Re: [gentoo-user] ~x86 emerge: broke emerge
i had this problem about a week ago. emerging gcc for the first time after switching to ~x86 resulted in something happening where many of the g++ libs that are usually (iirc - im not at my computer right now) put in /usr/lib/ were not there. basically i was getting the exact same error message you are. i did a locate for it (and then subsequently all the other libs it couldnt find) and copied them manually to /usr/bin. i didnt think to try ldconfig or anything like that so im not sure. but since then i have upgraded gcc again and everythings worked perfectly. one side affect you may notice if you tried starting gnome from the command prompt would be that nautilis didnt start even after you fixed the problem. im not sure why but just running it once from a terminal and everything went back to normal (normal being that nautilis starts with gnome again). ive been using ~x86 for maybe 2 weeks now and that was my first encounter. just remember. switching back to x86 isnt the easiest thing to do. and besides - there are too many benifits to living on the edge. constant heart failure is one of them ;) currently im having problems with openoffice-ximian but i think its related to trying to run 5 parallel makes (the changelog says something about parallel makes not playing nice) and pwlib wont compile either. other than that im much happier in ~x86. have fun, Frank J. Mattia Kurt Guenther wrote: It broke on a "emerge world". I haven't deleted anything for some time. I found the shlib, so I put it on my LD_LIBRARY_PATH. Much better! I'll go back to the relative safety of X86. --Kurt -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] courier-imap / maildir
please excuse me if this problems already been solved - or if im
answering the wrong question. i have only read the last two posts to
this thread.
my web host provides courier-imap service to me on a openbsd box and
recently i decided to take advantage of it. what i found was that i was
unable to delete any directories under trash. i brought it up to them
and here is what came out of it.
--
Here comes a patch to be included in your Courier IMAP 1.7.2 port. courier-imap cannot delete (empty) folders because it uses unlink() on Maildirs which are not emty (they contain cur/, tmp/ new/). courier-imap checks for a return code EISDIR on unlink and then recurses down to this dir (delsubdir() in maildir/maildirdelfolder.c).
On OpenBSD (3.4-current, at least) unlink returns EPERM, not EISDIR if unlink() is used on a directory.
This small patch changes the folder deletion routine accordingly.
Regards,
Marc
--090009000502030601020708
Content-Type: text/plain; name="patch-maildir_maildirdelfolder_c"
Content-Disposition: inline; filename="patch-maildir_maildirdelfolder_c"
Content-Transfer-Encoding: 7bit
--- maildir/maildirdelfolder.c.orig Mon Jan 26 16:24:17 2004
+++ maildir/maildirdelfolder.c Mon Jan 26 16:27:25 2004
@@ -60,7 +60,7 @@
strcat(strcat(strcpy(p, s), "/"), de->d_name);
- if (unlink(p) && errno == EISDIR)
+ if (unlink(p) && errno == EPERM)
{
if (delsubdir(p))
{
--090009000502030601020708--
hope this helps any.
- Frank
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-user] NPTL help
are you using ~x86. also, what version of glibc did you compile? Wayne Oliver wrote: Hi All I have been running the 2.6.x kernels for sometime now. I decided to give nptl a try. The machine is a P3 933 with 256Mb RAM running mm-sources(2.6.1) I emerged linux-headers... and then emerge glibc but I get an Error stating You have the nptl use flag set but your kernel or architecture does not support it. Do I need to recompile my kernel after emerging the 2.6 headers? Regards Wayne -- This message may contain information which is confidential, private or privileged in nature. If you are not the intended recipient , you may not peruse , use, disseminate, distribute or copy this message or file which is attached to this message. If you have received this message in error , please notify the sender immediately by e-mail , facsimile or telephone and thereafter return and /or destroy the original message. Any views of this communication are those of the sender except where the sender specifically states them to be those of Tsogo Sun Casino's . Please note that the recipient must scan this e-mail and any attached files for viruses and the like. Tsogo Sun Casino's accept no liability of whatever nature for any loss, liability, damage or expense resulting directly or indirectly from the access and/or downloading of any files which are attached to this e-mail message. -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] many apps fail after upgrading gcc
thanks for the response, First read go and very very carefully read: http://www.gentoo.org/doc/en/handbook/handbook.xml?part=2&chap=2#doc_chap4 i've read that and basically it explained everthing i needed to know beforehand. which, mostly i did. however, my questions come after making the move to 2.6.1-mm5 - not necessarily ~x86. Once you've done that, _try_ downgrading gcc again unless your confident that i cant downgrade anything. emerge right now fails on everything because it needs libstdc++.so.5... so ill symlink 6 to 5 and try again. Also realize that the most "bleeding-edge" thing you can do at the moment is to go with the 2.6 headers. Furthermore, they weren't actually required to use a 2.6 kernel. right, they weren't. but they were required to get nptl support compiled into glibc. now, im noticing that nptl support was stripped from the latest build of glibc and if i get portage working, ill have to resort to a pre-xxx version. but even still, shouldnt there be a linux-headers-2.6.1 so that i can have matching headers to compile glibc against? thanks again, Frank -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] many apps fail after upgrading gcc
sorry, i was in a rush to leave for work this morning and now that im rereading my message i can see that i was lacking in some clarity... so heres more of a detailed explanation: last week i had started my trip to the bleeding edge pretty gently. i had a 2.6.0 kernel and decided to get the linux-headers to match. got the headers, added nptl to my use flags, recompiled kernel, then glibc. all is well. nptl works. i passed on recompiling my whole system just yet because of the problems with X and svgalib. moving to the end of last week i started using the 2.6.0-mm series. impressed with the mm branch this week i moved to the 2.6.1-mm line. note, since the first time i recompiled glibc, i have not done it since. yesterday i noticed a glibc upgrade so i decided to go for it. however before the build process started i got an error about the kernel or headers or something being incompatible with nptl. from here i assumed that it was because my headers didnt match my kernel. to my dimay (or maybe i just dont know something that everyone else does) i could not find any 2.6.1 linux-headers. after some toying around i decided to go "~x86" and started upgrading packages... first portage... then gentoolkit.. then binutils... then gcc.. BAM... after gcc upgraded to 3.3.something (the lastest inportage as of yesterday).. most apps that i open say they cant find libstdc++.so.5 which isnt suprising because it doesnt exist anymore... so is there a sane way for me to go about fixing this problem or do i just never upgrade gcc? or is there a certain way of upgrading gcc as to not break things... thanks again, Frank whats the standard way of upgrading gcc? i just started using ~x86 last nite and upgrading gcc completed - but is now throwing more errors on libstdc++.so.5 even emerge throws errors on it. so i cant emerge anything else. any idea whats going on? im using a 2.6.1-mm5 kernel with 2.6.0 linux-headers (because 2.6.1 headers arent out in portage apparently) with NPTL in glibc (compiled when my kernel matched my headers, last week). thanks. -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
[gentoo-user] many apps fail after upgrading gcc
whats the standard way of upgrading gcc? i just started using ~x86 last nite and upgrading gcc completed - but is now throwing more errors on libstdc++.so.5 even emerge throws errors on it. so i cant emerge anything else. any idea whats going on? im using a 2.6.1-mm5 kernel with 2.6.0 linux-headers (because 2.6.1 headers arent out in portage apparently) with NPTL in glibc (compiled when my kernel matched my headers, last week). thanks. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] portage-2.0.50_pre19.tar.bz2
yeah, and i just noticed that it was only released a little bit ago... the package database says it was released today at 21:55 something or another. here in NJ, its just about hitting 21:30. btw, are you the same Ciaran who authored a few of the html-kit plugins? - Frank J. Mattia Mattia Industries, Inc. On Wed, 2004-01-21 at 22:25, Ciaran McCreesh wrote: > On Wed, 21 Jan 2004 22:26:04 -0500 "Frank J. Mattia" > <[EMAIL PROTECTED]> wrote: > | is it me or is that version of portage not on any mirror at all? ive > | been trying for over an hour between mirrorselect and whathaveyou.. > | trying to find it... but all ikeep getting are file not found > | errors... > > Portage _pre versions are generally only kept on Nick's box (this lets > him get an idea of the number of users). Unfortunately, his ISP is being > screwy just now, so you may have problems getting hold of it... -- [EMAIL PROTECTED] mailing list
[gentoo-user] portage-2.0.50_pre19.tar.bz2
is it me or is that version of portage not on any mirror at all? ive been trying for over an hour between mirrorselect and whathaveyou.. trying to find it... but all ikeep getting are file not found errors... -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] linux-headers-2.6.1?
my apologies for starting a new topic in an existing thread. it was very unintentional. (i did a reply-to-list in ximian and didnt realize i was clicking on the thread i was currently reading). On Wed, 2004-01-21 at 21:49, Frank J. Mattia wrote: > last week sometime i compiled a 2.6.0-mm kernel and glibc with NPTL. i > had my linux-headers emerged and all.. etc... worked fine. > > this week ive upgraded to a 2.6.1-mm5 kernel and (im guessing here) > because i still have 2.6.0 headers glibc complains and says i have a > problem with compatibility and wont do NPTL... is there any way around > this or do i have to wait for 2.6.1 headers to make it to portage... > > btw.. i just probably gave myself more grief than i started with. i > just did an ACCEPT_KEYWORDS="~x86".. heh.. now im emerging some > things... while not emerging glibc... any insight would be appreciated > > -Frank J. Mattia > Mattia Industries, Inc. > > > -- > [EMAIL PROTECTED] mailing list > -- [EMAIL PROTECTED] mailing list
[gentoo-user] linux-headers-2.6.1?
last week sometime i compiled a 2.6.0-mm kernel and glibc with NPTL. i had my linux-headers emerged and all.. etc... worked fine. this week ive upgraded to a 2.6.1-mm5 kernel and (im guessing here) because i still have 2.6.0 headers glibc complains and says i have a problem with compatibility and wont do NPTL... is there any way around this or do i have to wait for 2.6.1 headers to make it to portage... btw.. i just probably gave myself more grief than i started with. i just did an ACCEPT_KEYWORDS="~x86".. heh.. now im emerging some things... while not emerging glibc... any insight would be appreciated -Frank J. Mattia Mattia Industries, Inc. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Anyone interested in a fix compiling XFree86 with headers 2.6.0???
I just managed to compile my xfree 4.3.0-r3 with linux-headers 2.6 it needs i have always been a little unsure about what the purpose was to upgrade your linux-headers.. it seems like that should be something you do everytime they're upgraded but apparently people still use old ones... what needs to be done after you emerge them.. do they overwrite the existing ones or do they get another folder..etc.. -Frank -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Centralized user authentication
In a similar topic on slashdot i posed the same question. the general concensus is NIS but if there is an open alternative i would love to hear about it. my comment on slashdot (and all subsequent replys) are at http://slashdot.org/comments.pl?sid=90828&cid=7827715http://slashdot.org/comments.pl?sid=90828&cid=7827715 cheers, Frank J. Mattia i think thats the first time ive ever used cheers in a salutation. and if anyone knows of a better name for the closing part of a letter (other than salutation) - please, for the love of rediwhip... tell me. i've been racking my brain - and google - to no avail. lucas wrote: I am wanting to setup some form of Centralized user authentication for my home gentoo network and was wondering what ppl here recommend. I have had a bit of a look at NIS, and OpenLDAP. NIS from what i have read seams to be a bit insecure and outdated, and OpenLDAP seams to be over kill. Would really like to hear ppls recommendation and comments, thanks -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] "vga=791" no longer works with 2.6.0
therein lies the problem. im not using an nvidia anything. im using a matrox parhelia (which somehow i doubt is supported by the matroxfb. but anyway.. when i get home im going to replace my vga= line with video=matrox:vesa:0x1BF to see if that helps. vga= just looks like its asking for problems (in its simplicity. i mean come on.. who ever heard of any option in anything ever just taking a single paramter which controls screen res, color depth, and fb.. etc... its just tooo easy..). i do have all of the proper config options enabled. thank you for your advice though. if this doesnt work ill go back and rtfm some more. maybe find that one kernel opt i missed. on another note.. would you have any idea about the mouse probs, or what driver i should compile for a generic ac97 onboard soundcard? thanks. frank SN wrote: You have to use certain options for the kernel config, just do a search in the mailinglist archive for nvidia framebuffer and you will find the correct options you have to use. By the way I'm running 2.6 stable since yesterday, on my box everything is working X, nvidia drivers, sensors, alsa, sysfs. I just have one little issue at the moment, one programm while running needs about 20% of cpu time now with the old kernel it was between 5-10 percent, I'm just recompiling glibc and xfree to see if that helps. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] "vga=791" no longer works with 2.6.0
I have this same issue. I use 795 i think for 1600x1200 and i get the same blank screen. i know that someone mentioned something about their nvidia drivers or something but im using a plain jane matrox parheila 512 card and have never needed any external drivers.. not only that but when i take away the vga parameter (which im sure you must have tried) and gdm starts up.. if i ctl-alt-Fx anything.. i get a garbled output... its like all the vtermals are corrupted after any g/xdm starts... i would love to know a solution to this because its one of the few things hanging me up with the 2.6 kernel (that and i cant find the post on how to get 2.4 mouse behavior or how to get alsa working on generic ac97 onboard soundcard)... thanks, frank sorry for my poor english - im not usually that lazy. its just that its 3:30 am and i have work in 2 hours... and still need to sleep. Sebastian Bergmann wrote: I updated my GENTOO/Linux laptop to gentoo-dev-sources-2.6.0 -- and am quite happy with it. Except for one thing: The "vga=791" boot parameter that I used with the 2.4 kernel series to have a 1024x768 console during boot no longer works. If I pass it to the kernel I get a blank screen until gdm has been started. Any idea on what the correct parameter would be? Thanks in advance, Sebastian -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Evolution 1.5 on Gentoo
I do not recommend trying evolution's development versions, unless you have a red-carpet supported system, in my experience it is dependency hell (some of their test packages break other gnome programs). actually, they say on their site that using any dev version voids you right to support... so if thats the case.. i wouldnt use a dev version if you DO have a support contract... i could be wrong. - Frank On Wed, 2003-12-10 at 19:54, Goran Kavrecic wrote: I've downloaded Evolution 1.5 and tried to compile it on Gentoo 1.4 Gnome 2.4 Package Version Evolution 1.5 Gtkhtml 3.1.4 Gal 2.1.1 Evolution Data Server 0.0.3 Libsoup 2.1.2 It compiled libsoup ok, the configure failed on gtkhtml due missing enough new gal, but gal stopped due missing XML parser for perl: configure: error: XML::Parser perl module is required for intltool Or maybe it misses perl for XML-parser? How do I get it? Regards, Goran -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Windows partitions ro
check the man pages for mount and fstab... id say that its a very easy thing to do... that is, unless the windows partitions are on separate computers. [EMAIL PROTECTED] wrote: Is there any way to let non-root users to read/write to Windows partitions? Thanx. :o) -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
[gentoo-user] kde and qt required for gnome?
Good day, i was wondering why kde-base/arts and x11-libs/qt were required for gnome? i have USE ="-qt -kde" in my make.conf file yet emerge -p gnome it lists the aforementioned packages. Thanks, Frank -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] ebuild error on ncurses
Please forgive me ;) im an idiot.. I completely forgot to bootstrap the system (the light came on as i was trying to cd to the /usr/portage dir and i ls'd and saw the scripts dir... - Frank Frank J. Mattia wrote: I just started my second installation of a gentoo system and i noticed that i picked the right time too (with all of the portage problems going on ;) . However, im not sure if my problem is related directly to portage or just this package. im doing an emerge system and i consistantly get this error. Does anyone know what it could mean? What i do know is that it happens during the configure script for the ncurses library and it happens right after it checks for how to run the C++ preprocessor configure: error: C++ preprocessor /lib/cpp failed sanity check !!! ERROR: sys-libs/ncurses-5.3-r1 failed. !!! Function econf, Line 273, Exitcode 1. !!! econf failed Any ideas or suggestions anyone? Thank you, Frank J. Mattia -- [EMAIL PROTECTED] mailing list
[gentoo-user] ebuild error on ncurses
I just started my second installation of a gentoo system and i noticed that i picked the right time too (with all of the portage problems going on ;) . However, im not sure if my problem is related directly to portage or just this package. im doing an emerge system and i consistantly get this error. Does anyone know what it could mean? What i do know is that it happens during the configure script for the ncurses library and it happens right after it checks for how to run the C++ preprocessor configure: error: C++ preprocessor /lib/cpp failed sanity check !!! ERROR: sys-libs/ncurses-5.3-r1 failed. !!! Function econf, Line 273, Exitcode 1. !!! econf failed Any ideas or suggestions anyone? Thank you, Frank J. Mattia -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] HELP installation
Jason Gouger wrote: Code maturity level options ---> ~ [*] Prompt for development and/or incomplete code/drivers" i cant belive how many times i looked at the word [experimental] and saw that option and still didnt make the connection. Thanks - Frank -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] HELP installation
Jason Gouger wrote: I've also forgotten to include devfs. Ive been trying to get my stage1 install to boot but it keeps telling me that i didnt enable devfs... i know what it is but ive checked menuconfig MANY times and i cant find devfs as an option.am i missing something? apart from that i may be using the xfs sources (havent checked yet).. is devfs not in the xfs sources or am im just missing something else entirely... im planning on emerging the vanilla sources when i get home from work to see if devfs is an option there... - Frank -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Apache errors (/var/log/apache/error_log)
[Mon Jun 23 16:46:35 2003] [error] [client 127.0.0.1] File does not exist: /home/httpd/htdocs/favicon.ico To the best of my knowlege, this is just the UA trying to find the favorites icon. However, i dont know if its an automagic thing or if something has to actually trigger the UA to look for it. - Frank -- [EMAIL PROTECTED] mailing list
