Re: [gentoo-user] Re: Digest of gentoo-user@lists.gentoo.org issue 1422 (76278-76327)
Iain Buchanan wrote: On Mon, 2008-03-03 at 04:04 -0200, [EMAIL PROTECTED] wrote: Por las nuevas políticas de calidad ISO 9001 que la empresa está implementando, todos los temas relacionados con soporte técnico deben ser realizadas al correo electrónico [EMAIL PROTECTED] Muchas gracias y disculpe las molestías. Automáticamente este email será reenvio a [EMAIL PROTECTED] Is this spam or what? I seem to be getting the same message over and over to this list... I don't know the language but reading between the lines, this is my thoughts. This person has subscribed to the digest version of this list. They have set their email to send a "vacation reply". So, every time he/she gets a email of the digest version, it sends out a reply that he/she is not there to read the email. Not that we really care that he/she is not there to read it. ;-) [humor] I'm thinking we send him a email that has a attachment that shuts down his email program. That way it will not send all those out telling us he is on vacation plus his spam mail will not be getting confirmation that the address is valid. [/humor] Make sense to you? Dale :-) :-) :-) -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: Digest of gentoo-user@lists.gentoo.org issue 1422 (76278-76327)
On Mon, 2008-03-03 at 04:04 -0200, [EMAIL PROTECTED] wrote: > Por las nuevas políticas de calidad ISO 9001 que la empresa está > implementando, todos los temas relacionados con soporte técnico deben > ser realizadas al correo electrónico [EMAIL PROTECTED] > > Muchas gracias y disculpe las molestías. > > Automáticamente este email será reenvio a [EMAIL PROTECTED] Is this spam or what? I seem to be getting the same message over and over to this list... -- Iain Buchanan Life is like an analogy. -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] preprocess_ebuild_env error
Hi list, When I am trying to unmerge xorg-server, it complains as such: /var/tmp/binpkgs/x11-base/xorg-server-1.4-r2/temp/environment: line 4035: syntax error near unexpected token `(' /var/tmp/binpkgs/x11-base/xorg-server-1.4-r2/temp/environment: line 4035: ` done <<(eval ${command})' [31;01m*[0m [31;01m*[0m ERROR: x11-base/xorg-server-1.4-r2 failed. [31;01m*[0m Call stack: [31;01m*[0m ebuild.sh, line 1641: Called die [31;01m*[0m The specific snippet of code: [31;01m*[0m preprocess_ebuild_env || \ [31;01m*[0m die "error processing environment" [31;01m*[0m The die message: [31;01m*[0m error processing environment [31;01m*[0m [31;01m*[0m If you need support, post the topmost build error, and the call stack if relevant. [31;01m*[0m A complete build log is located at '/var/tmp/binpkgs/x11-base/xorg-server-1.4-r2/temp/build.log'. [31;01m*[0m The ebuild environment file is located at '/var/tmp/binpkgs/x11-base/xorg-server-1.4-r2/temp/environment'. [31;01m*[0m So, any idea? Thanks in advance. -- Guanqun -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] Re: Digest of gentoo-user@lists.gentoo.org issue 1422 (76278-76327)
Por las nuevas políticas de calidad ISO 9001 que la empresa está implementando, todos los temas relacionados con soporte técnico deben ser realizadas al correo electrónico [EMAIL PROTECTED] Muchas gracias y disculpe las molestías. Automáticamente este email será reenvio a [EMAIL PROTECTED] -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Apache 2 + portage
On Sun, 02 Mar 2008 15:39:39 +0100 dexters84 <[EMAIL PROTECTED]> wrote: > Hi > > I'm a little confused as long as new Apache use flags are concerned. > With old Apache versions ( pre 2.2.6) all valid flags shown in > *emerge -pv Apache* > could have been set in /etc/portage/package.use in following format > *www-servers/apache flag flag -foo -bar flag* > and life was great :) > > Unfortunately with the new version the concept changed. Some flags > contain *% *some contain* ** and some both (not sure what it is for) > - I only know that flags enclosed like this *(flag) *are the one I > have selected before. > man -P 'col -b' emerge | grep -A11 " --verbose" > My main problem is that emerge apache command is not affected by flag > settings in /etc/portage/package.use in any way. > How do I properly set new apache flags ? > Where do I define APACHE2_MODULES ? > Where do I define APACHE2_MPMS ? > > regards > dexter I don't believe you could get a proper answer w/o sending the exact output of the command(s) you're issuing and the relevant contents of the configuration file(s) in question. -- Best regards, Daniel -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Set a property on a file and have it remove when the file is modified?
On Sat, 01 Mar 2008 17:49:34 +0100 Erik <[EMAIL PROTECTED]> wrote: > Matthias Guede skrev: > > 2008/3/1, Erik <[EMAIL PROTECTED]>: > > > >> Is it possible to set a property on a file and have it remove > >> automatically when the file is modified? > >> > >> Suppose that we have a style checker that checks a lot of source > >> code files. Once it examined a file and found it to be clean, it > >> should set a property on the file ("style-clean"). Whenever the > >> style checker is executed it skips files with this property. > >> Whenever the file is modified, the filesystem removes the property. > >> > >> Is this possible? Which filesystems does it work on? > >> > > > > One solution would be using 'make'. With rules like the following > > only modified files will > > be proceeded: > > > > timestamp: myFile > > doSomthingWidth myFile > > touch timestamp > > > > We have thought about that, but we would like to avoid having a > parallel file hierarchy of timestamp files for our source tree. > Therefore something like the archive attribute (suggested by Etanoi > Shrdlu) would be better. Actually, if there are no other concerns, you'd have to keep only one file for reference. Then you could compare the modification times of all other files with this reference file. #!/bin/bash # initial/full scan touch timestamp.chk find hierarchy | while read some_file do [[ stylecheck("$some_file") -eq 0 ]] || touch "$some.file" done #EOF After this every time you run stylecheck(), you'd have to check only the files having newer time stamp than the one of "timestamp.chk". #!/bin/bash # incremental scan find hierarchy -newer timestamp.chk > modified.list touch timestamp.chk while read some_file do [[ stylecheck("$some_file") -eq 0 ]] || touch "$some.file" done < modified.list unlink modified.list #EOF -- Best regards, Daniel -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
> Are you using kde-4.0.x or why do you have kde-3.5.7. Tried to update to kde-4.0.x near the end of January. This was after almost two years of not updating anything. This led to a emerge -uD world which took about a week over my string-and-can modem. Many, many "failure to build" errors kept me returning to the list for advice. See ,for instance, "emerge -uD world: another obstacle" in the archive. It was like crawling through the desert pursuing a mirage. I was unmerging here remerging there. Maybe that's when qca-2.0.0-r2 slipped into my system. I never completely understood why kde-4 couldn't be emerged. I ran update-eix and eix-sync. Even now when I do emerge -u kde it just points to 3.5.8. > app-crypt/qca-2.0.0-r2 > unmasked? You can add "=app-crypt/qca-1.0-r3 > ~whateverarchyouhave" to > /etc/portage.keyords and try again "emerge -avuND > world". Ok, that worked. Thanks. mw Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
>> Now do emerge -avuND world <...> [blocks B ] So, localhost heathen # emerge -pvC qca These are the packages that would be unmerged: app-crypt/qca selected: 2.0.0-r2 protected: none omitted: none 'Selected' packages are slated for removal. 'Protected' and 'omitted' packages will not be removed. Seems like I've already got the package that's supposedly blocked. localhost heathen # emerge -vC app-crypt/qca-1.0-r3 --- Couldn't find 'app-crypt/qca-1.0-r3' to unmerge. No packages selected for removal by unmerge So I unmerged then remerged the one I already got and did emerge -auvND world. Same thing. The link to the gentoo manual says I can just ignore that particular package but it doesn't say how. Are you using kde-4.0.x or why do you have app-crypt/qca-2.0.0-r2 unmasked? You can add "=app-crypt/qca-1.0-r3 ~whateverarchyouhave" to /etc/portage.keyords and try again "emerge -avuND world". Maybe you have to do emerge -av --oneshot =app-crypt/qca-1.0-r3 before! -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
> > quickpkg =dev-lang/python-2.3.5-r3 > done > then run python-updater localhost heathen # python-updater * Can't determine any previous Python version(s). > > Now do > emerge -avuND world <...> [blocks B ] So, localhost heathen # emerge -pvC qca >>> These are the packages that would be unmerged: app-crypt/qca selected: 2.0.0-r2 protected: none omitted: none >>> 'Selected' packages are slated for removal. >>> 'Protected' and 'omitted' packages will not be removed. Seems like I've already got the package that's supposedly blocked. localhost heathen # emerge -vC app-crypt/qca-1.0-r3 --- Couldn't find 'app-crypt/qca-1.0-r3' to unmerge. >>> No packages selected for removal by unmerge So I unmerged then remerged the one I already got and did emerge -auvND world. Same thing. The link to the gentoo manual says I can just ignore that particular package but it doesn't say how. mw Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
On Mon, 3 Mar 2008 01:46:08 +0200, Alan McKinnon wrote: > emerge -avC =dev-lang/python-2.3.5-r3 > Unmerging python will leave you with an unusuable portage and you'll > need a backup copy or other voodoo to get it going again. What you > actually want is to remove just the SLOT you are no longer using Even so, I would quickpkg it first, just in case you accidentally remove too much. -- Neil Bothwick New sig wanted good price paid. signature.asc Description: PGP signature
Re: [gentoo-user] portage confusion
On Monday 03 March 2008, maxim wexler wrote: > > python. So you > > *should* be able to unmerge the old one. If > > something does break... > > then file a bug at b.g.o. > > It want's to grab both of them. How do I move 2.4.4.r6 > to the protected column? Or should I? Is this a good > place to use the && operator? > > localhost heathen # emerge -pC python emerge -avC =dev-lang/python-2.3.5-r3 Unmerging python will leave you with an unusuable portage and you'll need a backup copy or other voodoo to get it going again. What you actually want is to remove just the SLOT you are no longer using -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Ghosting a Ext3 partition
Rasmus Andersen wrote: FreeBSD's softupdates should make filesystem state always consistent, metadatawise. Or so I think I remember, its been a while. That might aleviate some of the problems noted on the dump page I referenced. Freebsd's dump -L (live option) uses ufs2 snapshot capability to ensure consistency, so its dump is reasonably smart. (FWIW / on Freebsd usually does not have softupdates specified, so usually won't help here). So yeah, it looks like some care is needed with plain old 'dump' on Linux - which is a bit of a pest! I use rdiff-backup for my backups but then again I have low requirements wrt. consistency outside file-level. I have considered LVM snapshots since I use LVM already but havent bothered so far. Right - what you intend the backup to protect against drives all this sort of stuff. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
maxim wexler schrieb: python. So you *should* be able to unmerge the old one. If something does break... then file a bug at b.g.o. It want's to grab both of them. How do I move 2.4.4.r6 to the protected column? Or should I? Is this a good place to use the && operator? try emerge -pC =python-2.3.5-r3 -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
> python. So you > *should* be able to unmerge the old one. If > something does break... > then file a bug at b.g.o. It want's to grab both of them. How do I move 2.4.4.r6 to the protected column? Or should I? Is this a good place to use the && operator? localhost heathen # emerge -pC python >>> These are the packages that would be unmerged: !!! 'dev-lang/python' is part of your system profile. !!! Unmerging it may be damaging to your system. dev-lang/python selected: 2.3.5-r3 2.4.4-r6 protected: none omitted: none >>> 'Selected' packages are slated for removal. >>> 'Protected' and 'omitted' packages will not be removed. And does this mean it won't remerge anything but MAY emerge something? If so, does that mean v2.4? I've already got that: localhost heathen # python-updater -p * Starting Python Updater from 2.3 to 2.4 : * Searching for packages with files in /usr/lib/python2.3 /usr/lib32/python2.3 /usr/lib64/python2.3 .. * No packages needs to be remerged. mw Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] Re: best circuit drawing software
On 2008-03-02, maxim wexler <[EMAIL PROTECTED]> wrote: >> The best format for line drawings is a vector format >> like svg. >> With a vector format the image can be scaled to any >> size and still >> stay sharp. > > Can it be viewed by someone who only has Explorer? > >> >> A bitmap with lossless compression like png is >> tolerable. >> >> Please don't use jpeg. It uses lossy compression >> that is designed for >> photos. They end up fuzzy. >> > Been working with xcircuit. It saves in PS but can be > made into a jpeg which looked just as sharp as the > original. > > I found this command online: > > gs -dBATCH -sOutputFile=4804SR-output.jpg -sDEVICE=jpeg 4804SR-output.ps > > Which I changed to this: > > gs -dBATCH -sOutputFile=4804SR-output.png -sDEVICE=jpeg 4804SR-output.ps > > The first file it created ends in .jpg, the second in > .png. But they're both the same size and look exactly > the same in the browser. Of course they're the same. All you did was change the filename name. The second command is still telling gs to create a jpeg file. > When I gave it -sDEVICE=png it replied "Unknown device: png". > So maybe it just makes jpegs but will call them whatever... gs doesn't care what you call files -- very few unix programs do. The file type is determined by the -sDEVICE= option. -- Grant Edwards grante Yow! Did you find a at DIGITAL WATCH in YOUR box visi.comof VELVEETA?? -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
On Sun, Mar 02, 2008 at 01:34:30PM -0800, Penguin Lover maxim wexler squawked: > Re-emerged gentoolkit. > > Now I get: > > [EMAIL PROTECTED] ~ $ equery depends > =dev-lang/python-2.3.5-r3 > [ Searching for packages depending on > =dev-lang/python-2.3.5-r3... ] > app-office/dia-0.95.1 (python? > sys-libs/libcap-1.10-r9 (python? > >=virtual/python-2.2.1) > [EMAIL PROTECTED] ~ $ > > Again, not "nothing". I can't unmerge all this stuff > can I? Should I still run python-updater etc? python-updater should've been run after you installed a new version of python (I'm pretty sure it is in the ebuild). equery depends is not smart enough to know that there is another python on your system that satisfies the dependencies. What you should look for in the output of equery depends is whether any of the dependencies (as listed in the paranthesis after the package name) is a hard dependency on a particular version of python. Basically, look for something that says =virtual/python-2.3 or <=virtual/python-2.3 or something like that. My cursory glance at the output you send suggests that nothing explicitly depends on that particular version of python. So you *should* be able to unmerge the old one. If something does break... then file a bug at b.g.o. W -- "Let me just make sure this is not on my list to avoid." ~DeathMech, S. Sondhi. P-town PHY 205 Sortir en Pantoufles: up 450 days, 20:31 -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] best circuit drawing software
On Sun, Mar 02, 2008 at 01:20:27PM -0800, Penguin Lover maxim wexler squawked: > Been working with xcircuit. It saves in PS but can be > made into a jpeg which looked just as sharp as the > original. > > I found this command online: > > gs -dBATCH -sOutputFile=4804SR-output.jpg > -sDEVICE=jpeg 4804SR-output.ps > > Which I changed to this: > > gs -dBATCH -sOutputFile=4804SR-output.png > -sDEVICE=jpeg 4804SR-output.ps > > The first file it created ends in .jpg, the second in > .png. But they're both the same size and look exactly > the same in the browser. When I gave it -sDEVICE=png > it replied "Unknown device: png". So maybe it just > makes jpegs but will call them whatever... Yes, it does just that: creates a jpg file with a .png extension. To make png files, you can try imagemagick (make sure to install it with the gs flag) W -- "Problems done right, which is not always the case, usually contain a clue." ~DeathMech, S. Sondhi. P-town PHY 205 Sortir en Pantoufles: up 450 days, 20:27 -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
On Sunday 02 March 2008, maxim wexler wrote: > Re-emerged gentoolkit. > > Now I get: > [EMAIL PROTECTED] ~ $ > > Again, not "nothing". I can't unmerge all this stuff > can I? Should I still run python-updater etc? No, you don;t need to unmerge all of it - most of those are virtuals or any version of python > x Run python-updater. That will remerge everything linked to 2.3 to 2.4 If all goes well then you can unmerge python-2.3.* and run revdep-rebuild as a last safety check. Be warned, python-updater iwll remerge a lot of stuff - it always does -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
--- Alan McKinnon <[EMAIL PROTECTED]> wrote: > On Sunday 02 March 2008, maxim wexler wrote: > > > You can probably get rid of python-2.3 safely, > but > > > first find out what > > > is using it: > > > > > > equery depends =dev-lang/python-2.3.5-r3 > > > > > > If nothing, then unmerge it, but first you might > > > want to make a backup > > > (just in case): > > > > [EMAIL PROTECTED] ~ $ equery depends > > =dev-lang/python-2.3.5-r3 > > [ Searching for packages depending on > > =dev-lang/python-2.3.5-r3... ] > > Traceback (most recent call last): > > File "/usr/bin/equery", line 1732, in ? > > cmd.perform(local_opts) > > File "/usr/bin/equery", line 1117, in perform > > deps = pkg.get_runtime_deps() + > > pkg.get_compiletime_deps() + > pkg.get_postmerge_deps() > > AttributeError: Package instance has no attribute > > 'get_postmerge_deps' > > > > Not exactly "nothing". Is this what you mean? > > No, that's an error. I've seen it somewhere before, > but can't recall > where. Can't recall what to do about it either. > > If no-one else offers an opinion, I'd start by > checking that all your > portage tools still work properly: emerge, > revdep-rebuild, the various > tools in gentoolkit and portage-utils http://forums.gentoo.org/viewtopic-p-3443611.html?sid=1e11270e20684fdc0eea90b188f26344 Re-emerged gentoolkit. Now I get: [EMAIL PROTECTED] ~ $ equery depends =dev-lang/python-2.3.5-r3 [ Searching for packages depending on =dev-lang/python-2.3.5-r3... ] app-office/dia-0.95.1 (python? >=dev-lang/python-1.5.2) app-pda/libopensync-0.22 (python? >=dev-lang/python-2.2) app-portage/esearch-0.7.1 (>=dev-lang/python-2.2) app-portage/gentoolkit-0.2.3-r1 (>=dev-lang/python-2.0) dev-java/java-config-1.3.7 (virtual/python) dev-java/java-config-2.0.33-r1 (dev-lang/python) (virtual/python) dev-lang/swig-1.3.31 (python? virtual/python) dev-libs/libxml2-2.6.30-r1 (python? dev-lang/python) dev-libs/libxslt-1.1.22 (python? dev-lang/python) dev-python/cddb-py-1.4 (virtual/python) dev-python/id3-py-1.2 (virtual/python) dev-python/numeric-24.2-r6 (virtual/python) (>=dev-lang/python-2.3) dev-python/pycairo-1.4.0 (virtual/python) (>=dev-lang/python-2.3) dev-python/pycrypto-2.0.1-r6 (virtual/python) dev-python/pygtk-2.12.0 (>=dev-lang/python-2.3.5) dev-python/pyid3lib-0.5.1-r1 (virtual/python) dev-python/pyogg-1.3-r1 (dev-lang/python) (virtual/python) dev-python/pyopengl-2.0.0.44 (virtual/python) dev-python/pysqlite-2.3.5 (virtual/python) (>=dev-lang/python-2.3) dev-python/python-fchksum-1.7.1 (dev-lang/python) dev-python/pyvorbis-1.4-r1 (dev-lang/python) (virtual/python) dev-python/wxpython-2.6.4.0 (>=dev-lang/python-2.1) dev-util/scons-0.96.1 (virtual/python) (>=dev-lang/python-2.0) gnome-base/libglade-2.6.2 (>=dev-lang/python-2.0-r7) kde-base/kdeutils-3.5.7 (dev-lang/python) media-libs/lcms-1.17 (python? >=dev-lang/python-1.5.2) media-libs/mutagen-1.12 (virtual/python) media-sound/dir2ogg-0.11 (virtual/python) media-sound/jack-3.1.1 (virtual/python) (>=virtual/python-1.5.2) sys-apps/file-4.21-r1 (virtual/python) sys-libs/cracklib-2.8.10 (python? dev-lang/python) sys-libs/libcap-1.10-r9 (python? >=virtual/python-2.2.1) [EMAIL PROTECTED] ~ $ Again, not "nothing". I can't unmerge all this stuff can I? Should I still run python-updater etc? mw Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] DPI is giving me a headache!
On Sonntag, 2. März 2008, Mick wrote: oh god. I hate this automatically generated xorg.confs. They are filled with rubbish. *sigh* hm, could you try without this? DisplaySize 360 290 #digital, oh wait, you said that doesn't change anything. Hm. You can set your DPI with the nvidia driver. But I am not convinced that it is X, the drivers or the monitors fault. The font antialising in linux is just bad ;) hm, oh, and try whatever Dan told you ;) Maybe enabling backingstore helps too. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] eselect fouled
On Sunday 02 March 2008, David Corbin wrote: > On Sunday 02 March 2008 03:03:15 pm Alan McKinnon wrote: > > /usr/lib/libGL.so is a symlink to > > /usr/lib/opengl/xorg-x11/lib/libGL.so on my machine, and that comes > > from media-libs/mesa. eselect updates that symlink. > > > > Do you have mesa correctly installed and does that target actually > > exist? > > Apparently not. media-libs/mesa wass emerged, but I did not have the > libGL.so you mentioned. Re-emerging mesa and then re-eselecting > seems to have fixed things. Cool. I take it your video is now OK and you don't need the ati or nvidia implementations? -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] best circuit drawing software
> The best format for line drawings is a vector format > like svg. > With a vector format the image can be scaled to any > size and still > stay sharp. Can it be viewed by someone who only has Explorer? > > A bitmap with lossless compression like png is > tolerable. > > Please don't use jpeg. It uses lossy compression > that is designed for > photos. They end up fuzzy. > Been working with xcircuit. It saves in PS but can be made into a jpeg which looked just as sharp as the original. I found this command online: gs -dBATCH -sOutputFile=4804SR-output.jpg -sDEVICE=jpeg 4804SR-output.ps Which I changed to this: gs -dBATCH -sOutputFile=4804SR-output.png -sDEVICE=jpeg 4804SR-output.ps The first file it created ends in .jpg, the second in .png. But they're both the same size and look exactly the same in the browser. When I gave it -sDEVICE=png it replied "Unknown device: png". So maybe it just makes jpegs but will call them whatever... mw Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild command doesn't fix broken libs it finds
Bob Young wrote: -Original Message- From: Dale [mailto:[EMAIL PROTECTED] Sent: Sunday, March 02, 2008 12:18 AM To: gentoo-user@lists.gentoo.org Subject: Re: [gentoo-user] revdep-rebuild command doesn't fix broken libs it finds Bob Young wrote: How do I determine if this is a case of "orphaned file, deep dependency, binary package or specially evaluated library" and, if it is one of those, how do I determine which one, and then how do I fix this...? Thanks for listening, Bob Young San Jose, CA This may help: http://bugs.gentoo.org/show_bug.cgi?id=125728 I'm not sure what changed but mine does not do this any more. I'm using app-portage/gentoolkit-0.2.3-r1 at the moment. Dale :-) :-) Thanks Dale, After a little thought and some investigation I'd already come up with the symlink solution on my own. However I do find it a little disturbing that this is exactly the same, as a bug that has a creation date of: 2006-03-10, nearly two years ago. I also know that I didn't have this problem until a recent new "stable" version of gcc was merged. That means somebody is re-introducing bugs that have already been fixed. Making such easily avoidable mistakes does not bode well... Thanks Again, Bob Young San Jose, CA. It has been around for a while but I don't think it actually breaks anything. I have never had a problem with my system and it did that for ages. My workaround has always been to just oneshot everything but gcc and then rerun revdep-rebuild -i -p again to make sure the other problems were fixed. Of course, it would be good if it worked to begin with. Then again, I would rather a bug that breaks something be fixed first too. o_O I have gcc-4.1.2 installed here and I do not get that error. It is masked but no problems for me so far. Could it be that specific version of gcc you have maybe? Maybe someone else can chime in on what version they have and if they get that error or not. Glad to have helped. Dale :-) :-) :-) -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] eselect fouled
On Sunday 02 March 2008 03:03:15 pm Alan McKinnon wrote: > /usr/lib/libGL.so is a symlink to /usr/lib/opengl/xorg-x11/lib/libGL.so > on my machine, and that comes from media-libs/mesa. eselect updates > that symlink. > > Do you have mesa correctly installed and does that target actually > exist? Apparently not. media-libs/mesa wass emerged, but I did not have the libGL.so you mentioned. Re-emerging mesa and then re-eselecting seems to have fixed things. Thanks > -- > Alan McKinnon > alan dot mckinnon at gmail dot com -- David Corbin Abolish the IRS - http://www.fairtax.org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] eselect fouled
On Sunday 2 March 2008, darren kirby wrote: > I suggest this because I use nVidia GL, and I > have '/usr/lib/opengl/nvidia/lib' listed in my ld.so.conf. Though, I > don't recall having to add it manually. I think either "env-update" or "eselect opengl set nvidia" does that. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] eselect fouled
On Sunday 02 March 2008, David Corbin wrote: > I have an ebuild saying that it cannot find -lGL. Everyone tells me > "use eselect". Well, I have, and although it gives me no errors, I > still have no /usr/lib/libGL.so > > > [EMAIL PROTECTED] /home/dcorbin]# eselect opengl set xorg-x11 > Switching to xorg-x11 OpenGL interface... done > [EMAIL PROTECTED] /home/dcorbin]# ls -l /usr/lib/libGL > libGL.la libGLU.so.1 libGLw.so > libGLw.so.1.0.0 > libGLU.la libGLU.so.1.3 libGLw.so.1 > libGLU.so libGLU.so.1.3.060502 libGLw.so.1.0 > > Suggestions for how to fix this please. /usr/lib/libGL.so is a symlink to /usr/lib/opengl/xorg-x11/lib/libGL.so on my machine, and that comes from media-libs/mesa. eselect updates that symlink. Do you have mesa correctly installed and does that target actually exist? -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] eselect fouled
quoth the David Corbin: > I have an ebuild saying that it cannot find -lGL. Everyone tells me "use > eselect". Well, I have, and although it gives me no errors, I still have > no /usr/lib/libGL.so > > > [EMAIL PROTECTED] /home/dcorbin]# eselect opengl set xorg-x11 > Switching to xorg-x11 OpenGL interface... done > [EMAIL PROTECTED] /home/dcorbin]# ls -l /usr/lib/libGL > libGL.la libGLU.so.1 libGLw.so > libGLw.so.1.0.0 > libGLU.la libGLU.so.1.3 libGLw.so.1 > libGLU.so libGLU.so.1.3.060502 libGLw.so.1.0 > > Suggestions for how to fix this please. > -- > David Corbin Just a guess: ensure '/usr/lib/libGL' is in /etc/ld.so.conf and then run 'ldconfig'... I suggest this because I use nVidia GL, and I have '/usr/lib/opengl/nvidia/lib' listed in my ld.so.conf. Though, I don't recall having to add it manually. HTH -d -- darren kirby :: Part of the problem since 1976 :: http://badcomputer.org "...the number of UNIX installations has grown to 10, with more expected..." - Dennis Ritchie and Ken Thompson, June 1972 -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] DPI is giving me a headache!
On Sun, 2 Mar 2008 18:10:58 + Mick <[EMAIL PROTECTED]> wrote: > Please find xorg.conf and Xorg.0.log attached. I am using the xorg > radeon driver. The font size is just right, would not like to > increase it. > A few thoughts based on your xorg.conf. 1) you set HorizSync 64 #31 - 80 #digital VertRefresh 60 #55 - 85 #digital you shouldn't need either. You might want to remove them and try without. 2) your ATI device has quite a lot of settings. If you don't need them all, you might want to remove or comment out what you can. They might affect the output in unexpected ways, although I'm no ati expert. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sunday 02 March 2008, Chris Walters wrote: > Alan McKinnon wrote: > | What you wrote doesn't make sense. depscan.sh is installed by > | baselayout and mktemp is installed by coreutils. You have > | depscan.sh Which package is blocking which? You don't have to guess > | which one, portage will tell you when an emerge fails. > > Well, apparently either the latest ~amd64 keyword masked version of > coreutils does not install /bin/mktemp, or makes changes so that > /sbin/depscan.sh cannot find it, because "/bin/mktemp missing" is a > part of the error message, I receive. When I mask the latest version > of coreutils, and merge the older one and the mktemp ebuild, the > problem disappears (yes, I was able to get emerge to work - finally). Ah. That's useful info. Are you saying that current coreutils does not supply mktemp (it should), so you have to use an older coreutils and a discrete mktemp ebuild? What's in the build log for the non-working coreutils regarding mktemp? -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] eselect fouled
I have an ebuild saying that it cannot find -lGL. Everyone tells me "use eselect". Well, I have, and although it gives me no errors, I still have no /usr/lib/libGL.so [EMAIL PROTECTED] /home/dcorbin]# eselect opengl set xorg-x11 Switching to xorg-x11 OpenGL interface... done [EMAIL PROTECTED] /home/dcorbin]# ls -l /usr/lib/libGL libGL.la libGLU.so.1 libGLw.so libGLw.so.1.0.0 libGLU.la libGLU.so.1.3 libGLw.so.1 libGLU.so libGLU.so.1.3.060502 libGLw.so.1.0 Suggestions for how to fix this please. -- David Corbin -- David Corbin Learn more about the FairTax - http://www.fairtax.org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sun, 02 Mar 2008 14:20:08 -0500, Chris Walters wrote: > Well, apparently either the latest ~amd64 keyword masked version of > coreutils does not install /bin/mktemp, or makes changes so > that /sbin/depscan.sh cannot find it, because "/bin/mktemp missing" is > a part of the error message, I receive. % eix -c -e coreutils [I] sys-apps/coreutils ([EMAIL PROTECTED]/01/08): Standard GNU file utilities % qlist coreutils | grep mktemp /usr/share/man/man1/mktemp.1.bz2 /bin/mktemp /usr/bin/mktemp It's certainly there on this box, and the other ~amd64 and ~x86 boxes I have. -- Neil Bothwick Theory is when you know everything, but nothing works. Reality is when everything works, but you don't know why. However, usually theory and reality are mixed together : Nothing works, and nobody knows why not. signature.asc Description: PGP signature
[gentoo-user] 403 error on apache2
The new apache configuration style has got me all mixed up. I am running apache-2.2.8 and I am also trying to install drupal-5.6 (with vhosts). Pointing the browser to localhost works fine, I get the page that says "IT WORKS!", but not the Apache logo and picture of a feather as it used to be the case. Does this mean that my new apache installation is not properly configured? Pointing to mywebdomain.co.uk/drupal fails with error 403: === Forbidden You don't have permission to access /drupal/install.php on this server. Apache Server at mywebdomain.co.uk Port 80 === The only way I have found to get the pages accessible from a browser is to add this to /etc/apache2/modules.d/00_default_settings.conf file: Options FollowSymLinks AllowOverride All Order deny,allow Allow from all However, I am not sure that this the correct way to allow access to a vhost. Where should I define that this vhost (mywebdomain.co.uk) should be accessible from the Internet? This is my /etc/apache2/vhosts.d/00_default_vhost.conf file, just in case I am missing something essential: #Listen 12.34.56.78:80 Listen 80 # Use name-based virtual hosting. NameVirtualHost *:80 ServerName localhost Include /etc/apache2/vhosts.d/default_vhost.include ServerEnvironment apache apache ServerName mywebdomain.co.uk ServerAlias mywebdomain.co.uk *.mywebdomain.co.uk DocumentRoot /var/www/mywebdomain.co.uk/htdocs Can you please advise if I need to change anything in the default Gentoo apache configuration to set it up to work with vhosts and drupal? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Can anyone tell me?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Alan McKinnon wrote: | On Sunday 02 March 2008, Chris Walters wrote: |> Alan McKinnon wrote: | I don't -O3 can ever be considered "standard". Also you say you don't | think that's it, then admit -O3 changes the code substantially. I'm | having horrible visions that you are taking a shotgun approach to | fault-finding Say again? How am I "taking a shotgun approach to fault-finding"? |> The problem has to do the the Service Dependencies not being able to |> be scanned, and I am advised to run /sbin/depscan.sh |> |> When I run that, I just get the same error - which also involves a |> missing /bin/mktemp file. It seems that that package blocks that |> latest version of coreutils... | | What you wrote doesn't make sense. depscan.sh is installed by baselayout | and mktemp is installed by coreutils. You have depscan.sh Which package | is blocking which? You don't have to guess which one, portage will tell | you when an emerge fails. Well, apparently either the latest ~amd64 keyword masked version of coreutils does not install /bin/mktemp, or makes changes so that /sbin/depscan.sh cannot find it, because "/bin/mktemp missing" is a part of the error message, I receive. When I mask the latest version of coreutils, and merge the older one and the mktemp ebuild, the problem disappears (yes, I was able to get emerge to work - finally). | You really should supply more information so that we can help you. You | have now posted 4 times on this thread, and have not supplied any | relevant info at all apart from your arch is ~amd64 and you have a | problem. So let's do this the right way which involves you supplying | the following: | | - when your system "broke twice", what exactly does this mean? What no | longer works, and how does the system's behaviour differ from what you | expect? | - relevant logs | - command(s) run before the problem manifests | - console output that demonstrates a problem I asked for specific and general information in my original message to this list. That was what packages had others, using the "ACCEPT_KEYWORDS=~amd64", had trouble with. If you have no answer to that question, then you should just say so, or not have bothered to reply. I am not liking the attitude on this list one bit. I didn't ask you, or anyone else to solve a specific problem for me, just a simple general question. If I wanted specific help, I would have provided all that you are claiming I should provide. Chris -BEGIN PGP SIGNATURE- iD8DBQFHyv3mUx1jS/ORyCsRChqaAJsHpoz1bA6ry3id6SXVjdTY5YZasACaAsGE TVslqDzdm1KxKhJNI3t+xi4= =s/VG -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
RE: [gentoo-user] revdep-rebuild command doesn't fix broken libs it finds
> -Original Message- > From: Dale [mailto:[EMAIL PROTECTED] > Sent: Sunday, March 02, 2008 12:18 AM > To: gentoo-user@lists.gentoo.org > Subject: Re: [gentoo-user] revdep-rebuild command doesn't fix broken libs > it finds > Bob Young wrote: >> How do I determine if this is a case of "orphaned file, deep dependency, >> binary package or specially evaluated library" and, if it is one of >those, >> how do I determine which one, and then how do I fix this...? >> >> Thanks for listening, >> Bob Young >> San Jose, CA >> This may help: http://bugs.gentoo.org/show_bug.cgi?id=125728 I'm not sure what changed but mine does not do this any more. I'm using app-portage/gentoolkit-0.2.3-r1 at the moment. Dale :-) :-) Thanks Dale, After a little thought and some investigation I'd already come up with the symlink solution on my own. However I do find it a little disturbing that this is exactly the same, as a bug that has a creation date of: 2006-03-10, nearly two years ago. I also know that I didn't have this problem until a recent new "stable" version of gcc was merged. That means somebody is re-introducing bugs that have already been fixed. Making such easily avoidable mistakes does not bode well... Thanks Again, Bob Young San Jose, CA. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sun, 02 Mar 2008 12:03:44 -0500, Chris Walters wrote: > I find these paragraphs to be rude and insulting. I am not an idiot - > I know exactly what "testing" means, and what "unstable" means. Sorry if you feel that way, but many people confuse the various meanings of unstable and stable in the context of software branches. -- Neil Bothwick Oxymoron: Reagan memoirs. signature.asc Description: PGP signature
Re: [gentoo-user] Can anyone tell me?
On Sunday 02 March 2008, Chris Walters wrote: > Alan McKinnon wrote: > > On Sunday 02 March 2008, Dan Farrell wrote: > >> doesn't sound like a broken package to me. perhaps something else > >> got borked? > > > > Or maybe some "unusual" compiler settings? > > > > OP, please post your /etc/make.conf > > I don't think it is the compiler settings - they are fairly standard > "-O3 -march=athlon64 -pipe" That's it. I've never had any problems > with -O3, but it could still be a part of the problem, since it > substantially changes the code at compile time. I don't -O3 can ever be considered "standard". Also you say you don't think that's it, then admit -O3 changes the code substantially. I'm having horrible visions that you are taking a shotgun approach to fault-finding > The problem has to do the the Service Dependencies not being able to > be scanned, and I am advised to run /sbin/depscan.sh > > When I run that, I just get the same error - which also involves a > missing /bin/mktemp file. It seems that that package blocks that > latest version of coreutils... What you wrote doesn't make sense. depscan.sh is installed by baselayout and mktemp is installed by coreutils. You have depscan.sh Which package is blocking which? You don't have to guess which one, portage will tell you when an emerge fails. You really should supply more information so that we can help you. You have now posted 4 times on this thread, and have not supplied any relevant info at all apart from your arch is ~amd64 and you have a problem. So let's do this the right way which involves you supplying the following: - when your system "broke twice", what exactly does this mean? What no longer works, and how does the system's behaviour differ from what you expect? - relevant logs - command(s) run before the problem manifests - console output that demonstrates a problem -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] DPI is giving me a headache!
On Sunday 02 March 2008, Dan Farrell wrote: > On Sun, 2 Mar 2008 14:02:40 +0100 > > Volker Armin Hemmann <[EMAIL PROTECTED]> wrote: > > Essentially, the > > > > > antiailiasing seems to alter the consistency of fonts in an > > > irregular manner causing them to blur (differently) across the > > > screen, as if the monitor resolution is out of sync. > > hmm, that reminds me of the appearance of the screen when you're > running at a resolution that isn't native for the device. For example, > if you run a 1280x1024 at widescreen, each 6th pixel or something is > stretched to consist of 2 physical pixels to widen the screen out to > fill the entire display. I hope that explination makes sense - > clearly there are other, better ways to use widescreen in a case like > this. > > i would make sure you're running in the screen's native resolution if > it's an LCD, else make sure you're running at a resolution that fits > the aspect ratio of your screen. Yep, that's pretty much what I think it is. Stretches and blurs along the width of the screen, every few pixels. I thought I *was* using the native resolution. Of course DDC may not be picking up the right dimensions in the first place. This is not a wide screen, just a 1280x1024 vanilla resolution. The monitor is a NEC MultiSync LCD 1860NX. Screen 0: minimum 640 x 480, current 1280 x 1024, maximum 1280 x 1024 default connected 1280x1024+0+0 0mm x 0mm 1280x1024 60.0* 1024x768 60.0 800x60060.0 640x48060.0 832x62460.0 This is what the Internet tells me (because the manual is rubbish): Display Type: IPS TFT 46 cm (18.1 inch) Active Display Area: 360 x 290 mm Pixel Pitch: 0.281 mm Viewing Angle: 160° horizontal/160° vertical (at contrast ratio 10:1) Brightness:200 cd/m2 Contrast Ratio:350:1 Response Time: 30 ms (white to black 15 ms, black to white 15 ms) Number of Colours: 16.77 million Optimum Resolution:1280 x 1024 at 60 Hz (1.3 mega pixel) Other Resolutions: 1024 x 768; 832 x 624; 800 x 600; 640 x 400; 640 x 480; 720 x 400 Features/Adjust Functions: NTAA (Non-Touch-Auto-Adjustment); Auto adjust; Contrast; Brightness; Fine adjust (analog); OmniColor™: sRGB and 6 axis colour control; Colour temperature control; Monitor info; Language Select; Intelligent Power Management (VESA/EPA/NUTEK compliant); On-Screen Manager (OSM) lock out User Controls: On/Off; OSM menu; Interfaces Plug & Play; Asset Management: VESA DDC2B; DDC2Bi; DDC/CI and EDID standard Horizontal Frequency: 31–82 kHz Vertical Frequency:55–85 Hz Dot Clock Rate:135 MHz -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] portage confusion
On Sunday 02 March 2008, maxim wexler wrote: > > You can probably get rid of python-2.3 safely, but > > first find out what > > is using it: > > > > equery depends =dev-lang/python-2.3.5-r3 > > > > If nothing, then unmerge it, but first you might > > want to make a backup > > (just in case): > > [EMAIL PROTECTED] ~ $ equery depends > =dev-lang/python-2.3.5-r3 > [ Searching for packages depending on > =dev-lang/python-2.3.5-r3... ] > Traceback (most recent call last): > File "/usr/bin/equery", line 1732, in ? > cmd.perform(local_opts) > File "/usr/bin/equery", line 1117, in perform > deps = pkg.get_runtime_deps() + > pkg.get_compiletime_deps() + pkg.get_postmerge_deps() > AttributeError: Package instance has no attribute > 'get_postmerge_deps' > > Not exactly "nothing". Is this what you mean? No, that's an error. I've seen it somewhere before, but can't recall where. Can't recall what to do about it either. If no-one else offers an opinion, I'd start by checking that all your portage tools still work properly: emerge, revdep-rebuild, the various tools in gentoolkit and portage-utils -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can't boot Compaq Proliant 1600
Daniel da Veiga wrote: Tell me about it, gotta drag this old machine all the way up to the forth floor. It was an inexpensive choice for a file server, and I was happy to see the Gentoo Minimal CD booting (cause then I was sure it would work). Anyway, I had to go (I'm on the road right now) and left it compiling a new kernel (manually this time), that I'll test tomorrow. Is there any way to copy the LiveCD kernel and initrd so I can boot the EXACT kernel the CD is using? This way I can troubleshoot this damn thing without a LiveCD and chroot every 5 minutes... There may be a way to copy those setting and such but I'm not sure how. If Knoppix can do it there has to be a way. I read you can put Knoppix on a hard drive too. I can say one thing about the server, it has great cooling for a old rig. I think they made it big and heavy so the fans would not push it across the floor. LOL I wonder if I can copy that to the floppy? I think the floppy works. I'll check on that in a little bit. May save me from dragging that thing in the house. Dale :-) :-) -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [OT] beware ebay seller
On Sonntag, 2. März 2008, maxim wexler wrote: > > well, I buyed hardware over 40 times on ebay - and I > > was always a very > > satisfied costumer. From mainboards, cpus to > > tapedrives and libs, I always > > got the right stuff. > > Me too. This has been my first experience of outright > fraud. I've been shipped the wrong stuff or broken > stuff but the seller always made good in the end. once I bought a cheap soundcard. I got it, everything was fine. Ca one year later I got a letter from ebay - they would be investigating the seller because of fraud and if there had been any trouble with that person. ... -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
> > You can probably get rid of python-2.3 safely, but > first find out what > is using it: > > equery depends =dev-lang/python-2.3.5-r3 > > If nothing, then unmerge it, but first you might > want to make a backup > (just in case): [EMAIL PROTECTED] ~ $ equery depends =dev-lang/python-2.3.5-r3 [ Searching for packages depending on =dev-lang/python-2.3.5-r3... ] Traceback (most recent call last): File "/usr/bin/equery", line 1732, in ? cmd.perform(local_opts) File "/usr/bin/equery", line 1117, in perform deps = pkg.get_runtime_deps() + pkg.get_compiletime_deps() + pkg.get_postmerge_deps() AttributeError: Package instance has no attribute 'get_postmerge_deps' Not exactly "nothing". Is this what you mean? mw Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [Probably off-topic] How do I find out what is consuming the bandwidth?
On Sat, 1 Mar 2008 22:56:20 +0200 Alan McKinnon <[EMAIL PROTECTED]> wrote: > There's one other way that I just remembered (for future reference). > You don't *have* to use a linux machine as a gateway if you have a > decent managed switch - set it to route all traffic on all ports out > through the port that a monitoring machine is connected to. In other > words, that one part acts like a hub. Now that the monitoring machine > can see every bit on the entire Ethernet, it can count 'em :-) > > However, these switches cost a fortune and I very much doubt that the > el-cheapo ADSL routers on the market have this feature. Both of mine > certainly don't. You could also do this with a non-switching hub, if you can find one. They would be a whole lot cheaper, I'd imagine, than a managed switch, even used on ebay. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
Alan McKinnon wrote: On Sunday 02 March 2008, Dan Farrell wrote: doesn't sound like a broken package to me. perhaps something else got borked? Or maybe some "unusual" compiler settings? OP, please post your /etc/make.conf I don't think it is the compiler settings - they are fairly standard "-O3 -march=athlon64 -pipe" That's it. I've never had any problems with -O3, but it could still be a part of the problem, since it substantially changes the code at compile time. The problem has to do the the Service Dependencies not being able to be scanned, and I am advised to run /sbin/depscan.sh When I run that, I just get the same error - which also involves a missing /bin/mktemp file. It seems that that package blocks that latest version of coreutils... Regards, Chris -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] DPI is giving me a headache!
On Sun, 2 Mar 2008 14:02:40 +0100 Volker Armin Hemmann <[EMAIL PROTECTED]> wrote: > Essentially, the > > antiailiasing seems to alter the consistency of fonts in an > > irregular manner causing them to blur (differently) across the > > screen, as if the monitor resolution is out of sync. > > hmm, that reminds me of the appearance of the screen when you're running at a resolution that isn't native for the device. For example, if you run a 1280x1024 at widescreen, each 6th pixel or something is stretched to consist of 2 physical pixels to widen the screen out to fill the entire display. I hope that explination makes sense - clearly there are other, better ways to use widescreen in a case like this. i would make sure you're running in the screen's native resolution if it's an LCD, else make sure you're running at a resolution that fits the aspect ratio of your screen. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sun, 02 Mar 2008 12:03:44 -0500 Chris Walters <[EMAIL PROTECTED]> wrote: > I find these paragraphs to be rude and insulting. I am not an idiot > - I know exactly what "testing" means, and what "unstable" means. > Just because I ask a relatively simple question in this group does > not mean that I am "not prepared to deal with the occasional > problem". Were that the case, I would not be working with computers > at all, since all operating systems and distributions have an > "occasional problem" even in their "stable" branches. > > Chris If I may speak for Neil, he provides a lot of very useful information to the list and is a very courteous poster as well. In my mind, that little lemming that somehow appears along with his emails is the sign of a good addition to the thread. I'm sure he didn't mean to insult you. I hope that you agree that even though you started the thread, the information he gave could be useful to others reading it. I thought it was an informative and well-written post myself, not that yours aren't, but don't be too defensive. We're all here to learn (and perhaps to teach, occasionally at least ;) ) to answer your original question succinctly: > Can anyone tell me what packages you know of that will break your > system if you choose to put "ACCEPT_KEYWORDS=~amd64" in your make.conf > file? no, no one can tell until they are tested, and then they will be marked stable. If I may take a moment to make a few (friendly and respectful!) criticisms of your post, that may have given people the wrong impression, I think there are probably two things that may have done so: firstly, your subject line was 'Can anyone help me?' Sure, you're asking for help, but a more relevant subject line would have nicely synopsized your post. Most people that start a thread here _are_ looking for help, after all. Secondly, I think this: >I do know that the only way to fix the problem was to restore from >backup, or to try re-installing again. I just want to know which >packages are so unstable that I should mask them. definitely made my blood boil a little. It sounds as if, with your gawk case here, a careful analysis of the log files could have perhaps provided you with a few fundamentals from /usr/lib that were missing and only needed to be copied over to / before /usr or /usr/lib was mounted from it's seperate filesystem. (I am just guessing that's how Neil solved this particular problem, although I wouldn't know.) Saying that the only way to fix a particular problem is by replacing the software with a working version is very rarely the case. I hope you can understand how that could give us a little bit of a bad first impression here on the lists, because it consists of a lot of serious gentooers that all seem to share a dislike of reinstalls and backup restorations rather than responding to particular error messages and resolving their problems that way. Perhaps it's just the gentoo way - reinstalling seems to be very popular in ubuntu. Anyhow, my advice to you is to do what many, including myself do - save yourself the headache of running ~amd64, and only use package.keywords to unmask packages as necessary. Good luck, and may you withhold judgment of me as I have of you, Dan Farrell -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can't boot Compaq Proliant 1600
On Sat, Mar 1, 2008 at 11:48 PM, Dale <[EMAIL PROTECTED]> wrote: > > Alan McKinnon wrote: > > On Saturday 01 March 2008, Daniel da Veiga wrote: > > > >> Hello list, > >> > >> I am struggling with an old Compaq Proliant 1600 for a while. It > >> boots perfectly using the Minimal Install CD, and all the install > >> process goes well. When its time to boot its own kernel, GRUB loads > >> the files, goes to the message "OK, booting the kernel." and just > >> hangs there. I used genkernel to compile the kernel. I'm now trying > >> to compile it manually to see if it will work. > >> > >> Please note that this particular machine has no SMART-2 Array > >> controller, so the disks are seen as common SCSI devices (sda and > >> sdb), currently I'm installing the whole system at sda, leaving "b" > >> untouched for backup purposes when I get this beast up and running. > >> > >> Anyone has any experience booting one of these machines? > >> > > > > Not this specific model, but I would start by comparing the boot command > > and modules loaded from the Minimal CD with what you are trying to use > > when booting the installed OS. > > > > Supplying some error messages, command lines and even a relevant config > > file or two at this point might be found to be useful by those inclined > > to help you out of this predicament... > > > > > > > > I *think* I have one of these old boat anchors out in my shop. I'm not > sure if it still boots or not but I may could get you a .config file for > the kernel if you have no luck any other way. I say 'any other way' > cause I would have to drag that monster in the house and hook it back up > to the network to access it. > > Let me know if all else fails. I'll check the model before hooking the > tractor and chain to it. LOL > Tell me about it, gotta drag this old machine all the way up to the forth floor. It was an inexpensive choice for a file server, and I was happy to see the Gentoo Minimal CD booting (cause then I was sure it would work). Anyway, I had to go (I'm on the road right now) and left it compiling a new kernel (manually this time), that I'll test tomorrow. Is there any way to copy the LiveCD kernel and initrd so I can boot the EXACT kernel the CD is using? This way I can troubleshoot this damn thing without a LiveCD and chroot every 5 minutes... -- Daniel da Veiga -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Neil Bothwick wrote: | On Sat, 01 Mar 2008 14:09:45 -0500, Chris Walters wrote: | I run two completely ~amd64 systems here and have very few problems. I've run testing on Gentoo and other distributions. With Gentoo, for over a year, with few problems, and those were generally easily fixed. | The ~ in ~amd64 means the ebuilds are in testing, not that they, or the | software they install, are considered unstable in the "likely to crash" | meaning of the term. Because you are using bleeding edge ebuilds, there | is the odd occasion when things don't play nicely together, or mistakes | are made. The gawk problem one one such situation, where it depended on a | library in /usr/lib and broke any system with /usr on a separate | filesystem. It didn't require a reinstall to fix, I know because I was | hit by it and didn't reinstall. It was a one-off that was fixed quickly, | if you didn't sync and update each day you could easily have missed it. | | The testing ebuilds are for just that, it is only by people using them | and reporting problems that those problems are kept out of the stable | tree. If you are not prepared to deal with the occasional problem, | running a testing system is not for you. I find these paragraphs to be rude and insulting. I am not an idiot - I know exactly what "testing" means, and what "unstable" means. Just because I ask a relatively simple question in this group does not mean that I am "not prepared to deal with the occasional problem". Were that the case, I would not be working with computers at all, since all operating systems and distributions have an "occasional problem" even in their "stable" branches. Chris -BEGIN PGP SIGNATURE- iD8DBQFHyt3tUx1jS/ORyCsRCt5YAJ9yTa9Bz9zWJjgn9moyE2mi/0FIGgCfX3OY KWvc1mFs3pZiOOJIZuwE7dY= =m3jB -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [OT] beware ebay seller
> > From what you've told us it's possible that the > seller didn't look > inside the box and assumed it was a new PSU, after > their spouse > tidied the old one up into an empty box. Certainly, > I always keep > packaging in case I need to make a warranty return, > and in the case > that a PC is just misbehaving, with random reboots & > stuff, the first > thing I do is replace the PSU, keeping the old one > in case that makes > no difference. Nah, this guy sells lots of them. The power supply he did mail is brand new -- just not the one I thought I was buying. The box originally had been sealed by Best Buy. The box came with the original Dynex User Guide. My theory: these are cheap run of the mill PSs put into Dynex boxes acquired from Best Buy which used the original PSs in their PCs. The seller took a picture of the box and put that on eBay. When the order comes in he takes a shiny new PS "New" and puts it in the Dynex box "in Box!". So, technically he hasn't lied. Typical scamster M.O. > > I would email the seller in the first instance & see > what they have > to say. No reply > feedback - a scammer with several negs will be > quickly obvious from > their feedback score, This is puzzling. Maybe the seller gets his family and friends to "buy" stuff and give him props. Then he runs his scam, cancels his account and starts another. Where's my tinfoil hat! > personally > wouldn't have high expectations of the "buyer > protection" offered by > fleaBay or ThievingScumPal. > It should be in their best interest to supply good service. mw Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [OT] beware ebay seller
> > well, I buyed hardware over 40 times on ebay - and I > was always a very > satisfied costumer. From mainboards, cpus to > tapedrives and libs, I always > got the right stuff. Me too. This has been my first experience of outright fraud. I've been shipped the wrong stuff or broken stuff but the seller always made good in the end. mw Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] Re: Digest of gentoo-user@lists.gentoo.org issue 1421 (76228-76277)
Por las nuevas políticas de calidad ISO 9001 que la empresa está implementando, todos los temas relacionados con soporte técnico deben ser realizadas al correo electrónico [EMAIL PROTECTED] Muchas gracias y disculpe las molestías. Automáticamente este email será reenvio a [EMAIL PROTECTED] -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Apache 2 + portage
On Sunday 02 March 2008, dexters84 wrote: > Hi > > I'm a little confused as long as new Apache use flags are concerned. > With old Apache versions ( pre 2.2.6) all valid flags shown in > *emerge -pv Apache* > could have been set in /etc/portage/package.use in following format > *www-servers/apache flag flag -foo -bar flag* > and life was great :) > > Unfortunately with the new version the concept changed. Some flags > contain *% *some contain* ** and some both (not sure what it is for) > - I only know that flags enclosed like this *(flag) *are the one I > have selected before. That's not an Apache thing - it's an emerge display output thing. See 'man emerge' in the --verbose section for details. > > My main problem is that emerge apache command is not affected by flag > settings in /etc/portage/package.use in any way. > How do I properly set new apache flags ? > Where do I define APACHE2_MODULES ? > Where do I define APACHE2_MPMS ? Apache now pays attention to these flags: Installed versions: 2.2.8-r1(2)(14:59:42 02/03/08) (apache2_modules_actions apache2_modules_alias apache2_modules_auth_basic apache2_modules_auth_digest apache2_modules_authn_anon apache2_modules_authn_dbd apache2_modules_authn_dbm apache2_modules_authn_default apache2_modules_authn_file apache2_modules_authz_dbm apache2_modules_authz_default apache2_modules_authz_groupfile apache2_modules_authz_host apache2_modules_authz_owner apache2_modules_authz_user apache2_modules_autoindex apache2_modules_cache apache2_modules_dav apache2_modules_dav_fs apache2_modules_dav_lock apache2_modules_dbd apache2_modules_deflate apache2_modules_dir apache2_modules_disk_cache apache2_modules_env apache2_modules_expires apache2_modules_ext_filter apache2_modules_file_cache apache2_modules_filter apache2_modules_headers apache2_modules_ident apache2_modules_imagemap apache2_modules_include apache2_modules_info apache2_modules_log_config apache2_modules_logio apache2_modules_mem_cache apache2_modules_mime apache2_modules_mime_magic apache2_modules_negotiation apache2_modules_proxy apache2_modules_proxy_ajp apache2_modules_proxy_balancer apache2_modules_proxy_connect apache2_modules_proxy_http apache2_modules_rewrite apache2_modules_setenvif apache2_modules_speling apache2_modules_status apache2_modules_unique_id apache2_modules_userdir apache2_modules_usertrack apache2_modules_vhost_alias doc ldap ssl threads -apache2_modules_asis -apache2_modules_authn_alias -apache2_modules_cern_meta -apache2_modules_charset_lite -apache2_modules_dumpio -apache2_modules_log_forensic -apache2_modules_proxy_ftp -apache2_modules_version -apache2_mpms_event -apache2_mpms_itk -apache2_mpms_peruser -apache2_mpms_prefork -apache2_mpms_worker -debug -selinux -sni -static -suexec) -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
On Sat, 1 Mar 2008 22:10:39 -0800 (PST), maxim wexler wrote: > Notice in the following portage has nothing to say > about "blockers" following a warning from > revdep-rebuild. And the -u switch calls forth an > earlier version of python but without the -u switch is > content to re-emerge the newer package which I already > have on my system. > > This was all precipitated after emerge xcircuit I was > advised to upgrade tk and tclx and then > revdep-rebuild. > > emerge -u tk just said zero packages to emerge. Not > "no ebuilds to satisfy 'tk'", zero packages. That's because you already have the latest available version of tk installed, so there is nothing for emerge -u to do. > All prepared. Starting rebuild... > emerge --oneshot =dev-lang/python-2.4.4-r6 > =dev-lang/python-2.3.5-r3 > .. > Calculating dependencies... done! > > !!! Error: the conflicts with another package; > !!!the two packages cannot be installed on the > same system together. > !!!Please use 'emerge --pretend' to determine > blockers. You need to do what it says, run the emerge command from revdep-rebuild with the --pretend option. > localhost heathen # emerge -puv python This is not the same. Python is slotted and it appears that your system has, or at least needs, 2.3 and 2.4 so run the command it tells you to to see what blocks what. -- Neil Bothwick Beware of low-flying butterflies. signature.asc Description: PGP signature
Re: [gentoo-user] Apache 2 + portage
2008/3/2, dexters84 <[EMAIL PROTECTED]>: > My main problem is that emerge apache command is not affected by flag > settings in /etc/portage/package.use in any way. > How do I properly set new apache flags ? > Where do I define APACHE2_MODULES ? > Where do I define APACHE2_MPMS ? as defined in http://www.gentoo.org/doc/en/apache-upgrading.xml you can use APACHE2_MODULES an APACHE2_MPMS in /etc/make.conf if you upgrade from version > 2.2.6-r4 you can use this two shell line: echo APACHE2_MODULES=\"$(sed '/^mod_/s/mod_\(.*\)\s\+\(shared\|static\)/\1/;t n;d;:n' /etc/apache2/apache2-builtin-mods)\" >> /etc/make.conf rm /etc/apache2/apache2-builtin-mods -- "Per il giovane ribelle non c'è soluzione..." http://lobotomiatbm.wordpress.com http://www.pettinix.org http://pettinatoridibambole.blogspot.com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sat, 01 Mar 2008 14:09:45 -0500, Chris Walters wrote: > Can anyone tell me what packages you know of that will break your > system if you choose to put "ACCEPT_KEYWORDS=~amd64" in your make.conf > file? I have had my system break, twice now, from a package upgrade - > I think that one of the culprits is gawk, but can't be certain. I run two completely ~amd64 systems here and have very few problems. > I do know that the only way to fix the problem was to restore from > backup, or to try re-installing again. I just want to know which > packages are so unstable that I should mask them. TIA. The ~ in ~amd64 means the ebuilds are in testing, not that they, or the software they install, are considered unstable in the "likely to crash" meaning of the term. Because you are using bleeding edge ebuilds, there is the odd occasion when things don't play nicely together, or mistakes are made. The gawk problem one one such situation, where it depended on a library in /usr/lib and broke any system with /usr on a separate filesystem. It didn't require a reinstall to fix, I know because I was hit by it and didn't reinstall. It was a one-off that was fixed quickly, if you didn't sync and update each day you could easily have missed it. The testing ebuilds are for just that, it is only by people using them and reporting problems that those problems are kept out of the stable tree. If you are not prepared to deal with the occasional problem, running a testing system is not for you. -- Neil Bothwick There's an old proverb that says just about whatever you want it to signature.asc Description: PGP signature
[gentoo-user] Apache 2 + portage
Hi I'm a little confused as long as new Apache use flags are concerned. With old Apache versions ( pre 2.2.6) all valid flags shown in *emerge -pv Apache* could have been set in /etc/portage/package.use in following format *www-servers/apache flag flag -foo -bar flag* and life was great :) Unfortunately with the new version the concept changed. Some flags contain *% *some contain* ** and some both (not sure what it is for) - I only know that flags enclosed like this *(flag) *are the one I have selected before. My main problem is that emerge apache command is not affected by flag settings in /etc/portage/package.use in any way. How do I properly set new apache flags ? Where do I define APACHE2_MODULES ? Where do I define APACHE2_MPMS ? regards dexter -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
On Sunday 02 March 2008, Dan Farrell wrote: > > > Can anyone tell me what packages you know of that will break > > > your system if you choose to put "ACCEPT_KEYWORDS=~amd64" in your > > > make.conf file? I have had my system break, twice now, from a > > > package upgrade - I think that one of the culprits is gawk, but > > > can't be certain. > > > > > > I do know that the only way to fix the problem was to restore > > > from backup, or to try re-installing again. I just want to know > > > which packages are so unstable that I should mask them. TIA. > > doesn't sound like a broken package to me. perhaps something else > got borked? Or maybe some "unusual" compiler settings? OP, please post your /etc/make.conf -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] portage confusion
On Sunday 02 March 2008, maxim wexler wrote: > Hi group, > > Notice in the following portage has nothing to say > about "blockers" following a warning from > revdep-rebuild. And the -u switch calls forth an > earlier version of python but without the -u switch is > content to re-emerge the newer package which I already > have on my system. Um, yes it does have something to say about blockers. It explicitly states that python-2.3.6-r2 conflicts with something else (probably python-updater): > !!! Error: the conflicts with another package; > !!!the two packages cannot be installed on the > same system together. > !!!Please use 'emerge --pretend' to determine > blockers. You have a python-2.3 and python-2.4 on your system,probably left over from ages ago, as emerge will not by default unmerge an older SLOT when you install a newer SLOT of a package. You can probably get rid of python-2.3 safely, but first find out what is using it: equery depends =dev-lang/python-2.3.5-r3 If nothing, then unmerge it, but first you might want to make a backup (just in case): quickpkg =dev-lang/python-2.3.5-r3 then run python-updater Now do emerge -avuND world to make sure everything is up to date followed by emerge -av --depclean and revdep-rebuild -- Alan McKinnon alan dot mckinnon at gmail dot com -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] what's PMS?
On Sonntag, 2. März 2008, Chuanwen Wu wrote: > In GMN20080218, there is a section: > > EAPI=1 (Where is the specification?): The general agreement was that > any new EAPIs should not be added until EAPI=0 is fully approved. > However, there wasn't any consensus on changing anything about EAPI=1. > Mark Loeser agreed to work on PMS for EAPI=0, and will provide an > update at the next meeting. > > I just don't know what PMS and EAPI means. in gentoo: PMS = Package Manager Specification. What a gentoo package manager has to be able to do, how it does certain steps etc pp. Portage is compatible by definition. EAPI = afaik ebuild api. Current ebuilds are all eapi = 0. If there are incompatible changes to the ebuild structure in the future, this ebuilds will use eapi 1, 2, it is mostly a sign for the package manager telling it how to deal with the ebuild. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Ghosting a Ext3 partition
On Sun, Mar 02, 2008 at 09:51:47PM +1300, Mark Kirkwood wrote: > Understood - I have seen that article too. I must say, I've mainly had > experience with 'dump' on Freebsd and 'xfsdump' on Linux, and never had > restore issues with *either* of these. Now I'm not sure whether these are > supposed to be better than 'dump' on Linux aimed at ext2|3 filesystems - > certainly Freebsd's 'dump' has an option to tell it that it is dumping a > 'live' filesystem, and the man pages for xfsrestore have notes concerning > what happens when restoring an (xfs)dump from a 'live' filesystem - so they > may well be! FreeBSD's softupdates should make filesystem state always consistent, metadatawise. Or so I think I remember, its been a while. That might aleviate some of the problems noted on the dump page I referenced. > On the other hand I've certainly routinely seen cases of people using dd > (rsync, cpio, tar etc) and coming to grief at restore time. I am reluctant > to suggest that folks use xfs and hence get access to xfsdump, as one of > the nice things about Linux is the choice of a variety of filesystems - > but it is pretty important to get able to backup of (for instance ) / ... > and you usually don't have much option other than doing it live! I use rdiff-backup for my backups but then again I have low requirements wrt. consistency outside file-level. I have considered LVM snapshots since I use LVM already but havent bothered so far. Cheers, Rasmus -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] what's PMS?
In GMN20080218, there is a section: EAPI=1 (Where is the specification?): The general agreement was that any new EAPIs should not be added until EAPI=0 is fully approved. However, there wasn't any consensus on changing anything about EAPI=1. Mark Loeser agreed to work on PMS for EAPI=0, and will provide an update at the next meeting. I just don't know what PMS and EAPI means. Anyone can explain? Thanks in advanced! -- wcw -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [OT] beware ebay seller
On 1 Mar 2008, at 00:37, maxim wexler wrote: ... If you're looking for PC power supply on ebay and you come across a Dynex, big quiet fan, PCI-E, SATA, 24 pin for cheap from 2213Joseph. "New in Box!" Don't buy it. You'll get the Dynex box alright and a new PS but it's just a run of the mill, noisy, substandard unit. No SATA, no PCI-E etc. This is such an obvious fraud it makes me wonder at his 100% positive feedback. I've started a dispute with Paypal against him. Just to add to your other replies, it is just possible this is an honest mistake. From what you've told us it's possible that the seller didn't look inside the box and assumed it was a new PSU, after their spouse tidied the old one up into an empty box. Certainly, I always keep packaging in case I need to make a warranty return, and in the case that a PC is just misbehaving, with random reboots & stuff, the first thing I do is replace the PSU, keeping the old one in case that makes no difference. I would email the seller in the first instance & see what they have to say. If, as a seller, I found I had made such a mistake, I would send you a refund or replacement immediately. As a buyer I wouldn't be prepared to pay the cost of shipping the duff component back to the seller, and wouldn't expect the seller to insist on that. If the seller isn't prepared to accommodate you then give them a negative feedback - a scammer with several negs will be quickly obvious from their feedback score, and will quickly lose sales. I personally wouldn't have high expectations of the "buyer protection" offered by fleaBay or ThievingScumPal. Stroller. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] DPI is giving me a headache!
On Sonntag, 2. März 2008, Mick wrote: > Hi All, > > I changed the monitor on a box that runs kdm/KDE and the fonts on KDE apps > are giving me a headache from eye strain. This is particularly bad when > working at a console (white letters on black background) and KDE text > editors (with black letters on white background). Essentially, the > antiailiasing seems to alter the consistency of fonts in an irregular > manner causing them to blur (differently) across the screen, as if the > monitor resolution is out of sync. > > The resolution in xorg.conf is set at 360 290 mm, but the DDC picks this up > on its own anyway, whether I define it in xorg.conf, or not. I have no > modeline set. All I can think that is amiss is the DPI. I seem to be > getting a DPI of 90x89: > > $ xdpyinfo | grep resolution > resolution:90x89 dots per inch > > Could this be the cause of my headache and how should I fix it? post your xorg.conf? drivers? increase font size? -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] [Probably off-topic] How do I find out what is consuming the bandwidth?
On 1 Mar 2008, at 20:56, Alan McKinnon wrote: ... There's one other way that I just remembered (for future reference). You don't *have* to use a linux machine as a gateway if you have a decent managed switch - set it to route all traffic on all ports out through the port that a monitoring machine is connected to. In other words, that one part acts like a hub. Now that the monitoring machine can see every bit on the entire Ethernet, it can count 'em :-) However, these switches cost a fortune and I very much doubt that the el-cheapo ADSL routers on the market have this feature. Both of mine certainly don't. FYI: these are actually really cheap secondhand on eBay, if you're happy with 10/100, or 10/100 with 2 x gigabit ports. Stroller. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Can anyone tell me?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Hello Mark, I was under the impression that packages with the ~amd64 (or ~x86) keywords are in testing, but no serious instabilities had been found, or they would be hard masked. I have had non-testing packages break my system before, as well. What generally happens is the the environment variables become messed up, or the service dependencies do - either way, I can't use emerge or any of the available utilities to fix the problem, or even find out what it does. About a year ago, I went to a testing system and haven't had any problems, except when I re-install Gentoo. I guess it is always a choice - either go with the "stable" version of a distribution or you go with the testing version (some people really push the envelope and go for the unstable version). I read up on the keywords, and found out that having the ~mad64 keyword on a package just means that it hasn't been adequately tested on that architecture. That's how people like me help move things along - by testing those packages on our systems, and reporting any problems we find. Regards, Chris Mark Knecht wrote: | On Sat, Mar 1, 2008 at 11:09 AM, Chris Walters <[EMAIL PROTECTED]> wrote: |> Hello, |> |> Can anyone tell me what packages you know of that will break your system if you |> choose to put "ACCEPT_KEYWORDS=~amd64" in your make.conf file? I have had my |> system break, twice now, from a package upgrade - I think that one of the |> culprits is gawk, but can't be certain. |> |> I do know that the only way to fix the problem was to restore from backup, or |> to try re-installing again. I just want to know which packages are so unstable |> that I should mask them. TIA. |> |> Regards, |> Chris |> | | Hi Chris, |I don't think your question can be answered as phrased. | |*Any* package marked with '~' is 'new', 'in testing', 'unstable', | etc. Very few (in my experience) 'break' my machine, but I have a rule | that any package energed as part of emerge system must be stable and I | personally add ~x86 or ~amd64 only for specific packages that I want | or need some new feature. | | Hope this helps, | Mark -BEGIN PGP SIGNATURE- iD8DBQFHypMsUx1jS/ORyCsRCg3SAJ9RjQk0hCUPo1oLfGRJR0gBYOdEmACfaoL+ 0Vb2FeuvF/RoA2MWEZjAs9U= =Gp0G -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] DPI is giving me a headache!
Hi All, I changed the monitor on a box that runs kdm/KDE and the fonts on KDE apps are giving me a headache from eye strain. This is particularly bad when working at a console (white letters on black background) and KDE text editors (with black letters on white background). Essentially, the antiailiasing seems to alter the consistency of fonts in an irregular manner causing them to blur (differently) across the screen, as if the monitor resolution is out of sync. The resolution in xorg.conf is set at 360 290 mm, but the DDC picks this up on its own anyway, whether I define it in xorg.conf, or not. I have no modeline set. All I can think that is amiss is the DPI. I seem to be getting a DPI of 90x89: $ xdpyinfo | grep resolution resolution:90x89 dots per inch Could this be the cause of my headache and how should I fix it? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] revdep-rebuild command doesn't fix broken libs it finds
Bob Young skrev: How do I determine if this is a case of "orphaned file, deep dependency, binary package or specially evaluated library" and, if it is one of those, how do I determine which one, and then how do I fix this...? There are 2 commands: % qfile /path/to/file % equery b /path/to/file qfile is in app-portage/portage-utils and equery is in app-portage/gentoolkit. They both show which package a file belongs to. If the file does not belong to a package it is orphaned. Then it is up to you whether to keep or remove it. If it belongs to a package, check if it is binary. -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Ghosting a Ext3 partition
Rasmus Andersen wrote: If you do backup live filesystems/data then dump is on par with dd; both read from the underlying device and might bypass the kernel's page cache. Ie., there might be unwritten data cached thats not on disk yet. Tar/rdiff-backup/etc reads through the pagecache and avoids this problem. The dump people talk a bit about this themselves on http://dump.sourceforge.net/isdumpdeprecated.html Note I dont want to dis dump, backing up live filesystems is just tricky (depending on your consistency requirements :) and dump adds another level to that. Understood - I have seen that article too. I must say, I've mainly had experience with 'dump' on Freebsd and 'xfsdump' on Linux, and never had restore issues with *either* of these. Now I'm not sure whether these are supposed to be better than 'dump' on Linux aimed at ext2|3 filesystems - certainly Freebsd's 'dump' has an option to tell it that it is dumping a 'live' filesystem, and the man pages for xfsrestore have notes concerning what happens when restoring an (xfs)dump from a 'live' filesystem - so they may well be! On the other hand I've certainly routinely seen cases of people using dd (rsync, cpio, tar etc) and coming to grief at restore time. I am reluctant to suggest that folks use xfs and hence get access to xfsdump, as one of the nice things about Linux is the choice of a variety of filesystems - but it is pretty important to get able to backup of (for instance ) / ... and you usually don't have much option other than doing it live! regards Mark -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild command doesn't fix broken libs it finds
Bob Young wrote: After a recent emerge -DuN world, messages for one of the packages stated that it was necessary to run revdep-rebuild after emerging the package, so I did. The revdep-rebuild ended up merging six packages, with one of them being gcc. Emerging all six packages took several hours, and I noticed that gcc by itself took a significant amount of time. The final message stated that I could re-run revdep-rebuild to verify that all inconsistencies had been resolved, unfortunately, I did not add a -p to the command and to my surprise it spent the next couple of hours or so emerging gcc again. After that finished, I again ran revdep-rebuild although this time with a -p and below is the output: << SNIP >> How do I determine if this is a case of "orphaned file, deep dependency, binary package or specially evaluated library" and, if it is one of those, how do I determine which one, and then how do I fix this...? Thanks for listening, Bob Young San Jose, CA This may help: http://bugs.gentoo.org/show_bug.cgi?id=125728 I'm not sure what changed but mine does not do this any more. I'm using app-portage/gentoolkit-0.2.3-r1 at the moment. Dale :-) :-) -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] revdep-rebuild command doesn't fix broken libs it finds
After a recent emerge -DuN world, messages for one of the packages stated that it was necessary to run revdep-rebuild after emerging the package, so I did. The revdep-rebuild ended up merging six packages, with one of them being gcc. Emerging all six packages took several hours, and I noticed that gcc by itself took a significant amount of time. The final message stated that I could re-run revdep-rebuild to verify that all inconsistencies had been resolved, unfortunately, I did not add a -p to the command and to my surprise it spent the next couple of hours or so emerging gcc again. After that finished, I again ran revdep-rebuild although this time with a -p and below is the output: __ Configuring search environment for revdep-rebuild Checking reverse dependencies... Packages containing binaries and libraries broken by a package update will be emerged. Collecting system binaries and libraries... done. (/root/.revdep-rebuild.1_files) Collecting complete LD_LIBRARY_PATH... done. (/root/.revdep-rebuild.2_ldpath) Checking dynamic linking consistency... broken /usr/lib/gcc/i686-pc-linux-gnu/4.1.2/libgcjawt.la (requires /usr/lib/lib-gnu-java-awt-peer-gtk.la) broken /usr/lib/gcc/i686-pc-linux-gnu/4.1.2/libgij.la (requires /usr/lib/libgcj.la) done. (/root/.revdep-rebuild.3_rebuild) Assigning files to ebuilds... done. (/root/.revdep-rebuild.4_ebuilds) Evaluating package order... done. (/root/.revdep-rebuild.5_order) All prepared. Starting rebuild... emerge --oneshot -p =sys-devel/gcc-4.1.2 These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] sys-devel/gcc-4.1.2 Now you can remove -p (or --pretend) from arguments and re-run revdep-rebuild. __ It wants to build gcc again. I don't want to have to build gcc every time I need to use revdep-rebuild. The final message from revdep-rebuild is: Build finished correctly. Removing temporary files... You can re-run revdep-rebuild to verify that all libraries and binaries are fixed. If some inconsistency remains, it can be orphaned file, deep dependency, binary package or specially evaluated library. How do I determine if this is a case of "orphaned file, deep dependency, binary package or specially evaluated library" and, if it is one of those, how do I determine which one, and then how do I fix this...? Thanks for listening, Bob Young San Jose, CA -- gentoo-user@lists.gentoo.org mailing list