- CRUX - lists.crux.nu

blocksize
by Matthias Pfeifer 08 Sep '06

08 Sep '06

Hello list, What would be a good number for the blocksize of an ext3 root-filesystem. My root-fs is 4.7 Gb and it keeps everything instead of home. After installation of crux and some additional packages (linux, gnuplot, tetex, prolog, xemacs, lmbench) and keeping the sources - there are 3.5 Gb in use (2.0 Gb for sources where only tetex takes 865 Mb). Are these common values? Is it possible that these numbers are too big and that this is due to my blocksize? I guess i have a default value... A directory node taks 4k. Is this my blocksize? What would be a good value? How do i figure out the exact value? How is the blocksize related to the pagesize? Do they have to be equal? Is there a problem with a too low blocksize, is there any? Please help.. matthias pfeifer PS: Is there a way to search the list-archive?

2 5

Server Migration documentation rsync update
by Clemens Koller 07 Sep '06

07 Sep '06

Hello! I've just tempted to follow http://crux.nu/Main/ServerMigration to get an pretty old CRUX2.1 server updated from the old ports collection to the new one. The ports.prt.tar.gz part works fine but rsync-2.6.7 is not available at samba.org at the mentioned place. It's 2.6.8 now. After updating the Pkgfile and the md5, everything seems to work fine. Can you please update your rsync.prt.tar.gz? Attached are my changes. Beside that, the document is fine! Next step would be a CRUX 2.1 to CRUX 2.2 migration. A link to some documentation for that would be fine... Best greets, -- Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19 --- Pkgfile~ 2006-03-13 22:15:05.000000000 +0100 +++ Pkgfile 2006-08-08 10:39:18.000000000 +0200 @@ -4,7 +4,7 @@ # Depends on: openssh name=rsync -version=2.6.7 +version=2.6.8 release=1 source=(http://samba.org/ftp/$name/$name-$version.tar.gz \ rsyncd.conf rsyncd rsync.driver) 082a9dba1f741e6591e5cd748a1233de rsync-2.6.8.tar.gz 7e01fb425c40a8c10c86f3a80aec9e41 rsync.driver a71995f22768c931c5649a1336d25ffb rsyncd b4e95fa8c8f3ae13cfdf616abd6a3960 rsyncd.conf

10 26

CRUX from LFS
by Clemens Koller 07 Sep '06

07 Sep '06

Hi There! I want to migrate a selfmade proprietary LFS embedded powerpc system to follow the CRUX look&feel by hosting port repositories for my architecture and use the crux package tools to keep several of those machines in sync and up to date. 1. Is there some documentation available for the lazy guys (i.e. from the CRUX-PPC folks) to avoid duplicate brainwork? Any pointers would be appreciated. 2. Is it possible / a good idea to build the ports db by package-by-package migration (installation / update) via my own ports tree just starting it, installing over the existing system? And once this is finished, is it a good idea to remove the files carefully from the filesystem which are not part of a package in the db? Is there a way to show/remove stale files from the system which are not needed (not in the db) anymore? Thanks, -- Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19

3 3

Re: CRUX from LFS
by Clemens Koller 07 Sep '06

07 Sep '06

Hi, Giorgio Giorgio Agrelli wrote: > Hello Clemens, > Unfortunately we don't have a mailing list... we might open one international > mailing list on slackware.it, i have to ask a friend... Do you think it could > be helpful?? I'm not sure about how many persons would subscribe to it, > probably very few... but if that could help the development and the users... > well, let's do it! ;-) Hmm, I prefer mailing lists because the main OpenSource development communication is going on in mailing lists. I am subscribed to about 15 lists. Checking 15 different places for information is just not efficient for me. >> (BTW: cruxppc.sunsite.dk is way too slow for serious work over here) > > > are you talking about the forum??? well the new look is really ugly i think, > i've been asking giulivo to change that but he became lazy lately... :-( > But it doesn't look so slow to me... I don't care much about the design, as long as the content is okay. But the speed is reallly far off comfortable! It takes about 10 seconds to load http://cruxppc.sunsite.dk/wp/index.php to see the content, about 10 more seconds to finish loading with the white background. Reading the forum means getting a coffee-addict first. I see no way to get me on that train. >> I think that my system - an MPC8540 embedded PPC (BookE, e500 core) >> should be able to get along with the default ppc stuff most of the time. > > > I hope it will! :) anyway if you want to join the development team, you're > welcome... Thanks! > we should change quite a bit of things (site, forum, cvs accounts > on sunsite) to manage working together, but i hope we will eventually manage > to have things fixed. I hardly work with cvs. I prefer git which is more powerful. As already mentioned, I think about getting a root-server for all my stuff. Hmm... pkgtools, prt-get, httpup are working fine - out of the x86 tree. I just have to manually get rid of my old out-of-db files first when I install packages. An example of the new ppc stuff as of today: root@ecam:/usr/ports/core/udev$ pkgmk -i =======> Package '/usr/ports/core/udev/udev#096-4.pkg.tar.gz' is up to date. =======> Installing '/usr/ports/core/udev/udev#096-4.pkg.tar.gz'. pkgadd /usr/ports/core/udev/udev#096-4.pkg.tar.gz etc/udev/rules.d/25-lfs.rules etc/udev/udev.conf sbin/udevcontrol sbin/udevd usr/bin/udevinfo usr/bin/udevtest usr/sbin/udevmonitor pkgadd: listed file(s) already installed (use -f to ignore and overwrite) =======> ERROR: Installing '/usr/ports/core/udev/udev#096-4.pkg.tar.gz' failed. root@ecam:/usr/ports/core/udev$ prt-get listinst httpup ports prt-get Is there already a way to get the conflicting files out of my LFS-tree moved to i.e. var/lib/pkg/removed/... and with the names written to a logfile? The -f is too destructive for me, I want to be able to go back to the previous version if I run into troubles. > The point is that i suspect giulivo is not interested any more in the > development of crux-ppc; in the past he destroyed big part of the work i did > on 64bit systems and on documentation (and of course i was mad for that so for > a long while i wasn't willing to do anything for crux-ppc) to have something > similar to crux 2.2 the quick and dirty way. Then he abandoned the thing > again, and left the cvs in a complete mess and i wasn't able to work with all > those screwed up files any more... > Well this stuff is irrelevant to you i'm telling just to explain our > "lazyness" lately... Okay, I understand. Time squeezes all men under it's foot like a bug. Just to give you a focus: For me, I am doing this (ppc stuff) at work for the company in this project (an embedded mobile high performance framegrabber). We have some Mac G4's here as well, with CRUX-2.0 on it. There was also a G5/64 but no CRUX on it yet due n/a of a distribution. I do have several x86 and x86-64 with crux on it at the company and at home. > The cvs tree is really messy now, anyway you should only look at ppc directory > now and gcc64 if your processor is 64bit (core and opt are obsolete); with the > new tree layout we only have ports that differ from the x86 tree (we used to > have all the ports, even the ones that were identical). I am 32bit only. We don't have a G5/64 here yet. I am just about to start a new tree... my toolchain (arch=ppc, -me500) (selfmade from scratch) will be in there as well in the future. > If you look at the cvs comments, sometimes there is an hint about the > differences... i proposed to add a field in the pkgfile to explain that but > none wanted to do that so we didn't do that any more... That's a pity. That's exactly the information which makes me happy. :-( > Anyway on the toolchain stuff and on other things you can ask me by mail if > you wish. Jep... currently, I have to get an efficient (hacked) solution as quick as possible as we want to go to a fair by sept. 26 and some new hardware should be up and running also. But after that and some vacation, I will be around to spend some time to crux, too, hopefully. :-) Best greets, Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19

1 0

stale 50-udev.rules in /etc/udev/rules
by Clemens Koller 06 Sep '06

06 Sep '06

Hi There, I updated one of my machines from the crux-2.2 installation. Since about a week, after some udev updates, udevtrigger started to yell with something like: udevd-event[8578]: run_program: 'udev_run_hotplugd input' failed udevd-event[8578]: run_program: '/lib/udev/udev_run_hotplugd' failed The reason was a stale /etc/udev/rules.d/50-udev.rules which are some outdated gentoo specific rules. Removing the file solved those problems. I'm not sure if this happens more frequently out there. If so, we should put that into the README or modify the Pkgfile to move that file to rejected / removed. Best Greets, -- Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19

1 0

[security] crux-2.2/ports/core/openssl/
by crux＠crux.nu 06 Sep '06

06 Sep '06

This is a notification of a security related commit in crux-2.2/ports/core/openssl/: update openssl to 0.9.8c: http://www.openssl.org/news/secadv_20060905.txt Commit information: Author: jw Date: 2006-09-06 11:12:20 +0200 (Wed, 06 Sep 2006) URL http://crux.nu/svnweb/CRUX/revision/?rev=1821

1 0

Re: VFS: Mounted root (reiserfs filesystem) readonly
by Theo G. Kanter 01 Sep '06

01 Sep '06

> The readonly mount from the kernel (VFS) is the expected behaviour. What > problems are you seeing you'd like to fix? > > Regards, Johannes I wasn't seeing any, but dmesg from basically the same CRUX2.2 on my other machine did not contain this same message. This I found suspicious combined with newsgroups mentioning quite a few struggling with such messages but no real clues. Thank you for clarifying this. best regards Theo

2 1

VFS: Mounted root (reiserfs filesystem) readonly
by Theo G. Kanter 01 Sep '06

01 Sep '06

I installed Crux 2.2 on a new machine, copying the configuration of the existing machine. The only difference being that this one is a double boot with Grub. Crux boots fine but at boot time the following message succeeds the ReiserFS messages but *only* for the root partition (/dev/hda6): VFS: Mounted root (reiserfs filesystem) readonly Would very much appreciate hints as to what might caused this and what I need to do to fix this. The only thing I can think of is that I went wrong somewhere with Grub, but I cannot see what would be wrong with (omitting non relevant parts): title CRUX kernel (hd0,5)/vmlinuz root=/dev/hda6 Nor can I see something particularly out of the ordinary in /etc/fstab: # # /etc/fstab: static file system information # # <file system> <dir> <type> <options> <dump> <pass> /dev/hda6 / reiserfs defaults 1 0 /dev/hda5 swap swap defaults 0 0 /dev/hda7 /var reiserfs defaults 1 0 /dev/hda8 /home reiserfs defaults 1 0 /dev/hda1 /boot reiserfs defaults 1 0 /dev/cdrom /mnt/cdrom iso9660 ro,user,noauto,unhide 0 0 devpts /dev/pts devpts defaults 0 0 sysfs /sys sysfs defaults 0 0 proc /proc proc defaults 0 0 usb /proc/bus/usb usbfs defaults 0 0 # End of file --theo

2 1

Setting up sendmail
by Helge Fredriksen 31 Aug '06

31 Aug '06

Hello again, I'm trying to setup sendmail on my dyndns address so that people in my company might use it as a smtp server. Is this possible with crux? How do I open up for externally use, possibly requiring some authentication to avoid misuse. I know this doesn't have directly with the crux kernel to do, but maybe you could point me to some "newbie" docs on this, I found the document http://www.sendmail.org/doc/sendmail-current/doc/op/op.pdf, but the document were just overwhelming... Helge

2 1

Trying to use crux in a Tho
by Helge Fredriksen 31 Aug '06

31 Aug '06

Hello, I've been trying lately to attach my crux server to a network with a Thompson SpeedTouch ADSL router as a backbone. I want to have a static IP-address to it, and use this /etc/rc.d/net $ more /etc/rc.d/net #!/bin/sh # # /etc/rc.d/net: start/stop network # case $1 in start) /sbin/modprobe 3c59x /sbin/ifconfig lo 127.0.0.1 /sbin/ifconfig eth0 10.0.0.50 netmask 255.255.255.0 /sbin/route add default gw 10.0.0.138 # /sbin/dhcpcd eth0 ;; stop) # killall -q /sbin/dhcpcd /sbin/ifconfig eth0 down /sbin/ifconfig lo down ;; restart) $0 stop $0 start ;; *) I'm managing to register internally with the SpeedTouch, and to ping all local machines on that network. I'm also able to access this machine from other PC's on the network, so things look promising. However, I'm not able to reach internet like I do from other Windows PC's on the same network. How come? Anything missing from my network setup? Best regards, Helge Fredriksen

2 1