select the first field, and in the drop-down menu, select "Custom" to make a new filter. Type 'List-Id' in the box, and click the Add button, then OK. You can now select 'List-Id' as a basis to filter. If you select any message from the list, and choose View=>Headers=All and then expand the header of the message, you will see that the List-Id header for this list is Gentoo Linux mail So you can just put [List-Id] [is] [Gentoo Linux mail ] in the three fields. If you want to treat certain messages from the list specially, add further filters, but don't forget to select "Match all" rather than "Match any". Or, if you want to do as I did, add [Subject] [contains] [[gentoo-user]], set whatever you want to do with the message, save it, then create another filter exactly the same except that this time set List-Id to [is not], and of course the name and the disposition. The result being that messages whose subject contains the string [gentoo-user] *and* which have the List-Id go to my ML folder, whereas messages that contain the string [gentoo-user] but do *not* contain the List-Id go into my 'off-topic' folder. Messages that contain neither go into my Inbox, of course, but that is as it should be, since I at least glance at anything that winds up in my Inbox unfiltered (I've got a fair lot of filters, both mail and junk). Anyway, this is just one example of how cool T-bird's mail filters are. Definitely one of it's best features if you take a little time to set it up. You should be able to work something out so that you can have your cake and eat it too (have a copy of your sent message and still get it from the list). Two things that come to mind are: 1) BCC yourself rather than relying on the Sent folder (so you will get two messages from Gmail, the BCC that you can filter into your Sent folder, and the list post which you can filter into your List folder) 2) Set up GMail as an account and send the mail via Thunderbird; instructions are on the GMail site, but you probably know that. Hope this info is useful to you. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] So how does one write a CD in 2.6?
Govind Chandra schreef: > Just installed Gentoo 2005.0. > > cdrecord -scanbus says there are issues with kernel 2.5 and newer. > > Is there any way of writing CDs in Gentoo 2005.0? > > Govind > > Of course there is... don't you think you'd have heard about it if none of us could write CD's and DVDs? I mean, that would be *news* ;-) . That message is... well, not untrue, but misleading. We've been beyond the 2.5 stage long enough that while the "issues" are not "fixed" (due to the developer of cdrecord being in philosophical conflict with the kernel devs), they have long ago had workarounds developed for them. So CD recording works fairly normally under Linux-- certainly under X, where there are several frontends available that record such media reliably. Under the console, a read-through of the man pages is in order due to 1) specific settings needed to run cdrecord using an ATAPI recorder rather than a SCSI one; 2) additional programs needed to record DVDs (rather than CDs) or special formats (like ISOs). But it's all pretty straightforward, and works well (I haven't burned a coaster in months). What precisely is the problem you're having? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] CDRW media source
Grant schreef: > Does anyone have a good online source for cheap and reliable CDRW media? > > - Grant > I'm very fond of Opus Supplies, but again they only are available to you if you're in Western Europe (and shipping gets fairly pricey if you're not in NL or BE): www.opus.nl Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge curiosity
reg hughson schreef: > As shown below, why wouldn't "emerge -u world" pick up the update available > for gdm? > > Actually, I think it is probably because it is not listed in > /var/lib/portage/world so I guess I am actually wondering why it wouldn't be > listed there? > > Obviously my system knows gdm is installed but how does it know this? I guess > I was always under the impression that everything I installed would be placed > in /var/lib/portage/world. Obviously not. I know I can edit that file and add > gdm dut that doesn't really answer my question. > It knows this because GDM is a dependency of the gnome meta-package. If you installed that, then GDM was pulled in as a dependency, but dependencies are not placed in the world file. Had you installed (for example) the gnome-light package, which does not install GDM as a dependency, in order to get GDM you would have had to explicitly emerge it, which would have placed it in your world file. If you want to be sure that dependencies will be updated as well as listed packages, use emerge -uaD(tv) world instead of emerge -ua(v) world The --deep switch makes Portage also check the update availability of dependencies of the packages in your world file. Hope this helps. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] So how does one write a CD in 2.6?
Rob schreef: > At 10:35 AM 6/9/2005, Rob wrote: > >> At 04:50 AM 6/9/2005, you wrote: >> >>> Govind Chandra schreef: >>> > Just installed Gentoo 2005.0. >>> > >>> > cdrecord -scanbus says there are issues with kernel 2.5 and newer. >>> > >>> > Is there any way of writing CDs in Gentoo 2005.0? >>> > >>> > Govind >>> > >>> > >>> >>> Of course there is... don't you think you'd have heard about it if none >>> of us could write CD's and DVDs? I mean, that would be *news* ;-) . >>> >>> That message is... well, not untrue, but misleading. We've been beyond >>> the 2.5 stage long enough that while the "issues" are not "fixed" (due >>> to the developer of cdrecord being in philosophical conflict with the >>> kernel devs), they have long ago had workarounds developed for them. >>> >>> So CD recording works fairly normally under Linux-- certainly under X, >>> where there are several frontends available that record such media >>> reliably. Under the console, a read-through of the man pages is in order >>> due to 1) specific settings needed to run cdrecord using an ATAPI >>> recorder rather than a SCSI one; 2) additional programs needed to record >>> DVDs (rather than CDs) or special formats (like ISOs). But it's all >>> pretty straightforward, and works well (I haven't burned a coaster in >>> months). >>> >>> What precisely is the problem you're having? >>> >>> Holly >>> -- >>> gentoo-user@gentoo.org mailing list > > Dear list, > > Here is my experience: I followed the kernel warnings that ide-scsi was > depricated, then found out that most of the burner programs would not > work or worked lousy with an ide ATAPI interface. So I went back to the > hdc=ide-scsi kernel option. I would like to, however, follow the > correct procedure, that is, use an ide atapi interface, only it didn't > work out for me the first time. Any help or advice would be appreciated. > > Sincerely, Rob. > What kernel was this under, and what kernel are you using now? In the very early days of 2.6, one really had to explicitly run ide-cd (the replacement for ide-scsi) in order to get things working, but again, we are far beyond that now. Ide-cd is automatically compiled into the kernel (you don't even get to choose it, it's just there), and /dev/hdc (assuming that that's where your CD/DVD burner is) will use it when it is detected to be a writeable device during the boot process. So again, I ask what precisely your issue is what goes wrong when you try to burn a CD or DVD? What commands are you using to do so, and what type of disk are you trying to create (a bootable ISO? a data disk? a SVCD? an audio disk)? If we knew that, maybe we'd know where to start looking for a misconfiguration or missing backend subsystem. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] So how does one write a CD in 2.6?
Rob schreef: > At 10:55 AM 6/9/2005, you wrote: > > >> --- Rob <[EMAIL PROTECTED]> wrote: >> >> > procedure, that is, use an ide atapi interface, only >> > it didn't work out for >> > me the first time. Any help or advice would be >> > appreciated. >> >> Hi Rob, >> >> What do you mean "it didn't work out"? More specific >> please. Instead of using the scsibus address like >> -dev=0,0,0 or whatever you simply substitute >> -dev=/dev/hdc. >> >> Zac > > > Hi Zac, > > Sorry for not being specific enough. What I mean was that by specifying > /dev/hdc as the cd burner, the only program I could get to work was > x-cdroast. None of the other burner programs that I emerged and tested > would work with the ATAPI interface. So that is why I went back to the > deprecated ide-scsi interface. > > So again I am asking, did I do something wrong? Or am I just caugh in a > temporary transition to /dev/hdc? > > Thanks alot!! Rob > > > Rob, I regret to inform you that your syntax seems to be all wrong, and this stands a good chance of being your problem. I don't myself use cdrecord directly all that often, but I did manage to remember cdrecord dev=ATAPI -scanbus which produced this output: Cdrecord-Clone 2.01 (i686-pc-linux-gnu) Copyright (C) 1995-2004 Jörg Schilling on-the-fly encryption (version 1.0-rc1) built-in, (C) 2004,2005 Maximilian Decker NOTE: this version of cdrecord is an inofficial (modified) release of cdrecord and thus may have bugs that are not present in the original version. Please send bug reports and support requests to . For more information please see http://burbon04.gmxhome.de/linux/CDREncryption.html. The original author should not be bothered with problems of this version. cdrecord: Warning: Running on Linux-2.6.11-gentoo-r8 cdrecord: There are unsettled issues with Linux-2.5 and newer. cdrecord: If you have unexpected problems, please try Linux-2.4 or Solaris. scsidev: 'ATAPI' devname: 'ATAPI' scsibus: -2 target: -2 lun: -2 Warning: Using ATA Packet interface. Warning: The related Linux kernel interface code seems to be unmaintained. Warning: There is absolutely NO DMA, operations thus are slow. Using libscg version 'schily-0.8'. scsibus0: 0,0,0 0) '_NEC' 'DVD_RW ND-2510A ' '2.15' Removable CD-ROM 0,1,0 1) * 0,2,0 2) * 0,3,0 3) * 0,4,0 4) * 0,5,0 5) * 0,6,0 6) * 0,7,0 7) * so the correct syntax to access my burner would be cdrecord dev=0,0,0 whatever comes after that. The whole /dev/hdc thing is just not correct (it's "dev=" whatever). Check the man page for more info. Hope this helps get you pointed in the right direction. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] So how does one write a CD in 2.6?
Rob schreef: > At 10:55 AM 6/9/2005, you wrote: > > >> --- Rob <[EMAIL PROTECTED]> wrote: >> >> > procedure, that is, use an ide atapi interface, only >> > it didn't work out for >> > me the first time. Any help or advice would be >> > appreciated. >> >> Hi Rob, >> >> What do you mean "it didn't work out"? More specific >> please. Instead of using the scsibus address like >> -dev=0,0,0 or whatever you simply substitute >> -dev=/dev/hdc. >> >> Zac > > > Hi Zac, > > Sorry for not being specific enough. What I mean was that by specifying > /dev/hdc as the cd burner, the only program I could get to work was > x-cdroast. None of the other burner programs that I emerged and tested > would work with the ATAPI interface. So that is why I went back to the > deprecated ide-scsi interface. > > So again I am asking, did I do something wrong? Or am I just caugh in a > temporary transition to /dev/hdc? > > Thanks alot!! Rob > > > Actually (since I don't use cdrecord often), I noted down the command I must use to generate and burn an iso file, so here the 'burn' section: cdrecord -v -dao -eject dev=ATAPI:0,0,0 driveropts=burnfree /wherever/the/file/is.iso for example. So the syntax is dev=ATAPI=#,#,# ,depending on what scanbus told you your device was. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fstab entry
Martins Steinbergs schreef: > hi, > > what's wrong with these, users cant access partitions (Access denied to > /mnt/win_j.), only root can go there. > > fstab > > /dev/hda1/mnt/win_cntfsdefaults,ro,user0 0 > /dev/hdb8/mnt/win_jvfatdefaults,rw,user0 0 > > > Martins Noting in particular as this is not so much an fstab issue but a permissions issue, which are normally not controlled by fstab. However, for vfat (and probably ntfs), you can actually control permissions via fstab (I think the Windows fses are the only ones you can use fstab to control their permissions). The vfat mount needs: 1) that the mount folder /mnt/win_j should have its permissions manually adjusted so that users (the group users, if nothing else) has rights to read (and write to if desired) the folder itself; 2) the fstab entry should include at least the gid= option (see man mount), so that the group to which your users belong has some ability to own the files. You may also want to use the uid= option if you desire a particular user to own the contents of the folder when mounted. I only specify this for vfat, because I don't know if ntfs takes these options (probably does). Check man mount for options specific to ntfs mounts, and to learn more about the options for vfat mounts. Man mount is well worth a good read through. You can do quite a lot with it, once you know how. Optional: You may want to get rid of that 'defaults' option; it restricts the mount from being automounted, and restricts any executables on the mount from being executed (afaik). 'default' means, among other options, noauto, nosuid, and noexec; you have managed to avoid the default setting of 'defaul root-only mounting' by adding the 'user' flag after the 'default' flag (so the user flag supercedes the default setting), but that has always seemed redundant to me. I prefer just to get rid of 'defaults' and explicitly set the options the way I want them. YMMV. You may also want to use the umask= option to specify read/write/execute permissions for the files mounted. Again, see man mount. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] So how does one write a CD in 2.6?
Richard Fish schreef: > Holly Bostick wrote: > > >>so the correct syntax to access my burner would be >> >>cdrecord dev=0,0,0 whatever comes after that. >> >>The whole /dev/hdc thing is just not correct (it's "dev=" whatever). >>Check the man page for more info. >> >> >> > > > Actually Holly, both methods "should" work. I use dev=/dev/cdrw without > any trouble at all. > >>From what I remember reading, the kernel devs prefer this, because to > them, creating an artificial numbering system (bus, id, lun) for a bus > topography (IDE, SATA) that doesn't use it to be just silly. They > consider the whole ide-scsi thing to be an unnecessary hack. Plus, > identifying things by their dynamically-assigned address on a bus > doesn't work well in a hot-pluggable world. > > The cdrecord author prefers the dev=x,x,x syntax, because then the same > syntax is used to access all CD-R[W] drives on all platforms. > > Neither side seems willing to accept the other's viewpoint as valid. > > -Richard > Fair enough, Richard. I'm sure that dev=/dev/hdc would likely work. However, it would have to be /dev/hdc (the original device name), because I cannot necessarily be sure that I will always have /dev/cdrw, for example. Atm, my DVD burner as hdc is symlinked to /dev/cdrom, /dev/cdrw, /dev/dvd and /dev/dvdr, so I could probably use any of those. But that's all udev's doing, and under other Linuxes, other versions of udev, or even with other drives, I'd have to check the symlinks every time to make sure that my DVD+R wasn't now *only* linked to /dev/dvd and /dev/dvdr, and no more to /dev/cdrom, and /dev/cdrw. Of course I'm sure that /dev/hdc would always work under all circumstances unless I physically moved the drive. But if I scan the bus and see what cdrecord thinks, then that will also always work. So I agree that both viewpoints are valid, but not once you start adding in the symlinks to the real device, because that's a distribution-specific config issue and not a 'base' kernel or program issue... and if you're distro-hopping or multi-booting, the most irritating thing is to have to remember that under X distro the command is one thing, and under Y distro, the command is another. To me, anyway. I have a hard enough time remembering CLI commands as it is :-) . Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Hardware problem
Bill Six schreef: > Hi, > > This isn't a Gentoo related problem, but a hardware > one (I know next to nothing about hardware). > > First of all I have a Micron computer with a floppy > drive, DVD-read drive (/dev/hdc), CDR drive > (/dev/hdd), a master harddrive (/dev/hde), and a slave > harddrive (/dev/hdf). > > Yesterday my computer was running fine. Today my dad > put in a DVD burner where the CDR was previously > (/dev/hdd). He also took both harddrives out and put > them back in because the ribbon (whatever it's called, > the grey cable that goes into each drive) was twisted. > > Now, when I try to start up my computer, it doesn't > start at all, it just goes "Beep (wait 3 > seconds)...Beep (wait 3 seconds)...Beep (wait > 3 seconds)..." over and over. It doesn't show the > splash screen for the video card, or for the > manufacturer (Micron). > > Any idea why it would do that? What do the beeps mean? The beeps tell you why it would do that. Depending on your BIOS type (AMI, Award) the pattern of the beeps is meant to tell you what has failed. The information as to what pattern means what is available on the Internet, but you may not have access to the web, or know what kind of BIOS you have, and researching all of that now would be a distraction. > I put the CDR back in instead of the DVD burner, and > it still gives the same error. Since replacing the DVD with the original CD didn't help, that's probably not the problem-- and I wouldn't have thought it was anyway. I've assembled a fair number of computers in my time, so from experience I can take an educated guess that the problem is also not the removal and replacement of the hdd cables (if it was, the system would probably boot up to the point where it needed to access a drive and *then* fail). The fact that you don't even get a POST (that's the part where you count the memory and that shows the logos you aren't seeing), strongly suggests that what has happened is that your dad "misattached"* one or more *power* cables when attaching them to the drives. The motherboard seems to have power, or else you wouldn't even have beeps. *I say "misattached", but it's probably not his fault, if this is in fact the problem. [Inexact explanation follows] The thing is, current power supplies have a certain number of "leads" (maybe 3 or 4), and then these leads are split into multiple connector types (floppy, hdd, motherboard), of which you maybe have 10 in total. What one (and I don't mean just your dad, I've been bitten by this more than once myself) fails to realize is that just because there may be 5 connectors on one lead, this is not the same as 5 separate leads, and all of the connectors on any given lead are sharing the power output of that particular lead. If the group of devices connected to a single lead via the various split connectors demands more power than the lead as a whole is designed to output, the lead won't output anything (to prevent damage to itself or to your components). So it often happens that one has to fiddle with which set of power connectors must be connected to which devices so that the system will boot. If you bought this computer pre-assembled, that was all done by the OEM; it's not surprising that your dad wouldn't have thought that any old hdd power connector couldn't be connected to whatever hdd it was closest to, regardless of what other devices were already on that particular splitter. It's an extremely tedious operation, switching around power connectors, but worth a try. While you're back there, make sure that the IDE ribbon cables are correctly plugged into the hard drives (it's quite easy to reattach them "crooked"-- i.e., not correctly pin-matched, with some pins not in their holes, despite the fact that this is not supposed to be possible). Anyway, that's what I'd check first, but maybe somebody else has a better idea. In any case, hope it's helpful. > > Last time I let my dad touch my computer haha. Awww, give him a break. How else is he going to learn? Don't discourage him just because he didn't get it quite right :-) . Holly > > Thanks for any help, > > Bill Six > > > > > > __ > Discover Yahoo! > Use Yahoo! to plan a weekend, have fun online and more. Check it out! > http://discover.yahoo.com/ -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge error "Bad interpretor"
Ognjen Bezanov schreef: > Zac Medico wrote: > > >>--- Ognjen Bezanov <[EMAIL PROTECTED]> wrote: >> >> >> >> >>>Hello, >>> >>>I set the portage_tmpdir in my make.conf file, and >>>have tried emerging >>>some packages. But there is a problem, because i >>>keep getting errors no >>>matter what I emerge, the error looks like this: >>> >>>..etc/ebuild.sh ./configure: /bin/sh: bad >>>interpreter: Permission >>>denied >>> >>>and yes, I am running as root. Anyone got any tips? >>> >>>P.S please CC me, i dont trust my email address >>> >>> >>> >> >>Hi Ognjen, >> >>Somebody already said that the problem must be that >>you have the noexec option on that partition because >>of the way it is mounted. See the manpage for mount. >> >> >> > > Sorry I didnt get that reply. Either way to answer that question, no - i > do not have the noexec option in my fstab (where the drive is > automounted). This is the line in my fstab: > > /dev/hda6 /mnt/storage reiserfs notail,exec,user > > > What could the problem be? It is strange because if I type: > > mount /dev/hda6 /mnt/storage -o exec > > it works. > > What do I need to add to my fstab to get it working normally? > > thanks. > Maybe nothing-- what are the permissions of the /mnt/storage *folder*? It doesn't matter much whether the files inside the folder are executeable if the folder itself does not allow files to be executed from within it. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fix_libtool_files.sh question
Grant schreef: > I've run into a problem trying to emerge gnome-vfs on one of my > systems. It complains about libstdc++.la but there is a bug report > that indicates running fix_libtool_files.sh will fix it. How do I > know which to use? > > - Grant > It's in the error message. Just before the whole "exit here", "exit there" message, there's a line detailing some failure to find some file in gcc-version.something.or.other. That's the old GCC version which is no longer valid (which is why the files aren't there, they're in the folder of the new GCC version), and so needs to be fixed with fix_libtool_files.sh. See, it all makes sense :-). -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] A better USE flags documentation?
Colin schreef: > I've been Gentooing for some time now, but there are still USE flags > that confound me. Like "ftp" and "ssl," for instance--do you need those > to make FTP/HTTPS connections in a web browser, or are those flags just > for incoming connections, as with FTP/Web servers? If a USE flag is explicitly listed when running emerge with the --verbose switch, it's optional. For example, the ability to make FTP connections is obviously not optional for an FTP server, so there is no ftp USE flag for such programs. Looking at the emerge -pv output for Firefox, there is no ftp or ssl USE flag for that either, because the functionality is part of the program (not optional). So the meaning of the flag depends on the program to some extent-- for what application is ssl an optional configuration? If you encounter such an application, it's usually pretty clear what enabling or disabling the option does within the context of what the program as a whole does. > Does "-mozilla" > block Firefox- and Thunderbird-related stuff? It "blocks" Firefox related stuff as well as Mozilla related stuff, since they're built on the same engine; for example, if you compile Liferea (the RSS news reader) -mozilla, when you click on a news story, the page will open up using the program's internal HTML rendering engine rather than the external Mozilla or Firefox engine that may be present on your system. Basically, you told Liferea not to look for such a thing when compiling, so it doesn't. But of course the use of any such external engine is *optional*, so you may turn it off. On the other hand, the Galeon web browser requires Mozilla to run, so you *will* install Mozilla if you try to install Galeon, no choice about it (no matter how daft it seems to install a web browser you don't want in order to run another web browser you do want). As far as I know, there is no "Thunderbird-related stuff" that other applications might optionally use. So no worries there And what happens to use > flags that aren't specified--are they treated as "flag" or "-flag?" USE flags that aren't specified in what way? If you mean not present within the list of options displayed by the -v (or --verbose) switch (the text editor Gedit, for example does not have any plus or minus mozilla USE flag associaed with it), then of course it's not used. If you mean not specified by you to be used in either /etc/make.conf or /etc/portage/package.use, then it's going to be off unless you've turned it on globally (in /etc/make.conf) or locally just for the one package (in /etc/portage/package.use). Gentoo would never be so impolite as to just turn stuff on without asking you-- except in /etc/make.profile/make.defaults, but even stuff there you can turn off in /etc/make.conf. > > I've got a lot of questions, but gentoo.org's USE flag documentation > really isn't all that great. More than one line about a flag would > help. Is there any better documentation out there? global USE flags are very contextual; their meaning depends on the application they're being associated with, so they can only be described in general terms. Local USE flags, specific to a particular application, like "moznomail" for mozilla alone, are fairly well described by one line, being much more limited in scope. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Can't su - to root, etc
Michael Sullivan schreef: > My network updates its software every night at midnight. There was a > problem with a couple of blocking packages night before last, so I > unmerged the two blocking packages and did the emerge -avuD world in a > screen. Now they're done and I can no longer su - to root from my > personal account. I have to log completely out and then log in as root > from the login screen. I use users-admin for my user maintenance, but > now there's a tab that says "User Privileges" where the "User's Groups" > tab used to be, so I can't figure out how to make sure that my personal > account is in the right group(s) to su - to root. A lot of things are > different now. I keep a gnome-terminal open at all times and it the > prompt now shows "[EMAIL PROTECTED] ~ $" when I'm in my home directory > instead of "[EMAIL PROTECTED] michael $" which I preferred. I could su - to > root yesterday while the upgrades were taking place and root's prompt > was the same. And then my Inbox Monitor applet is gone, so I have to > click on the running evolution button every time I want to see if I have > any email (which is quite often actually). Is there any way I can fix > all this, or at least get my su - privileges back? > > Obviously the su to root issue is more important, but I know the cause of the other two issues (the change in the prompt and the loss of the mail checking tool). The change from "[EMAIL PROTECTED] explicit home directory name" to "[EMAIL PROTECTED] ~", is, I believe, a change in Bash, which only affects (afaik) the display name of the user's home directory. First of all, in Bash 3, the usage of /w and /W have been reversed; the lowercase now expands the full path display and the upper case truncates to just the cwd (current working directory). Somewhere in the revisions to Bash 2.05, the expansion of the /w and /W variables when in your /home was changed so that /home/username (or /root), which the /w variable already truncated (showing just the directory without the path to it), has been shortened from multiple characters (whose username is just one character?) to one character: ~, which is the commonly acknowledged abbreviation for /home/username. This affects the display no matter how deep you go into the user's /home directory (at least in Bash 3; I just upgraded, and I don't think Bash 2.05 -r9 did this)-- /home/username/.mozilla/firefox is now displayed as ~/.mozilla/firefox. I'm almost sure that yesterday (when I was still using Bash 2.05) it would have been /home/motub/.mozilla/firefox. But actually, I like it; I currently have my prompt split over 3 lines to prevent long cwds from destroying my prompt; this might mean I could lower that number to 2 lines, since the display of the cwd is much shorter by default. Other than that, I see no way to change it (it's a feature, not a bug), other than changing shells, which is also a possibility, of course-- but I, at least would need a better reason than that to go to the troule of learning a new shell, when I hardly know bash. -- As for the mail checker-- you must have upgraded from <=GNOME 2.8.x to GNOME 2.10. The previously included mail-notification utility does not work with GNOME 2.10, and in fact should have been removed. Try mail-notification (emerge mail-notification); it's a nice email checker for the GNOME panel whose benefit is that you can set it up to check both POP mail and GMail (possibly only if you have POP download enabled), as well as other types of accounts, such as IMAP or system mail. I have also used GBiff (emerge gnubiff), which is much cuter with a Povray Tux icon, but I don't believe it checks GMail. But for general use, it's fine, and has much the same featureset as mail-notification. And of course, there's GkrellM's mail notification plugin, gDesklets may a mail checking utility, and various dockapps provide this functionality as well, if you happen to use a dock. So you just have to use a different program of your choice, rather than relying on a pre-installed tool. --- As to the inability to su to root, assuming that the user is in the wheel group-- don't I remember hearing that this is some issue with the update to 'shadow'(or pam and shadow), and something has to be fixed in /etc/login.defs oh right, it's something about the SU_WHEEL_ONLY=yes, which sounds like it means that only the wheel group should be able to su, but in fact means that only members the ROOT group (who are also members of the wheel group) can su (due to another setting in another file which also restricts the ability to su)... I read here that the solution is to change this setting to NO. But that results, as I understand it, in everybody and their little sister being able to su to root, which is not really optimal. A better proposed solution is to reset SU_WHEEL_ONLY to 'no' and also to "create /etc/suauth with this sing
Re: [gentoo-user] emerge -pv world error
Walter Dnes schreef: > On Wed, Jun 08, 2005 at 06:17:06PM +0200, Holly Bostick wrote > > >>PAM is an *optional* dependency; you can compile the program "-pam" and > > > PAM is an *OPTIONAL PROGRAM*; build your system with... > > [m450][root][~]cat /etc/portage/package.mask > sys-libs/pam Yes, I know (my system is built without PAM, and I didn't mask it, I removed it, re-emerged shadow and then the system -pam in the first place), but removing PAM if your system is built from default settings is a fairly complex task (though not as complex as it used to be), and a distraction from the original question. Therefore I suggested a quicker solution to solving the original problem (getting one program installed and proceeding with the world update), which was removing the pam dependency for that program only (you can then decide about the rest later). If one wants to remove PAM totally (which is not a bad idea, but opinions vary), there's lots of threads on the forums that provide adequate guidance, from how to remove PAM on a running system, to how to install without it in the first place. > > in addition to "-pam" in use, and you will find Gentoo a lot more > pleasant. One question... why does OpenOffice *DEMAND* PAM to build > from source??? The binary ebuild installs OK without PAM. > There's a bug about this. http://bugs.gentoo.org/show_bug.cgi?id=71898 and the article in the Wiki http://gentoo-wiki.com/HOWTO_Remove_PAM also discusses this, as well as the similar (previous) issue with GDM, whose ebuild has been updated to make PAM an optional dependency. Myself, I use the binary OOo build anyway; I just didn't have the patience to deal with that long compile, no-pam patch or not (but I did compile OOo-ximian once with the patch, and it built, anyway. Apparently there's some delay in upstream testing and including the patch to make PAM optional. But since they're so close to finishing 2.0, I guess I'm not surprised that they might have other priorities. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] problem with emerge
Tim Igoe schreef: > > Allan Spagnol Comar wrote: > >>Hi all I am having some problem with emerge, I had emerged bootsplash, >>then I discover in this list that for newer kernels we should use >>splashutils, then I unmerged bootsplash and emerged splashutils, now I >>am trying to update my system and I am receiving this message: >> >>[blocks B ] media-gfx/splashutils (is blocking >>media-gfx/bootsplash-0.6.1-r7) >>[blocks B ] media-gfx/bootsplash (is blocking >>media-gfx/splashutils-0.9.1) >>.2] >>[ebuild N] media-gfx/bootsplash-0.6.1-r7 >> >>what should I do to fix it ? >> >> > > Try unmerging the pair of them, then emerge splashutils > > If not, emerge -uDpt world - look at the tree produced to work out what > pulls in bootsplash > Cookie bet that it's a bootsplash-themes package... if so, uninstall said package (since you can't use it if you don't have bootsplash anyway), and then try the update again. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Can't su - to root, etc
Michael Sullivan schreef: > I emerged mail-notification awhile back (someone on this list was > talking about it.) I opened a terminal and typed in mail-notification > and I get this message: > > [EMAIL PROTECTED] ~ $ mail-notification > > (mail-notification:15971): Gnome-WARNING **: Accessibility: failed to > find module 'libgail-gnome' which is needed to make this application > accessible > GTK Accessibility Module initialized > > (mail-notification:15971): Gnome-WARNING **: Accessibility: failed to > find module 'libatk-bridge' which is needed to make this application > accessible > mail-notification-Message: Mail Notification is already running > > It claims to be already running, but I don't see it anywhere on the > panel. I have a notification area applet running, but it's not showing > me anything... > > On Sat, 2005-06-11 at 11:22 +0200, Holly Bostick wrote: > If you don't have any mail, you don't see anything (the icon appears and disappears when mail is detected or downloaded). And of course you don't have any mail, as far as the applet is concerned, because you haven't configured it. Now, the fact that you haven't configured it suggests that the package did not install properly, because the first time you run it, you should get the default settings, which include the popup window-- which has a menu so you can configure the program to know where your mail is. So if you don't see the popup window, on firstrun, something is wrong, and the only errors you've mentioned are in the accessibility area, which 1) doesn't seem related and 2) I turned off in gnome anyway, so I wouldn't know how deeply related this might be to the application in question. The first thing I would do is open gnome-system-monitor and see if it's still running. If it is, kill it. Then, rather than running the app from the terminal, use the GNOME panel menu to add it to your panel. Then it should hopefully start correctly-- or at least give more info as to why it isn't. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fglrxconfig
Mark Shields schreef: > after you emerged ati-drivers, did you modprobe the driver and/or add > it to /etc/modules.autoload.d/kernel-2.x (where x is your minor > version, e.g. 4 or 6, as in 2.4 or 2.6)? I just replaced my ati card, > so I'm not sure what the module name is. > Unsurprisingy, the module name is fglrx. Antoine, the "no screens found" error usually means that some section of xorg.conf is misconfigured (often the "monitor" section, whether that is a CRT, LCD, or television). But since the X server's screen is composed of 4 parts: - input device (mouse) - input device (keyboard) - video card - monitor the misconfiguration of any of these will result in a "no screens found" if there is no alternative configuration that works. The fact you get a "no devices found" error suggests that Mark is right and the fglrx driver isn't loaded (so the device is not "on", as it were). So try that first, and if it still doesn't work or gives a different error, post your xorg.conf and maybe somebody can spot the issue (I use the fglrx drivers, but not TV-out, so I don't know anything about it). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] problem with emerge
YoYo Siska schreef: > Holly Bostick wrote: > >>Tim Igoe schreef: >> >> >>>Allan Spagnol Comar wrote: >>> >>> >>> >>>>Hi all I am having some problem with emerge, I had emerged bootsplash, >>>>then I discover in this list that for newer kernels we should use >>>>splashutils, then I unmerged bootsplash and emerged splashutils, now I >>>>am trying to update my system and I am receiving this message: >>>> >>>>[blocks B ] media-gfx/splashutils (is blocking >>>>media-gfx/bootsplash-0.6.1-r7) >>>>[blocks B ] media-gfx/bootsplash (is blocking >>>>media-gfx/splashutils-0.9.1) >>>>.2] >>>>[ebuild N] media-gfx/bootsplash-0.6.1-r7 >>>> >>>>what should I do to fix it ? >>>> >>>> >>> >>>Try unmerging the pair of them, then emerge splashutils >>> >>>If not, emerge -uDpt world - look at the tree produced to work out what >>>pulls in bootsplash >>> >> >> >>Cookie bet that it's a bootsplash-themes package... if so, uninstall >>said package (since you can't use it if you don't have bootsplash >>anyway), and then try the update again. >> >>Holly > > > don't bootsplash themes work with fbsplash and splashutils? > i think i'm using my old bootsplash theme right now... > but i may be wrong ;) > > yoyo No, they don't-- or in any case, such themes have to be converted in order to run (since fbsplash is now so bootsplash compatible). These are different programs that use different syntax to do the same type of thing. They use a different backend as well; bootsplash doesn't know anything about vesafb-tng, either. In any case, any bootsplash-themes package you have installed doesn't know anything about fbsplash or splashutils-- it thinks it is dependent on bootsplash, therefore it would drag it in. Since it's about the only thing that would, as far as I can tell, a simple 'eix bootsplash' will confirm if I've won a cookie today (or not). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 3 - ERRORS
Rumen Yotov schreef: > Joseph wrote: > > >>How do you replace revdep-rebuild, it is not in ebuild? >>I solved the problem by recompiling OO from source instead of binary. >> >> >> > > Hi, > Revdep-rebuild is in portage-package, as '/usr/bin/revdep-rebuild'. > It's a shell-script, no need to compile it, just to unpack/place it. > You could copy the new one over the old one (make a backup first). > Later check the flags/permissions etc. > HTH. Rumen Thanks for the info, but I notice that there's also a patch to portage.py attached to the bug. Is it correct to just patch it with the standard "patch -p1 blah blah blah" (patch syntax doesn't roll trippingly off my typing finger, but I'll look it up before proceeding)? Is this "safe" (insofar as it's patching a portage file, unlike revdep-rebuild itself)? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] music.raw?
Michael Sullivan schreef: > I've discovered that whenever I run monsterz the file is created. I > don't know why. monsterz is a python script, so I opened it up in vi > and searched for the string "music.raw" but did not find it > in /usr/games/monsterz. Maybe something the pygame module or python > itself, but I don't know how to check for that... > Does the game play music? Maybe the file is how it converts the music from whatever form it's in, wherever it's kept, to something that is played during the course of the game. Maybe it's the music itself. I don't know a lot about this, and certainly not about monsterz specifically, never having played it, but I've seen games get up to all kinds of hijinks in order to play music or sound effects, under both Win and Lin. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Java java java, I miss my java
Wade Brown schreef: > So far I've been able to run eclipse, azureus, and mozilla java > with no problems using the latest sun-jdk, but do be warned several > programs are expected to fail building. > I use a variation on this; my system jre and jdk are blackdown-1.4.2, but azureus really really prefers sun-1.5.whatever (or at least 2.3.0.0 did, just upgraded to 2.3.0.2--which finally made it into Portage-- but haven't used it yet; it's supposed to solve this issue, afaik). Since 1) azureus crashed alot with blackdown, and 2) I didn't want to change my entire system to sun 1.5, since I had no idea of the effects and 3) not changing my entire system results in having only the sun jre, but not the sun jdk (so a mismatched system), my solution was to write a small script to run azureus: #!/bin/sh java-config -s sun-jre-bin-1.5.0.03 source /home/motub/.gentoo/* azureus Works fine, and saves me having to remember to do it manually every time I want to start the program (and being reminded when I tried to open the Configuration tab and the program crashed with a Java HotSpot error). So, for what it's worth, it is possible to manage 'stable' and 'unstable' java versions on the system, at least as far as pre-compiled java-based executables go. I'm not a Java developer (or any kind of developer), so I don't really have much interest in testing the compilation of programs against the new Java, which, as I understand it, may work, but may not. What I'm wondering is whether blackdown is ever going to update to a 1.5 version. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Sync only installed and dependence packages with portage.
Qian Qiao schreef: > Hi, > > As the title suggested, I've got a box with very limited disk space, > is it possible to sync only the packages currently installed and their > dependencies with the portage tree and leave out the rest? > > TIA. > > -- Joe > Hi, I don't think so, but I don't think it makes such a huge amount of difference. When you sync the Portage tree, you are only adding and deleting ebuilds, which are very small. On my system, 98,135 files in the Portage tree take up 296.5 MB-- which is actually more than I expected; have to look into that. Oh... I did a couple of "qpkg"s when I was unmerging that extra version of Python. Without counting /usr/portage/packages, 97,919 files take up 93.5MB of space, which is more like what I would have figured. IMHO, if you don't even have that much room available, it's time to start clearing out space, repartitioning, or considering getting a new HDD anyway. However, this is without counting /usr/portage/distfiles, which holds the tarballs... if you don't mind re-downloading them if you decide to reinstall a program previously installed (or installed, then uninstalled), a more effective space-saving measure might be to delete the contents of that folder, as well as backing any packages you may have used qpkg to create off the drive. IIrc, there's a setting in /etc/make.conf that will allow you to dump the tarballs automatically, but you could always do so as a cron job. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: rc.conf is very short now?
Richard Fish schreef: > > I think PROTOCOLS was voted off the island. Watching a bit too much Survivor/Big Brother/The Real World (any city)/The Farm/The Bus/Temptation Island/Expeditie Robinson/Celebrity Survivor/Idols/American Idols/America's Next Top Model lately? :D Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Reboot when GRUB starts
Colin schreef: >>>On Mon, 13 Jun 2005, Colin wrote: >>> >>> /boot/grub/grub.conf === default 0 timeout 10 splashimage=(hd0,0)/grub/splash.xpm.gz title=Gentoo Linux 2.6.11-r9 root (hd0,0) kernel=/kernel-2.6.11-gentoo-r9 root=/dev/hde3 video=vesafb:mtrr,ywrap vga=0x31B >>> >>> >>>Shouldn't that be: >>> >>>kernel /kernel-2.6.11-gentoo-r9 root=/dev/hde3 video=vesafb:mtrr,ywrap >>>vga=0x31B >>> >> >>Good catch! Yes, fix this first. Also, there is no '=' for the title. > > > The handbook shows an equal sign. It's worked before on another system. No, it does not: Code Listing 3: grub.conf for non-genkernel users # Which listing to boot as default. 0 is the first, 1 the second etc. default 0 # How many seconds to wait before the default listing is booted. timeout 30 # Nice, fat splash-image to spice things up :) # Comment out if you don't have a graphics card installed splashimage=(hd0,0)/grub/splash.xpm.gz title=Gentoo Linux 2.6.11-r3 # Partition where the kernel image (or operating system) is located root (hd0,0) kernel /kernel-2.6.11-gentoo-r3 root=/dev/hda3 # The next four lines are only if you dualboot with a Windows system. # In this case, Windows is hosted on /dev/hda6. title=Windows XP rootnoverify (hd0,5) makeactive chainloader +1 Code Listing 4: grub.conf for genkernel users default 0 timeout 30 splashimage=(hd0,0)/grub/splash.xpm.gz title=Gentoo Linux 2.6.11-r3 root (hd0,0) kernel /kernel-2.6.11-gentoo-r3 root=/dev/ram0 init=/linuxrc ramdisk=8192 real_root=/dev/hda3 udev initrd /initrd-2.6.11-gentoo-r3 # Only in case you want to dual-boot title=Windows XP root (hd0,5) makeactive chainloader +1 Perhaps you're thinking of the LiLO config: Code Listing 11: Example /etc/lilo.conf boot=/dev/hda # Install LILO in the MBR prompt# Give the user the chance to select another section timeout=50# Wait 5 (five) seconds before booting the default section default=gentoo# When the timeout has passed, boot the "gentoo" section # For non-genkernel users image=/boot/kernel-2.6.11-gentoo-r3 label=gentoo# Name we give to this section read-only # Start with a read-only root. Do not alter! root=/dev/hda3 # Location of the root filesystem # For genkernel users image=/boot/kernel-2.6.11-gentoo-r3 label=gentoo read-only root=/dev/ram0 append="init=/linuxrc ramdisk=8192 real_root=/dev/hda3 udev" initrd=/boot/initrd-2.6.11-gentoo-r3 # The next two lines are only if you dualboot with a Windows system. # In this case, Windows is hosted on /dev/hda6. other=/dev/hda6 label=windows GRUB does not take an equal sign in that position, but Istr that Lilo does. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Java java java, I miss my java
Zac Medico schreef: > Kevin O'Gorman wrote: > >>Here's what it looks like: >> >>treat 1.5-bundles # emerge -av sun-jdk sun-sdk-docs >> >>These are the packages that I would merge, in order: >> >>Calculating dependencies - >>!!! All ebuilds that could satisfy "=dev-java/java-sdk-docs-1.5.0*" >>have been masked. >>!!! One of the following masked packages is required to complete your request: >>- dev-java/java-sdk-docs-1.5.0 (masked by: package.mask, ~x86 keyword) >># <[EMAIL PROTECTED]> >># lotsa things in the tree don't compile with 1.5 yet >># 1.5 defaults too -target 1.5 making downgrading to a 1.4(/1.3) >># impossible, see bug 65937 for more information/discussion >> >> >>For more information, see MASKED PACKAGES section in the emerge man page or >>section 2.2 "Software Availability" in the Gentoo Handbook. >>!!!(dependency required by "dev-java/sun-jdk-1.5.0.03" [ebuild]) >> >> > > > They're hard masked inside /usr/portage/profiles/package.mask. For the full > story see http://bugs.gentoo.org/show_bug.cgi?id=65937. > > Short story: > > echo ">=dev-java/sun-jdk-1.4.99" >> /etc/portage/package.unmask > echo ">=dev-java/java-sdk-docs-1.4.99" >> /etc/portage/package.unmask > emerge sun-jdk sun-sdk-docs You also have to echo to /etc/portage/package.keywords-- both packages are twice-masked. So in addition to the above: echo "dev-java/sun-jdk ~x86" >> /etc/portage/package.keywords echo "dev-java/java-sdk-docs ~x86" >> /etc/portage/package.keywords You can unmask just the specific version using Zac's syntax, but you can just unmask this and all future versions on general principle. > > After that you may want to use java-config to set java-1.4 as the default > system compiler. > > Zac On my system, the previously-installed system default VM did not change when I installed a new one-- what you might want to do is change the *user* VM with java-config, in order to actually use the newly-installed one. I'm sure there's a way to set the user VM permanently, but I don't know what it is, and it doesn't seem to 'stick' (once I close the program using Java, the user VM seems to go back to the default). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Problems with Internationalization [Gnome]
Rafael Fernández López schreef: > Hi ! > > Everything worked perfectly, until I right-clicked on an audio/video > file and clicked on Properties -> Audio/Video tab. > > I'm spanish, all Gnome is compiled with LINGUAS="es" and everything is > in spanish. I've some folders like "Música" that means "Music" and > inside I've got my music files. > > All right, when I right click and go to Audio/Video, it says me > "Bitrate", "Name" IN ENGLISH, and then nautilus replaces my folder > "Música" with "M?sica (Invalid encoding)", so I assume that the one that > is breaking everything is that tab. > > I need to know what provides that tab, what ebuild, to study it better > and re-emerge if necessary with the appropiate flags. > > Thank you. To the best of my knowledge, GNOME doesn't use the LINGUAS variable-- afaik, that's for OpenOffice.org (if you compile it, that tells it what language to display as default). GNOME is actually very good in using the LANG variable to decide what language should be used (unlike, for example, KDE, where you have to install a whole separate package to get another language, and then choose that language from within the KDE Control Center for it to be used). For me to have a GNOME desktop in Dutch, all I have to do is choose Dutch in GDM's Language menu. However, I also have export [EMAIL PROTECTED] in my ~/.bashrc, in the event that I start from startx rather than GDM. So one issue is that your LANG variable may not be correctly set, because in my experience, GNOME is very well translated, certainly for a "common" language such as Spanish. You shouldn't be seeing any English, honestly. I don't (except sometimes in the terminal and always in the man pages), and Dutch is not so common a language as something like Spanish, French, or German. The second issue is that "(unknown encoding)". It could be that, because you're likely using ISO8859-1 (US English), which doesn't contain the accented characters you need, that that's why the display is all messed up... but I don't like that "unknown". That's just a bit weird. Did you compile only limited locales, as discussed in the Gentoo Localization Guide at http://www.gentoo.org/doc/en/guide-localization.xml and somehow did not include the one you need (either 8859-15 or UTF8, or both)? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Java java java, I miss my java
Zac Medico schreef:
> Holly Bostick wrote:
>
>
>>You also have to echo to /etc/portage/package.keywords-- both packages
>>are twice-masked.
>>
>>So in addition to the above:
>>
>>echo "dev-java/sun-jdk ~x86" >> /etc/portage/package.keywords
>>echo "dev-java/java-sdk-docs ~x86" >> /etc/portage/package.keywords
>>
>
>
> Silly me, I have ACCEPT_KEYWORDS="~x86" :-).
Silly me, I'm so cautious these days I forgot one might have it set that
way :-) .
>
>
>>I'm sure there's a way to set the user VM permanently, but I don't
>>know what it is
>
> It works permanently for me with "source ${HOME}/.gentoo/java-env" in
> ~/.bashrc.
>
Ooh, thanks for the tip! Edited-- I've got an alias to edit .bashrc,
that's how much I do it, but never thought of doing that. Now I can set
azureus back to the binary in my menu.xml. Thanks a bunch.
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Java java java, I miss my java
Kevin O'Gorman schreef: > As I pointed out before, I already did those. Again, here are 'cat's > of the files, > which nevertheless do not allow an emerge: > > treat 1.5-bundles # cat /etc/portage/package.keywords > app-office/gnucash quotes > dev-java/sun-jdk ~x86 > dev-java/sun-jre-bin ~x86 > treat 1.5-bundles # cat /etc/portage/package.unmask > >>=dev-java/sun-jdk-1.4.99 >>=dev-java/sun-sdk-docs-1.4.99 > > treat 1.5-bundles # > > Did I miss something? > Looks like it; look again at your post of the error message: - dev-java/java-sdk-docs-1.5.0 (masked by: package.mask, ~x86 keyword) The docs package is masked both by package mask and ~arch keyword. But you've only unmasked the docs' hard mask in /etc/portage/package.unmask. You haven't unmasked the docs in /etc/portage/package.keywords. Holly -- gentoo-user@gentoo.org mailing list
