Re: [gentoo-user] Firefox Keeps Hanging on some pages
On Tue, 2006-09-12 at 00:05 -0700, Drew wrote: And related to these sites that firefox chokes on, are there any common elements they all share? Perhaps a plugin or something thats called? I'm not sure. I'm seeing this issue from basically clicking on Liferea (RSS reader) links and it's supposed to open up firefox browsers and I just noticed it from there onwards. Do the errors crop up only when you open firefox by clicking through liferea, or do the errors also appear when you visit the pages manually? Like the earlier page - Zedomax, it's choking up even when it's opened manually. (i even got to the point of mv ~/.mozilla ~/,mozilla-old to test things out. I did some digging with google and there does seem to be some problems with using liferea mozilla/firefox together. Is liferea opening a hmm.. didn't realise that but.. anyway.. it's not only happening _together_ firefox window within itself (similar to those annoying apps with a builtin MSIE browser) or is firefox opening the URL itself? no.. it' opening it as a windows (or a new tab) I suspect the problem may be due to the loading of flash (flash 9??) and I only have flash 7 installed. I've (re)installed flashblock. and it seems it can open www.zedomax.com properly. (But it still hangs halfway for like 2 secs before it the circle thingy moves again. -Drew -- gentoo-user@gentoo.org mailing list
[gentoo-user] postfix with openLDAP
Hi, I am trying to configure postfix and courier-imap with openLDAP. But i am not getting any good documents on this topic. I have posted my problem 4 days before also. But didnt get any reply. Please please help me. I will be very thankful to you. Send instant messages to your online friends http://uk.messenger.yahoo.com
Re: [gentoo-user] Simplified apache2
Michael Crute wrote: net-www/apache mpm-prefork threads As a side note from the Apache maintainer: This USE-flag combination is pointless. The threads USE-flag is only used if you don't specify a MPM USE-flag, and caused the mpm to be mpm-worker. (If threads is not set, the MPM defaults to mpm-prefork) -- Michael Stewart [EMAIL PROTECTED] Gentoo Developerhttp://dev.gentoo.org/~vericgar GnuPG Key ID 0x08614788 available on http://pgp.mit.edu -- signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Purging Gnome from the system
On Wednesday 13 September 2006 06:27, Dale wrote: With these USE flags in place I do not have gnome on my system: -eds -esd -gnome -gtk Proof: # equery l gnome [ Searching for package 'gnome' in all categories among: ] * installed packages # Cool. I need to mark this one important so I can find it again. Do you notice anything missing like in OOo, Mozilla/Seamonkey or anything like that?? FWIW I've now gotten my system much closer to where I want it, with the same 4 USE flags removed. The only packages still left with some remnant of gnome in them are seamonkey, firefox, openoffice and xine-lib, all of which I know function just fine without gnome. Big compiles so I'm saving them for the weekend alan -- gentoo-user@gentoo.org mailing list
[gentoo-user] Where can I find a list of supported VIDEO_CARD's?
Hi, I've set up a new computer w/ gentoo which uses a newer graphics card than my old one (an Intel 945 on board card). I installed modular xorg, and AFAIK the drivers which are getting installed are determined by the VIDEO_CARDS flag in make.conf. Is there some list of supported values for VIDEO_CARDS? Ciao, Wolfgang Liebich -- Wolfgang Liebich PSE ECT PSS7 Tel: +43 51707 47734 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] changing CHOST
Timothy A. Holmes wrote: Hi folks: In the course of learning gentoo, I managed to create several systems using the wrong stage 3 tarballs (or something) They all have a CHOST setting of i386 Should I change this? What benefits will it bring me, and Since the new glibc - yes How do I do it? http://dev.gentoo.org/~amne/temp/change-chost.txt -- Naga -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: borked gcc
David Talkington wrote: Richard Fish wrote: On 9/12/06, Richard Fish [EMAIL PROTECTED] wrote: http: //article.gmane.org/gmane.linux.gentoo.user/169677 http: //article.gmane.org/gmane.linux.gentoo.user/169844 Bah, 169844 is not the one I wanted to link to. I wanted this one: http://article.gmane.org/gmane.linux.gentoo.user/169894 You should use this one :) http://dev.gentoo.org/~amne/temp/change-chost.txt -- Naga -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Where can I find a list of supported VIDEO_CARD's?
2006/9/13, Wolfgang Liebich [EMAIL PROTECTED]: Hi, I've set up a new computer w/ gentoo which uses a newer graphics card than my old one (an Intel 945 on board card). I installed modular xorg, and AFAIK the drivers which are getting installed are determined by the VIDEO_CARDS flag in make.conf. Is there some list of supported values for VIDEO_CARDS? Hi, isn't there a list of useflags when you do 'emerge -pv xorg-x11' ? regards, Boris. Ciao, Wolfgang Liebich -- Wolfgang Liebich PSE ECT PSS7 Tel: +43 51707 47734 -- gentoo-user@gentoo.org mailing list -- Quiconque me parle de Dieu en veut à ma bourse ou à ma liberté. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Installing 2006.1 from the livedvd - is it really possible?
Hi, GenToo (at least the 2006.1 livedvd) is driving me nuts. I have tried several installations (more than 6) and they all fail by e.g. EmergePackageError :FATAL: emerge: Could not emerge mail-mta/ssmtp it cannot build a kernel (just copy the kernel on the dvd) it cannot build EmergePackageError :FATAL: emerge: Could not emerge media-libs/xine-lib-1.1.2-r2 Furthermore I cannot enter the URIs for the ftp-proxie and the rsync-proxie. The fields are too short. I have been using the command line installer in 'advanced' mode. So, what can I do? Many thanks for life-saver, Helmut Jarausch Lehrstuhl fuer Numerische Mathematik RWTH - Aachen University D 52056 Aachen, Germany -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] changing CHOST
On 9/12/06, Richard Fish [EMAIL PROTECTED] wrote: On 9/12/06, darren kirby [EMAIL PROTECTED] wrote: system. The cpu instructions are forward (but not backward) compatible. Thus, I did not think it necessary for a wholesale rebuild of the entire system on the spot. Indeed it seems this is not as simple as I thought. I may have been thinking changing CHOST was as simple as changing -march or -mcpu... Well I've brought up the issue on -dev, and it seems that something official is in the works on this, so we'll see what comes of that. Maybe your answer will turn out to be correct! And indeed, the emerge -e world may in fact be optional. A quite from the current draft of the guide [1]: - You may want to run # emerge -e world now. In theory it should not be necessary to do so, but it can not be 100% guaranteed that this is actually the case. ;-) I think I'd still like to see people run an emerge -e system regardless...just as an extra safety step to make sure the system packages are sane. But my understanding of what can be affected by a change in CHOST is definitely changed. -Richard [1] http://dev.gentoo.org/~amne/temp/change-chost.txt -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Compile failure: fox-1.2.6-r3
On Wednesday 13 September 2006 04:56, Richard Fish wrote: echo ~x11-libs/fox-1.2.18 ~x86 /etc/portage/package.keywords Replace ~x86 with ~amd64 if you are on the amd64 arch. Or just remove it all together: echo ~x11-libs/fox-1.2.18 /etc/portage/package.keywords If nothing is specified it will be interpreted by portage as ~YOURARCH. -- Bo Andresen pgpZkn7ZjyGbn.pgp Description: PGP signature
Re: [gentoo-user] Compile failures: my system
On Wednesday 13 September 2006 05:14, Meino Christian Cramer wrote: to not include my system specifications in any of the compile failure mails I will summarize it in separate mail: Oh, ffs! Was it really necessary to clutter this list with 7 independent threads out of which 6 requires the exact same answer?? Never do that again! If you really have to clutter the list with useless mails stating your inability to search bugzilla at least keep it in one thread! vamps is the only one of those packages that doesn't have a bug already which contains all the info there is about this for you to find it. So file a bug about vamps [3] and search for the rest of the packages [1] or [2]! Never file a bug without searching bugzilla first though [1] and [2]! [1] http://bugs.gentoo.org/query.cgi?format=specific [2] http://bugs.gentoo.org/query.cgi?format=advanced [3] http://bugs.gentoo.org/enter_bug.cgi?product=Gentoo%20Linux -- Bo Andresen pgpIwgN9nmlWJ.pgp Description: PGP signature
Re: [gentoo-user] Where can I find a list of supported VIDEO_CARD's?
On Wednesday 13 September 2006 08:17, Wolfgang Liebich wrote: I've set up a new computer w/ gentoo which uses a newer graphics card than my old one (an Intel 945 on board card). I installed modular xorg, and AFAIK the drivers which are getting installed are determined by the VIDEO_CARDS flag in make.conf. Is there some list of supported values for VIDEO_CARDS? Boris has already answered how to get a list of valid values for VIDEO_CARDS. When in doubt about which driver to use, however, I believe xorg.freedesktop.org is the best resource: http://xorg.freedesktop.org/wiki/IntelGraphicsDriver VIDEO_CARDS=i810 vesa (I always add vesa as a fallback video driver. i810 is what you need). -- Bo Andresen pgpfmIksI6byx.pgp Description: PGP signature
Re: [gentoo-user] Outputing 15.9kHz?
Hi, On Tue, 12 Sep 2006 18:21:44 -0700 Grant [EMAIL PROTECTED] wrote: I'm using the modeline from this link: http://www.sput.nl/hardware/tv-x.html and my xorg.conf looks like this: Section Monitor Identifier monitor1 Modeline 736x485i 14.16 736 760 824 904 485 491 496 525 interlace -hsync -vsync EndSection [...] When I try to open an xfce4 desktop like this, the image is somewhat scrambled and constantly rolls on the TV. It looks normal on a monitor with the same settings. OK, that rolling is bad sync. /var/log/Xorg.0.log reports this: (II) I810(0): monitor1: Using default hsync range of 28.00-33.00 kHz That's the culprit. You need to override defaults (TVs don't usually have DDC, such no automatic detection) for hsync/vsync. Add a HorizSync line to your xorg.conf: Section Monitor ... HorizSync 19.00-20.00 ... EndSection Note that VertRefresh probably hasn't to be set, since its default is sufficient. (II) I810(0): Not using mode 736x485i (unknown reason) Hm, I can not explain why this would be unknown, though. For me, this seems like another case of hsync out of range... -hwh -- gentoo-user@gentoo.org mailing list
[gentoo-user] BUG 2.6.17-gentoo-r8 does not compile
Hi, gentoo doesn't seem to like me. Trying to install gentoo, I did USE=-doc symlink emerge gentoo-sources After menuconfig make terminates due to a shell syntax error. Are the gentoo kernel patches buggy? Many thanks for hint, Helmut Jarausch Lehrstuhl fuer Numerische Mathematik RWTH - Aachen University D 52056 Aachen, Germany -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] BUG 2.6.17-gentoo-r8 does not compile
Hi, I don't know if it's relevant but as the installation guide says, I do 'make make modules_install'. Otherwise you could try to use genkernel, just to see if kernel sources are corrupted. On 9/13/06, Helmut Jarausch [EMAIL PROTECTED] wrote: Hi,gentoo doesn't seem to like me.Trying to install gentoo, I didUSE=-doc symlink emerge gentoo-sources After menuconfigmaketerminates due to a shell syntax error.Are the gentoo kernel patches buggy?Many thanks for hint,Helmut JarauschLehrstuhl fuer Numerische MathematikRWTH - Aachen University D 52056 Aachen, Germany--gentoo-user@gentoo.org mailing list-- /JM
Re: [gentoo-user] BUG 2.6.17-gentoo-r8 does not compile
After menuconfig make terminates due to a shell syntax error. Are the gentoo kernel patches buggy? Many thanks for hint, hint: posting an exact copy of the output from your terminal would probably get you more help ;) -- Iain Buchanan iaindb at netspace dot net dot au Does the name Pavlov ring a bell? -- gentoo-user@gentoo.org mailing list
[gentoo-user] coax spdif sound output on Intel 82801EB/ER audio
Hi all, I used to (rarely) use the coax output on my Dell Inspiron 9100 laptop to feed into my hifi. Now that I watch a lot more movies / tv from my laptop, I wanted to get it going again, but I no longer get any output from the coax out. Investigating google a bit, I found that this works with mplayer: mplayer -ao alsa:device=spdif -srate 48000 blah.mp3 giving me crystal clear audio (I had to mute it to make sure there wasn't someone in my house, cause the sound effects on the audio track were so realistic!) how do I get it to work by Default for everything alsa? (gnome, rhythmbox, etc) I tried playing around with all the options I could see in gnome-volume-control, no luck. google suggested setting IEC958 playback to 0 (yes, zero) which worked for others, but firstly this setting was never saved with gnome-volume-control or even alsamixer, and secondly, it didn't seem to have any effect. Unfortunately, I can't tell you anything about my system when it did work like this (it was a while ago), only that it doesn't anymore. Here are the specs: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02) pci-id 8086:24d5 suspend2 sources 2.6.17-r5, with alsa built as a module (snd_intel8x0). Can anyone shed any light on the issue? many thanks, -- Iain Buchanan iaindb at netspace dot net dot au Price's Advice: It's all a game -- play it to have fun. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] BUG 2.6.17-gentoo-r8 does not compile
On Wed, 13 Sep 2006 13:04:40 +0200 (CEST), Helmut Jarausch wrote: After menuconfig make terminates due to a shell syntax error. Are the gentoo kernel patches buggy? Who knows? Maybe if the error message wasn't considered Top Secret, someone would be able to shed some light on the cause. I've compiled and am running -r8 on x64, amd64 and PPC systems with no such messages, secret or otherwise. -- Neil Bothwick Don't let your mind wander, it's too little to be let out alone. signature.asc Description: PGP signature
[gentoo-user] /etc/profile.d/ and non-root users
Hello, I have created a file /etc/profile.d/aliases.sh containing aliases.sh and when I log in as root it is clearly loaded. But when I log in as my normal user, it is not. Any ideas ? -- Pupeno [EMAIL PROTECTED] (http://pupeno.com) pgp8V9wgM5BgU.pgp Description: PGP signature
RE: [gentoo-user] changing CHOST
On 9/12/06, Richard Fish [EMAIL PROTECTED] wrote: On 9/12/06, darren kirby [EMAIL PROTECTED] wrote: system. The cpu instructions are forward (but not backward) compatible. Thus, I did not think it necessary for a wholesale rebuild of the entire system on the spot. Indeed it seems this is not as simple as I thought. I may have been thinking changing CHOST was as simple as changing -march or -mcpu... Well I've brought up the issue on -dev, and it seems that something official is in the works on this, so we'll see what comes of that. Maybe your answer will turn out to be correct! And indeed, the emerge -e world may in fact be optional. A quite from the current draft of the guide [1]: - You may want to run # emerge -e world now. In theory it should not be necessary to do so, but it can not be 100% guaranteed that this is actually the case. ;-) I think I'd still like to see people run an emerge -e system regardless...just as an extra safety step to make sure the system packages are sane. But my understanding of what can be affected by a change in CHOST is definitely changed. -Richard [1] http://dev.gentoo.org/~amne/temp/change-chost.txt -- gentoo-user@gentoo.org mailing list [Timothy A. Holmes] Hi folks -- I really appreciate all the input on this -- It turns out - the system borked and wouldn't let me login from ssh or console, so I dropped in a live cd and am in the process of rebuilding it (with the proper stage) Thanks again for all the help -- I really appreciate it and I have learned a great deal TIM Timothy A. Holmes IT Manager / Network Admin / Web Master / Computer Teacher Medina Christian Academy A Higher Standard... -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: Simplified apache2
Ryan Tandy tarpman at gmail.com writes:
Michael Crute wrote:
USE=-* hardened pic ncurses ssl crypt berkdb tcpd pam perl python
readline
Hello Ryan,
glibc croaked during an upgrade/recompile and told me to add:
'nptl nptlonly' to make.conf. I did and the sytem completed a deep recompile
late last night.
Even this is a bit more bloated than it needs to be. I have never used
'tcpd' or 'berkdb' on any system I run, and 'perl' and 'python' are
*much* more useful (IMO) as local flags (in package.use) than as global
ones. Even 'ssl' doesn't *have* to be there, especially in the global
scope - 'www-client/links ssl' in package.use should be more than
sufficient. One flag missing from that line that I like to have is
'bzip2' - tar just isn't quite the same without bz2 support. ;)
Ok,
So I'll test your suggestions.
The more minimized the global flags are, the more secure the server.
Also, be careful using the hardened flag without running the hardened
profile. The hardened profile masks out a couple of packages and flags
that don't work so well on a hardened system.
H,
Not sure I fully grasp what you mean by a 'hardened system'. If you mean
running a hardened kernel with only necessary software installed, then
yes, I run hardened kernels on most servers {dns, web, mail, firwalls}
If running a hardened system means more than that, please explain,
or point me to some docs.
BTW, the flags with underscores in them (kernel_linux, userland_GNU,
elibc_glibc, video_cards_radeon and such) are known as USE_EXPAND or
expanded USE flags.
This is nice to know.
I did not get the memo on this.
Any docs for further reading you can point me to?
thanks for all of the information,
thanks to everyone for help on this,
James
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] /etc/profile.d/ and non-root users
On Wednesday 13 September 2006 14:26, Pupeno wrote: Hello, I have created a file /etc/profile.d/aliases.sh containing aliases.sh and when I log in as root it is clearly loaded. But when I log in as my normal user, it is not. Any ideas ? I have absolutely no idea and most likely neither will anyone else. But if you post the results of these commands, we will be in a better position to help you: ls -al /etc/profile.d/aliases.sh cat /etc/profile.d/aliases.sh alan -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] changing CHOST
On 9/13/06, Timothy A. Holmes [EMAIL PROTECTED] wrote: On 9/12/06, Richard Fish [EMAIL PROTECTED] wrote: On 9/12/06, darren kirby [EMAIL PROTECTED] wrote: system. The cpu instructions are forward (but not backward) compatible. Thus, I did not think it necessary for a wholesale rebuild of the entire system on the spot. Indeed it seems this is not as simple as I thought. I may have been thinking changing CHOST was as simple as changing -march or -mcpu... Well I've brought up the issue on -dev, and it seems that something official is in the works on this, so we'll see what comes of that. Maybe your answer will turn out to be correct! And indeed, the emerge -e world may in fact be optional. A quite from the current draft of the guide [1]: - You may want to run # emerge -e world now. In theory it should not be necessary to do so, but it can not be 100% guaranteed that this is actually the case. ;-) I think I'd still like to see people run an emerge -e system regardless...just as an extra safety step to make sure the system packages are sane. But my understanding of what can be affected by a change in CHOST is definitely changed. -Richard [1] http://dev.gentoo.org/~amne/temp/change-chost.txt -- gentoo-user@gentoo.org mailing list [Timothy A. Holmes] Hi folks -- I really appreciate all the input on this -- It turns out - the system borked and wouldn't let me login from ssh or console, so I dropped in a live cd and am in the process of rebuilding it (with the proper stage) Thanks again for all the help -- I really appreciate it and I have learned a great deal The thread is kinda dead now that the OP borked the system and is reinstalling, but anyway, I'll report to it for future reference. Old threads in this list and foruns talk about this with different ways/advices, but what ended up working for me (changind from i386 to i686) was: some quickpkg's (gcc, glibc, python, portage, libstdc++) (just to be sure I would NOT need a livecd) edit make.conf emerge gcc (as it compiles itself with the new compiler, ended up with i686-pc-linux-gnu) emerge glibc python portage libstdc++ (just to make sure all tools for emerge were already compiled with the new GCC, I don't know wich order the emerge -e system uses for packages, it upgraded my glibc, wich was the whole point of the chost change) emerge -e system (just to make sure most tools were already compiled with the new chost before any reboot/retry) emerge -e world (I guess recompiling stuff with the new chost will take advantage of features present in new processors, that's the whole point of HAVING this flag, isnt? Anyway, its more like a bug hunting command, also, broken dynamic linking that could be hanging will appear, specially libstdc++, but I kinda ask myself if it was indeed needed.) Up and running here... -- Daniel da Veiga Computer Operator - RS - Brazil -BEGIN GEEK CODE BLOCK- Version: 3.1 GCM/IT/P/O d-? s:- a? C++$ UBLA++ P+ L++ E--- W+++$ N o+ K- w O M- V- PS PE Y PGP- t+ 5 X+++ R+* tv b+ DI+++ D+ G+ e h+ r+ y++ --END GEEK CODE BLOCK-- -- gentoo-user@gentoo.org mailing list
[gentoo-user] php-5.1.6-r4 failed
Hi, Does anyone know how to work this around. I have no apache, I use lighttpd. Here is some of the output of emerge: * Determining SAPI(s) to build * Enabled SAPI: cli * Enabled SAPI: cgi * Disabled SAPI: apache * Disabled SAPI: apache2 --snip Compiling source in /tmp/portage/php-5.1.6-r4/work/php-5.1.6 ... * Enabling bcmath * Enabling bz2 * Enabling calendar * Enabling ctype * Enabling curl * Enabling curlwrappers * Disabling dbase * Enabling dom * Disabling exif * Disabling fbsql * Disabling fdftk * Disabling filepro * Enabling ftp * Enabling gettext * Enabling gmp * Disabling hash * Disabling hwapi * Enabling iconv * Disabling informix * Disabling ipv6 * Disabling kerberos * Enabling libxml * Enabling mbstring * Enabling mcrypt * Enabling memory-limit * Enabling mhash * Disabling ming * Disabling msql * Disabling mssql * Enabling ncurses * Enabling openssl * Enabling openssl-dir * Enabling pcntl * Enabling pcre-regex * Disabling pdo * Disabling pgsql * Enabling posix * Enabling pspell * Disabling recode * Disabling reflection * Disabling simplexml * Disabling shmop * Disabling snmp * Disabling soap * Enabling sockets * Disabling spl * Disabling sybase * Disabling sybase-ct * Enabling sysvmsg * Enabling sysvsem * Enabling sysvshm * Enabling tidy * Disabling tokenizer * Disabling wddx * Enabling xml * Disabling xmlreader * Disabling xmlwriter * Enabling xmlrpc * Disabling xsl * Enabling zlib * Disabling debug * Enabling cdb * Disabling db4 * Disabling flatfile * Disabling gdbm * Disabling inifile * Disabling qdbm * Enabling freetype-dir * Enabling t1lib * Disabling gd-jis-conv * Enabling gd-native-ttf * Enabling jpeg-dir * Enabling png-dir * Enabling xpm-dir * Enabling gd * Enabling imap * Enabling imap-ssl * Enabling mysql * Enabling mysql-sock * Disabling mysqli * Enabling readline * Disabling libedit * Disabling mm * Enabling sqlite * Enabling sqlite-utf8 * Enabling PIC support --snip checking for Berkeley DB2 support... no checking for DB1 support... no checking for DBM support... no checking for cdb_read in -lcdb... no checking for cdb_read in -lc... no configure: error: DBA: Could not find necessary library. !!! ERROR: dev-lang/php-5.1.6-r4 failed. Call stack: ebuild.sh, line 1546: Called dyn_compile ebuild.sh, line 937: Called src_compile php-5.1.6-r4.ebuild, line 173: Called src_compile_normal php-5.1.6-r4.ebuild, line 323: Called php5_1-sapi_src_compile php5_1-sapi.eclass, line 575: Called die !!! configure failed !!! If you need support, post the topmost build error, and the call stack if relevant. -- Best regards, Daniel -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] /etc/profile.d/ and non-root users
On Wednesday 13 September 2006 14:26, Pupeno wrote: Hello, I have created a file /etc/profile.d/aliases.sh containing aliases.sh and when I log in as root it is clearly loaded. But when I log in as my normal user, it is not. Any ideas ? Looking at the shell initialization files, seems that the only script which checks the contents of /etc/profile.d and sources some files from there is /etc/csh.cshrc (unless I missed some). So, is your root account using csh as its default shell and the normal user instead uses bash? -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] /etc/profile.d/ and non-root users
On Wednesday 13 September 2006 15:34, Etaoin Shrdlu wrote: Looking at the shell initialization files, seems that the only script which checks the contents of /etc/profile.d and sources some files from there is /etc/csh.cshrc (unless I missed some). You are missing /etc/profile. Only thing this requires is the file is located in /etc/profile.d, ends on .sh and is readable and sourceable by the user. Probably just a permissions problem. -- Bo Andresen pgpGeV2Aimuvb.pgp Description: PGP signature
Re: [gentoo-user] /etc/profile.d/ and non-root users
On Wednesday 13 September 2006 15:15, Bo Ørsted Andresen wrote: You are missing /etc/profile. Only thing this requires is the file is located in /etc/profile.d, ends on .sh and is readable and sourceable by the user. Probably just a permissions problem. This is strange: my /etc/profile doesn't have any references to /etc/profile.d (that's why I wrote the previous email). In the past, I used to customize that file a lot, so at some point I probably told etc-update to keep my version. Indeed, looking at /etc/profile on another box, I found the /etc/profile.d stuff. Thanks for the correction. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
Hi,
On Wed, 13 Sep 2006 12:36:45 + (UTC)
James [EMAIL PROTECTED] wrote:
Ryan Tandy tarpman at gmail.com writes:
Michael Crute wrote:
USE=-* hardened pic ncurses ssl crypt berkdb tcpd pam perl
python readline
You could omit pic here IIRC (on a hardened profile) hardened
includes -fpic -fpie CFLAGS, plus SSP in GCC-4.1.1 (a default).
If using a vanilla (desktop server) profile you'll need 'pie' as well.
Maybe (if not using a hardened profile) you'll also need some LDFLAGS.
Ok,
So I'll test your suggestions.
The more minimized the global flags are, the more secure the server.
+1
Could also check the flags in hardened profile.
Also, be careful using the hardened flag without running the
hardened profile. The hardened profile masks out a couple of
packages and flags that don't work so well on a hardened system.
+1
H,
Not sure I fully grasp what you mean by a 'hardened system'. If you
mean running a hardened kernel with only necessary software
installed, then yes, I run hardened kernels on most servers {dns,
web, mail, firwalls}
If running a hardened system means more than that, please explain,
or point me to some docs.
Check hardened docs page on w.g.o, in short hardened means a kernel
with PaX (+ -fpie for packages) some sort of RBAC system - grsec, RSBAC
or SELinux and all user-land build with SSP,pic,pie (IMHO).
BTW, the flags with underscores in them (kernel_linux,
userland_GNU, elibc_glibc, video_cards_radeon and such) are known
as USE_EXPAND or expanded USE flags.
This is nice to know.
I did not get the memo on this.
Any docs for further reading you can point me to?
...SKIP...
James
HTH.Rumen
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Simplified apache2
On 9/13/06, Michael Stewart (vericgar) [EMAIL PROTECTED] wrote: Michael Crute wrote: net-www/apache mpm-prefork threads As a side note from the Apache maintainer: This USE-flag combination is pointless. The threads USE-flag is only used if you don't specify a MPM USE-flag, and caused the mpm to be mpm-worker. (If threads is not set, the MPM defaults to mpm-prefork) Hmm... I seem to remember some problem with PHP only compiling if threads was set to on. Maybe not, will have to test it to see if there is any good reason that I still have that in my package.use. Thanks for the tip. -Mike -- Michael E. Crute http://mike.crute.org I may not have gone where I intended to go, but I think I have ended up where I intended to be. --Douglas Adams -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
On 9/13/06, James [EMAIL PROTECTED] wrote:
Not sure I fully grasp what you mean by a 'hardened system'. If you mean
running a hardened kernel with only necessary software installed, then
yes, I run hardened kernels on most servers {dns, web, mail, firwalls}
If running a hardened system means more than that, please explain,
or point me to some docs.
I guess I should have clarified when I made my initial suggestion. A
hardened system is one that is running the hardened profile. All my
server systems are built from the hardened stage 1 tarball. So
basically, you should not use the hardened useflag if your system was
not built with the hardened profile.
Note that there is more to hardening a system than just using a
certain profile or a combination of useflags but its a good start.
-Mike
--
Michael E. Crute
http://mike.crute.org
I may not have gone where I intended to go, but I think I have ended
up where I intended to be. --Douglas Adams
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] /etc/profile.d/ and non-root users
On Wed, Sep 13, 2006 at 03:49:51PM +0200, Etaoin Shrdlu wrote: On Wednesday 13 September 2006 15:15, Bo ??rsted Andresen wrote: You are missing /etc/profile. Only thing this requires is the file is located in /etc/profile.d, ends on .sh and is readable and sourceable by the user. Probably just a permissions problem. This is strange: my /etc/profile doesn't have any references to /etc/profile.d (that's why I wrote the previous email). In the past, I used to customize that file a lot, so at some point I probably told etc-update to keep my version. Indeed, looking at /etc/profile on another box, I found the /etc/profile.d stuff. Just remember that those aliases will stay only in login shell, any other shell level wont keep them. There should be something like /etc/shrc.d/ sourced by bashrc for such things. -- Regards Lukasz Pawelczyk -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] php-5.1.6-r4 failed [SOLVED]
Just for information in case somebody else finds him/herself in the same situation. PHP emerged successfully after disabling cdb use flag with: euse -D cdb Daniel Iliev wrote: Hi, Does anyone know how to work this around. I have no apache, I use lighttpd. Here is some of the output of emerge: * Determining SAPI(s) to build * Enabled SAPI: cli * Enabled SAPI: cgi * Disabled SAPI: apache * Disabled SAPI: apache2 --snip Compiling source in /tmp/portage/php-5.1.6-r4/work/php-5.1.6 ... * Enabling bcmath * Enabling bz2 * Enabling calendar * Enabling ctype * Enabling curl * Enabling curlwrappers * Disabling dbase * Enabling dom * Disabling exif * Disabling fbsql * Disabling fdftk * Disabling filepro * Enabling ftp * Enabling gettext * Enabling gmp * Disabling hash * Disabling hwapi * Enabling iconv * Disabling informix * Disabling ipv6 * Disabling kerberos * Enabling libxml * Enabling mbstring * Enabling mcrypt * Enabling memory-limit * Enabling mhash * Disabling ming * Disabling msql * Disabling mssql * Enabling ncurses * Enabling openssl * Enabling openssl-dir * Enabling pcntl * Enabling pcre-regex * Disabling pdo * Disabling pgsql * Enabling posix * Enabling pspell * Disabling recode * Disabling reflection * Disabling simplexml * Disabling shmop * Disabling snmp * Disabling soap * Enabling sockets * Disabling spl * Disabling sybase * Disabling sybase-ct * Enabling sysvmsg * Enabling sysvsem * Enabling sysvshm * Enabling tidy * Disabling tokenizer * Disabling wddx * Enabling xml * Disabling xmlreader * Disabling xmlwriter * Enabling xmlrpc * Disabling xsl * Enabling zlib * Disabling debug * Enabling cdb * Disabling db4 * Disabling flatfile * Disabling gdbm * Disabling inifile * Disabling qdbm * Enabling freetype-dir * Enabling t1lib * Disabling gd-jis-conv * Enabling gd-native-ttf * Enabling jpeg-dir * Enabling png-dir * Enabling xpm-dir * Enabling gd * Enabling imap * Enabling imap-ssl * Enabling mysql * Enabling mysql-sock * Disabling mysqli * Enabling readline * Disabling libedit * Disabling mm * Enabling sqlite * Enabling sqlite-utf8 * Enabling PIC support --snip checking for Berkeley DB2 support... no checking for DB1 support... no checking for DBM support... no checking for cdb_read in -lcdb... no checking for cdb_read in -lc... no configure: error: DBA: Could not find necessary library. !!! ERROR: dev-lang/php-5.1.6-r4 failed. Call stack: ebuild.sh, line 1546: Called dyn_compile ebuild.sh, line 937: Called src_compile php-5.1.6-r4.ebuild, line 173: Called src_compile_normal php-5.1.6-r4.ebuild, line 323: Called php5_1-sapi_src_compile php5_1-sapi.eclass, line 575: Called die !!! configure failed !!! If you need support, post the topmost build error, and the call stack if relevant. -- Best regards, Daniel -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Flashing the BIOS
On Wednesday 13 September 2006 05:03, Brett I. Holcomb wrote: Most BIOS programs save the existing BIOS so they need somewhere to write to. I never encountered a bios flasher that forced me to save the bios somewhere. A lot of them did not even had to save the bios, if you wanted to. But even if you have such a crazy flasher - that is, what the dos-partition is made for. Just let it save the stuff to the harddisk. No floppy, no cry ;) -- gentoo-user@gentoo.org mailing list
[gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this?-- David Granthttp://www.davidgrant.ca
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On Wednesday 13 September 2006 18:24, David Grant wrote: revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this? So why don't you show us the output of # revdep-rebuild -i -vp ? That way we would be able to give a qualified answer instead of guessing... Anyhow, it's probably bug #125728 [1] again. [1] http://bugs.gentoo.org/show_bug.cgi?id=125728 -- Bo Andresen pgpSn2at0z76D.pgp Description: PGP signature
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this? I have the same problem. I ignore that and everything works well ;) -- Petr Uzel [EMAIL PROTECTED] Jabber : [EMAIL PROTECTED] ICQ : 101606095 Gentoo Linux -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Compile failure: fox-1.2.6-r3
On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: On Wednesday 13 September 2006 04:56, Richard Fish wrote: echo ~x11-libs/fox-1.2.18 ~x86 /etc/portage/package.keywords Replace ~x86 with ~amd64 if you are on the amd64 arch. Or just remove it all together: Thanks Bo. I keep forgetting about that default option! ;-) -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Emacs - (I think [the]) Fonts [are] Broken
On Mon, 11 Sep 2006 16:24:11 -0700 Lord Sauron [EMAIL PROTECTED] wrote: quoth the Lord Sauron: [...] What should I do to fix Emacs? I hear it's a very powerful and [...] Much more confused than before. I get why I'd want to use unicode, but the instructions on how to do that were outdated or for a setup other than mine and they threw me off real quickly. Did you try ran emacs in X also? Is there the same behaviour? In X you can pres Shift + Left mouse click to select fonts. You can also play with locale setting in your system. For example try to start emacs with default/none locale: $ LC_ALL=C emacs To set emacs to use UTF-8 put this into your ~/.emacs file: (setq locale-coding-system 'utf-8) (set-terminal-coding-system 'utf-8) (set-keyboard-coding-system 'utf-8) (set-selection-coding-system 'utf-8) (set-clipboard-coding-system 'utf-8) (prefer-coding-system 'utf-8) (modify-coding-system-alist 'file 'utf-8) (setq process-coding-system-alist '((.* . utf-8))) (setq x-select-request-type '(UTF8_STRING COMPOUND_TEXT TEXT STRING)) But it should not be necessary if you use some UTF-8 locale in your system (emacs will use UTF-8 by default then). You can also try to change the 'utf-8' string to some other (non utf-8) enconding, for example iso-8859-1. Try to emerge some basic X.Org fonts, for example media-fonts/font-misc-misc. Also emerge media-fonts/intlfonts. I hope it helps. Robert -- Robert Cernansky E-mail: [EMAIL PROTECTED] Jabber: [EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On Wednesday 13 September 2006 18:24, David Grant wrote: revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this? Yes, usually I recreate two links by hand. I can't remember which files are to be linked, neither can now find in bash history the commands used... Please post output of revdep-rebuild. Ciao Francesco -- Linux Version 2.6.17-gentoo-r8, Compiled #2 PREEMPT Tue Sep 12 23:42:20 CEST 2006 One 1GHz AMD Athlon 64 Processor, 2GB RAM, 2004.02 Bogomips Total aemaeth -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
On Wednesday 13 September 2006 14:36, James wrote: The more minimized the global flags are, the more secure the server. Were I the only one who wasn't quite convinced by that statement? -- Bo Andresen pgp90yjwqq0Ib.pgp Description: PGP signature
[gentoo-user] [OT] print to printer on winxp via cups samba smb
I need some help getting printing to work from my gentoo machine to an HP 7410xi (one of those multi-function jobs). No one is responding to a similar query on the cups list. So bringing it here although it may be a little off topic. The printer works from windows but not from gentoo through cups. What is kind of baffling me is that the logs don't really show a clear error and in fact appear to be saying that the print job was successful, but nothing comes out of the printer. My /etc/cups/printers.conf uses an smb:// addressing (xx'ed out in this post): DefaultPrinter chub-print Info chub-print Location chub DeviceURI smb://reader:[EMAIL PROTECTED]/chub-print) State Stopped StateMessage /usr/libexec/cups/backend/smb failed StateTime 1157936264 Accepting Yes Shared Yes JobSheets none none QuotaPeriod 0 PageLimit 0 KLimit 0 OpPolicy default ErrorPolicy stop-printer /Printer Log level in cupsd.conf is set to debug ... here is the error_log output from one print attempt where user reader calls: lp file request id is chub-print-18 (1 file(s)) The only thing that looks like it might indicate an error is the line saying: cupsdAuthorize: No authentication data provided. 28 lines from error_log produced by above print attempt: === D [13/Sep/2006:11:53:26 -0500] cupsdAcceptClient: 5 from localhost:631 (IPv4) D [13/Sep/2006:11:53:26 -0500] cupsdReadClient: 5 POST / HTTP/1.1 D [13/Sep/2006:11:53:26 -0500] cupsdAuthorize: No authentication data provided. D [13/Sep/2006:11:53:26 -0500] CUPS-Get-Printers D [13/Sep/2006:11:53:26 -0500] cupsdProcessIPPRequest: 5 status_code=0 (successful-ok) D [13/Sep/2006:11:53:26 -0500] cupsdReadClient: 5 POST / HTTP/1.1 D [13/Sep/2006:11:53:26 -0500] cupsdAuthorize: No authentication data provided. D [13/Sep/2006:11:53:26 -0500] CUPS-Get-Classes D [13/Sep/2006:11:53:26 -0500] cupsdProcessIPPRequest: 5 status_code=0 (successful-ok) D [13/Sep/2006:11:53:26 -0500] cupsdReadClient: 5 POST / HTTP/1.1 D [13/Sep/2006:11:53:26 -0500] cupsdAuthorize: No authentication data provided. D [13/Sep/2006:11:53:26 -0500] CUPS-Get-Default D [13/Sep/2006:11:53:26 -0500] cupsdProcessIPPRequest: 5 status_code=0 (successful-ok) D [13/Sep/2006:11:53:26 -0500] cupsdCloseClient: 5 D [13/Sep/2006:11:53:26 -0500] cupsdAcceptClient: 5 from localhost:631 (IPv4) D [13/Sep/2006:11:53:26 -0500] cupsdReadClient: 5 POST /printers/chub-print HTTP/1.1 D [13/Sep/2006:11:53:26 -0500] cupsdAuthorize: No authentication data provided. D [13/Sep/2006:11:53:26 -0500] Print-Job ipp://localhost/printers/chub-print D [13/Sep/2006:11:53:26 -0500] print_job: auto-typing file... D [13/Sep/2006:11:53:26 -0500] print_job: request file type is text/plain. D [13/Sep/2006:11:53:26 -0500] add_job: requesting-user-name=reader I [13/Sep/2006:11:53:26 -0500] Adding start banner page none to job 18. D [13/Sep/2006:11:53:26 -0500] Discarding unused job-created event... I [13/Sep/2006:11:53:26 -0500] Adding end banner page none to job 18. I [13/Sep/2006:11:53:26 -0500] Job 18 queued on chub-print by reader. D [13/Sep/2006:11:53:26 -0500] Job 18 hold_until = 0 D [13/Sep/2006:11:53:26 -0500] cupsdProcessIPPRequest: 5 status_code=0 (successful-ok) D [13/Sep/2006:11:53:26 -0500] cupsdCloseClient: 5 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
Francesco Talamona wrote: On Wednesday 13 September 2006 18:24, David Grant wrote: revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this? Yes, usually I recreate two links by hand. I can't remember which files are to be linked, neither can now find in bash history the commands used... Please post output of revdep-rebuild. Ciao Francesco It's on one of the roach reports. I saw it the other day. You may can search for revdep-rebuild and gcc and find it. Dale :-) :-) -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Problem with compiling OpenOffice 2.0.3
On 9/13/06, Xavier MOGHRABI [EMAIL PROTECTED] wrote: Dear All, I try to compile OpenOffice and the compilation failed with that message : /usr/bin/ar rv uconvmsg/libuconvmsg.a uconvmsg/uconvmsg_dat.o /usr/bin/ar: création de uconvmsg/libuconvmsg.a a - uconvmsg/uconvmsg_dat.o Inconsistency detected by ld.so: dl-minimal.c: 137: realloc: Assertion `new == ptr' failed! This is the second time [1] in recent weeks this has come up. For some reason, the first time I wasn't able to find [2]. -Richard [1] http://www.mail-archive.com/gentoo-user%40lists.gentoo.org/msg42319.html [2] http://bugs.gentoo.org/show_bug.cgi?id=130837 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Problem with compiling OpenOffice 2.0.3
From: Richard Fish [EMAIL PROTECTED] Subject: Re: [gentoo-user] Problem with compiling OpenOffice 2.0.3 Date: Wed, 13 Sep 2006 10:38:33 -0700 On 9/13/06, Xavier MOGHRABI [EMAIL PROTECTED] wrote: Dear All, I try to compile OpenOffice and the compilation failed with that message : /usr/bin/ar rv uconvmsg/libuconvmsg.a uconvmsg/uconvmsg_dat.o /usr/bin/ar: création de uconvmsg/libuconvmsg.a a - uconvmsg/uconvmsg_dat.o Inconsistency detected by ld.so: dl-minimal.c: 137: realloc: Assertion `new == ptr' failed! This is the second time [1] in recent weeks this has come up. For some reason, the first time I wasn't able to find [2]. -Richard [1] http://www.mail-archive.com/gentoo-user%40lists.gentoo.org/msg42319.html [2] http://bugs.gentoo.org/show_bug.cgi?id=130837 -- gentoo-user@gentoo.org mailing list Same here mcc -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
Bo Ørsted Andresen wrote: On Wednesday 13 September 2006 14:36, James wrote: The more minimized the global flags are, the more secure the server. Were I the only one who wasn't quite convinced by that statement? No. Stefan -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
On Wed, 13 Sep 2006 19:01:18 +0200, Bo Ørsted Andresen wrote: The more minimized the global flags are, the more secure the server. Were I the only one who wasn't quite convinced by that statement? If that means leaving GAPING_SECURITY_HOLE out of USE, then it holds some truth ;-) -- Neil Bothwick New Intel opcode #007 PUKE: Put unmeaningful keywords everywhere signature.asc Description: PGP signature
Re: [gentoo-user] Simplified apache2
Can one covert a non-hardended machine to use the hardended-profile, or
do you have to start from scratch?
Michael Crute wrote:
On 9/12/06, James [EMAIL PROTECTED] wrote:
I used 2006.1 livecd to install a pII machine. It's going
to become a (minimalistic) apache2 server. I just let the
installation
set the flags for the install so I have these flags currently:
snip
Those look a bit excessive for a minimalist machine. I would start
over ;-)
Some of these flag look questionable, such as the one with
underscores (kernel_linux userland_GNU) as I only found
information on them, where they are describe as 'undocumented
use flags'. What's up with these flags?
My understanding is that these are set in the profile and simply tell
portage that you are using Linux. I don't think there is any way
(short of profile hacking) to change them. So don't worry about it.
Where do I look to discern the minimal list of (necessary) system
flags that
must be kept? (I want to avoid negating any flags that are critical).
These are my proposed list of flags:
snip
Still a little excessive in my opinion. The approach that I would (do)
take is to put only the bare minimum use flags in make.conf and
override the rest on a per-package level in /etc/portage/package.use.
So can I just use this list, or do I have to include a -{flag} for
each one?
IS there simpler syntax to globally remove unwanted flags [-*], but,
not any
critical system flags? (Is this the same as just leaving the flag out
of the USE param. setting in make.conf?
-* will work but be careful it can break things if you don't know what
your doing.
Are there default system flag settings that I can safely remove?
Where is the list and how do I know which ones can be removed or
negated?
My (limited) understanding of flags are that the highest priority are
those set in /etc/portage/package.use, then /etc/make.conf then
the system default flags which may be located in several locations.
Is there any docs or listing of all of these location and details
on precedence?
http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=2chap=2
OK, my advice to you would be to start over with a hardened profile.
While hardened is not specifically required I highly recommend it if
this is just going to be a headless server machine.
You probably want to set your machine up with a similar USE= string in
make.conf
USE=-* hardened pic ncurses ssl crypt berkdb tcpd pam perl python
readline
I believe that is the bare minimum if you use -*. Now you can compile
your system and you have a blank slate to start working with. As you
start emerging packages just make sure you use the -pv flags for
emerge and check out the available use flags and add the ones you want
to /etc/portage/package.use. Here is an example of my package.use line
for apache2
net-www/apache mpm-prefork threads
This setup works smashingly for me on my production servers by YMMV.
Best of luck.
-Mike
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
David Grant wrote: On 9/13/06, *Bo Ørsted Andresen* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote: On Wednesday 13 September 2006 18:24, David Grant wrote: revdep-rebuild keeps wanting to re-emerge gcc-4.1.1 over and over again. Is anyone else seeing this? So why don't you show us the output of # revdep-rebuild -i -vp ? That way we would be able to give a qualified answer instead of guessing... I didn't include it originally because it is long. :-) I've attached it, hopefully this mailing list accepts attachments. If it doesn't I will resend with a shorterned version. Thanks for the link to the bug. -- David Grant http://www.davidgrant.ca Hi, The same Bug Richard spoke of. Three solutions i'm ware of: remove java USE flag from GCC recompile; edit toolchain.eclass in an overlay as pointed in the Bug and recompile or create symlinks to point to the true .la files (easiest of all). PS:i've done the second option, waiting for new toolchain.eclass HTH.Rumen smime.p7s Description: S/MIME Cryptographic Signature
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On Wednesday 13 September 2006 19:53, David Grant wrote: I didn't include it originally because it is long. :-) If you think it's too long compress it. If you still think it's too long don't send the mail at all... Just omitting it leaving us to guess at the issues is simply pointless. I've attached it, hopefully this mailing list accepts attachments. If it doesn't I will resend with a shorterned version. Thanks for the link to the bug. Well, what have you actually done with this? I mean it tells you to recompile all of the following: # emerge -va1 \ =sys-devel/gcc-4.1.1 \ =dev-libs/libpqxx-2.5.1\ =media-libs/xine-lib-1.1.2-r2 \ =media-plugins/libvisual-plugins-0.2.0 \ =media-video/vlc-0.8.5-r5 Did you not try to recompile them? Or did they all fail for you? Or did you just give up when libpqxx failed? You know nothing prevents you from keywording libpqxx-2.5.5 (don't you know how to do that?) or running: # emerge --skipfirst to compile the rest...? In short: If you want help you really should provide more info about what you have done and where you didn't know what to do...! -- Bo Andresen pgpyuJJVqy5Xo.pgp Description: PGP signature
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: to compile the rest...? In short: If you want help you really should provide more info about what you have done and where you didn't know what to do...! In addition to what Bo said, you have a lot of broken .la files. More than likely, these are files that fix_libtool_files.sh modified, so portage didn't remove them when the associated packages got upgraded or removed. So for each of the broken .la files, you should do a equery belongs foo.la. If nothing owns the file, remove it. -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: On Wednesday 13 September 2006 14:36, James wrote: The more minimized the global flags are, the more secure the server. Were I the only one who wasn't quite convinced by that statement? No... I think it is a little rush to state something like that... -- Daniel da Veiga Computer Operator - RS - Brazil -BEGIN GEEK CODE BLOCK- Version: 3.1 GCM/IT/P/O d-? s:- a? C++$ UBLA++ P+ L++ E--- W+++$ N o+ K- w O M- V- PS PE Y PGP- t+ 5 X+++ R+* tv b+ DI+++ D+ G+ e h+ r+ y++ --END GEEK CODE BLOCK-- -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Purging Gnome from the system
On Wednesday 13 September 2006 05:27, Dale wrote: Mick wrote: With these USE flags in place I do not have gnome on my system: -eds -esd -gnome -gtk Proof: # equery l gnome [ Searching for package 'gnome' in all categories among: ] * installed packages # Cool. I need to mark this one important so I can find it again. Do you notice anything missing like in OOo, Mozilla/Seamonkey or anything like that?? Nothing that is apparent from the small exposure that I have had to the Gnome DE (mostly on FreeBSD). If you prefer the GTK2 icons on OOo instead of KDE you can specify that in your .bashrc and OOo will be launched with the desired GUI. PS. I have not used Seamonkey so far, but Mozilla suite and FF look and behave identically (both with source compiles and bin ebuilds). -- Regards, Mick pgptV2N0SNENB.pgp Description: PGP signature
Re: [gentoo-user] [OT] print to printer on winxp via cups samba smb
On Wednesday 13 September 2006 17:59, [EMAIL PROTECTED] wrote: I need some help getting printing to work from my gentoo machine to an HP 7410xi (one of those multi-function jobs). No one is responding to a similar query on the cups list. So bringing it here although it may be a little off topic. I'm afraid I can't answer your question directly because I do not use Samba, but if you search this list you'll find an answer as to how to set your WinXP box as a server for *nix machines and share the printer that way. The Samba interface adds one more layer of complexity and probability for error. Make sure you only allow the specific Linux box IP address to connect to your WinXP box (set it up in the WinXP firewall). If you get no Samba specific replies and you can't find the previous post then ask again and I will spend sometime searching for it. -- Regards, Mick pgpHottY4OWWo.pgp Description: PGP signature
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
David Grant wrote: I didn't include it originally because it is long. :-) You should have bzipped it: your mail would have been a tenth of its size. When sending logs to a mailing list, zip them, only when attaching them to bugs in Bugzilla leave them unzipped. Benno -- Cetere mi opinias ke ne ĉio tradukenda estas. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] print to printer on winxp via cups samba smb
On 9/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: State Stopped I'm guessing that this is your problem. What does lpc status report? Does cupsenable chub-print help? -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: Well, what have you actually done with this? I mean it tells you to recompileall of the following:# emerge -va1\=sys-devel/gcc-4.1.1 \=dev-libs/libpqxx- 2.5.1\=media-libs/xine-lib-1.1.2-r2\=media-plugins/libvisual-plugins-0.2.0 \=media-video/vlc-0.8.5-r5Did you not try to recompile them? Or did they all fail for you? Or did you just give up when libpqxx failed? You know nothing prevents you fromkeywording libpqxx-2.5.5 (don't you know how to do that?) or running:# emerge --skipfirstYeah I know how to do all these things. I have done emerge -1 gcc (still shows up in revdep-rebuild) after emerging, I have rebuilt vlc and libvisual-plugins and xine-lib, libpgqxx does indeed fail. -- David Granthttp://www.davidgrant.ca
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On 9/13/06, Richard Fish [EMAIL PROTECTED] wrote: On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: to compile the rest...? In short: If you want help you really should provide more info about what you have done and where you didn't know what to do...! In addition to what Bo said, you have a lot of broken .la files.Morethan likely, these are files that fix_libtool_files.sh modified, soportage didn't remove them when the associated packages got upgraded or removed.So for each of the broken .la files, you should do a equery belongsfoo.la.If nothing owns the file, remove it.Looks like most of these are not owned by anything. I'll start deleting them. -- David Granthttp://www.davidgrant.ca
[gentoo-user] Re: Simplified apache2
On Wednesday 13 September 2006 21:13, Daniel da Veiga wrote: On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: On Wednesday 13 September 2006 14:36, James wrote: The more minimized the global flags are, the more secure the server. Were I the only one who wasn't quite convinced by that statement? No... I think it is a little rush to state something like that... There is a little truth in it. If you decrease the amount of packages installed, you also decrease the amount of packages installed that contain a security hole. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Installing 2006.1 from the livedvd - is it really possible?
On Wednesday 13 September 2006 03:49, Helmut Jarausch wrote: Hi, GenToo (at least the 2006.1 livedvd) is driving me nuts. I have tried several installations (more than 6) and they all fail by e.g. EmergePackageError :FATAL: emerge: Could not emerge mail-mta/ssmtp it cannot build a kernel (just copy the kernel on the dvd) it cannot build EmergePackageError :FATAL: emerge: Could not emerge media-libs/xine-lib-1.1.2-r2 Furthermore I cannot enter the URIs for the ftp-proxie and the rsync-proxie. The fields are too short. I have been using the command line installer in 'advanced' mode. So, what can I do? If you have broadband, then fall back to installing gentoo manually. I have had nothing but problems, similar t yours, with both the live-cd and live-dvd. Neither one finished to completion, however, the failed live-dvd install left just enough system on the target harddrive that I was able to finish up the install manually. Who ever thought these live disks were good for installing from... well... it isn't. That said, the Knoppix Live-DVD installs flawlessly to hard drives. Jerry. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Emacs - (I think [the]) Fonts [are] Broken
On Wednesday 13 September 2006 09:53, Robert Cernansky wrote: On Mon, 11 Sep 2006 16:24:11 -0700 Lord Sauron [EMAIL PROTECTED] wrote: quoth the Lord Sauron: [...] What should I do to fix Emacs? I hear it's a very powerful and [...] Much more confused than before. I get why I'd want to use unicode, but the instructions on how to do that were outdated or for a setup other than mine and they threw me off real quickly. Did you try ran emacs in X also? Is there the same behaviour? In X you can pres Shift + Left mouse click to select fonts. You can also play with locale setting in your system. For example try to start emacs with default/none locale: $ LC_ALL=C emacs To set emacs to use UTF-8 put this into your ~/.emacs file: (setq locale-coding-system 'utf-8) (set-terminal-coding-system 'utf-8) (set-keyboard-coding-system 'utf-8) (set-selection-coding-system 'utf-8) (set-clipboard-coding-system 'utf-8) (prefer-coding-system 'utf-8) (modify-coding-system-alist 'file 'utf-8) (setq process-coding-system-alist '((.* . utf-8))) (setq x-select-request-type '(UTF8_STRING COMPOUND_TEXT TEXT STRING)) But it should not be necessary if you use some UTF-8 locale in your system (emacs will use UTF-8 by default then). You can also try to change the 'utf-8' string to some other (non utf-8) enconding, for example iso-8859-1. Try to emerge some basic X.Org fonts, for example media-fonts/font-misc-misc. Also emerge media-fonts/intlfonts. I tried that, and emerge --pretend promptly told me that xorg 6.9 or something was blocking virtually everything x-related. It was really weird. It's also blocking xemacs (the package I think will fix this mess). This probably means that I have to go find and fix my package masks. The difficulty there is that there's a ton of junk I'm still using (like Eclipse 3.1) which is masked. I think this is going to end up being a battle between me and portage to sort out my package masks. Is there a untampered copy of packages.mask (or whatever it's called) that I can use to compare with mine to try and fix this mess? -- http://lordsauronthegreat.googlepages.com/ pgp35ZssuPswn.pgp Description: PGP signature
Re: [gentoo-user] coax spdif sound output on Intel 82801EB/ER audio
Iain Buchanan wrote: [...] mmm. I'm rather fuzzy on exactly how it's done, but what you're going to have to do is set up a ~/.asoundrc that sets spdif as the default output device. http://alsa.opensrc.org/intel8x0 http://gentoo-wiki.com/HOWTO_Dolby_Digital_Out_(AC3,_SPDIF) HTH. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] firewall minimized make.conf flags
james wrote: USE= -* hardened pic ncurses ssl acl crypt berkdb tcpd pam perl pcre python readline bzip2 zlib apm krbr kerberos nptl nptlonly lm_sensors syslog Try tossing logrotate in there for kicks. It's an absolute joy, especially on systems that you don't intend to interact directly with too often. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] coax spdif sound output on Intel 82801EB/ER audio
On Wed, 2006-09-13 at 15:53 -0700, Ryan Tandy wrote: Iain Buchanan wrote: [...] mmm. I'm rather fuzzy on exactly how it's done, but what you're going to have to do is set up a ~/.asoundrc that sets spdif as the default output device. really? It used to just go to both outputs (spdif and headphone jack) and I used to just move a volume slider depending on which one I wanted... I wonder why it's changed... I don't particularly want to send everything over the spdif by default, cause a lot of the time I don't have the laptop plugged into the hifi. Thanks for the links though! cya, -- Iain Buchanan iaindb at netspace dot net dot au There are worse things in life than death. Have you ever spent an evening with an insurance salesman? -- Woody Allen -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] postfix with openLDAP
bijayant kumar wrote: Hi, I am trying to configure postfix and courier-imap with openLDAP. But i am not getting any good documents on this topic. I have posted my problem 4 days before also. But didnt get any reply. Please please help me. I will be very thankful to you. For a postfix problem, rather go here http://www.postfix.org/documentation.html and for LDP in particular try this http://www.postfix.org/LDAP_README.html. Send instant messages to your online friends http://uk.messenger.yahoo.com -- gentoo-user@gentoo.org mailing list
[gentoo-user] kdm xorg-7.1 problem
After finally upgrading to xorg-7.1 cause I finally decided nvidia-drivers were working well enough, now my kde-3.5.4 that was already installed won't give me a start new user new login for multiple logins. What did I miss? Mike -- Michael W. Holdeman Powered by Gentoo Linux www.gentoo.org | Kernel 2.6.15-ck2 | VMWare Workstation 5.5.1 vmware.com | Win4LinPro 6.1.1-03 win4lin.com | | -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: [OT] print to printer on winxp via cups samba smb
Mick [EMAIL PROTECTED] writes: I'm afraid I can't answer your question directly because I do not use Samba, but if you search this list you'll find an answer as to how to set your WinXP box as a server for *nix machines and share the printer that way. The Samba interface adds one more layer of complexity and probability for error. Make sure you only allow the specific Linux box IP address to connect to your WinXP box (set it up in the WinXP firewall). I did find a previous post of yours alluding to an earlier post but that post of yours does give sort of brief howto. http://thread.gmane.org/gmane.linux.gentoo.user/166004/focus=166061 I'm going to pursue samba solutions first since I have done that several times and it mostly just worked for me when I made a few config settings. I've never really had trouble getting printing to work thru samba until now and have setup it up at least 5-6 times. The method you propose in the URL above does sound interesting and I may give it a try just to compare it to the samba approach. Thanks -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: firewall minimized make.conf flags
Ryan Tandy tarpman at gmail.com writes: USE= -* hardened pic ncurses ssl acl crypt berkdb tcpd pam perl pcre python readline bzip2 zlib apm krbr kerberos nptl nptlonly lm_sensors syslog Try tossing logrotate in there for kicks. It's an absolute joy, especially on systems that you don't intend to interact directly with too often. YES, I've been meaning to come up to speed on using logrotate and some scripts I have found Thanks! James -- gentoo-user@gentoo.org mailing list
[gentoo-user] SOLVED Re: installing an amd k6
Dirk Heinrichs ext-dirk.heinrichs at nokia.com writes: Hello Dirk, I posted a resonse early today, but, it never made it.. Are you suggesting I use this in my make.conf file? CFLAGS=-O2 -march=i586 -pipe -march=k6 (or even k6-2) is OK, as listed above. I used this: CFLAGS=-O2 -march=i586 -pipe Because when I untarred the i586 file it was what was written by default into the make.conf file by: stage3-i586-2006.0.tar.bz2 When booting I gave these options gentoo-nofb acpi=no nosmp For this type of machine, I'd go with the stage 3. It might recompile some of the packages, depending on your use flags upon next emerge -DNuvp world/system, baut that's still faster than bootstrapping from stage 1. I did use the latest portage I could find: portage-latest.tar.bz2 The K6 is now compiling a hardened kernel. Thanks to all, as it was a lot of little things that helped! James -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: [OT] print to printer on winxp via cups samba smb
Richard Fish [EMAIL PROTECTED] writes: On 9/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: State Stopped I'm guessing that this is your problem. What does lpc status report? Does cupsenable chub-print help? lpc status first reported: chub-print: printer is on device 'smb' speed -1 queuing is enabled printing is disabled no entries daemon present As you see, `disabled' but why would that be? My config doesn't mention enable/disable and in the past I simply ran the cupsd service or not to turn off/on. Something new? So I thought it might just start working after: cupsenable chub-print But I guess it is not going to. After running `cupsenable chub-print' and attempting to print a job the log output in error_log contains thousands of lines of output from just one attempt. Among them grepping for the Error lines I see stuff indicating a connection was not allowed. That output in a moment. First let me show that a connection is definately in working order: smbclient -L //chub -Ureader |grep print WARNING: The printer admin option is deprecated Password: Domain=[CHUB] OS=[Windows 5.1] Server=[Windows 2000 LAN Manager] Domain=[CHUB] OS=[Windows 5.1] Server=[Windows 2000 LAN Manager] print$ Disk Printer Drivers chub-print Printer HP Officejet 7400 series smbclient //chub/chub-print -U reader WARNING: The printer admin option is deprecated Password: Domain=[CHUB] OS=[Windows 5.1] Server=[Windows 2000 LAN Manager] smb: \ The share is there and can be accessed with smbclient After attempting to print, a few minutes pass and a new look with lpc status: Shows printing is now disabled again. Partial Error_log output follows: grep ^E /var/log/cups/error_log (Note: Many in-between lines are skipped and those printed are wrapped for mail) E [13/Sep/2006:19:44:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:44:53 -0500] [Job 24] No ticket cache found for userid=1000 E [13/Sep/2006:19:44:53 -0500] [Job 24] Can not get the ticket cache for reader E [13/Sep/2006:19:44:53 -0500] [Job 24] Session setup failed: NT_STATUS_LOGON_FAILURE E [13/Sep/2006:19:44:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:44:53 -0500] [Job 24] Unable to connect to CIFS host, will retry in 60 seconds... E [13/Sep/2006:19:45:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:45:53 -0500] [Job 24] No ticket cache found for userid=1000 E [13/Sep/2006:19:45:53 -0500] [Job 24] Can not get the ticket cache for reader E [13/Sep/2006:19:45:53 -0500] [Job 24] Session setup failed: NT_STATUS_LOGON_FAILURE E [13/Sep/2006:19:45:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:45:53 -0500] [Job 24] Unable to connect to CIFS host, will retry in 60 seconds... E [13/Sep/2006:19:46:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:46:53 -0500] [Job 24] No ticket cache found for userid=1000 E [13/Sep/2006:19:46:53 -0500] [Job 24] Can not get the ticket cache for reader E [13/Sep/2006:19:46:53 -0500] [Job 24] Session setup failed: NT_STATUS_LOGON_FAILURE E [13/Sep/2006:19:46:53 -0500] [Job 24] Tree connect failed (NT_STATUS_BAD_NETWORK_NAME) E [13/Sep/2006:19:46:53 -0500] [Job 24] Unable to connect to CIFS host, will retry in 60 seconds... E [13/Sep/2006:19:47:53 -0500] [Job 24] Unable to connect to CIFS host after (tried 3 times) E [13/Sep/2006:19:47:53 -0500] PID 15252 (/usr/libexec/cups/backend/smb) stopped with status 1! E [13/Sep/2006:19:47:54 -0500] PID 15251 (/usr/libexec/cups/filter/foomatic-rip) stopped with status 9! -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild wants to re-emerge gcc-4.1.1 endlessly
On 9/13/06, David Grant [EMAIL PROTECTED] wrote: On 9/13/06, Richard Fish [EMAIL PROTECTED] wrote: On 9/13/06, Bo Ørsted Andresen [EMAIL PROTECTED] wrote: to compile the rest...? In short: If you want help you really should provide more info about what you have done and where you didn't know what to do...! In addition to what Bo said, you have a lot of broken .la files.Morethan likely, these are files that fix_libtool_files.sh modified, soportage didn't remove them when the associated packages got upgraded or removed.So for each of the broken .la files, you should do a equery belongsfoo.la.If nothing owns the file, remove it. Looks like most of these are not owned by anything. I'll start deleting them. Doing that and also re-emerging gcc without the gcj use flag seemed to fix everything. Now my revdep-rebuild runs clean. And I accidentally deleted lipqxx so that doesn't show up either. :-) -- David Granthttp://www.davidgrant.ca
Re: [gentoo-user] Re: [OT] print to printer on winxp via cups samba smb
On 9/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: First let me show that a connection is definately in working order: Hmm, do you get something similar if you do: DEVICE_URI=smb://reader:[EMAIL PROTECTED]/chub-print export DEVICE_URI /usr/bin/smbspool 0 reader test 1 0 /etc/fstab -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Simplified apache2
Even this is a bit more bloated than it needs to be. I have never used 'tcpd' Doesn't tcpd add security, thus suggesting it's use for a more secure system? Thanks, Brian
Re: [gentoo-user] Simplified apache2
[EMAIL PROTECTED] wrote: Even this is a bit more bloated than it needs to be. I have never used 'tcpd' Doesn't tcpd add security, thus suggesting it's use for a more secure system? Thanks, Brian It can. I haven't yet had a need to implement host-based security, and if I ever do, I plan to merge it into my existing iptables stuff. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] kdm xorg-7.1 problem
I didn't get what your problem is, but at home, after emerge -e world, GDM decided not to work. upon further investigation, nvidia did not load. so I had to recompile my kernel and then recompile nvidia-drivers to make it work again. I hope it helps. regards On 9/13/06, Michael W. Holdeman [EMAIL PROTECTED] wrote: After finally upgrading to xorg-7.1 cause I finally decided nvidia-drivers were working well enough, now my kde-3.5.4 that was already installed won't give me a start new user new login for multiple logins. What did I miss? Mike -- Michael W. Holdeman Powered by Gentoo Linux www.gentoo.org | Kernel 2.6.15-ck2 | VMWare Workstation 5.5.1 vmware.com | Win4LinPro 6.1.1-03 win4lin.com | | -- gentoo-user@gentoo.org mailing list -- gentoo-user@gentoo.org mailing list
[gentoo-user] prioritzed Ethernet traffic?
Hi, I'm just wondering how one goes about creating higher/lower priority Ethernet traffic on a home network. Is it possible? Difficult? When I'm in my office working on my Gentoo box I have two main things I'm doing across a wireless network: 1) Watching MythTV. This traffic is going wireless to the router and then wired to the backend server. I'd like this traffic to be high priority. 2) Browsing the web, using GMail, etc. In general I'd like this traffic to be low priority. With a default network setup, when I change web pages to something that sends in a lot of data I get delays in MythTV making it less compelling to watch. Is there something I can do about this? Thanks, Mark -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: Simplified apache2
Brian Davis bridavis at comcast.net writes: Can one covert a non-hardended machine to use the hardended-profile, or do you have to start from scratch? Hello Brian, The short answer is YES. The correct answer is you have to read quite a lot (I'm in the middle of that) and decide which 'path/technology' you want to follow. Here's docs you should start looking at: http://www.gentoo.org/proj/en/hardened/primer.xml http://www.gentoo.org/proj/en/hardened/ I choose 'SElinux' as the path to follow for me that makes most sense. Since the NSA was the prime motivator, it's an easy path to convince my clients to follow. Although SElinux is not a complete solution, other complementary software combined with SElinux does provide for a complete (security) solution, almost. http://www.gentoo.org/proj/en/hardened/selinux/ http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml?part=2 hth, http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml?part=2 James -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Simplified apache2
Good point, I suppose iptables can do pretty much everything tcp_wrappers can do. -- Original message -- From: Ryan Tandy [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:Even this is a bit more bloated than it needs to be. I have never used'tcpd' Doesn't tcpd add security, thus suggesting it's use for a more secure system? Thanks, Brian It can. I haven't yet had a need to implement host-based security, and if I ever do, I plan to merge it into my existing iptables stuff. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Simplified apache2
Thanks James! -- Original message -- From: James [EMAIL PROTECTED] Brian Davis comcast.net writes: Can one covert a non-hardended machine to use the hardended-profile, or do you have to start from scratch?Hello Brian, The short answer is YES. The correct answer is you have to read quite a lot (I'm in the middle of that) and decide which 'path/technology' you want to follow. Here's docs you should start looking at: http://www.gentoo.org/proj/en/hardened/primer.xml http://www.gentoo.org/proj/en/hardened/ I choose 'SElinux' as the path to follow for me that makes most sense. Since the NSA was the prime motivator, it's an easy path to convince my clients ! ; to follow. Although SElinux is not a complete solution, other complementary software combined with SElinux does provide for a complete (security) solution, almost.http://www.gentoo.org/proj/en/hardened/selinux/ http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml?part=2 hth, http://www.gentoo.org/proj/en/hardened/selinux/selinux-handbook.xml?part=2 James -- gentoo-user@gentoo.org mailing list
[gentoo-user] Update or Install again? Testing vs. stable?
I have questions (for one of my machines, running ~amd64) whether it is worth it to update/upgrade or whether to reinstall from scratch. Another machine that melted down three months ago is being rebuilt with a new motherboard, and new technology; I am installing from scratch for obvious reasons in this case; but I would like to solicit comments about my plan to use only a straight stable version (with single packages unmasked only). I'll try to be brief, but I beg in advance the indulgence of the list for an overlength posting. And I apologize for going over territory once again that has been covered elsewhere. My ~amd64 machine has an extremely large number of packages, many of them for multimedia viewing, DVD burning, editing DVDs, graphics, latex---a large variety of applications many of them which can be used simultaneously at any given time. The system is amazing, and I suspect that any other distro would fall over under this diversity of processes. In particular, I have had good success editing videos with avidemux---but only with the most up to date overlay! The machine has gotten tangled into a knot that seems to be almost impossible to get all the way up to date at one time. At least for me. I received amazingly clear and lucid advice on how to get up to date, upgrade the compiler, and rebuild, revdep-rebuild, etc., etc.; but with my unreliable connectivity over the summer, I was unable to get the machine to an absolutely clean state to start that process. It took a tremendous amount of time, months, to setup all of these packages; maintenance of the system also demands not only a fast network connection, but also a lot of time. It's not going to be easy to reinstall this system in it's current form. Perhaps it's necessary, however. The usual advice about updating is something like it isn't necessary; there is no such thing as versions with Gentoo. But with at least three major upgrade bugs to overcome (xorg-x11 / nvidia-drivers; compiler upgrades; revdep-rebuild failures; changes in network administration, to name a few issues) I wonder whether it isn't just as easy to install afresh. Because my time is taken up with teaching and science, I don't have unlimited time for system maintenance. I need to have access to recent updates of some software packages, like avidemux, which only in recent incarnations has been able to deal with the avi files my video capture device generates. So I am thinking to run a stable (amd64 or x86) version, and only install the necessary packages as ~arch. I want to ask the intelligentsia how much difference will that make. I am afraid I have not been as useful to the group as I might have been at reporting bugs, but I found a large number of packages able to run amd64. The same goes for the newly cobbled together machine with an amd64 processor, and a gateway laptop that has been running gentoo but I've been running on Ubuntu because of trouble understanding the docs on how to use WPA authentication on the school wireless network. I am going to either reinstall gentoo or (if I haven't buggered it yet) upgrade, so the same questions apply. I might add that my experience with Ubuntu, after running gentoo only for 9 months, has been interesting: Ubuntu is pretty amazing, but it has nowhere near the polish of Gentoo. The newest betas are really amazing---edgy eft knot2, and the new Gnome very interesting; however, at almost every turn the Gentoo packages, out of the box with fairly mundane USE flags, are MUCH more serviceable, the details have been taken care of, and I believe Gentoo to be much more stable. TIA, Alan Davis -- Alan Davis, Kagman High School, Saipan [EMAIL PROTECTED] 1-670-256-2043 I consider that the golden rule requires that if I like a program I must share it with other people who like it. Richard Stallman -- gentoo-user@gentoo.org mailing list
[gentoo-user] Update or Install again? Testing vs. stable?
I have questions (for one of my machines, running ~amd64) whether it is worth it to update/upgrade or whether to reinstall from scratch. Another machine that melted down three months ago is being rebuilt with a new motherboard, and new technology; I am installing from scratch for obvious reasons in this case; but I would like to solicit comments about my proposal to use only a straight stable version. I'll try to be brief, but I beg in advance the indulgence of the list for an overlength posting. And I apologize for going over territory once again that has been covered elsewhere. My ~amd64 machine has an extremely large number of packages, many of them for multimedia viewing, dvd burning, editing dvds, graphics, latex---a large variety of applications many of them which can be used simultaneously at any given time. The system is amazing, and I suspect that any other distro would fall over under this diversity of processes. In particular, I have had good success editing videos with avidemux---but only with the most up to date overlay! The machine has gotten tangled into a knot that seems to be almost impossible to get all the way up to date at one time. At least for me. I received amazingly clear and lucid advice on how to get up to date, upgrade the compiler, and rebuild, revdep-rebuild, etc., etc.; but with my unreliable connectivity over the summer, I was unable to get the machine to the absolutely clean state to start that process. It took a tremendous about of time to setup all of these packages; maintainence of the system also demands not only a fast network connection, but also a lot of time. It's not going to be easy to reinstall this system in it's current form. Perhaps it's necessary, however. The usual advice about updating is something like it isn't necessary; there is no such thing as versions with Gentoo. But with at least three major upgrade bugs to overcome (xorg-x11 / nvidia-drivers; compiler upgrades; revdep-rebuild failures; changes in network administration, to name a few issues) I wonder whether it isn't just as easy to install afresh. Because my time is taken up with teaching and leftover time should be spent on science, I don't have unlimited time for system maintainence Because one of my machines melted down in June, I am preparing to reinstall Gentoo. My connectivity has been sporadic over the summer; maybe it's improving; we might get a DSL soon at home. My home machine was installed and maintained almost exclusively over a dialup connection. -- Alan Davis, Kagman High School, Saipan [EMAIL PROTECTED] 1-670-256-2043 I consider that the golden rule requires that if I like a program I must share it with other people who like it. Richard Stallman -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: [OT] print to printer on winxp via cups samba smb
On 9/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hmm, do you get something similar if you do: DEVICE_URI=smb://reader:[EMAIL PROTECTED]/chub-print export DEVICE_URI /usr/bin/smbspool 0 reader test 1 0 /etc/fstab Well, that tried to print something at least. Somekind of stepping problem seems to have prevented a real print. I got 2 pages. one page with the first and second line of fstab. The second line was indented about half a page. Then a blank page. Ok, that's good. The staircasing is easily explained/fixed, but not really important. Mostly we are just interested in the fact that that command did connect and try and print. First, do an ls -l /usr/libexec/cups/backend/smb It should be a symbolic link pointing to /usr/bin/smbspool. If it is, then we need to see what the differences are between running that command directly vs the way cups runs it. We can use some shell script trickery to do this: 1. mv /usr/bin/smbspool /usr/bin/smbspool.bin 2. $EDITOR /usr/bin/smbspool Make a new script that contains: -- #!/bin/bash tmpfile=/tmp/smbspool.$$ echo $0:$1:$2:$3:$4:$5:$6:$7 $tmpfile env $tmpfile # this part is just a guess.. export DEVICE_URI=$0 /usr/bin/smbspool $1 $2 $3 $4 $5 $6 $tmpfile -- 3. chmod 755 /usr/bin/smbspool Try printing again through cups. After the failure, you should have one or more /tmp/smbspool.* files. Post one of those here, although you may want to inspect it first and mask out anything that reveals your password! -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Emacs - (I think [the]) Fonts [are] Broken
quoth the Lord Sauron: I tried that, and emerge --pretend promptly told me that xorg 6.9 or something was blocking virtually everything x-related. It was really weird. Not weird, it wants you to install Xorg 7.0 which is modular. Ie: you must unmerge Xorg 6.x altogether and then reinstall. xorg-x11 is now a virtual package which brings in all the now modular parts that you need. See: http://www.gentoo.org/proj/en/desktop/x/x11/modular-x-howto.xml -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@gentoo.org mailing list
Re: [gentoo-user] Update or Install again? Testing vs. stable?
On 9/13/06, Alan E. Davis [EMAIL PROTECTED] wrote: It took a tremendous amount of time, months, to setup all of these packages; maintenance of the system also demands not only a fast network connection, but also a lot of time. It's not going to be easy to reinstall this system in it's current form. Perhaps it's necessary, however. I think at this point you have two options: 1. Reinstall the system using the curernt 2006.1 AMD media. Advantages: - you can use the GRP binary packages to do a quick update. - you avoid recompiling everything for the gcc update - updating to current should be a fairly small change Disadvantages: - You need to be sure to backup your configuration files, or you may lose something that took a long time to get right. - If you have/make a lot of changes to USE, you may have a lot of things that need to be rebuilt after you are done with the install. If you do this, be sure to backup your configuration files, especially /etc/portage, /etc/make.conf. Plus you probably want to keep /var/lib/portage/world as well. 2. Remove ~amd64 from keywords, and basically follow the gcc upgrade guide, since you probably need to udpate to gcc 4.1 anyway. Advantages: - It's relatively easy to see what things are going to be downgraded, so you can decide what needs to be added to package.keywords. - Your configuration files will be protected by CONFIG_PROTECT. Disadvantages: - You end up doing an emerge -e world, which is going to take quite a while to execute. From a general perspective on running some ~arch packages on an otherwise stable system, how successful that is usually depends on how many ~arch packages you have. ~arch packages tend to depend on other ~arch things, so there is a viral effect that leads some people to give up and use ~arch for the entire system. Using the ~cate-gory/package-ver.s.ion syntax can help here, as it only allows the ~arch keyword for specific versions or -r releases. HTH, -Richard -- gentoo-user@gentoo.org mailing list
