[gentoo-user] bash-2.05b-r9 emerge fails
portage wants to build build bash-2.05b-r9: current = -r7. Unfortunately this it fails to link: print_cmd.o(.text+0x1739): In function __builtin_va_start' print_cmd.o(.text+0x1927): In function __builtin_va_start' error.o(.text+0x108): In function __builtin_va_start' error.o(.text+0x1c3): In function __builtin_va_start' error.o(.text+0x243): In function __builtin_va_start' etc. # gcc --version gcc (GCC) 3.3.2 20031218 (Gentoo Linux 3.3.2-r5, propolice-3.3-7) Copyright (C) 2003 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Any else seen this? Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] You cann't get there from here
Look at the following: lws root # ACCEPT_KEYWORDS='~x86' emerge -pv subversion These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild N] net-www/apache-2.0.48-r4 +berkdb +gdbm -ldap [blocks B ] =sys-devel/autoconf-2.58* (from pkg dev-util/subversion-0.37.0) [ebuild N] net-misc/neon-0.24.4 +ssl +zlib -expat [ebuild N] dev-util/subversion-0.37.0 +ssl +apache2 +berkdb +python -emacs Note that autoconf blocks neon. If we unmerge autoconf (lots fo stuff breaks...) including our ability to build apache. Of course I can always build apache then unmerge autoconf, then build neon and subversions, then reemerge autoconf? (maybe), it seems twisted. Any comments? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] You cann't get there from here
On Mon, 2004-02-02 at 20:45, Lincoln A. Baxter wrote: Look at the following: lws root # ACCEPT_KEYWORDS='~x86' emerge -pv subversion These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild N] net-www/apache-2.0.48-r4 +berkdb +gdbm -ldap [blocks B ] =sys-devel/autoconf-2.58* (from pkg dev-util/subversion-0.37.0) [ebuild N] net-misc/neon-0.24.4 +ssl +zlib -expat [ebuild N] dev-util/subversion-0.37.0 +ssl +apache2 +berkdb +python -emacs Note that autoconf blocks neon. If we unmerge autoconf (lots fo stuff breaks...) including our ability to build apache. Of course I can always build apache then unmerge autoconf, then build neon and subversions, then reemerge autoconf? (maybe), it seems twisted. Any comments? To answer my own post... Downgrading autoconf to 1.57* seems to work. Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] Emerge bootsplash syntax error ??
Has anyone seen this: Calculating world dependencies ...done! emerge (1 of 70) media-gfx/bootsplash-0.6-r3 to / /usr/sbin/ebuild.sh: /usr/portage/media-gfx/bootsplash/bootsplash-0.6-r3.ebuild: line 121: syntax error: unexpected end of file !!! ERROR: media-gfx/bootsplash-0.6-r3 failed. !!! Function , Line 1238, Exitcode 1 !!! error sourcing ebuild -- [EMAIL PROTECTED] mailing list
[gentoo-user] New glibc and gnome 2.4.1 unmasked
Whenever there is a HUGE unmasking of 50-100 packages or more, I get nervous about doing and emerge -Uun world. I saw several messages go by earlier in the week about problems with the new glibc. Have these been resolved? Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] smbfs and 2.6.0 kernel
On Thu, 2004-01-15 at 21:44, Jens Mayer wrote: * On Thu, Jan 15, 2004 at 00:58:04 -0500, Lincoln A. Baxter wrote: I have managed to get everything working in 2.6.0 kernel, EXCEPT, for smbfs. It seems to hang... and eventually time out. Are you using NPTL (Native Posix Thread Library)? For what I do know, there was a known bug in this context triggered by Samba, which caused smbmounts to freeze quite frequently. Interesting question... How would I know? Cann't find it in the kernel configs, or by searching with emerge. Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] smbfs and 2.6.0 kernel
Hi, I have managed to get everything working in 2.6.0 kernel, EXCEPT, for smbfs. It seems to hang... and eventually time out. I can niether mount nor umount the filesystem Anybody got any ideas? Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] quicktime
I see we have quicktime libraries in portage, can any recommend viewers that use it. In particular I am interested in view quicktime videos on a website. If no mozilla plugin exists, I'd be happy with the standalone viewer. -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] java-config ebuild is broken in portage
lws log # emerge -pv java-config These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild U ] dev-java/java-config-1.2.2 [1.1.5] lws log # emerge java-config Calculating dependencies ...done! emerge (1 of 1) dev-java/java-config-1.2.2 to / Downloading http://gentoo.oregonstate.edu/distfiles/java-config-1.2.2.tar.bz2 --21:45:22-- http://gentoo.oregonstate.edu/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving gentoo.oregonstate.edu... 128.193.0.3 Connecting to gentoo.oregonstate.edu[128.193.0.3]:80... connected. HTTP request sent, awaiting response... 404 Not Found 21:45:22 ERROR 404: Not Found. Downloading http://distro.ibiblio.org/pub/Linux/distributions/gentoo/distfiles/java-config-1.2.2.tar.bz2 --21:45:22-- http://distro.ibiblio.org/pub/Linux/distributions/gentoo/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving distro.ibiblio.org... 152.2.210.109 Connecting to distro.ibiblio.org[152.2.210.109]:80... connected. HTTP request sent, awaiting response... 404 Not Found 21:45:22 ERROR 404: Not Found. Downloading http://gentoo.gnukorea.org/distfiles/java-config-1.2.2.tar.bz2 --21:45:22-- http://gentoo.gnukorea.org/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving gentoo.gnukorea.org... 211.202.2.240 Connecting to gentoo.gnukorea.org[211.202.2.240]:80... connected. HTTP request sent, awaiting response... 401 Authorization Required Authorization failed. Downloading http://gentoo.linux.no/distfiles/java-config-1.2.2.tar.bz2 --21:45:23-- http://gentoo.linux.no/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving gentoo.linux.no... 193.201.220.90 Connecting to gentoo.linux.no[193.201.220.90]:80... connected. HTTP request sent, awaiting response... 404 Not Found 21:45:23 ERROR 404: Not Found. Downloading http://mirrors.sunsite.dk/gentoo/distfiles/java-config-1.2.2.tar.bz2 --21:45:23-- http://mirrors.sunsite.dk/gentoo/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving mirrors.sunsite.dk... 130.225.247.92 Connecting to mirrors.sunsite.dk[130.225.247.92]:80... connected. HTTP request sent, awaiting response... 404 Not Found 21:45:24 ERROR 404: Not Found. Downloading http://gentoo.oregonstate.edu/distfiles/java-config-1.2.2.tar.bz2 --21:45:24-- http://gentoo.oregonstate.edu/distfiles/java-config-1.2.2.tar.bz2 = `/usr/portage/distfiles/java-config-1.2.2.tar.bz2' Resolving gentoo.oregonstate.edu... 128.193.0.3 Connecting to gentoo.oregonstate.edu[128.193.0.3]:80... connected. HTTP request sent, awaiting response... 404 Not Found 21:45:24 ERROR 404: Not Found. Downloading ftp://ftp.dale.ro/pub/mirrors/ftp.ibiblio.org/pub/Linux/distributions/gentoo/distfiles/java-config-1.2.2.tar.bz2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Spam from gentoo-lists
On Mon, 2004-01-05 at 12:42, Andrej Kacian wrote: (Mon, 05 Jan 2004 18:25:39 +0100) And Norbert Kamenicky [EMAIL PROTECTED] said: PS: Mostly I receive spam with these headers: Undelivered Mail Returned To Mailer (from Admin [EMAIL PROTECTED]) always contains EXE files Last Internet Upgrade (from Microsoft Corporation Security Support [EMAIL PROTECTED]) Current Internet Security Patch (from Commercial Partner [EMAIL PROTECTED]) Microsoft Critical Upgrade (from Program Security Center [EMAIL PROTECTED] ) I have received all of these... I just recently modified spamassassin config files to score MICROSOFT executables in mail messages as SPAM. No one should mail around M$ exe's these days. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] problem with portage and alsa-lib
On Sun, 2004-01-04 at 01:09, Glenn Johnson wrote: I am maintaining an ~x86 system and have alsa-lib-1.0.0_rc2 installed. When doing an 'emerge -Dup world', portage wants to downgrade alsa-lib to 0.9.8. What is strange is that it seems that alsa-lib itself is causing this. I did an 'emerge -C alsa-lib' to remove alsa-lib-1.0.0_rc2. Then upon doing 'emerge -Dup world' I get the following: ===begin output=== These are the packages that I would merge, in order: Calculating world dependencies ...done! [ebuild N] media-libs/alsa-lib-1.0.0_rc2 ===end output=== So then I do 'emerge alsa-lib', which installs alsa-lib-1.0.0_rc2 Then, after the installation of alsa-lib-1.0.0_rc2, I do the 'emerge -Dup world' again and get the following: ===begin output=== These are the packages that I would merge, in order: Calculating world dependencies ...done! [ebuild UD] media-libs/alsa-lib-0.9.8 [1.0.0_rc2] ===end output=== Are you using the -U (capital U) switch? which says, ONLY do UPGRADES (no downgrades). Hence: emerge -pvUun world emerge -Uun It never wants to down grade a package I have emerged with ACCEPT_KEYWORDS='~x86'. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] how long does it take to commit a patch to portage
It could be FOREVER! I am coming to the conclusion that the gentoo folks are not really very interested in user contributions. I think you have to be part of the club. I have had distinctly mixed experiences: 1.) Last May I contributed this: http://bugs.gentoo.org/show_bug.cgi?id=21580 It has yet to show up in portage. And not official comments are in the bug. Now I know there are MANY who think qconfirm is a bad idea, I don't even use myself any more, especially since spam assassin has gotten so much better. Even if one disagrees with using this, is this a reason to make it hard for others, when some user has taken the time to build a perfectly good ebuild, that makes it easier for the next guy?. That, it seems to me is DICTATING POLICY!. At the very least they should put a comment into the bug that says that AREN't going to accept it and WHY! 2.) About 1.5 months ago they broke the qmail pop3 server. (Atleast with its interaction with Evolution) BAD. I submitted a bug, which they DID pay attention too. I worked with the team who asked me to generate some evolution debug output. Which enabled ME to find what they broke, I submitted the fix, and it was fairly quickly incorporated into the package. See http://bugs.gentoo.org/show_bug.cgi?id=34646 for the full history. 3.) I have developed an ebuild for the perl interface to the Xerces-c XML parser. Which is a bit of a pain to build. The Gentoo guides for submitting ebuilds says that the perl herders need to approve such ebiulds. Well, they do not even tell you how to send mail to the perl herders. (No link, no nothing). So I have a perfectly good ebuild for perl's XML-Xerces module in my portage overlay dir, that I have not submitted, because of my experience with bug 21580, and my inability to find the perl herders. I am waiting to see if the Zynot folks (http://www.zynot.org/) produce anything usable, and whether they will be a little more accepting of user contributions. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] how long does it take to commit a patch to portage
On Fri, 2004-01-02 at 15:31, Ciaran McCreesh wrote: On Fri, 02 Jan 2004 15:11:40 -0500 Lincoln A. Baxter [EMAIL PROTECTED] wrote: [snip] | 1.) Last May I contributed this: | http://bugs.gentoo.org/show_bug.cgi?id=21580 | | It has yet to show up in portage. And not official comments are in the | bug. Now I know there are MANY who think qconfirm is a bad idea, I | don't even use myself any more, especially since spam assassin has | gotten so much better. Even if one disagrees with using this, is this | a reason to make it hard for others, when some user has taken the time | to build a perfectly good ebuild, that makes it easier for the next | guy?. That, it seems to me is DICTATING POLICY!. At the very least | they should put a comment into the bug that says that AREN't going to | accept it and WHY! That hasn't been rejected. It's just that no developer is willing to accept responsibility for it. Hmmm... not that I'm interested anymore but, why not invite the contributor to maintain it? That would do TWO things: Let him know the status... and inform him of whats required... He could then decide to pass if he felt he could not do it, or was not interested enough. Isn't the contributor the best person to maintain it anyway? | 3.) I have developed an ebuild for the perl interface to the Xerces-c | XML parser. Which is a bit of a pain to build. The Gentoo guides for | submitting ebuilds says that the perl herders need to approve such | ebiulds. Well, they do not even tell you how to send mail to the perl | herders. (No link, no nothing). So I have a perfectly good ebuild | for perl's XML-Xerces module in my portage overlay dir, that I have | not submitted, because of my experience with bug 21580, and my | inability to find the perl herders. This should probably be handled by g-cpan.pl, How can you say that? Have you built XML-Xerces successfully with g-cpan.pl? I doubt it. It has a dependency on dev-libs/xerces-c, AND, it requires specific EVs be set prior to building. BTW: here is the output of g-cpan.pl as far is it gets: lws root # g-cpan.pl XML-Xerces mkdir /tmp/perl-modules_28142 mkdir /tmp/perl-modules_28142/dev-perl CPAN: Storable loaded ok CPAN: LWP::UserAgent loaded ok Fetching with LWP: ftp://ftp.saintjoe.edu/pub/CPAN/authors/01mailrc.txt.gz LWP failed with code[400] message[FTP return code 000] Fetching with Net::FTP: ftp://ftp.saintjoe.edu/pub/CPAN/authors/01mailrc.txt.gz etc... works real well I'd say... Doesn't even move on to another server on failure. My /root/.cpan/MIRRORED.BY file has over 200 lines that would qualify as mirrors for this file... but if not, a google search for Gentoo perl herd will tell you everything you need. Did you try that? I get: Your search - Gentoo perl herd - did not match any documents. Other searches take one back the the Guidelines docs that have no links or addresses. | I am waiting to see if the Zynot folks (http://www.zynot.org/) produce | anything usable, and whether they will be a little more accepting of | user contributions. I hope you enjoy long waits :) You are probably right... but you never know, sometimes (often?) small groups of people can do amazing things... look at gentoo :-) -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] how long does it take to commit a patch to portage
On Fri, 2004-01-02 at 17:56, Ciaran McCreesh wrote: No , sorry, should have stated that explicitly. Ok that enabled me to find: http://www.gentoo.org/proj/en/metastructure/herds/herds.xml Which has some email addresses on it... I'll see where that goes. Is there a way to get to this from the gentoo.org site and I just didn't see it? It seems pretty will hidden. Thanks. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Resizing root
On Thu, 2004-01-01 at 02:29, Jeremy Maitin-Shepard wrote: lucas [EMAIL PROTECTED] writes: [snip] What about the underlying partition, how do i resize it without killing the data on it? You should resize the filesystem first, then change the partition using fdisk by first deleting it and then recreating it. You have to be very careful when doing this though. Unless you are _REAL_ careful, you could be starting over. Best to make a backup, first, to another file system. Then you can boot with that filesystem as root, and resize the first... Of course, if the reason you are doing this is you are so strapped for space, that you cannot find room for a root file system, then I would strongly suggest that you should spend the $$ on another disk... and when you rearrange things, always allocate two partitions for root: one which you use, and one which you save as backup. That way, you always have the space to do something like this. 80GB ide drives are about $80. (US). Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] devfs problem
On Thu, 2004-01-01 at 14:13, Collins wrote: On Thu, 1 Jan 2004 13:51:50 -0500 Jerry McBride [EMAIL PROTECTED] wrote: On Thursday 01 January 2004 01:34 am, Ben Munat wrote: Just wanted to follow up in case anyone was wondering. [ snipped ] Call me OLD FASHION, but devfs is the first thing I dump after installing gentoo on a new box. The steps I take are; recompile the kernel without devfs support, add gentoo-nodevfs to the append line in lilo.conf, correct any defvs associations iin lilo.con and fstab, run lilo, rc-update del devfs and then reboot. Once booted, move to /dev and run MAKEDEV. If you are running alsa, also run snddevices anad add in any symlinks you might need for various applicaitons... The thing I like least about devfs is that fact that one cannot make a working backup of a root partition. On has to do a difference root first. I have working around this by making backups (excluding the /dev directory, and then (from a LiveCD boot), making a tarball of /dev which I can restore into a /dev directory on the copy before I try to boot the copy. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] devfs problem
On Thu, 2004-01-01 at 19:37, Collins wrote: On Thu, 01 Jan 2004 14:23:10 -0500 Lincoln A. Baxter [EMAIL PROTECTED] wrote: On Thu, 2004-01-01 at 14:13, Collins wrote: On Thu, 1 Jan 2004 13:51:50 -0500 Jerry McBride [EMAIL PROTECTED] wrote: On Thursday 01 January 2004 01:34 am, Ben Munat wrote: Just wanted to follow up in case anyone was wondering. [ snipped ] Call me OLD FASHION, but devfs is the first thing I dump after installing gentoo on a new box. The steps I take are; recompile the kernel without devfs support, add gentoo-nodevfs to the append line in lilo.conf, correct any defvs associations iin lilo.con and fstab, run lilo, rc-update del devfs and then reboot. Once booted, move to/dev and run MAKEDEV. If you are running alsa, also run snddevices anad add in any symlinks you might need for various applicaitons... The thing I like least about devfs is that fact that one cannot make a working backup of a root partition. On has to do a difference root first. I have working around this by making backups (excluding the /dev directory, and then (from a LiveCD boot), making a tarball of /dev which I can restore into a /dev directory on the copy before I try to boot the copy. I'm not sure I understand. When I create a backup (done from init 1 environment), I have a script that creates the / directory structure, then copies(cp-a) each needed directory (obviously /tmp, /proc, etc. aren't needed) including/dev. All I have to change is /etc/fstab and add a new grub boot stanza to boot from the backup. Ah, but you are doing from init 1 env. I used to do it from a system at init 3 or init 5 (prior to gentoo). With gentoo the result is un-bootable, because if you use tar's --one-file-system switch, it would stop at /dev (as expected), but in reality, there is stuff under there that devfs needs. Its kind of funky. In an init 1 env, it would find that stuff (assuming /dev had not been mounted). What I did, to work around the problem of wanting to take a backup, without droping to init 1 or booting a live CD, is Do once: 1 boot the live CD, 2 mount the root partition, 3 tar up the root partitions /dev directory. 4 save it in /root/ on the root partition. Now I can backup the root partition with out taking the system down. Cause I have a tarball of what /dev looks like before devfs is mounted on it. If I have to boot my backup root partition, i just unpack the /dev tarball before I try to boot it. It was a little painfull to have to figure out this busines with devfs as procedures I had used for years stopped working what I switched gentoo. Once I figured out this work around, everything is fine. Actually, I currenly do not use tar to backup partitions, rather, I use rsync, (first I take down services that need to be down (like oracle, mq series, qmail etc), in order to get a clean copy. Rsync keeps me with good image of each file system that is no more that 24 hours old. Disk space is cheap, time is not. While I can always restart from scratch, I'd much rather not. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] mass-replacing text
On Tue, 2003-12-30 at 12:04, Ciaran McCreesh wrote:
On Tue, 30 Dec 2003 10:14:26 -0600 Andrew Gaffney
[EMAIL PROTECTED] wrote:
| Although, there is one case this doesn't work for. In some of the HTML
| files, the text I'm looking to strip is split over 2 lines like:
|
| a href=someurlimg src=button.gif border=0 width=51
| height=20/a
|
| How would I strip the text in this case?
find . -name '*.html' -exec perl -pi.bak -e '$/=;' -e \
's- width=51\s+height=20--g' '{}' \;
I would make two changes to this:
1) change $/=; - undef $/;
2) add s to the end of the regex (this tells perl run the regex accross
the entire string crossing new line boundaries. So the result would be:
find . -name '*.html' -exec perl -pi.bak -e 'undef $/;' -e \
's- width=51\s+height=20--gs' '{}' \;
Lincoln
--
Lincoln A. Baxter [EMAIL PROTECTED]
--
[EMAIL PROTECTED] mailing list
[gentoo-user] devfs and udev (kernel 2.6 and beyond)
The 2.6.0 series kernels and later consider devfs obsolete. Does anyone know the gentoo road map for moving away from devfs and toward udev? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] using gentoo to see partions
On Tue, 2003-12-23 at 13:57, Chris wrote: [snip] i guess i'll just keep trying dif combos until this thing finally mounts winxp-vfat Are you aware that the default XP partition is and NTFS file system? vfat will never do it unless you have explicitly created the partition as FAT. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] kernel 2.6.0
On Thu, 2003-12-18 at 00:01, Bill Kenworthy wrote: just saw kernel 2.6.0 has been released. Is there a roadmap on what is happening with kernels in gentoo? i.e., will gs-sources move to 2.6 soon, or will there be a gs-sources-2.6.n and a gs-sources-2.4.n Its more of an issue as to what to look out for than to keep doing a emerge sources -s in the hope somethings changed. BillK Actually, I think the build instructions need some overhauls, or atleast some clear addenda added, before 2.6 kernels should be unmasked. Especially since system setup is actually different, interms of obsoleted devices. Isn't devfs obsoleted in this kernel? If so, thats a big change, I would think. -- [EMAIL PROTECTED] mailing list
[gentoo-user] perl XML::Xerces
Has anyone successfully build and tested the perl XML::Xerces module on Gentoo? No ebuild exists... I am willing to create one if someone can give me a clue as to what works. The current CPAN perl module is XML::Xerces-2.3.0-4 which will compile against xercesc built with /usr/portage/dev-libs/xerces-c/xerces-c-2.3.0-r6.ebuild but not 2.4.0. Tests fail to load the perl Xerces.so with undefined symbol: _ZN11xercesc_2_49XMLString9transcodeEPKt xercesc does not build with gcc-2.95. I have tried building the module with both gcc-2.95 and with gcc-3.2.3, and it compiles and links with both compilers, but both builds failure to load with undefined symbol errors. -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] Setting the ethernet speed on at 3c59x ethernet card.
Does any know how to force the speed (10Mbit vs 100Mbit) on the 3c59x ethernet card? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] make.conf ambiguity - i486 or i686
On Fri, 2003-12-05 at 17:23, Jason Newquist wrote:
In the CHOST section at the top, my default value (before entering
stage 1) is i486-pc-linux-gnu. In reading the comments, it says
to leave this alone for a Pentium Pro or higher processor. My P4
is indeed = PPro. However, then it says that any modern machine
should use i686.
A 486 is NOT a modern machine in this context.
Later on, in CFLAGS, the file has an -mCPU=i686 declaration.
I have a PPro... (586-200mHX) as my gentoo firewall. I use:
CHOST=i586-pc-linux-gnu
CFLAGS=-mcpu=i586 -O2 -pipe
CXXFLAGS=${CFLAGS}
If you have a 486, I believe you are a pentium (NOT a pentium pro). I
would change all the i586's above to i486s.
You definitely do NOT want to compile with -mcpu=i686
It will generate code that will not work. I think that is true for
-mcpu=i586 as well... but I may be wrong. Play it safe, though.
--
[EMAIL PROTECTED] mailing list
[gentoo-user] gentoo.org IP address change?
Did this happen? It seems that the gentoo ezmlm was rejected by smtp at comcast.net as in invalid IP address. Did anyone else get this? -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] emerge of latest qmail seems to have broken mail delivery
It would appear, that upon emerging the latest qmail (1.03-r13) has broken delivery of mail. It seems the messages are being accepted and sent out. Messages for local deliver are also being accepted, but are not available when one connects with a POP client Upon examination of my maildir, all the messages have been written to the maildir/new/ directory. They are not being moved over to the maildir/cur/ directory. Does anyone know what changed in this new qmail (besides the fact that the ebuild wants you to generate ssl keys)... I did this, and it seems to have made no difference. The other major thing I emerged this evening was an new version of gcc, but I doubt this would have broken things. Had any one else had the same problem? and/or know of a fix. Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] emerge of latest qmail seems to have broken mail delivery
On Fri, 2003-11-28 at 08:35, Mike Williams wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Friday 28 November 2003 13:08, Lincoln A. Baxter wrote:
It would appear, that upon emerging the latest qmail (1.03-r13) has
broken delivery of mail.
It seems the messages are being accepted and sent out.
Messages for local deliver are also being accepted, but are not
available when one connects with a POP client
Upon examination of my maildir, all the messages have been
written to the maildir/new/ directory. They are not being moved over to
the maildir/cur/ directory.
Messages are supposed to be left in the new directory until they are
read/viewed, then they are moved to cur.
Yea, that seems right, I noticed files in my wifes cur directory, which
confused me. I think they are left over from when her MUA was
configured to 'leave messages on the server'.
Does anyone know what changed in this new qmail (besides the fact that
the ebuild wants you to generate ssl keys)... I did this, and it seems
to have made no difference. The other major thing I emerged this evening
was an new version of gcc, but I doubt this would have broken things.
Had any one else had the same problem? and/or know of a fix.
Was your pop server updated at the same time?
Qmail is still doing the right thing.
Sending mail works just fine. I agree that the problem is with the pop
server. I am using qmail-pop3d. Have been for several years. And bin
through several gentoo updates without a problem. This update replaced
all the run scripts in the supervise directory, and created a bunch of
new config files in the control directory, and left qmail-pop3d broken.
It appears (based on a troll of the logs) that it is looking for
/etc/tcp.pop3.cdb. Now I need to figure out how to build that file.
Also new with this update, were /etc/tcp.{smtp,qmtp,qmqp}{,.cdb}
It appears that what ever script built these has overlooked the build of
the tcp.pop3 and tcp.pop3.cdb. So I am off to figure this out (by
reading through the ebuild... I suspect the code the built these files,
is probably in the package config function.
I will post again when I figure this out.
But there is little doubt that the developer must not use qmail-pop3d,
or this would not have been released way more broken than any previous
update of qmail, since I have been using gentoo. (just about a year
now).
Thanks...
- --
Mike Williams
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)
iD8DBQE/x081InuLMrk7bIwRAvMaAJ9MZz86WbMvWMjQAlMtoz+XX51bIgCgqQ0+
i5CICClYjwvECNR6PlP/6lI=
=EuyH
-END PGP SIGNATURE-
--
[EMAIL PROTECTED] mailing list
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-user] emerge of latest qmail seems to have broken mail delivery
On Fri, 2003-11-28 at 09:23, Lincoln A. Baxter wrote:
On Fri, 2003-11-28 at 08:35, Mike Williams wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Friday 28 November 2003 13:08, Lincoln A. Baxter wrote:
It would appear, that upon emerging the latest qmail (1.03-r13) has
broken delivery of mail.
It seems the messages are being accepted and sent out.
Messages for local deliver are also being accepted, but are not
available when one connects with a POP client
Upon examination of my maildir, all the messages have been
written to the maildir/new/ directory. They are not being moved over to
the maildir/cur/ directory.
Messages are supposed to be left in the new directory until they are
read/viewed, then they are moved to cur.
Yea, that seems right, I noticed files in my wifes cur directory, which
confused me. I think they are left over from when her MUA was
configured to 'leave messages on the server'.
Does anyone know what changed in this new qmail (besides the fact that
the ebuild wants you to generate ssl keys)... I did this, and it seems
to have made no difference. The other major thing I emerged this evening
was an new version of gcc, but I doubt this would have broken things.
Had any one else had the same problem? and/or know of a fix.
Was your pop server updated at the same time?
Qmail is still doing the right thing.
Sending mail works just fine. I agree that the problem is with the pop
server. I am using qmail-pop3d. Have been for several years. And bin
through several gentoo updates without a problem. This update replaced
all the run scripts in the supervise directory, and created a bunch of
new config files in the control directory, and left qmail-pop3d broken.
It appears (based on a troll of the logs) that it is looking for
/etc/tcp.pop3.cdb. Now I need to figure out how to build that file.
Also new with this update, were /etc/tcp.{smtp,qmtp,qmqp}{,.cdb}
It appears that what ever script built these has overlooked the build of
the tcp.pop3 and tcp.pop3.cdb. So I am off to figure this out (by
reading through the ebuild... I suspect the code the built these files,
is probably in the package config function.
I will post again when I figure this out.
But there is little doubt that the developer must not use qmail-pop3d,
or this would not have been released way more broken than any previous
update of qmail, since I have been using gentoo. (just about a year
now).
Thanks...
Oh, we've made it a little farther. the tcprules program is used to
build the /etc/tcp.pop3.cdb file as follows:
tcprules /etc/tcp.pop3.cdb /etc/tcp.pop3.tmp /etc/tcp.pop3
Once one creates this file, connections to the pop3 server are accepted.
But we hang on the connection to the pop3 server.
In the new configuration files the POP3 server is configured to call
/bin/checkpassword. I am pretty sure was was before too. I have tested
by telneting to port 110, and authenticating with
user NAME
pass PASSWORD
and
list
quit
which returns a list of mails and sizes, and quits just fine. So now I
think pop3d IS working correctly. And it is evolution that is
misbehaving.
So I reconfigure EVO a 100 times, starting from scratch, and it just
HANGS talking to the qmail-pop3d. I configure EVO to talk to my ISP,
and it works first time. Its not the EVO config, its gotta be the new
qmail-pop3d server.
Lincoln
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-user] emerge of latest qmail seems to have broken mail delivery
I just submitted the following bug:
http://bugs.gentoo.org/show_bug.cgi?id=34646
Lincoln
On Fri, 2003-11-28 at 14:14, Lincoln A. Baxter wrote:
On Fri, 2003-11-28 at 09:23, Lincoln A. Baxter wrote:
On Fri, 2003-11-28 at 08:35, Mike Williams wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Friday 28 November 2003 13:08, Lincoln A. Baxter wrote:
It would appear, that upon emerging the latest qmail (1.03-r13) has
broken delivery of mail.
It seems the messages are being accepted and sent out.
Messages for local deliver are also being accepted, but are not
available when one connects with a POP client
Upon examination of my maildir, all the messages have been
written to the maildir/new/ directory. They are not being moved over to
the maildir/cur/ directory.
Messages are supposed to be left in the new directory until they are
read/viewed, then they are moved to cur.
Yea, that seems right, I noticed files in my wifes cur directory, which
confused me. I think they are left over from when her MUA was
configured to 'leave messages on the server'.
Does anyone know what changed in this new qmail (besides the fact that
the ebuild wants you to generate ssl keys)... I did this, and it seems
to have made no difference. The other major thing I emerged this evening
was an new version of gcc, but I doubt this would have broken things.
Had any one else had the same problem? and/or know of a fix.
Was your pop server updated at the same time?
Qmail is still doing the right thing.
Sending mail works just fine. I agree that the problem is with the pop
server. I am using qmail-pop3d. Have been for several years. And bin
through several gentoo updates without a problem. This update replaced
all the run scripts in the supervise directory, and created a bunch of
new config files in the control directory, and left qmail-pop3d broken.
It appears (based on a troll of the logs) that it is looking for
/etc/tcp.pop3.cdb. Now I need to figure out how to build that file.
Also new with this update, were /etc/tcp.{smtp,qmtp,qmqp}{,.cdb}
It appears that what ever script built these has overlooked the build of
the tcp.pop3 and tcp.pop3.cdb. So I am off to figure this out (by
reading through the ebuild... I suspect the code the built these files,
is probably in the package config function.
I will post again when I figure this out.
But there is little doubt that the developer must not use qmail-pop3d,
or this would not have been released way more broken than any previous
update of qmail, since I have been using gentoo. (just about a year
now).
Thanks...
Oh, we've made it a little farther. the tcprules program is used to
build the /etc/tcp.pop3.cdb file as follows:
tcprules /etc/tcp.pop3.cdb /etc/tcp.pop3.tmp /etc/tcp.pop3
Once one creates this file, connections to the pop3 server are accepted.
But we hang on the connection to the pop3 server.
In the new configuration files the POP3 server is configured to call
/bin/checkpassword. I am pretty sure was was before too. I have tested
by telneting to port 110, and authenticating with
user NAME
pass PASSWORD
and
list
quit
which returns a list of mails and sizes, and quits just fine. So now I
think pop3d IS working correctly. And it is evolution that is
misbehaving.
So I reconfigure EVO a 100 times, starting from scratch, and it just
HANGS talking to the qmail-pop3d. I configure EVO to talk to my ISP,
and it works first time. Its not the EVO config, its gotta be the new
qmail-pop3d server.
Lincoln
--
[EMAIL PROTECTED] mailing list
--
[EMAIL PROTECTED] mailing list
[gentoo-user] Storage of Kernel turning params...
I have recently installed Oracle 9.2.0 on a current Gentoo system (I had to use gcc-2.95 for one of the links, and two other programs (ctx/lib) did not link .. but they are not needed). Oracle (and other vendor software packages required that certain turning parameters be set... as in: echo 100 32000 100 100 /proc/sys/kernel/sem echo 1 /proc/sys/kernel/shmmax echo 4096 /proc/sys/kernel/shmmni echo 2097152/proc/sys/kernel/shmall echo 65536 /proc/sys/fs/file-max Now currently have have this in my init.d/dbora script. But this strikes me as wrong. What if I had another package with its own requirements which I wanted run on the same server. So, I look around for a standard location or script for setting kernel tuning parameters during system startup, where this kind of information should really be set. That what one systems requirement would not overwrite anothers. I think we are talking about /etc/sysctl.conf. I presume the above values would be stored as follows in this file: kernel.sem = 100 32000 100 100 kernel.shmmax = 2147483648 kernel.shmmax = 1 kernel.shmmni = 4096 kernel.shmall = 2097152 fs.file-max = 65536 Comments anyone? Lincoln -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] RivaTV
On Mon, 2003-11-24 at 11:27, Francisco Andrades wrote: Greetings, I just finished installing Gentoo 1.4 from stage 1 and I'm trying to emerge RivaTV (that's the main reason why I switched to Gentoo). I'm getting these errors in the emerge 'configure' execution: Checking for CONFIG_VIDEO_DEV... no *** ERROR: *** Your kernel is not correctly configured, please enable CONFIG_VIDEO_DEV Checking for CONFIG_I2C... no *** ERROR: *** Your kernel is not correctly configured, please enable CONFIG_I2C Checking for CONFIG_I2C_ALGOBIT... no *** ERROR: *** Your kernel is not correctly configured, please enable CONFIG_I2C_ALGOBIT I installed the kernel the easy way: # emerge -k sys-kernel/gentoo-sources # emerge -k genkernel # genkernel So I guess I have to recompile the kernel and activate those features. Is this true? Do I need anything else (beside those features) activated in the kernel to use RivaTV? Thanks genkernel --config #set the options the emerge is complaining about -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] New software in portage
Looks like an interesting package. Create a bug in bugs.gentoo.org requesting the addition. Or even better, create the ebuild yourself. Its not that hard. There are good docs for how to do it, and you can of course look at the ebuilds on your system as samples, and then submit it. God knows how long it will take to get into portage. An ebuild I submitted, still hasn't made it (more than 6 months later). But it might be because it was for a package that could be said to have political problems... and I don't even use it any more, so I don't really care. Lincoln On Sat, 2003-11-22 at 00:07, SMS WebMaster wrote: Hi I have found a very nice program at http://www.agata.org.br/index.php I want to send this link to protage team so they add it in to portage , can anyone tell me what is there email ? Thank you -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] openvpn configuration (the gentoo way)
Does anyone have a working openvpn setup that was built using the init.d/openvpn start script that the gentoo ebuild creates? This script appears to walk directories in /etc/openvpn using local.conf in each directory. Presumably each directory represents are remote system to which a tunnel has been created. If you have a working configuration using the gentoo ebuild created structure, would you be willing to share the config files? Thanks, -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] iptables firewall+nat problem
I have been running my own personally developed IPTABLES ruleset since I converted from ipchains to iptables. My topology is is pretty simple: WAN (cable modem) --- eth1 [FW] eth0 --- [HUB] -- [LAN boxes] Note that I am forwarding port 25 from the FW to an internet mail server. This thread caused me to take a closer look at both shorewall, and gsheild (I think it was). I actually emerged shorewall, and attempted to configure it. In the end I found it more confusing than my own custom built script. Which I have pretty extensively tested. (and which I will be happy to share if any one is interested). Frankly, I like understanding what is going on under the covers... so I unmerged shorewall, and went back to using my script. On Sat, 2003-11-01 at 19:17, Joshua Banks wrote: --- Simon_Khling [EMAIL PROTECTED] wrote: I wonder if your firewall is blocking ping scans. Disable the firewall and see if you can ping google. well, you are right - disabling the firewall makes ping work again. maybe it is easier to build my own script from scratch instead of using the one from gentoo-security-guide. If you insist. Your making allot of extra work for yourself. Shorewall already has all of the scripts that you need. All you need to do is simply modify them. Trust me. Try it, and you will understand. If you don't like it go back to writing everything from scratch. http://www.shorewall.net JBanks __ Do you Yahoo!? Exclusive Video Premiere - Britney Spears http://launch.yahoo.com/promos/britneyspears/ -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] 2.4.20-gentoo-r7 and Adaptec 2940
I had the same problem with gentoo-sources. Went back to generic sources. On Tue, 2003-10-28 at 20:35, Stewart C. Russell wrote: Are there any known problems with Athlon XP kernels and a plain old Adaptec 2940? I have one driving a slide scanner and a CD-ROM, and it fails to initialise. After successful POST, and the Adaptec screen identifies the devices hanging off the chain, the init procedure gets knocked for a loop. It gets stuck producing messages like: host 1 abort timed out - bus is being reset (scsi:0:-1:-1) Disconnected list inconsistency ... Yikes! There is a loop in the free list ... and this goes on and on, bumping the pid up by one as it goes. It worked fine under Redhat 7.3 (kernel 2.4.20-20.7). I'm a little suspicious that the card might be borked, as Knoppix 3.2 didn't pick up any devices on the chain. I tried setting acpi=no in the kernel parms, but that didn't help. I haven't yet tried to restart under RH7.3; that would involve a bit of hard-drive shuffling. Ideas, suggestions, replacement cards welcomed. thanks, Stewart -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] RE: problem cloning root filesystem to alternate partition
On Sun, 2003-10-19 at 03:17, Martin Polley wrote: The way I did this was to boot from CD (e.g. LiveCD) and then do the copy operation. That way /dev comes out the way it should. HTH, Well that was the ticket. Its a bummer that I have to take a system off line to do this. I used to be able to do it hot. (/var is NOT on my root partition for this reason). I wonder if a script could be developed that would read /dev, and create the correct /dev on the copy? If I knew the rules I would contribute such a script. Sigh... at least I have now managed to get off of reiserfs which was already beginning to eat itself up -- a few weeks back my system became unbootable until I did a reiserfsck --rebuild-tree which fortunately worked. I'm back on ext3... which is nothing if not stable. I do use NFS, and it seems (based on recent posts to this list) that NFS and reiserfs do not get along so well). My other system which was also reiserfs I am converting now. Thanks again, Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] Has anyone successfully built OO 1.1 (recently unmasked)
It runs for serveral hours and croaks with an undefined symbol error. Here are that (last few lines upto and including the error) Making: ../unxlngi4.pro/lib/libfwe645li.so /usr/bin/ccache gcc -c -fPIC -o ../unxlngi4.pro/slo/fwe_dflt_version.o -DUNX -I../unxlngi4.pro/inc /var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/src/version.c /usr/bin/ccache gcc -z combreloc -z defs -Wl,-rpath,'$ORIGIN' -shared -L../unxlngi4.pro/lib -L../lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/unxlngi4/lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solver/645/unxlngi4.pro/lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/unxlngi4/lib -L/opt/blackdown-jdk-1.4.1/lib -L/opt/blackdown-jdk-1.4.1/jre/lib/i386 -L/opt/blackdown-jdk-1.4.1/jre/lib/i386/client -L/opt/blackdown-jdk-1.4.1/jre/lib/i386/native_threads - ion.o ../unxlngi4.pro/slo/menudocumenthandler.o ../unxlngi4.pro/slo/saxnamespacefilter.o ../unxlngi4.pro/slo/statusbarconfiguration.o ../unxlngi4.pro/slo/statusbardocumenthandler.o ../unxlngi4.pro/slo/toolboxconfiguration.o ../unxlngi4.pro/slo/toolboxdocumenthandler.o ../unxlngi4.pro/slo/toolboxlayoutdocumenthandler.o ../unxlngi4.pro/slo/imagesconfiguration.o ../unxlngi4.pro/slo/imagesdocumenthandler.o ../unxlngi4.pro/slo/xmlnamespaces.o ../unxlngi4.pro/slo/actiontriggerpropertyset.o ../unxlngi o ../unxlngi4.pro/slo/acceleratorinfo.o -lvcl645li -lsvl645li -lutl645li -ltl645li -lcomphelp3gcc3 -lcppuhelpergcc3 -lcppu -lvos3gcc3 -lsal -ldl -lpthread -lm -Wl,-Bdynamic -lstlport_gcc -lstdc++ rm -f ../unxlngi4.pro/lib/check_libfwe645li.so mv ../unxlngi4.pro/lib/libfwe645li.so ../unxlngi4.pro/lib/check_libfwe645li.so /var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/bin/checkdll.sh -L../unxlngi4.pro/lib -L../lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/unxlngi4/lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solver/645/unxlngi4.pro/lib -L/var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solenv/unxlngi4/lib -L/opt/blackdown-jdk-1.4.1/lib -L/opt/blackdown-jdk-1.4.1/jre/lib/i386 -L/opt/blackdown-jdk-1.4.1/jre/lib/i386/client -L/opt/blackdown-jdk-1.4.1/jre/lib/i386/native_thread Checking DLL ../unxlngi4.pro/lib/check_libfwe645li.so ...: ERROR: /var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/solver/645/unxlngi4.pro/lib/libvcl645li.so: undefined symbol: _ZN10MenuButton8KeyInputMRK8KeyEvent dmake: Error code 1, while making '../unxlngi4.pro/lib/libfwe645li.so' ---* TG_SLO.MK *--- ERROR: Error 65280 occurred while making /var/tmp/portage/openoffice-1.1.0/work/oo_1.1_src/framework/util !!! ERROR: app-office/openoffice-1.1.0 failed. !!! Function src_compile, Line 450, Exitcode 1 !!! Build failed! -- [EMAIL PROTECTED] mailing list
[gentoo-user] problem cloning root filesystem to alternate partition
In the past (with Redhat systems) I have made backups of root filesystems to alternate partitions and booted them. Typically I after making the alternate FS, I mount it on /mnt/altroot and cd to / and do the following: tar -cplf - | ( cd /mnt/altroot tar -xvf - ) Then (after adding a stanza in /boot/grub/grub.conf), I boot into the alternate filesystem. This is real useful for being able to do upgrades, and quickly back it out, by just booting into the original FS. Recently I tried this with my Gentoo system and the boot into the copy crash back to the bios after printing unable to open console device. So I looked at /dev on the alternate root FS. And sure enough it is empty. My kernel is configured with /devfs support. (I think this was a gentoo default with genkernel). I understand, why it is empty, I told tar to stay on the same filesystem. But was does the kernel not populate it when I boot the alternate. Any suggestions on how I can get my alternate root filesystem bootable? When I try this on my GEN -- [EMAIL PROTECTED] mailing list
[gentoo-user] Trouble attaching .doc files to Evolution mail after gnome 2.4 upgrade
After gnome 2.4 upgrade (it was unmasked in portage), I have not been able to attach many files to Evolution email. Has anyone else experienced this problem. Also, evolution appears to crash and freeze randomly. I am going to try reemerge evolution, but again, has anyone else experienced this. Seems like the unmasking of gnome 2.4 was a leeetle premature. Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] ICONs in Nautilus (after gnome 2.4 upgrade)
It would appear, that after upgrading to the newly unmasked gnome 2.4, all of the icons in nautilus have disappeared. Now all files and folders appear to have a generic icon. (perhaps they were never there I am just noticing. Has anyone else experienced this? Does anyone know where defaults for icons by file extension could be set? Say, by mime type? -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] ICONs in Nautilus (after gnome 2.4 upgrade)
Never mind, I found it. Its in the Themes. Select a theme an I files get icons based on theme selected. On Sun, 2003-10-12 at 21:07, Lincoln A. Baxter wrote: It would appear, that after upgrading to the newly unmasked gnome 2.4, all of the icons in nautilus have disappeared. Now all files and folders appear to have a generic icon. (perhaps they were never there I am just noticing. Has anyone else experienced this? Does anyone know where defaults for icons by file extension could be set? Say, by mime type? -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Blocks introduced into portage
Thats what I did... but it seems that epiphany want Mozilla compiled with gtk2 and galeon wants Mozilla compiled with gtk1.2 Is seem the Mozilla is not slotted so it looks like galeon is busted, unless I recompile moz with gtk1.2 will will bust gnome (I think). Catch 22 it seems. Lincoln On Sat, 2003-10-11 at 22:10, HvR wrote: happened to me also when gentoo wanted to go to gnome 2.4, i just did and unmerge of these two packages followed by a emerge -u world and all was well... On Fri, 2003-10-10 at 20:00, Lincoln A. Baxter wrote: Some time early this week the following blocks (from emerge -unp world) showed up after a sync. [blocks B ] gnome-base/bonobo-activation (from pkg gnome-base/libbonobo-2.4.0) [blocks B ] gnome-extra/gnome-utils-2.3 (from pkg gnome-extra/gucharmap-1.0.0) It looks like it was created by a MAJOR unmasking of a bunch of new package versions and it does not seem to be going away. Would some one like to share the solution that does not break anything? -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] Blocks introduced into portage
Some time early this week the following blocks (from emerge -unp world) showed up after a sync. [blocks B ] gnome-base/bonobo-activation (from pkg gnome-base/libbonobo-2.4.0) [blocks B ] gnome-extra/gnome-utils-2.3 (from pkg gnome-extra/gucharmap-1.0.0) It looks like it was created by a MAJOR unmasking of a bunch of new package versions and it does not seem to be going away. Would some one like to share the solution that does not break anything? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] CD Writer and Dell Inspirion 8200
Thanks everyone for the responses. I got it working with the hdc=ide-scsi on the kernel line in grub.conf, and then modprobe ide-scsi. This latest post interests me however. (no need for ide-scsi). Is anyone else using this? Perhaps cdrecord should be modified to default to this, so that it just works in the most common case, without requiring custom configuration. I have it working now with ide-scsi... I wonder if I should back that out, and try the dev=ATAPI:0,0,0 solution. It sure would have been nice if that had been documented in the man page somewhere. (may be it was and I missed it). At the risk of starting a religious war, does anyone have opinions on the relative merits of the ATAPI interface vs. the ide-scsi interface? Thanks Lincoln On Mon, 2003-10-06 at 05:11, Mike Williams wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Monday 06 October 2003 02:11, Lincoln A. Baxter wrote: The device appears to be /dev/hdc. cat /proc/ide/ide1/hdc/model shows: cdrecord -scanbus Cdrecord 2.01a14 (i686-pc-linux-gnu) Copyright (C) 1995-2003 Jrg Schilling cdrecord: No such file or directory. Cannot open '/dev/pg*'. Cannot open SCSI driver. cdrecord: For possible targets try 'cdrecord -scanbus'. Make sure you are root. cdrecord: For possible transport specifiers try 'cdrecord dev=help'. Do I need to create some /dev/pgN devices? (presumably with mknod) ? cdrecord -scanbus dev=ATAPI: NO NEED FOR SCSI EMULATION ANYMORE! :) cdrecord dev=ATAPI:0,0,0 -speed=16 driveropts=burnfree -v -data \ downloads/gentoo/x86-1.4-20030911-cd2.iso Works just fine for me, on my stable x86 :) - -- Mike Williams -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQE/gTHJInuLMrk7bIwRAj2UAJ9HWmw10J0Qxilxqv7YkzjDpDaHywCeMR5a YIODiOz78ZhbKRs3sIEeuuA= =KiNw -END PGP SIGNATURE- -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] Getting Evolution to recognize shar files
Some time ago (before the latest upgrade of Evolution), Evolution was able to recogize the contents of shar files as attachments. This enabled me to send myself an shell archive of a compressed tarball, and I would see it in Evolution as a tar.gz attachment. The upgrade to 1.4 broke this. (Or I did and do not know what I changed). Can any shed some light on this? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] NFS vs. samba
Why not just use the automounter? emerge -s autofs On Sun, 2003-09-21 at 11:33, Marshal Newrock wrote: On Sun, 21 Sep 2003, Janne Johansson wrote: If the laptop's not plugged in, after about 10-20 seconds rpc gives up with destination unreachable and the filesystem just doesn't mount. Yup. If you are worried about the delay, you can always take it out of the fstab and mount it manually. It seems to me that a speedup would be to first attempt to ping the host, and if that succeeds, then try to mount the NFS share. This would have to be optional, of course, since some hosts may be configured not to reply to pings. -- Lincoln A. Baxter [EMAIL PROTECTED]
Re: [gentoo-user] Re: Please confirm your message to me
This is very strange. Some thing has changed in the last several days, and it was not here. I am curious why qmail at comcast is complaining about a misconfigured sender why it get the message from gentoo. (see below). I am going to turn it off for a few days... Lincoln On Mon, 2003-09-08 at 00:07, bob bob wrote: LOL This has virtually eliminated SPAM from my inbox, and I thank you in advance for helping to keep SPAM from my inbox. Oh I bet it has.. I bet that its elimiated many MANY non spam e-mails as well.. *rolls eyes* Imagine a school with children that can read and write, but with teachers who cannot, and you have a metaphor of the Information Age in which we live. - Peter Cochrane From: qconfirm [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Please confirm your message to me Date: 8 Sep 2003 03:58:22 - Received: from sccrmhc11.comcast.net ([204.127.202.55]) by mc5-f27 with Microsoft SMTPSVC(5.0.2195.5600); Sun, 7 Sep 2003 20:59:55 -0700 Received: from baxter.homelinux.com ([68.83.245.204]) by comcast.net (sccrmhc11) with SMTP id 2003090803595401100ksu8ve; Mon, 8 Sep 2003 03:59:54 + Received: (qmail 6220 invoked by uid 500); 8 Sep 2003 03:58:22 - X-Message-Info: JGTYoYF78jF8X58JtF3VliqBwANNRsqo Message-ID: [EMAIL PROTECTED] Return-Path: [EMAIL PROTECTED] X-OriginalArrivalTime: 08 Sep 2003 03:59:55.0948 (UTC) FILETIME=[AA1816C0:01C375BD] Hi, I (lab=Lincoln A Baxter) am now using the the qconfirm mail-handling program, to quarentine mail from all senders whose email addresses are unknown to me, or whose email addresses I have not explicitly enabled. One or more messages from you are being held in my pending queue because your address was not recognized. To release your pending message(s) for delivery, please reply to this request. Your reply will not be read, so an empty message is fine. If you do not reply to this request, your message(s) will eventually be returned to you, and will never be delivered to me. Your confirmation verifies to me (and qconfirm) that your message(s) is legitimate and not junk-email, and you should never have to do this again as long as you use the same reply to email address. If you sent a message to me by doing a reply all to a message I sent to an email list we have both subscribed to, rest assured that I have read your message from the list. If you wish(ed) to communicate with me personally (off list) please reply to this message as indicated above, and your message will be delivered. This has virtually eliminated SPAM from my inbox, and I thank you in advance for helping to keep SPAM from my inbox. The reason this works is as follows: SPAMMERS tend not to use real reply to addresses. If they did, we would be able to find them and sue them for illegally trespassing on our equipment. Even if they did use real reply to addresses, they would have to construct autoresponders (like this program) which would intelligently do the right thing to get their message through. SPAM works because it is cheap for the sender, this breaks that business model. Eventually qconfirm bounces the unconfirmed messages, and I never have to look at them. More information about the qconfirm program can be found at http://smarden.org/qconfirm/ Note: this URL is just a website not an information gathering click through so RELAX! Lincoln --- Below this line is the top of the message from you. Received: (qmail 6214 invoked from network); 8 Sep 2003 03:58:22 - Received: from unknown (HELO localhost) (127.0.0.1) by localhost with SMTP; 8 Sep 2003 03:58:22 - Received: from mail.comcast.net by localhost with POP3 (fetchmail-6.2.3) for [EMAIL PROTECTED] (multi-drop); Sun, 07 Sep 2003 23:58:22 -0400 (EDT) Received: from sccrgxc02.comcast.net ([204.127.202.62]) by sccrmxc13.comcast.net (sccrmxc13) with ESMTP id 20030908035802s1300ai2pee; Mon, 8 Sep 2003 03:58:02 + Received: from gentoo.org (mail.gentoo.org[204.126.2.42](misconfigured sender)) by comcast.net (sccrgxc02) with SMTP id 20030908035800e0200hdfgse; Mon, 8 Sep 2003 03:58:00 + Received: (qmail 26141 invoked by uid 1002); 8 Sep 2003 03:57:35 - Mailing-List: contact [EMAIL PROTECTED]; run by ezmlm Precedence: bulk List-Post: mailto:[EMAIL PROTECTED] List-Help: mailto:[EMAIL PROTECTED] List-Unsubscribe: mailto:[EMAIL PROTECTED] List-Subscribe: mailto:[EMAIL PROTECTED] List-Id: Gentoo Linux mail gentoo-user.gentoo.org Reply-To: [EMAIL PROTECTED] X-BeenThere: [EMAIL PROTECTED] Delivered-To: mailing list [EMAIL PROTECTED] Received: (qmail 4434 invoked from network); 8 Sep 2003 03:57:35 - X-Originating-IP: [203.52.232.1] X-Originating-Email: [EMAIL PROTECTED] From: bob bob [EMAIL PROTECTED] To: [EMAIL PROTECTED] Bcc: Date: Mon, 08 Sep 2003 08:57:34 +0500 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: [EMAIL PROTECTED] X-OriginalArrivalTime: 08 Sep 2003 03:57:34.0641 (UTC
Re: [gentoo-user] [OT] Mr. Baxter strikes again?
OK, You guys got my attention. And I am confused. The way this list is setup, if you reply your mail gets sent to the list. I have set qconfirm to ACCEPT mail from the list. If you addressed mail directly to me, you would have, and should have received a confirmation request. Can one of you explain what exactly you did (or I did), so that we can straighten this out? I wonder if it might have to do with the fact that I am registered to the list with the address I sent this from, and qconfirm is using my personal smtp address (which it must do, in order to implement the automatic confirmation. I am willing to fix ASAP, if I we con figure out what is wrong. Lincoln On Sun, 2003-09-07 at 18:38, Sylvain wrote: i also had a such mail. Hoppefully, nobody else here use qconfirm, or people who do so know how to set it up to accept mail from this list :-) sylvain Le Sun, 7 Sep 2003 21:24:03 +0100 (WEST) Jorge Almeida [EMAIL PROTECTED] a crit: Hello everyone. Just thought I should mention I received a mail from Mr. Baxter [EMAIL PROTECTED], who shares with me (with us all?) the joys of qconfirm. I send him a message classifying his mail as SPAM, and, of course, just received another one manifesting appreciation for helping keeping *his* mailbox clean. I recall that someone complained about the same a few days ago... I know I can procmail him off, but isn't there some list policy about such cases? Regards, -- Jorge Almeida -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED]
Re: [gentoo-user] The modem device on a DELL 8200 laptop
Thanks I found that. I did not try to do it outside of X, so I was not bit by conflict you spoke of. Thanks for noting that however, I would have pulled my hair out trying to figure out what I broke, if you hadn't and you hadn't meantioned it. Lincoln On Thu, 2003-08-28 at 03:12, Andrea Gagliardi wrote: Used it a lot before GPRS PCMCIA card come to my slot... simply emerge net-dialup/hsflinmodem, possibly last masked version and after a hsfconfig -a (configure automatically everything) hsconfig -c ITALY (to localize your telephone lines) and hsfconfig -s at each boot (i put it in /etc/conf.d/local.start) and you have a /dev/ttySHSF0 (strange device) Just one annotation, I was not able to start the connection with pppd and then start X11, all crashed, but everything goes fine when the connection is started from inside X11... strange thing... Bye Andrea Bill Kenworthy wrote: Running lspci on my 8200 gives 00:1f.6 Modem: Intel Corp. 82801CA/CAM AC'97 Modem Controller (rev 02) Never used it though ... BillK On Thu, 2003-08-28 at 10:43, Lincoln A. Baxter wrote: Hi, I have a DELL Inspiron 8200 laptop. Gentoo is up and running fine. But I cannot figure our what the modem device is. It does not appear to be /dev/ttyS0. Can someone shed a clue or two? Do I need to enable something in the Kernel to see it? gentoo-sources kernel built with genkernel Thanks. -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED]
[gentoo-user] The modem device on a DELL 8200 laptop
Hi, I have a DELL Inspiron 8200 laptop. Gentoo is up and running fine. But I cannot figure our what the modem device is. It does not appear to be /dev/ttyS0. Can someone shed a clue or two? Do I need to enable something in the Kernel to see it? gentoo-sources kernel built with genkernel Thanks. -- Lincoln A. Baxter [EMAIL PROTECTED]
[gentoo-user] Completely trashed root filesystem
Look that this command tar -clf - | ( cd /mnt/altroot/ tar -xvf - ) See what's wrong? Yep... on one amper... Result: completely fubared root file system. (this command makes empty files of every file in the root file system). I did kill it before it completed... but I know the system will not boot. I was able recover things (using the RESCUE tar ball in /usr/portage/sys-apps/portage/file...) I was able to restore /usr/bin and /lib from an other system so that emerge would run. I was able to emerge portage I was able to emerge -e world... and it completed about 34 of the 482 packages to be rebuilt before it traced back on file that should have been a directory in /var/tmp/portage... while emerging xfree-4.3. Then emerge was broken again: It just segfaults with no information. So I did the resume tar ball again... But to now avail. it just segfaults: # emerge portage Segmentation fault # Any ideas on where to go from here? Am I stuck? and starting over with a fresh install in my alternate root? -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] gcc 3
At least one person on this list reported that this broke qmail. There could be other packages not unmasked, that could break as well. Approach with caution. On Fri, 2003-07-18 at 00:58, blade- wrote: Hi all, When I do emerge -Up world it lists the following [ebuildU ] sys-devel/gcc-config-1.3.3-r1 [1.3.1-r1] [ebuildU ] sys-devel/gcc-3.2.3-r1 [3.2.2] [ebuildU ] sys-libs/glibc-2.3.2-r1 [2.3.1-r4] Will I have to recompile everything again or is it just a minor change the will not require me to do much at all. Thanks -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] Setting up a remote mailhost
Hi all, I want to be about to send mail from a gentoo box that does not have an MTA running on it. Does anyone know how to setup a gentoo box, so that mailx will send to a remote MAILHOST. I.E, on another box? I know this is possible on other various unixes, and I presume it is possible on gentoo, I just can not figure out how to do it. Thanks, Lincoln -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] [Q] bashrc
You missed the point of the first person to respond to you. Bash, when called as sh acts like sh. I.E. it does NOT look for .bashrc or .bash_profile, it looks for .profile. If you insist on keeping you shell as sh, then create a .profile (which can be a soft link to your .bash_profile if you want., but you are also going to want a .shrc. Why not just fix your shell to be bash in /etc/passwd? Lincoln On Sun, 2003-07-06 at 04:23, YOON, Joo-Yung wrote: Hi, Yes, it worked to type bash. It reads .bash_profile which calls .bashrc. And bash -l also worked. I looked up into the man. It says the option means that bash acts as a login shell. Then does it also mean that my $SHELL is not the login shell? Again, echo $SHELL says /bin/sh, and it is symlinked to /bin/bash. I'm so confused as to poke into this and that, but do not have the right place to refer to. I forgot telling that I telnet to my server, and it happens so. But this headache is not related to the telnet login. On Sat, Jul 05, 2003 at 11:37:15PM -0700, Luke Ravitch wrote: On 2003-07-05 23:25, YOON, Joo-Yung [EMAIL PROTECTED] wrote: It exists for sure, and contains many settings therein, but the thing is it is not called or executed automatically upon logging in. (I must do source .bashrc to realize it.) I think what he was trying to say was you need to put a command in ~/.bash_profile to source ~/.bashrc. Bash reads ~/.bashrc if the shell is interactive and it's NOT a login shell. If it's a login shell, bash sources ~/.bash_profile or ~/.profile, but not ~/.bashrc. So, unless you have setting for non-login shells that you don't want login shells to share, the thing to do is add a line like: #This file is sourced by bash when you log in interactively. [ -f ~/.bashrc ] . ~/.bashrc to your ~/.bash_profile. To test if this the problem, run bash again from within your first bash (i.e., type bash at the command prompt). Is your ~/.bashrc being sourced now? (I'll go ahead and guess yes.) Now run a login bash (type bash -l at the command prompt). This time, no ~/.bashrc. Is that what's happening? Hope this helps. -- Luke -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] kmail to evolution
If Kmail stores mail in mbox format you can import these into Evolution from: File - Import menu. On Fri, 2003-07-04 at 01:07, Reg HUGHSON wrote: Does anyone know how I can transfer my email already received by Kmail to Evolution? -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] OT: tabs in vi[m]
You can also:set expandtab or :set et Which replaces tabs with spaces at whatever ts (tabstop) distance you prefer. Note that whenever you set ts you also want to set sw. (shiftwidth). If you use et, you never have to worry about what someone else has his ts set to, the file will ALWAYS read with the original indentation. (at the expense of making the file just a little bigger). I use the following in my .vimrc file: set ts=3 sw=3 set et and at the bottom I add (for ebuild files) because that is what the gentoo folks want: if (getcwd() =~ 'gentoo-x86\|gentoo-src\|portage') set tabstop=4 shiftwidth=4 noexpandtab endif This requires that I be IN or below one of the named directories when I edit the ebuild, for it to work. And I usually am, if I am not just inspecting things. Lincoln On Tue, 2003-06-24 at 21:18, [EMAIL PROTECTED] wrote: On Tue, Jun 24, 2003 at 09:07:32PM -0400, Mike Principito wrote: This is a bit off topic, but I'm curious if anyone knows how to set up a tab to be 3 spaces instead of a tab character in vim. I'm assuming I'd have to add something to my .vimrc, but I don't know what. Any help would be appreciated. Well, the answer is :set tabstop=8, but I'd recommend :set shiftwidth=3 and leaving tabs hard set to 8 (then anyone viewing your file will still see your desired indentation). Just use ^T and ^D in insert mode to tab/backtab, and to indent/un-indent in command mode. Regards, -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Get Msgs button in Mozilla greyed out after emergeto latest
Well I answer my one question again. The differences between /usr/portage/net-www/mozilla/mozilla-1.3-r1.ebuild /usr/portage/net-www/mozilla/mozilla-1.3-r2.ebuild are HUGE! The new ebiuld is broken IMO, because it does not respect use flags. I got it to work by emerging as follows: USE=-ldap -crypt emerge $eargs mozilla I have -ldap in my USE flags in /etc/make.conf but it was not respected. It is possible I could have gotten it to work with just -ldap, but I do not care about mail encryption of this MUA anyway, and a quick perusal of the ebuild lead me to beleive there were likely interactions between these flags. Lincoln On Tue, 2003-06-17 at 23:43, Lincoln A. Baxter wrote: Yesterday I was testing a script I posted to the list earlier today, and reemerged mozilla (several times actually). I did not notice (because only my wife uses mozilla) (I cann't seem to get here to switch to evolution), that the Get Msgs button in mozilla is greyed out in mozilla mail. Does anyone know what would cause this? or How to fix it? Lincoln -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
RE: [gentoo-user] Can Linux web browsing be a complete experience?
How about this: (/usr/local/bin/emerge.browsers.bsh) It sets up the old compiler and the java environment and a /etc/make.conf that will compile the browsers, then emerges mozilla, flash, realplayer, and galeon. Then it puts it all back. NOTE: I have never been able to build these tools with the new compiler. If someone has that incantation please share. = #!/usr/bin/bash eargs=$1 #for passing args like '-unp' to emerge cconf=$(gcc-config -c) #gcc-config i686-pc-linux-gnu-3.2.2 gcc-config i686-pc-linux-gnu-2.95.3 #java-config --set-system-vm=blackdown-jdk-1.4.1 java-config --set-system-vm=sun-jdk-1.4.1.03 #java-config --set-system-vm=ibm-jdk-1.3.1 dt=$(date +%Y.%m.%d.%H.%M.%S) cp -p /etc/make.conf /etc/make.conf.save.$dt cat /etc/make.conf.save.$dt \ | sed 's/^CFLAGS=.*/CFLAGS=-O2 -pipe/' \ | sed 's/^MAKEOPTS=.*/MAKEOPTS=-j1/' \ | sed 's/^FEATURES=/#FEATURES/' \ /etc/make.conf source /etc/profile echo sourced /etc/profile cd / if true; then emerge $eargs mozilla netscape-flash realplayer galeon fi mv /etc/make.conf /etc/make.conf.otw mv /etc/make.conf.save.$dt /etc/make.conf echo gcc-config $cconf gcc-config $cconf -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] Get Msgs button in Mozilla greyed out after emerge to latest
Yesterday I was testing a script I posted to the list earlier today, and reemerged mozilla (several times actually). I did not notice (because only my wife uses mozilla) (I cann't seem to get here to switch to evolution), that the Get Msgs button in mozilla is greyed out in mozilla mail. Does anyone know what would cause this? or How to fix it? Lincoln -- [EMAIL PROTECTED] mailing list
[gentoo-user] LCD flickers with XFree 4.3
This weekend I upgraded to XFree86 (4.3) (the latest unmasked in portage). I have a Matrox 450 dual head video card: Screen0 is a 17 CRT, and Screen1 is a generic 15 LCD display. After the upgrade, the LCD display (which used to be a pleasure to use), now flickers horribly. Its like all the pixels a jumping around constantly. Does anyone have any ideas for how get my LCD monitor (which was rock steady in 4.2) to stop flickering? Could this have to do with anti aliased fonts that we have been hearing so much about? Thanks. Lincoln -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] LCD flickers with XFree 4.3
Well, in answer to my own post: I edited the Monitor section of the XF86Config for this monitor, to more acurately reflect the H/V scan rates per the spec for the monitor. and presto... no more flicker. It must of been that the older version of X was less sensitive to these configurations? Thanks, I hope this helps someone else. Lincoln On Sun, 2003-06-15 at 18:36, Lincoln A. Baxter wrote: This weekend I upgraded to XFree86 (4.3) (the latest unmasked in portage). I have a Matrox 450 dual head video card: Screen0 is a 17 CRT, and Screen1 is a generic 15 LCD display. After the upgrade, the LCD display (which used to be a pleasure to use), now flickers horribly. Its like all the pixels a jumping around constantly. Does anyone have any ideas for how get my LCD monitor (which was rock steady in 4.2) to stop flickering? Could this have to do with anti aliased fonts that we have been hearing so much about? Thanks. Lincoln -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Removing old packages from /usr/portage/packages
Very nice! Thanks, I had started to think some more about the problem, and you have solved it most elequently. Lincoln On Fri, 2003-06-13 at 18:16, [EMAIL PROTECTED] wrote: On Thu, Jun 12, 2003 at 11:00:58AM -0700, Zack Gilburd wrote: Content-Description: signed data On Thursday 12 June 2003 07:58, keanu wrote: On Wednesday 11 June 2003 20:51, Jose Romildo Malaquias wrote: Does anybody has a script for removing old files from /usr/portage/packages ? I have been accumalating package files (*.tbz2) produced by emerge whith the --buildpkg option. Now I want to (automatically) remove the older ones if there are different versions for the same package. There are a few scripts on the gentoo forums (some better then others) under the Documentation, Tips Tricks section if i'm not mistaken (you'll find alot of usefull stuff there) i don't have time to look for the actual script, but it shouldn't be hard to find Unfortunately, the only scripts that I could find that would be even close to fitting the bill aren't for /usr/portage/packages, but rather they're for /usr/portage/distfiles. In the topic discussed at http://forums.gentoo.org/viewtopic.php?t=23027 there is two scripts for cleaning both the distfiles and the packages subdirectories of the portage directory, based on which packages are not currently installed. In the topic http://forums.gentoo.org/viewtopic.php?t=3011 one can find also a script for cleaning the distfiles subdirectory, based on the existence of a related ebuild in the portage hierarchy. I have adapted it for cleaning the packages subdirectory. Those scripts are more conservative as they keep any distfile or package file for which there is an ebuild. Those scripts are attached. They do not remove the files itself, but outputs their name to the standard output. Romildo -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Bayesian spam filtering
If you are using qmail, consdier qconfirm: It requires you to expose your qmail to the net but as long as you have an always on connection, it is easy to do with dyndns.org. It virtual elminates SPAM, because only those that have explicitly responded to a confirmation request, or you have explicitly placed in a whitelist, are allowed to delivered to your inbox. You can find an ebuild compressed tarball here: http://bugs.gentoo.org/attachment.cgi?id=12442action=view Put it in a PORTAGE_OVERLAY_DIR and just emerge. Lincoln On Thu, 2003-06-05 at 09:31, Larry Wright wrote: I am currently running qmail + fetchmail + procmail + spamassassin. Everything works pretty well, but I'm a little dissapointed in spamassassin's accuracy. I'd ideally like something similar to spambayes, which is trainable. Unfortunately spambayes does not appear to work with maildir. I'd really like to keep qmail, does anyone know of a bayesian filter that works with qmail/maildir? Thanks, Larry -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] logwatch ebuild 4.2.1-r1
Yea... no output... On Wed, 2003-05-28 at 11:16, Mike Morrell wrote: I noticed the missing file too. I upgraded to the latest ebuild and crontab no longer complains. But, I still get no output from logwatch at all. I have tried running logwatch.pl and having the results sent by email or by sending it to stdout (using the options in /etc/log.d/conf/logwatch.conf) but I get nothing. Have you actually tried running logwatch.pl to see if it does anything? Mike I saw some discussion on this list about logwatch. That reminder me that I wanted to install it. So i did. So I submitted bug# 21812. The problem is that root's crontab is updated with a reference to logwatch.pl in the /usr/sbin, and the script is not there, it IS in /usr/share/logwatch/scripts. This can be fixed by fixing root's crontab. Or... When I went to fix the ebuild, I noticed that this is fixed in the masked 4.3.2 ebuild. So you could just load that. Although I searched for a bug for logwatch (and did not find one), I probably should have looked at newer ebuilds as well, before submitting... Sigh... -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] ebuild of qconfirm
Hi All, Yesterday, I added an compressed tarball to gentoo bug# 21580, which contains an ebuild for qconfirm. (a little package that works with qmail to virually eliminate spam by using delivery confirmation requests, the same way the this list works). I was just curious, if anyone has any experience with how long it takes for the ebuild to be noticed and integrate into portage proper. -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] logwatch ebuild 4.2.1-r1
I saw some discussion on this list about logwatch. That reminder me that I wanted to install it. So i did. So I submitted bug# 21812. The problem is that root's crontab is updated with a reference to logwatch.pl in the /usr/sbin, and the script is not there, it IS in /usr/share/logwatch/scripts. This can be fixed by fixing root's crontab. Or... When I went to fix the ebuild, I noticed that this is fixed in the masked 4.3.2 ebuild. So you could just load that. Although I searched for a bug for logwatch (and did not find one), I probably should have looked at newer ebuilds as well, before submitting... Sigh... -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] fontconfig emerge errors
I had the same problem, and fixed it the same way. On Wed, 2003-05-28 at 00:49, Larry Meadors wrote: I had a similar problem and removed DOC from my USE flags. Seems to have done the trick for me, but YMMV. Larry [EMAIL PROTECTED] 05/27/03 9:39 PM Hi again :) I've been trying to install fontconfig for a day now (its in my update world list). However I'm running into some difficulty. Here's what my console looks like when it barfs. -- Installing fontconfig-user.txt -- Installing fontconfig-devel.txt -- Installing fontconfig-user.html for i in fontconfig-devel; do \ /bin/sh ../mkinstalldirs /var/tmp/portage/fontconfig-2.2.0-r2/image//usr/share/doc/fontconfig-2.2.0-r2/$i ; \ for f in $i/*; do \ echo '-- Installing '$f ; \ /bin/install -c -m 644 $f /var/tmp/portage/fontconfig-2.2.0-r2/image//usr/share/doc/fontconfig-2.2.0-r2/$f; \ done \ done mkdir /var/tmp/portage/fontconfig-2.2.0-r2/image/usr/share/doc/fontconfig-2.2.0-r2/fontconfig-devel -- Installing fontconfig-devel/* /bin/install: cannot stat `fontconfig-devel/*': No such file or directory make[2]: *** [install-data-local] Error 1 make[2]: Leaving directory `/var/tmp/portage/fontconfig-2.2.0-r2/work/fontconfig-2.2.0/doc' make[1]: *** [install-am] Error 2 make[1]: Leaving directory `/var/tmp/portage/fontconfig-2.2.0-r2/work/fontconfig-2.2.0/doc' make: *** [install-recursive] Error 1 !!! ERROR: media-libs/fontconfig-2.2.0-r2 failed. !!! Function einstall, Line 326, Exitcode 2 !!! einstall failed That directory ../fontconfig-devel is indeed there, however it is empty. Does anyone have any idea how I can fix this? Thanks, Mike -- And don't tell me there isn't one bit of difference between null and space, because that's exactly how much difference there is. :-) --Larry Wall in [EMAIL PROTECTED] SDF Public Access UNIX System - http://sdf.lonestar.org -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] naive question about distcc
It is obvious you have never built any of these packages by hand. Almost all OSS packages today use the GNU configure scripts to configure them for building on a BUNCH of difference platforms. Linux being just one. These scripts build the actual makefiles that are used to compile the package on the fly. All gentoo's ebuild system does really is put a bunch of wrappers around the standard build scripts that each package uses. To do what you ask would require that the EVERY package (or most), to be modified to check a system database of available functions. No such database exists, on all unices, and if it did using it would not be as reliable as actually testing to see if the required function is available. I would not expect this to change. It works amazingly well across an amazing large set of platforms, and frankly, is once of the great portability achievements of the FSF. Lincoln On Sun, 2003-03-30 at 22:14, Collins Richey wrote: On Sun, 30 Mar 2003 22:09:24 -0500 Jeremy Schneider [EMAIL PROTECTED] wrote: I've noticed that when I do makes, a lot of time is spent by the system checking lots of stuff: ... checking for a BSD-compatible install... /bin/install -c checking whether build environment is sane... yes checking for gawk... gawk checking whether make sets $(MAKE)... yes checking for i586-pc-linux-gnu-strip... no checking for strip... strip checking for i586-pc-linux-gnu-gcc... gcc ... It occurs to me that these things don't change very often on my system, and that the answer to these checks could be cached, perhaps associated with a hash or date of certain config files, such as make.conf. Does this make any sense, or is it too unworkable and/or risky? I've always wondered about that myself. It's a lot of repetitive work for every install. -- Collins -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] gnome 2.2...
I did it the safe way... (see below) and it seems good. No ~x86. I like the gnome 2.2... yea! Here is how I would do all major upgrades (I consider an emerge of 50 packages a major upgrade): mkFS /dev/unusedpartition #I use reiserfs (call it /dev/new) #edit fstab adding a /mnt/new mount of the new partition... mount /dev/new /mnt/new cd / tar -lpf - . | ( cd /mnt/new tar -xvf - ) cd /mnt/new chroot . mount proc -t proc /proc #emerge wants these file systems avail. mount none -t devfs /dev emerge -un world #edit the fstab (in the chrooted world) changing the root partition to /dev/new Now add boot configurations to the grub.conf in your boot partitions, and boot the new partition as the root partition. If every thing is ok, your old root partition will be your new root partition the next time you want to do a big upgrade. This is safe because if the emerge is craps out you can always boot the orginal root partition, and you can use your system safely while it is happening. Best, Lincoln On Mon, 2003-02-10 at 00:53, Alan wrote: On Sun, Feb 09, 2003 at 09:36:24PM -0800, Alan wrote: On Mon, Feb 10, 2003 at 12:33:43AM -0500, Lincoln A. Baxter wrote: Hi, Gnome 2.2 ebuilds were added (and apparently not masked), in portage over the weekend. emerge -unp world yeilds a huge list of modules to be updated. Has anyone had any trouble with this? Just thought I'd ask before I trashed me-self. Seems like a pretty big upgrade, especially if 1.4 is not final yet... perhaps they really wanted this in 1.4? Don't know about the gnome 2.2 builds (yay though), but when I try to update world now I get: phoenix alan # emerge -up world These are the packages that I would merge, in order: Calculating world dependencies \ !!! all ebuilds that could satisfy =gnome-extra/libgsf-1.6.0 have been masked. !!!(dependency required by gnome-base/librsvg-2.2.2.1 [ebuild]) A little bit more investigation reveals that the gnome package is updated to 2.2 stable, but a lot of dependancies aren't (gst-plugins, gstreamer, mpeglib, etc etc). I'm going to brave it and try an ~x86 update to gnome, and see what happens alan -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Grep woes
The reason is that you were using a double quoted string. Variables and Escape chars are interpolated into double quoted strings. Not so, with single quoted strings Lincoln On Mon, 2003-02-10 at 00:13, Mike Williams wrote: On Mon, 2003-02-10 at 05:09, Cedric Veilleux wrote: Hi, What is the correct regular expression for a simple $ character? I want to filter a file for any lines containing a $ char, but it does not work.. man grep tells me $ is a special regular expression char.. I try to escape it using \$ but it doesn't work.. Oddly a double backslash (\\) escape works, with or without quotes, but don't ask me why! grep \\$ afile -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-user] gnome 2.2...
Hi, Gnome 2.2 ebuilds were added (and apparently not masked), in portage over the weekend. emerge -unp world yeilds a huge list of modules to be updated. Has anyone had any trouble with this? Just thought I'd ask before I trashed me-self. Seems like a pretty big upgrade, especially if 1.4 is not final yet... perhaps they really wanted this in 1.4? Lincoln -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] mother board recommendations
FWIW: I have an Abit KT7-raid MB. It has run RedHat 7.1, 7.2, 7,3, 8.0 and now Gentoo 1.4 with no problems. (No overclocking) I think it is a VIA133 chipset. My only complaint is that I have had to replace the chipset fan (a 486 fan actually) about 3 times. Very annoying. I would agree with the people who said, if you buy an over-clocker's MB, and then NOT overclock you should get a pretty stable system. Lincoln On Fri, 2003-02-07 at 13:04, Sundance wrote: I heard brett holcomb said: Yes, but I don't want to stack the deck against myself by using VIA G. For the records, I've been using a VIA-based MB for over a year without a single problem. It would seem that the KT333 chipsets are a considerable improvement over the previous series. Additionally, VIA has been showing some rather interesting Linux awareness, too -- when it turned out ALSA didn't support the new VT8233a sound chipset that came on KT333-based boards, VIA took the initiative of contacting the ALSA developpers to provide them with the specs they needed to make it work. That was just way cool of them. In short -- VIA sucked, but they seem to suck less of late. That's a good thing. I hope it wasn't just me being lucky, but based on reviews of KT333 based boards, I think not. :) -- S. -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Galeon crashes on FLASH pages
Ok, I put up the 2.95.3 gcc with: ebuild /usr/portage/sys-devel/gcc/gcc-2.95.3-r8.ebuild merge I suppose one would use gcc-config to set the gcc environment prior to (re)-emerging these packages? I will then, (hopefully reemerge java, mozilla, netscape flash, and galeon), I will let everyone know the results. This does make me wonder if the default gentoo configuration should really be gcc-2.95.3.x rather than 3.2.1.x, or maybe BOTH compilers? What does 3.2.1 provide the 2.95.3 does not? Lincoln On Sat, 2003-02-01 at 17:42, Paul de Vrieze wrote: On Saturday 01 February 2003 23:23, Lincoln A. Baxter wrote: Gentoo 1.4 rc2 All current packages emerged, no masked packages emerged. net-www/mozilla 1.2.1-r5 net-www/galeon 1.2.7-r1 net-www/netscape-flash 6.0.69-r2 Does anyone else have a problem with this? Does anybody have any idea how to fix it? Compile galeon and mozilla with gcc-2.95.3 instead of 3.2. This also solves your java problems at the same instance. Paul -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Galeon crashes on FLASH pages
After ebuilding gcc-2.95.3 gcc-config i686-pc-linux-gnu-2.95.3 emerge mozilla ends with: * Configuring Xft2.0... checking for gcc... gcc checking for C compiler default output... configure: error: C compiler cannot create executables check `config.log' for details. !!! ERROR: net-www/mozilla-1.2.1-r5 failed. !!! Function src_compile, Line 427, Exitcode 77 !!! (no error message) Checking the referenced config.log file reveals that this version of gcc does not like -march=athlon-tbird -mcpu=athlon-tbird gcc switches... so I guess I'm off to my /etc/make.conf to back these out and try again... Lincoln On Sun, 2003-02-02 at 16:30, Paul de Vrieze wrote: On Sunday 02 February 2003 21:53, Lincoln A. Baxter wrote: I suppose one would use gcc-config to set the gcc environment prior to (re)-emerging these packages? Yep I will then, (hopefully reemerge java, mozilla, netscape flash, and galeon), I will let everyone know the results. Binary java doesn't need to be remerged. only mozila and galeon. Also don't forget to set your compiler back to 3.2.1 afterwards. This does make me wonder if the default gentoo configuration should really be gcc-2.95.3.x rather than 3.2.1.x, or maybe BOTH compilers? What does 3.2.1 provide the 2.95.3 does not? Optimizations, prelink, future, glibc-2.3.x Paul -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] How to write an init.d script
One thing that is REALLY COOL about gentoo, is its inovative dependancy based init system, where you do not have to find a place (ie a numeric value) which is appropriate for K and S scripts. As much as I liked sysV style initscripts, when I first encountered then, the gentoo way, is really supperior. Yiu just declare your dependant services, and gentoo will call the start script in the right place automatically after you have run rc-update add INITSCRIPT RUNLEVEL to add the script to the appropriate run level (usually just default). You can look at any of the scripts in /etc/init.d to get the idea. It took me less than an hour to convert the oracle dbora init script from my red hat partition and convert it to gentoo. This is one part of the gentoo system it is WORTH getting to know. And it is not hard. Lincoln On Thu, 2003-01-30 at 22:39, Matt Neimeyer wrote: Hey All, The box I'm running Gentoo on came with Redhat 7 (something) on it. The reason this matters is that it has an LCD panel on the front that displayed lots of useful information (like uptime, memory usage, etc) and some buttons that let you, among other things, shut down the computer. I've got qmail running on the box and it's going to be moving to a colo facility. Now the Redhat install had an init script and a daemon (binary only) and some other tools. On an identical box that I installed Redhat 8 on (for someone else) I was able to transplant those binaries and scripts and everything worked the way it should. So the question is... what's the likelihood that I can run those binaries on Gentoo without corrupting anything? And how would I rewrite the init script so that I can run it on Gentoo? Thanks! Matt P.S. I've already tried to identify the hardware with no luck but if someone knows what the LCD unit is that Gallentry installed on their GW500 boxes and a better / more direct way to control it... -- [EMAIL PROTECTED] mailing list -- Lincoln A. Baxter [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
