Re: [gentoo-user] Strange grub problem
Ivan Alden wrote: Hi, My laptop ran out of battery and shut off while I was using it and it seems to have done some damage to the bootloader. When I reboot I can't see the grub splashscreen any more but if I press enter I does boot into my kernel. As the computer is booting the output looks all messed up (but you can make that its initializing devices) until it reaches around the networking devices which then corrects and works properly. I've tried reconfiguring grub with root (hd0,0) setup (hd0) Does anybody have any suggestions of how to fix this? Thanks. Hello Ivan, I had just yesterday exactly the same problem, not because my laptop runs out of battery, but because of full system update. I do it only once a week. Anyway. I could fix it in the following way: - Switch off your laptop and wait some seconds. - Start your laptop - Boot your kernel, you have to imaging on which position in the boot menu this one is - after the machine is up and running - emerge -av grub - reboot and it should be fine That's it. Hope I could help W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Strange grub problem
Oh sorry, _very_ _important_ : Mount your boot partition _before_ you emerge grub. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Strange grub problem
Mick wrote:
On Sunday 20 July 2008, Wolf Canis wrote:
Oh sorry,
_very_ _important_ :
Mount your boot partition _before_ you emerge grub.
I believe that the grub ebuild mounts it for you, messes up your grub.conf
and
carries on with its business . . .
So far I that now, is that now changed. You have to mount it before
you emerge grub. The following is from the ebuild grub-0.97-r6.ebuild:
-- Quote begin -
pkg_postinst() {
if [[ -n ${DONT_MOUNT_BOOT} ]]; then
elog WARNING: you have DONT_MOUNT_BOOT in effect, so you must
apply
elog the following instructions for your /boot!
elog Neglecting to do so may cause your system to fail to boot!
elog
else
setup_boot_dir ${ROOT}/boot
# Trailing output because if this is run from pkg_postinst, it
gets mixed into
# the other output.
einfo
fi
elog To interactively install grub files to another device such as
a USB
elog stick, just run the following and specify the directory as
prompted:
elogemerge --config =${PF}
elog Alternately, you can export GRUB_ALT_INSTALLDIR=/path/to/use
to tell
elog grub where to install in a non-interactive way.
}
-- Quote end --
W. Canis
signature.asc
Description: OpenPGP digital signature
Re: [gentoo-user] Strange grub problem
Mick wrote: On Sunday 20 July 2008, Wolf Canis wrote: Mick wrote: On Sunday 20 July 2008, Wolf Canis wrote: Oh sorry, _very_ _important_ : Mount your boot partition _before_ you emerge grub. I believe that the grub ebuild mounts it for you, messes up your grub.conf and carries on with its business . . . So far I that now, is that now changed. You have to mount it before you emerge grub. The following is from the ebuild grub-0.97-r6.ebuild: Good! That's the preferred behaviour. It shouldn't really mess things up without asking. Yeh, you are right, but that seems not always to be true. :-( signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] chroot problem
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Peter Humphrey wrote: [...] # cd /mnt/rescue # mount -tproc proc proc # mount -obind /dev dev I mean that the mount commands should be: # mount -tproc proc /mnt/rescue/proc # mount -obind /dev /mnt/rescue/dev I just build a mini chroot environment. My working directory is /root. I create under /root a directory x. The contents under x is: # ls -R x: bin dev lib proc x/bin: bash x/dev: x/lib: ld-linux.so.2 libc.so.6 libdl.so.2 libncurses.so.5 x/proc: Then my mount commands: # mount -tproc proc x/proc # mount -obind /dev x/dev Then chroot: # chroot /root/x /bin/bash wolf-di6400 0(0) 10:38 AM / # Hope that helps. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkhBKyMACgkQKT9zBKF0twWttgCffzjUSQZAxNBZcAwf9avjvZYa YDoAn1Rw5y18equ4b+27hAhCnboyfF0x =Um4X -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] chroot problem
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Wolf Canis wrote: Peter Humphrey wrote: [...] # cd /mnt/rescue # mount -tproc proc proc # mount -obind /dev dev I mean that the mount commands should be: # mount -tproc proc /mnt/rescue/proc # mount -obind /dev /mnt/rescue/dev Ooops, I overlooked your cd command. Therefore the mount command is of course correct. :-[ W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkhBSZgACgkQKT9zBKF0twXnmgCeKd9BcrcinpSFZYlYHH6JkYmJ TAUAnjVNmHArsqLbx3nclUPDhIZqQzbW =nt4b -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Daniel Iliev wrote: Come ooon! :) The whole bet thing was of course a joke. What I had in mind is that you'd have to hack Gmail which I believe won't classify as relatively easy. Not to mention that even just for proof of concept this would be illegal, so I'd never expect you to do it. No problem. :-) Alright, the most important thing in this discussion appears that we all agree that signing mails to ML or not, either way there's no harm. So, I think we'd better stop at this point and let it go. Agreed? I have no problem with that. :-) -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkhAIVgACgkQKT9zBKF0twVWLQCfWd/4i0XgyOTuHuJIAxv8pq8D Ug0An0q7/0FB909Ox7SMu3qWAtndAQbL =6TZf -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Daniel Iliev wrote: Unfortunately many times one cannot control the reverse records, because the IP address pool belongs to the ISP. Nevertheless the SMTP server logs the IP address which the message came from. It doesn't matter if the message would be bounced or accepted because of the (in)correct reverse resolving. Additionally there's the SPF [1] and I believe the email system at gentoo.org uses it. If that's so and my poor abused address :) was at a domain with SPF record imposing fail policy, that message shouldn't be accepted at all. At best you'd get something like: Domain of [EMAIL PROTECTED] does not designate 192.0.2.25 as permitted sender. Anyways the right thing to do is to ban the IP address which the offencive message came from, not the email address. So, signatures don't come to play here. [1] http://www.openspf.org/ But you see it isn't that difficulty to abuse a email address. That what happened to your address and what P. S. Ziegler described was what I meant with relatively easy. ;-) Have fun, W. Canis :-) -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkhAIscACgkQKT9zBKF0twXUNACfdOnkosO99d8JqV0+JsYynrhP 0hkAoJgZzmfQAMcTpg8hehBhbZ/frb4M =XD5e -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 »Q« wrote: Wolf Canis [EMAIL PROTECTED] wrote: Would know a message reach the ML with my Name but no signature or a different signature, could one relatively be sure about the fact that this particular message is not from the original Wolf Canis. No, we'd have absolutely no way of telling whether or not it came from the original Wolf Canis. You could post using your usual signature, telling us the other one wasn't from you, but we'd have nothing to go on but your word. I think most of us /would/ take your word for it, but I doubt the signatures make a difference in that. That would mean that Wolf Canis is a bad boy and would have more than one signature, one for normal use and one or more for evil use. OK, if it's that what you mean, I understand it that way, then you are right. But I'm pretty sure that, if Wolf Canis comes with different signatures then it would be at least questionable and would probably lead to a ban, I think. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg+XMgACgkQKT9zBKF0twX1VwCeKc82++vshXFlOky3K0KEWVq1 Wd8AoII2bi7Ap7Za01PyetP+pMrCqYXM =OFkQ -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Daniel Iliev wrote: [...] Absolutely. I just wonder how many people will choose not to use such kind of list in order not to sacrifice their anonymity. Exactly. [...] It also might be the same person signing with different keys or sometimes signing somtimes - not. What's the difference for the other guys on the list - in both cases they will get some junk before the offending account is stopped. What's the difference for the sender - guilty or not, his address gets blacklisted. Correct. Signing makes only sense if you do it consistently. [...] Forgot, choosed not to, didn't renew... I believe it's the majority, but I may be wrong. OK, I forgot the human factor. ;-) [...] Relatively easy? Well, hereby I give you my blessing and dare you to send a proof of concept message to this list imposing as me. Additional condition: you must have no other access to Gmail than what is granted to everyone outside the company. If you succeed I promise to sign every single email I send from that point on. :) OK, I can't bring myself a proof of concept. I'm not a evil hacker. But I said relatively easy, I meant that if you have your own server running (with for example sendmail) and enough criminal energy, know how, I'm pretty sure that it's possible. And I'm also pretty sure that my thinking is much to complicated. Because e-mail abuse is not new and your proof of concept is probably since a long time ago produced. ;-) W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg+YNYACgkQKT9zBKF0twVe2QCfZJtt/Squj33IROJMnRNwDk4A 5ZEAn1mTDiyAa6bA7JYKiFE+9ZuaucIi =l5vv -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] chroot problem
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Peter Humphrey wrote: I have no problem chrooting into a system on the hard disk if I've booted from an installation CD, but every time I try it after booting from another HD partition I get e.g. this: # chroot /mnt/rescue /bin/bash chroot: cannot run command `/bin/bash': Permission denied Ls shows the same permissions in each case, and I always make sure to: # cd /mnt/rescue # mount -tproc proc proc # mount -obind /dev dev ...first. What am I doing wrong? Only for verification, have you under /mnt/rescue /bin/bash? Or with other words have this /mnt/rescue/bin/bash? And with the appropriate permissions? W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg+eZ0ACgkQKT9zBKF0twWTtwCdHIkXGHwaas50Zy2leKo5g6iU gP8AnRuiWCgemE/GFja4RaduEfcWp/9g =hplz -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Norberto Bensa wrote: Signed messages doesn't make any sense on a mailing list. I may ask you for a explanation, please? I think they make a lot of sense, because you or the mailing system are able to verify the message or rather the origin, if implemented. One would very easily see whether the person is the person who has subscribed to the list. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg5ldcACgkQKT9zBKF0twWpugCfXeAs+rrt1PkJSBcKFh8kEscb nMMAoIImyFjrBJ8rC39htY7FYCWnXDby =ccQk -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mick wrote: This is a nice list with helpful people. No doubt about that. :-) There are other lists however, when it is not that rare for malicious (or unhinged) individuals to impersonate someone else and hijack their email address to publish offensive content. After a while using a digital signature (GnuPG or x509) becomes a habit. That's exactly the case. ;-) It doesn't really add that much overhead anyway (197 Bytes for gpg to 3.1k Bytes for s/mime). That's what I thought. :-) W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg5qioACgkQKT9zBKF0twUtLACeIKqDkUvBYAMbdN8ZFVB4ujfi 4aMAn1KuvGPgRRNAleEZ2CyKAP5YK4lJ =wov3 -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Need help with a regex
Robin Atwood wrote: grep -e ^\s+provide\s+\w /etc/init.d but, as usual, nothing is matched. What am I doing wrong? You have to use back slashed versions of metacharacters. Following how would do that: $ grep -e '^[[:space:]]\+provide[[:space:]]\+[a-z]\+' /etc/init.d/* /etc/init.d/syslog-ng: provide logger /etc/init.d/vixie-cron: provide cron W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Need help with a regex
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Robin Atwood wrote: grep -e ^\s+provide\s+\w /etc/init.d but, as usual, nothing is matched. What am I doing wrong? You have to use back slashed versions of meta characters. Following how would do that: $ grep -e '^[[:space:]]\+provide[[:space:]]\+[a-z]\+' /etc/init.d/* /etc/init.d/syslog-ng: provide logger /etc/init.d/vixie-cron: provide cron W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg4dvkACgkQKT9zBKF0twURhwCZAQ/vK1la7X0mTuA2yTyWavqc rEIAn1yr+ytPejcRXMr/d6f/Wkg4SgSc =NQmh -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello all, it seems that sometimes mails of mine doesn't go to the list. :-( I had this problem just a couple of hours ago. I send a reply to the thread Need help with a regex but the mail doesn't reach the list. I looked in the archive and it doesn't reach there too. These mail was send with PGP/MIME. I send this message at 6:03 PM CET. At 10:13 PM CET I send the mail again but this time without PGP/MIME - and this time the mail reached the list. =-0 Now I'm wondering whether it could be that the list server has problems with those mails or perhaps those mails are simply blocked. Is there a problem with signed messages? Thanks in advance. W. Canis PS: Send at 11:12 PM CET without PGP/MIME -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg4hY8ACgkQKT9zBKF0twVwTQCfXhXFfWr4xhszNsXp/Y7tr842 h/wAn2yblYfRQ2hXqe7EhO86e3tJAGD+ =uC1C -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Mailing list and PGP/MIME
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mick wrote: I don't think that there is. I can see both of your messages in the Need help with a regex thread. Somewhat strange is it. On archives.gentoo.org the mentioned mails aren't, only the second, but on gmane they are. I just looked there. [...] Kmail shows this in its GnuPG header on the first message: Message was signed by [EMAIL PROTECTED] (Key ID: 0x293F7304A174B705). The signature is valid, but the key's validity is unknown. and the second message: Message was signed by Wolf Canis (Common) [EMAIL PROTECTED] (Key ID: 0xA174B705). The signature is valid, but the key is untrusted. That looks good. :-) Notwithstanding delays with googlemail, Gmane also takes some time before it shows posted messages. It seems that that is the case. But how know one that the mail is actually arrived? The list delivers not to sender of a post. This is normally absolutely correct on the one hand, on the other hand, if it would, one would know whether a mail has the list arrived or not. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkg4t3gACgkQKT9zBKF0twV97gCgh8xZ2IOQSCkRUMOKD8EEIePD Wq4AoI7uc1035kGSpwPNZKPJiMqG68nr =Ezrh -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] How to set package.use for layman overlays
[EMAIL PROTECTED] wrote: I've just started to use layman tools and wondered if setting such things as /etc/portage/package.use would still be done in that same place and same way? I think yes. I want to install an overlay of emacs-cvs but with different use flags I would say there is no problem, because in /etc/portage/package.use you can decide what version with what flags. I'm following along with the instruction at: http://www.enigmacurry.com/2007/05/24/multi-tty-emacs-on-gentoo-and-ubuntu/ On my laptop I run Gentoo Linux. Getting the latest version of Emacs on Gentoo was a breeze! : * Setup Layman * Add the emacs overlay: sudo layman -a emacs * Add the following USE flags for app-editors/emacs-cvs: sudo flagedit app-editors/emacs-cvs X Xaw3d alsa gif gzip-el jpeg lesstif png sound spell tiff toolkit-scroll-bars xpm -gtk -hesiod -motif -source. * GTK support is explicitly turned off as it causes problems with multi-TTY. This is no biggie for me as I always have (menu-bar-mode -1) and (tool-bar-mode -1) set. * Emerge: sudo emerge emacs-cvs -va * Tell the system to use the new emacs: sudo eselect emacs set emacs-23-multi-tty To get emacs-multitty set up. It isn't really clear what role layman plays in those instructions since the final command is emerge emacs-cvs That's, so far I that understand, the advantage for using overlays. Or will that automatically use the layman overlays. Yes. Because emacs-cvs is in the portage-tree but masked. Or maybe the author assumes I don't already have emacs installed from /usr/portage. If you want to install emacs-cvs from the portage tree, you have to unmask this package: $ emerge -atv emacs-cvs These are the packages that would be merged, in reverse order: Calculating dependencies / !!! All ebuilds that could satisfy app-editors/emacs-cvs have been masked. !!! One of the following masked packages is required to complete your request: - app-editors/emacs-cvs-23.0. (masked by: ~x86 keyword) - app-editors/emacs-cvs-23.0.50_pre20080201 (masked by: ~x86 keyword) - app-editors/emacs-cvs-22.2. (masked by: ~x86 keyword) For more information, see MASKED PACKAGES section in the emerge man page or refer to the Gentoo Handbook. I realize this is a little offhanded since its asking advice about 2nd party instructions. But I have no experience whatever with layman or using overlays at all. So thought maybe better to ask here than directly to the author of those instructions. That's the way I see it. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Gnome CD automount fails without error
Ian Hilt wrote: On Fri, 9 May 2008 at 5:53pm +0200, Wolf Canis wrote: Hello, I just experienced the same problem. Could you solve your problem? If so could you post the solution? Some data of mine: I'm in the groups: groups=10(wheel),11(floppy),18(audio),19(cdrom),27(video),35(games),85(usb),100(users),250(portage),443(plugdev),1000(rh),1001(wireshark) CD/DVD device: $ ls -l /dev/cdrom lrwxrwxrwx 1 root root 3 2008-05-09 04:45 /dev/cdrom - sr0 $ ls -l /dev/sr0 brw-rw 1 root cdrom 11, 0 2008-05-09 04:45 /dev/sr0 dbus and hal are running. USB disks, cameras or MP3-players are correctly mounted. Only CDs are not. Tips, comments highly appreciated. Thanks in advance. W. Canis What does /etc/fstab look like? If you're using hald to auto-mount, the cdrom should *not* be in /etc/fstab. Hello, thanks. /dev/cdrom was with the option noauto in fstab. I disabled this entry and it just worked fine. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Gnome CD automount fails without error
Hello, I just experienced the same problem. Could you solve your problem? If so could you post the solution? Some data of mine: I'm in the groups: groups=10(wheel),11(floppy),18(audio),19(cdrom),27(video),35(games),85(usb),100(users),250(portage),443(plugdev),1000(rh),1001(wireshark) CD/DVD device: $ ls -l /dev/cdrom lrwxrwxrwx 1 root root 3 2008-05-09 04:45 /dev/cdrom - sr0 $ ls -l /dev/sr0 brw-rw 1 root cdrom 11, 0 2008-05-09 04:45 /dev/sr0 dbus and hal are running. USB disks, cameras or MP3-players are correctly mounted. Only CDs are not. Tips, comments highly appreciated. Thanks in advance. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Which openoffice
Volker Armin Hemmann wrote: On Montag, 5. Mai 2008, Wolf Canis wrote: Volker Armin Hemmann wrote: On Montag, 5. Mai 2008, Wolf Canis wrote: Volker Armin Hemmann wrote: extremly long. So long that you have to start ooo several times a day for a year so that the saved startup time equalizes the time spent compiling it. ccache in make.conf is enabled and MAKEOPTS has a reasonable value, I have set it to -j2. I follow the rule MAKEOPTS=number CPUS. But in the case of openoffice, the ebuild overwrite this value with -j1. For the version 2.3.x I had set the variable WANT_MP but with version 2.4 it breaks the build. But how you can see in the following, that's only a minor problem. or not. So everything bigger than -j1 breaks the built. Which makes dual core cpus useless to speed up compilation. Not really, because if you have set -pipe in CFLAGS than you can easily, with top, check how the cpus are used. But that's it, of course. How I mentioned earlier with version 2.3.x I had set WANT_MP=true and MAKEOPTS=-j2 (and with my first builds -j4 and -j5 but that was pretty much useless, because the processes are hinder them self but they don't break the build) and that works for me. The only problem which occurred was this https://bugs.gentoo.org/show_bug.cgi?id=210065 wolf-di6400 0(0) 03:04 PM ~ # qlop -gH openoffice openoffice: Fri May 2 16:22:23 2008: 1 hour, 20 minutes, 38 seconds openoffice: Sat May 3 04:06:11 2008: 1 hour, 19 minutes, 12 seconds openoffice: 2 times emerge -p openoffice-bin|genlop -p These are the pretended packages: (this may take a while; wait...) [ebuild R ] app-office/openoffice-bin-2.4.0 Estimated update time: 2 minutes. Yeh, of course is that faster but why we use Gentoo? Because of the fast binary install? ;-) with packages that are only needed once in a while (ooo, frickelfox) binaries might be the right thing to do. How I said, everyone's own decision. I have compiled ooo in the past - on much, much slower machines. Ever compiled it on a 900mhz thunderbird? I did (and later faster cpus, of course). I don't know a machine with the name thunderbird :-[ . But I started with Gentoo on a Toshiba Tecra 8100, that's a PIII Copermine 800MHz and 512 MB RAM. In this respect, I can say: Yes, I did. :-) An emerge -e world lasted 11 hours, without OOO, OOO alone needs 16 hours to build, _but_ that, for me, was the fascinating thing - The build runs faultless, not even this strange segfaults of typesconfig. :-D Inclusive seeing it fail after 8h because the wrong java version was installed. It took less time to emerge ALL of kde than ooo. And one day I compared the differences. ooo started maybe 3 seconds faster than ooo-bin. As soon as started, no difference at all. That are bad experiences, but those things don't happened to me. Perhaps God has an eye on me. :-D For me isn't the start time of a program that important, but that all fits perfect together. That was not worth the trouble. In your case, maybe. Although I conduct all emerges at the console _not_ in X. Perhaps that's it. However, every user should do how he/she likes. it does not matter where - ooo is huge - bloated. And whereever you emerge it, it is the package needing the most time. That's absolutely right. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Which openoffice
Zdenek Travnicek wrote: I don't know a machine with the name thunderbird :-[ . But I started with Gentoo on a Toshiba Tecra 8100, that's a PIII Copermine 800MHz and 512 MB RAM. In this respect, I can say: Yes, I did. :-) An emerge -e world lasted 11 hours, without OOO, OOO alone needs 16 hours to build, _but_ that, for me, was the fascinating thing - The build runs faultless, not even this strange segfaults of typesconfig. :-D Cool! Yeah, and all couple of hours, I very carefully looked at the progress. ;-) One of my first compilations of OOo was on old Intel Celeron 400 for my parents and it took 44hours, and whole system (w/ X, FF, Tb, OOo) from stage1 exactly 5days (nearly 5x24 hours ;) And all the time the fear that the machine breaks or the build. ;-) Sweet old times :D It's loosing it's magic, when u can make it in 3 hours now ;-) Yup, but today we have to do other things too. My Gentoo box is my working machine too, therefore I'm really happy about the shorter build times. On my laptop (Dell Inspiron 6000) with [EMAIL PROTECTED], it still takes me around 13hours though... I guess that encrypted root (with /var/tmp) and swap does take it's price ;-) But even though I need to compile it overnight, it's still worth it. It's just the Right Gentoo Way (tm) :-D That's what I'm talking about. 8-) Although I conduct all emerges at the console _not_ in X. Perhaps that's it. However, every user should do how he/she likes. it does not matter where - ooo is huge - bloated. And whereever you emerge it, it is the package needing the most time. That's absolutely right. +1 P.S. 389.9cm :-)) 114.2cm Have fun, W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Re: grub weirdness
Sven Köhler wrote:
When you emerged grub-0.97-r5, this was displayed on your console:
WARN: postinst
*** IMPORTANT NOTE: you must run grub and install
the new version's stage1 to your MBR. Until you do,
stage1 and stage2 will still be the old version, but
later stages will be the new version, which could
cause problems such as an unbootable system.
Yes, the ebuild writes that to the screen.
But silently, in the background (because every output is piped to
/dev/null - how evil!), the ebuild calls grub with some commands
inside your grub.conf.
I just updated grub to version 0.97-r5 and this was,
at the end, displayed:
To avoid automounting and autoinstalling with /boot,
just export the DONT_MOUNT_BOOT variable.
Your boot partition was not mounted as /boot, but portage
was able to mount it without additional intervention.
Files will be installed there for grub to function correctly.
*** IMPORTANT NOTE: you must run grub and install
the new version's stage1 to your MBR. Until you do,
stage1 and stage2 will still be the old version, but
later stages will be the new version, which could
cause problems such as an unbootable system.
Copying files from /lib/grub and /usr/lib/grub to //boot/grub
To install grub files to another device (like a usb stick), just run:
emerge --config =grub-0.97-r5
If there's a setup-command in your grub.conf, it is indeed executed.
So if that command is outdated (something you won't notice, since that
command is not used by grub in any situation i know), the ebuild will
execute that setup-command and write to some device's boot sector. How
evil, again!
Regards,
Sven
P.S.: here's the code from grub-0.97-r5.ebuild:
if [[ -e ${dir}/grub.conf ]] ; then
egrep \
-v
'^[[:space:]]*(#|$|default|fallback|initrd|password|splashimage|timeout|title)'
\
${dir}/grub.conf | \
/sbin/grub --batch \
--device-map=${dir}/device.map \
/dev/null
fi
And following the code of the functions which does the job:
found in ebuild: /usr/portage/sys-boot/grub/grub-0.97-r5.ebuild
setup_boot_dir() {
local boot_dir=$1
local dir=${boot_dir}
[[ ! -e ${dir} ]] die ${dir} does not exist!
[[ ! -L ${dir}/boot ]] ln -s . ${dir}/boot
dir=${dir}/grub
if [[ ! -e ${dir} ]] ; then
mkdir ${dir} || die ${dir} does not exist!
fi
# change menu.lst to grub.conf
if [[ ! -e ${dir}/grub.conf ]] [[ -e ${dir}/menu.lst ]] ; then
mv -f ${dir}/menu.lst ${dir}/grub.conf
ewarn
ewarn *** IMPORTANT NOTE: menu.lst has been renamed to grub.conf
ewarn
fi
if [[ -e ${dir}/stage2 ]] ; then
mv ${dir}/stage2{,.old}
ewarn *** IMPORTANT NOTE: you must run grub and install
ewarn the new version's stage1 to your MBR. Until you do,
ewarn stage1 and stage2 will still be the old version, but
ewarn later stages will be the new version, which could
ewarn cause problems such as an unbootable system.
ebeep
fi
einfo Copying files from /lib/grub and /usr/lib/grub to ${dir}
for x in ${ROOT}/lib*/grub/*/* ${ROOT}/usr/lib*/grub/*/* ; do
[[ -f ${x} ]] cp -p ${x} ${dir}/
done
if [[ -e ${dir}/grub.conf ]] ; then
egrep \
-v
'^[[:space:]]*(#|$|default|fallback|initrd|password|splashimage|timeout|title)'
\
${dir}/grub.conf | \
/sbin/grub --batch \
--device-map=${dir}/device.map \
/dev/null
fi
# the grub default commands silently piss themselves if
# the default file does not exist ahead of time
if [[ ! -e ${dir}/default ]] ; then
grub-set-default --root-directory=${boot_dir} default
fi
}
How you can see isn't the message piped to /dev/null, only
the command /sbin/grub -batch -device-map
Have fun,
W. Canis
signature.asc
Description: OpenPGP digital signature
Re: [gentoo-user] Which openoffice
Volker Armin Hemmann wrote: extremly long. So long that you have to start ooo several times a day for a year so that the saved startup time equalizes the time spent compiling it. I have to disagree. On my laptop Dell Inspiron 6400, Dual Core Pentium (T2130) 1.8 GHz, 2 GB RAM, 160 GB HD. The compile time is absolutely OK. Important is that the feature ccache in make.conf is enabled and MAKEOPTS has a reasonable value, I have set it to -j2. I follow the rule MAKEOPTS=number CPUS. But in the case of openoffice, the ebuild overwrite this value with -j1. For the version 2.3.x I had set the variable WANT_MP but with version 2.4 it breaks the build. But how you can see in the following, that's only a minor problem. wolf-di6400 0(0) 03:04 PM ~ # qlop -gH openoffice openoffice: Fri May 2 16:22:23 2008: 1 hour, 20 minutes, 38 seconds openoffice: Sat May 3 04:06:11 2008: 1 hour, 19 minutes, 12 seconds openoffice: 2 times wolf-di6400 0(0) 03:04 PM ~ # emerge -avt openoffice These are the packages that would be merged, in reverse order: Calculating dependencies... done! [ebuild R ] app-office/openoffice-2.4.0 USE=binfilter cups dbus firefox gnome gstreamer gtk java ldap mono odk opengl pam -debug -eds -kde -seamonkey -webdav -xulrunner LINGUAS=-af -ar -as_IN -be_BY -bg -bn -br -bs -ca -cs -cy -da -de -dz -el -en -en_GB -en_US -en_ZA -eo -es -et -fa -fi -fr -ga -gl -gu_IN -he -hi_IN -hr -hu -it -ja -km -ko -ku -lt -lv -mk -ml_IN -mr_IN -nb -ne -nl -nn -nr -ns -or_IN -pa_IN -pl -pt -pt_BR -ru -rw -sh -sk -sl -sr -ss -st -sv -sw_TZ -ta_IN -te_IN -tg -th -ti_ER -tn -tr -ts -uk -ur_IN -ve -vi -xh -zh_CN -zh_TW -zu 0 kB Total: 1 package (1 reinstall), Size of downloads: 0 kB If you are even able to. Openoffice is a bitch to compile. Even the slightest change might break the compilation. It really, really sucks. IMHO openoffice is a nice example for everything that is wrong. I can't this confirm. Go with openoffice-bin. See above. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Which openoffice
Volker Armin Hemmann wrote: On Montag, 5. Mai 2008, Wolf Canis wrote: Volker Armin Hemmann wrote: extremly long. So long that you have to start ooo several times a day for a year so that the saved startup time equalizes the time spent compiling it. ccache in make.conf is enabled and MAKEOPTS has a reasonable value, I have set it to -j2. I follow the rule MAKEOPTS=number CPUS. But in the case of openoffice, the ebuild overwrite this value with -j1. For the version 2.3.x I had set the variable WANT_MP but with version 2.4 it breaks the build. But how you can see in the following, that's only a minor problem. or not. So everything bigger than -j1 breaks the built. Which makes dual core cpus useless to speed up compilation. Not really, because if you have set -pipe in CFLAGS than you can easily, with top, check how the cpus are used. But that's it, of course. How I mentioned earlier with version 2.3.x I had set WANT_MP=true and MAKEOPTS=-j2 (and with my first builds -j4 and -j5 but that was pretty much useless, because the processes are hinder them self but they don't break the build) and that works for me. The only problem which occurred was this https://bugs.gentoo.org/show_bug.cgi?id=210065 wolf-di6400 0(0) 03:04 PM ~ # qlop -gH openoffice openoffice: Fri May 2 16:22:23 2008: 1 hour, 20 minutes, 38 seconds openoffice: Sat May 3 04:06:11 2008: 1 hour, 19 minutes, 12 seconds openoffice: 2 times emerge -p openoffice-bin|genlop -p These are the pretended packages: (this may take a while; wait...) [ebuild R ] app-office/openoffice-bin-2.4.0 Estimated update time: 2 minutes. Yeh, of course is that faster but why we use Gentoo? Because of the fast binary install? ;-) If you are even able to. Openoffice is a bitch to compile. Even the slightest change might break the compilation. It really, really sucks. IMHO openoffice is a nice example for everything that is wrong. I can't this confirm. go to b.g.o and see the countless reports. Or the forums. Or have a look at the ebuild for all the crap that is there just to get the POS ooo compiled. I can only repeat that this doesn't apply to me and I'm pretty sure that I'm not the only one who don't use the bin-pkg. Although I conduct all emerges at the console _not_ in X. Perhaps that's it. However, every user should do how he/she likes. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Which openoffice
Mick wrote: You people don't know what pain means! :-)) -- [ebuild R ] app-office/openoffice-bin-2.4.0 Estimated update time: 5 minutes. -- [ebuild N] app-office/openoffice-2.4.0 Estimated update time: 23 hours, 12 minutes. -- The funny thing is that this PIII laptop is *significantly* faster than the PIII desktop that had its MoBo blow up on me. The irony of course is that the compiled from source is most needed on those machines that take the longest to emerge. Yeh, I know that. My first install was on a laptop Toshiba Tecra 8100. But it _always_ works fine. W. Canis signature.asc Description: OpenPGP digital signature
[gentoo-user] Only a - in top column WCHAN since downgrade to gcc 4.1.2
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello, since I gcc downgraded to version 4.1.2 I have in the top column WCHAN only a -. This applies to all processes. The System.map is available in /usr/src/linux. It works after installation, that was late 2007, with gcc 4.1.2 and it works with gcc 4.2.2 but I had to downgrade to gcc 4.1.2 because of build problems of firefox 2.0.0.14. I tried a symbolically link of System.map to the root directory. I also tried it with the boot partition permanently mounted and I re-emerged sys-process/procps-3.2.7, where top belongs to, too. But the behaviour of top is always the same, WCHAN shows only a -. Has someone encountered a similar behaviour with top? Or does someone know what I'm missing? Thanks in advance W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgd4t4ACgkQKT9zBKF0twXdgwCbBPY5so2UTKxWmblmS8WNvV7d YoIAnREiSL1QuDcDQYLMyP9r+G86zeUM =XxFw -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] man pages not displaying right - SOLVED
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] wrote: | I searched for NROFF in /etc/man-conf and found a note saying to add | -c if something had a specific version. I tried that and it works | now. There may be otehr fixesm but that works for me. Just edit it | and look for NROFF, it's in a comment. Hello, you are right it's in the comment. After adding -c to the commands NROFF, TROFF and JNROFF, the man pages are looking fine again (with vimmanpager). Thanks W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgazokACgkQKT9zBKF0twW8NQCbBqBlZUKdJYLLZQwBoiBytka/ DpoAoJXxmilxSU0LCKb/iKZ0zJOJo7Qx =rDlQ -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] checking for.....
[EMAIL PROTECTED] wrote: In the middle of doing a major upgrade from very old pkgs to current 2008 and compiling lots and lots of stuff. Seeing that line `checking for WHATEVER' go by 486,211 times so far makes me wonder if there wouldn't be someway to cache all those answers somewhere so whatever test is done for each line could be dispensed with for most of them. Probably would need more than 2-3 compiles to have all but rare ones answered. Some items really check a lot of things. I think it would be a major time saver when discussing huge numbers of compiles. Hello, ccache does caching, I use it and I'm very satisfied. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] checking for.....
Brandon Mintern wrote: ccache caches the compile step. I believe the OP was specifically looking for something that would cache the answers to the checking for lines (the configuration step). Yes, you are right, but I thought that ccache cached parts of the configuration too. That's what I noticed in outputs during the build process. Perhaps my conclusion is wrong. W. Canis signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] man pages not displaying right - SOLVED
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Michael George wrote: | | It was /etc/man-conf. A change in the NROFF definition caused the | problem. Running dispatch-conf didn't prompt me for the config change, | so I ran etc-update this time and found it. | | Hopefully this will be helpful for someone. Hello, yes it's helpful, but I used etc-update and was to fast and can't remember what are the changes. Could you give me an idea of the working NROFF definition? For now, I solved it by changing the MANPAGER variable to less. I had set MANPAGER to vimmanpager. Perhaps I have a additional problem? Thanks in advance. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgZ69gACgkQKT9zBKF0twU2owCeIpC+MkeXvQ7CJYzA7zIBHifk 6RsAnAteCGq9NKoeH9+El/OzzzoX90XV =o3kA -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] -fomit-frame-pointer switch
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 James wrote: | Hello, | | On an amd64, If I want to add -fomit-frame-pointer to a system's | CFLAGS setting, I can just add it and eventually all of the | executables will be recompile (willing to wait) | or do I have to rebuild system (all packages) or such to switch? Hello James, you have to rebuild the entire system, if the new CFLAGS settings shall have effect. # emerge --emptytree system emerge --emptytree world That's the recommend procedure, according to the handbook, to do that. But there are in the forums a big thread whether that is necessary or not. Some argue - is not, it's sufficient to rebuild the toolchain and than emerge -e world. Others say one should follow the recommend procedure. I follow the recommend procedure. Hope that helps. W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgZ9lwACgkQKT9zBKF0twU8hgCgmSyN7hTtp+FWizv/yVZXaTyL 2u8An0QMe2QaJUsxymGNTPo8QuDN+MUy =BHyv -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: -fomit-frame-pointer switch
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 James wrote: | | But in order to avoid recompiling all of those packages (for now) | I can just add it to my CFlags and wait a few months, as another option? Yes, there shouldn't any problems appear. | | Or is there real peril with this approach to slowly converting a system? I think no. But I'm not that Guru, I'm not sure whether there is package or dependency which if new compiled failed or lead to any instability. If you need that system for something important, I wouldn't recommend to experiment with CFLAGS. I would do that in a time window where I would have a couple of days time. That's what I would do. Because you can never know for sure whether it comes to problems or not. ;-) W. Canis -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgaAz0ACgkQKT9zBKF0twXFHACfTBNKv0j9s6Bo3cPmChf72X4x +5sAmwcqUWggUQhC98o1duN0r/ylXsBH =93sI -END PGP SIGNATURE- -- gentoo-user@lists.gentoo.org mailing list
