[Cooker-firewall] Framework Programmation Document
Hello guys, For those interested in Administration tools, in particular the Network Appliance Administration Tool used in the Firewall, i provide the Framework Presentation and programming documentation. You can find it here: http://people.mandrakesoft.com/~philippe/ We are working on new version, implementing new features, new design... Main Evolutions: -migrate to full XML representation and communication.(Backend / Frontend) -distributed configuration. -new frontend generation. Feel free to ask questions, contribute and give us your feedback. Sincerely -- Philippe Libat [EMAIL PROTECTED] Linux-Mandrake http://www.linux-mandrake.com _ Think Different, Think Linux
Re: [Cooker-firewall] Cooker 8.2
Phil Connor [EMAIL PROTECTED] writes: Ok, what am I doing wrong? I've tried installing the snf from the cooker and have not yet gotten it to work. I'm choosing expert install, unselecting everything, and then selecting snf-en as the install! No matter how you select the options it still does not work. Can anybody point me in a semi right direction PLEASE! Hi there, what is your problem actually ? I'll need some details on your prb in order to be able to eventually help you :) cheers, -- Florin http://www.mandrakesoft.com
RE: [Cooker-firewall] New packages available
Is there any information on a the new version of SNF officially being release? Also, is there a feature list or list of new features compared to the new version. I saw a posting before where you said there is a version on Cooker, as well as a newer version than that. Can you give us an idea of a release schedule for these? Thanks. I'm using SNF 1.0 now, and it's great. Very easy to administer, although I have had trouble with running a DNS server behind it. I understand this will be fixed in the new version? Thanks, Michael Segulja LMD Computing|Solutions -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, February 20, 2002 12:38 PM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: [Cooker-firewall] New packages available Hello everyone, We have uploaded some new packages concerning the new Mandrake Security Firewall. You can find them in cooker (any mirror will do) as snf-language packages, or on http://people.mandrakesoft.com/~florin/www/rpms for the firewall packages only. We hope that you will give it a try. Have a nice day, -- Florin http://www.mandrakesoft.com
Re: [Cooker-firewall] REPOST: Linux Hardening
Sean Bybee [EMAIL PROTECTED] writes: Are there any plans to include kernel hardening products in future SNF/Mandrake releases besides Bastille Linux? Like the NSA's Security-Enhanced Linux kernel, LIDS, or something similar to these products? I might be totaling off here, but it would seem that using Mandatory Access Control (MAC) in a firewall would improve/enhance security. we're using some security kernel patchesthat you can find on http://www.grsecurity.net/features.htm -- Florin http://www.mandrakesoft.com
Re: [Cooker-firewall] IPsec VPN clients through a SNF
Brent Baude [EMAIL PROTECTED] writes: Does the new cooker version support a windows client initiating a IPSec VPN that goes through SNF to a VPN server? This is sometimes known as VPN Masquerade. I'm working on the VPN part right now ... It will be finished until the next week :) sincerely, -- Florin http://www.mandrakesoft.com
RE: [Cooker-firewall] Cooker 8.2
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Thursday, February 21, 2002 3:36 AM To: [EMAIL PROTECTED] Subject: Re: [Cooker-firewall] Cooker 8.2 what is your problem actually ? I'll need some details on your prb in order to be able to eventually help you :) cheers, -- Florin http://www.mandrakesoft.com Thanks Florin, I'm not able to bring up the login page at all, Not sure naat is working at this point, doing A status on httpd-naat says it not running although extendedstatus says it's waiting? Did I miss something else to be done after fixing bastille and running the change password? Phil Connor [EMAIL PROTECTED] Registered Linux User # 189889 IT Services - Making IT a success at The Emory Booty Co.
[Cooker] ftp in rpmdrake not working
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 It seems that I cannot use ftp in rpmdrake: added medium contribs retrieving description file of contribs... curl: (6) Couldn't resolve host 'ftp' retrieving source hdlist (or synthesis) of contribs... curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' retrieve of source hdlist (or synthesis) failed no hdlist file found for medium contribs unable to update medium contribs Oh..and ofcourse...I didn't type ftp in the URL bar. Tried both ftp://ftp.surfnet.nl/somepath/ and ftp.surfnet.nl/somepath/ Danny - -- `...and the Universe,' continued the waiter, determined not to be deflected on his home stretch, `will explode later for your pleasure.' Ford's head swivelled slowly towards him. He spoke with feeling. `Wow,' he said, `What sort of drinks do you serve in this place?' The waiter laughed a polite little waiter's laugh. `Ah,' he said, `I think sir has perhaps misunderstood me.' `Oh, I hope not,' breathed Ford. - - Ford in paradise. -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE8c/paaeiN+EU2vEIRAq3qAJ946knIOXzAkcoOel1OLTyr4ySWIQCeISop 1+qp4J3xcodWP4kEYrRWkBk= =Wu9L -END PGP SIGNATURE-
[Cooker] drakfont is still ignoring capitals+ntfs
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I had a look at drakfont to fix it's behaviour. And although I know what is wrong with
it, and were,
I have to little knowledge of perl to fix it (exept maybe in a crude way, but that
would lead to other errors).
Anyway, I decided perl is terrible and hope that this gets fixed:
sub is_a_font {
local $_ = $_[0];
/.ttf$/i || /.pfa$/i || /.pfb$/i || /.pcf$/i || /.pcf.gz$/i || /.pfm$/i || /.gsf$/;
}
what if the font is TTF?? My Verdana is! And it didn't get installed.
Ofcourse, if I add TTF (and others) here, the comparing to installed font goes wrong.
Who solves it?
also:
sub search_windows_font {
foreach my $fstab_line (grep { /vfat/ } cat_('/etc/mtab') ) {
I could imagine some winnt people use ntfs? which is readable so we can copy those
fonts as well.
Danny
- --
You're one hundred percent positive that the ship which is
crashed on the bottom of this ocean is the ship which you
said you were one hundred percent positive could one
hundred percent positively never crash?
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8c/8naeiN+EU2vEIRAmqpAJ94whq4KDkfUonVIsQM1g+EcaYcegCggGAH
GwxZmswFPLXG1e/XqIc03ms=
=UMEy
-END PGP SIGNATURE-
[Cooker] Tonights Install - DrakX 1.658
Found a few problems: GUI Install: a) Advanced screen has 'security' spelt wrong (stage1) b) When using DHCP for network settings, the installer (stage2) does not ask for a hostname if I simply accept on the initial panel. This results in a host name of localhost, and more importantly, the DHCP_HOSTNAME entry in /etc/sysconfig/network is missing. However, if I cancel, then select lan, dhcp then it asks and everything is OK. c) CUPS startup leaves a defunct modprobe process. It looks like the problem is occuring when probing for usb devices in progress (dynamic/script??). I do not have a local parallel printer and no usb devices, just a Windows shared printer else where on the network; however both printer and usb hardware are enabled in bios. AutoInstall: a) lots and lots and ... of rpmdb: var/lib/rpm/Basesnames: No such file or directory messages on the primary console during the install. b) During installation, msec is changing the the permissions of the lilio.conf file from 600 to 644 when I have the keyword 'password' is in it: DrakX sets it correctly. The result is a Warning message from lilo that lilio.conf should be readable by root only. b) lilo is reporting a warning that the 'menu-scheme' is not supported by the boot loader. Best regards Dave
[Cooker] Move RPM_CHECK to msec 4?
Hi, How insecure is your system without RPM_CHECK=yes? Would it be dangerous to move it to msec level 4? I think about all the discussions about rpmv. Regards, Mattias
[Cooker] strange kernel issue
I had originally compiled 2.4.17-20mdk using CONFIG_MPENTIUMIII=y. While using that kernel, I would boot up to a situation where the ram was practically used up after bootup (512mg), I then had X crash twice because ram and swap (another 518mg) were completely at their limits. Upon recompile using the CONFIG_M586=y it appears to have RAM usage back to normal. very odd = SI Reasoning [EMAIL PROTECTED] A requirement of creativity is that it contributes to change. Creativity keeps the creator alive. -FRANK HERBERT, unpublished notes __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com
Re: [Cooker] Install report
Le Mercredi 20 Février 2002 10:20, Stéphane Teletchéa a écrit :
I installed the beta 2, and as the mouse was not properly detected (a
PS/2). (the step was skipped)
So at the end, it selected the mouse selection step, and got this message :
'une erreur est survenue:
setstep selectMouse
install-gtk::__ANON__('Gtk::DrawingArea=HASH(0x9d017ac)',
'Gtk::Gdk::Event=HASH(0xa171b5c)') called.
So after install (again thanks for disk in lilo.conf), i tried to use mcc,
and fot the mouse selection i got :
Glib_CRITICAL**: file gmain.c : line 500 (g_source_remove): asertion
'tag0' failed.
Stef
Missing : XmGrace (grace) package.
Still in beta3..
Stef
[Cooker] INSTALL REPORT
About beta 3 : during install : Warning : loading /tmp/af_packet.o will taint the kernel : forced load Warning : loading /tmp/serial.o will taint the kernel : forced load Warning : loading /tmp/parport.o will taint the kernel : forced load Warning : loading /tmp/parport_pc.o will taint the kernel : forced load Warning : loading /tmp/imm.o will taint the kernel : forced load /tmp/imm.o : init_module : No such file or directory Warning : loading /tmp/ppa.o will taint the kernel : forced load /tmp/ppa.o : init_module : No such file or directory Warning : loading /tmp/fat.o will taint the kernel : forced load Warning : loading /tmp/vfat.o will taint the kernel : forced load Warning : loading /tmp/reiserfs.o will taint the kernel : forced load Furthermore, why does it complains about kernel tainting ? (what does that mean ?) Another error message (after boot) : Arrêt du serveur de cartes YP : [erreur] (same message in beta 2) At shutdown : gpm : oops from gpm.c (1002) : ioctl for mouse failed (not finished but no time to write all). Stef
RE: [Cooker] kernel-2.4.17.18 installation report
-Original Message- From: [EMAIL PROTECTED] [mailto:cooker-owner@linux- mandrake.com] On Behalf Of huug Sent: Thursday, February 21, 2002 5:00 AM To: [EMAIL PROTECTED] Subject: Re: [Cooker] kernel-2.4.17.18 installation report huug [EMAIL PROTECTED] writes: Borsenkow Andrej [EMAIL PROTECTED] writes: On óÂÔ, 2002-02-16 at 01:12, huug wrote: We've seen that loop stuff already :) But /boot is fish nor flesh (vmlinuz new, all others old). [root@huug2001 huug]# rpm -iv kernel-2.4.17.18mdk-1-1mdk.i586.rpm Preparing packages for installation... kernel-2.4.17.18mdk-1-1mdk mke2fs 1.26 (3-Feb-2002) mount: Could not find any loop device. Maybe /dev/loop# has a wrong major number? Update pam and restart devfsd. If it does not fix it ... I am interested in more information. Pam is 0.75.18: I follow Cooker pretty closely, sometime even urpmi --auto-select (when I had a working urpmi;) Well, the kernel is in place now, it just won't boot. When it's time to mount root, it complains about an unsupported feature 4 and that is that. BRS time :( Yeah, I'm disappointed: .17 was making progress (by skipping over the mount points in fstab instead of hanging), and now this.. *Blush*, there was no initrd= in lilo.conf, so it tried to boot with the one from 8.34mdk... Shouldn't rpm -i have made one? If loop failed to load, there is little chance for initrd to be created.
RE: [Cooker] kernel install again
[root@huug2001 huug]# rpm -iv kernel-2.4.17.20mdk-1-1mdk.i586.rpm Preparing packages for installation... kernel-2.4.17.20mdk-1-1mdk mke2fs 1.26 (3-Feb-2002) mount: Could not find any loop device. Maybe /dev/loop# has a wrong major number? ... [root@huug2001 huug]# rmmod loop;insmod loop;mkinitrd /boot/initrd-2.4.17- 20mdk.img 2.4.17-20mdk Using /lib/modules/2.4.8-34.1mdk/kernel/drivers/block/loop.o.gz mke2fs 1.26 (3-Feb-2002) /root/tmp/initrd.img.JW8mVu is mounted; mke2fs forced anyway. Hmm ... it looks like it has been mounted but mount returned error for some reason? Hope /etc/mtab is incorrect. [root@huug2001 boot]# ln -sf vmlinuz-2.4.8-34.1mdk vmlinuz [root@huug2001 boot]# mcedit /etc/lilo.conf (no initrd= line again) [root@huug2001 boot]# lilo -vecho OK ... OK --
Re: [Cooker] network.img tokenring
Thanks for the link, the olympic module is included in other.img, however it does not support dhcp. Is there any possibility of a build procedure.I would like to include pump. On Wed, 2002-02-20 at 19:54, Pixel wrote: trax [EMAIL PROTECTED] writes: I wish to create network boot disks which will support desktop installations of mandrake via Token Ring (pci olympic and ibmtr isa drivers. olympic is already available in other.img as for ibmtr, i suggest you use this: http://www.mandrakelinux.com/cgi-bin/cvsweb.cgi/~checkout~/gi/mdk-stage1/doc/UPDATEMODULES?rev=1.1 _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com
[Cooker] Re: beta3 - mcc i18n still not fixed
Borsenkow Andrej [EMAIL PROTECTED] writes: Well, mcc still is not usable with in Russian locale. Just garbage instead of text. Pablo, your diagnostic? -- dam's
[Cooker] r128 still not good with -19mdk kernel
I just tried kernel -19mdk and XFree -6mdk: the infamous error message [drm:r128_cce_indirect] has vanished fromlog messages, kdm initial display is OK, but the box still freeze when switching virtual console while X display is owned by uid O (display manager). Once logged as a user, it's OK. -- Guillaume Rousse [EMAIL PROTECTED] GPG key http://lis.snv.jussieu.fr/~rousse/gpgkey.html
Re: [Cooker] ftp in rpmdrake not working
Danny Tholen [EMAIL PROTECTED] writes: It seems that I cannot use ftp in rpmdrake: added medium contribs retrieving description file of contribs... curl: (6) Couldn't resolve host 'ftp' retrieving source hdlist (or synthesis) of contribs... curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' curl: (6) Couldn't resolve host 'ftp' retrieve of source hdlist (or synthesis) failed no hdlist file found for medium contribs unable to update medium contribs Oh..and ofcourse...I didn't type ftp in the URL bar. Tried both ftp://ftp.surfnet.nl/somepath/ and ftp.surfnet.nl/somepath/ Danny This is a bug in urpmi, please upgrade it. François.
[Cooker] kwintv
Is it supposed to be broken? See attached screen shot. Cheers, R.Fox attachment: snap.jpg
Re: [Cooker] INSTALL REPORT
Stéphane Teletchéa [EMAIL PROTECTED] writes: Warning : loading /tmp/vfat.o will taint the kernel : forced load Warning : loading /tmp/reiserfs.o will taint the kernel : forced load Furthermore, why does it complains about kernel tainting ? (what does that mean ?) it doesn't like using -f for insmod'ing modules
Re: [Cooker] Tonights Install - DrakX 1.658
David Eastcott [EMAIL PROTECTED] writes: a) lots and lots and ... of rpmdb: var/lib/rpm/Basesnames: No such file or directory messages on the primary console during the install. yeah, we know :-( b) During installation, msec is changing the the permissions of the lilio.conf file from 600 to 644 when I have the keyword 'password' is in it: DrakX sets it correctly. The result is a Warning message from lilo that lilio.conf should be readable by root only. fredl, can you make sure msec doesn't *add* any permission to lilo.conf? b) lilo is reporting a warning that the 'menu-scheme' is not supported by the boot loader. francois, what this about?
Re: [Cooker] About Packager?
Borsenkow Andrej [EMAIL PROTECTED] writes: In rpm -qpi ? Yes, it does. But since I was the first to submit the package, I should be listed in rpmmon -p none the less. The correct way would be to ask me for permission so that you can upload it. Permission is hereby granted ;) Sure. I never intended to actually _do_ it without asking first. I just used it as example to clarify some things, I'm still learning how things are running here. Maintainers that do want to maintain a package can add a packager tag with their name in the spec file, so that even rebuilt by somebody else the packager tag will not change, if there is no packager tag, just use yours, or something like qa.mandrakesoft.com or like. -- Warly
Re: [Cooker] Still broken via-rhine in latest kernel 2.4.17-18
bill == Bill Greenwood [EMAIL PROTECTED] writes: bill [root@localhost root]# insmod via-rhine bill Using /lib/modules/2.4.17-18mdksecure/kernel/drivers/net/via-rhine.o.gz bill /lib/modules/2.4.17-18mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_link_ok_Rsmp_ba4049b4 bill /lib/modules/2.4.17-18mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_nway_restart_Rsmp_dceb1f73 bill /lib/modules/2.4.17-18mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_ethtool_sset_Rsmp_6d966584 bill /lib/modules/2.4.17-18mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_ethtool_gset_Rsmp_a986f194 Will be fixed in 21mdk, I still don't know what is happening, kernel-rpm is supposed to break when this happen :((( Sorry for the inconveniences. Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
[Cooker] Problem touchpad of my HP omnibook 6000 not supported!!!
Hi , I ve some issue with my laptop it's a HP Omnibook 6000 PIII600 the touchpad did not work at all , I need to use an external mouse , Any solution?
Re: [Cooker] mini low latency patch in kernel?
martin == Martin Maok [EMAIL PROTECTED] writes:
Hi
Lack of time, will look at integrate it.
Later, Juan.
martin Hi,
martin why isn't mini-ll patch from Andrew Morton included in standard MDK
martin kernel? It should be harmless/secure/stable, it's a short and simple
martin patch which identifies largest (longest) kernel locks and breaks them
martin into smaller with conditional schedules. It helps desktop
martin interactivity and minimizing audio/video delays (thrashing).
martin Andrea Arcangeli uses it in his -aa tree (so it's probably in standard
martin SuSE kernels, but I haven't checked it) and Alan Cox told it should go
martin to vanilla 2.4 tree.
martin I'm attaching the message/patch. I don't of any problem this could
martin cause and I didn't see a single post on LKML telling something bad of
martin mini-ll.
martin Have a nice day
martin --
martin Martin Maok http://underground.cz/
martin [EMAIL PROTECTED]http://Xtrmntr.org/ORBman/
martin Date: Sat, 12 Jan 2002 12:23:09 -0800
martin From: Andrew Morton [EMAIL PROTECTED]
martin To: Ed Sweetman [EMAIL PROTECTED]
martin Cc: Andrea Arcangeli [EMAIL PROTECTED], [EMAIL PROTECTED],
martin [EMAIL PROTECTED], Robert Love [EMAIL PROTECTED],
martin Alan Cox [EMAIL PROTECTED], [EMAIL PROTECTED],
martin Rob Landley [EMAIL PROTECTED], [EMAIL PROTECTED]
martin Subject: Re: [2.4.17/18pre] VM and swap - it's really unusable
martin Ed Sweetman wrote:
If you want to test the preempt kernel you're going to need something that
can find the mean latancy or time to action for a particular program or
all programs being run at the time and then run multiple programs that you
would find on various peoples' systems. That is the feel people talk
about when they praise the preempt patch.
martin Right. And that is precisely why I created the mini-ll patch. To
martin give the improved feel in a way which is acceptable for merging into
martin the 2.4 kernel.
martin And guess what? Nobody has tested the damn thing, so it's going
martin nowhere.
martin Here it is again:
martin --- linux-2.4.18-pre3/fs/buffer.c Fri Dec 21 11:19:14 2001
martin +++ linux-akpm/fs/buffer.c Sat Jan 12 12:22:29 2002
martin @@ -249,12 +249,19 @@ static int wait_for_buffers(kdev_t dev,
martin struct buffer_head * next;
martin int nr;
martin - next = lru_list[index];
martin nr = nr_buffers_type[index];
martin +repeat:
martin + next = lru_list[index];
martin while (next --nr = 0) {
martin struct buffer_head *bh = next;
martin next = bh-b_next_free;
martin + if (dev == NODEV current-need_resched) {
martin + spin_unlock(lru_list_lock);
martin + conditional_schedule();
martin + spin_lock(lru_list_lock);
martin + goto repeat;
martin + }
martin if (!buffer_locked(bh)) {
martin if (refile)
martin __refile_buffer(bh);
martin @@ -1174,8 +1181,10 @@ struct buffer_head * bread(kdev_t dev, i
martin bh = getblk(dev, block, size);
martin touch_buffer(bh);
martin - if (buffer_uptodate(bh))
martin + if (buffer_uptodate(bh)) {
martin + conditional_schedule();
martin return bh;
martin + }
martin ll_rw_block(READ, 1, bh);
martin wait_on_buffer(bh);
martin if (buffer_uptodate(bh))
martin --- linux-2.4.18-pre3/fs/dcache.c Fri Dec 21 11:19:14 2001
martin +++ linux-akpm/fs/dcache.c Sat Jan 12 12:22:29 2002
martin @@ -71,7 +71,7 @@ static inline void d_free(struct dentry
martin * d_iput() operation if defined.
martin * Called with dcache_lock held, drops it.
martin */
martin -static inline void dentry_iput(struct dentry * dentry)
martin +static void dentry_iput(struct dentry * dentry)
martin {
martin struct inode *inode = dentry-d_inode;
martin if (inode) {
martin @@ -84,6 +84,7 @@ static inline void dentry_iput(struct de
martin iput(inode);
martin } else
martin spin_unlock(dcache_lock);
martin + conditional_schedule();
martin }
martin /*
martin --- linux-2.4.18-pre3/fs/jbd/commit.c Fri Dec 21 11:19:14 2001
martin +++ linux-akpm/fs/jbd/commit.c Sat Jan 12 12:22:29 2002
martin @@ -212,6 +212,16 @@ write_out_data_locked:
martin __journal_remove_journal_head(bh);
martin refile_buffer(bh);
martin __brelse(bh);
martin + if (current-need_resched) {
martin + if
(commit_transaction-t_sync_datalist)
martin +
commit_transaction-t_sync_datalist =
martin + next_jh;
martin + if (bufs)
martin + break;
martin + spin_unlock(journal_datalist_lock);
martin + conditional_schedule();
martin + goto write_out_data;
martin +
Re: [Cooker] Laptop Problems
This email is subject to the disclaimer set out below * I concur with this sentiment. My Acer Extensa laptop also hangs when trying to come back from a resume of any kind and when the mains power is switched off. However, I still think there is a broader issue here - why are apmd and swapper throwing, presumably untrapped, NULL pointer exceptions that bring down the entire machine ? Owen A simple kernel oops logged some where would be sufficient ;-) or perhaps a java or C# kernel to trap exceptions properly ! LOL !! ;-) On Thu, 2002-02-21 at 07:28, SI Reasoning wrote: --- Randy Welch [EMAIL PROTECTED] wrote: Teemu Torma wrote: On Thursday 21 February 2002 02:04, SI Reasoning wrote: The kernel in 8.1 did not have APIC enabled. I am also using a Dell laptop. It seems that any BIOS event locks the laptop completely, whether it is unplug the power cord or BIOS display timeout or manual display blanking. Something like 2.4.13 works, which I am using right now, but all 2.4.17 and after kernels are completely unusable. What is it with Dells and others laptops? ThinkPads ( or at least the A2X series ) seem to resist any attempts by the kernel to enable it: syslog:Feb 19 21:13:10 randyspc kernel: Local APIC disabled y BIOS -- reenabling. syslog:Feb 19 21:13:10 randyspc kernel: Could not enable APIC! This issue reminds me of the 8.0 trackpoint mouse issue on the thinkpads. ( it was inoperative... ) Why not build a 'laptop' kernel with APIC turned off? Especially with the changes done to make those folks with r128 based laptops. -randy I couldn't agree more. It would be a lighter kernel to boot since you could disable alot of options that are just not used on a laptop (such as raid, ham radio, most pci/isa network cards, etc) It would also lighten the load on the standard kernel since you could remove laptop related issues such as pcmcia, etc. I think it is an obvious choice and Mandrake could brag about being the only Distro that caters to laptops! = SI Reasoning [EMAIL PROTECTED] A requirement of creativity is that it contributes to change. Creativity keeps the creator alive. -FRANK HERBERT, unpublished notes __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com THIS E-MAIL MAY CONTAIN INFORMATION THAT IS CONFIDENTIAL AND/OR LEGALLY PRIVILEGED. IF YOU ARE NOT THE INTENDED RECIPIENT, PLEASE DO NOT READ IT, DELETE IT IMMEDIATELY AND INFORM CAMBRIDGE POSITIONING SYSTEMS LIMITED BY TELEPHONING +44 (0)1223 326900. This Email has been checked for viruses but it is your responsibility to conduct your own virus checks on all parts of it. Cambridge Positioning Systems Limited, 62-64 Hills Road, Cambridge, CB2 1LA, is registered in England under company number 2808344. ***
Re: [Cooker] koffice has problems with kde3beta2
OK that worked . thanks :) Le Jeudi 21 Février 2002 03:25, Lonnie Borntreger a écrit : As root: $- cd /opt/kde3/share/servicetypes $- cp /usr/share/servicetypes/ko* . On Wed, 2002-02-20 at 14:14, pascal wrote: for example : kword does not start $ kword koffice (lib kofficecore): ERROR: The serviceType KOfficePart is missing. Check that you have a kofficepart.desktop file in the share/servicetypes directory. koffice (lib kofficecore): ERROR: Couldn't find the native MimeType in kword's desktop file. Check your installation ! surely a problem with non standard paths used in kde3beta Pascal
Re: [Cooker] Tonights Install - DrakX 1.658
Pixel [EMAIL PROTECTED] writes: b) lilo is reporting a warning that the 'menu-scheme' is not supported by the boot loader. francois, what this about? We can remove the warning but this is just a warning saying the menu-scheme entry in /etc/lilo.conf will not be supported by stage2 of lilo (of course, this is a picture wich is displayed). François.
Re: [Cooker] Beta 3 Install report
David Eastcott [EMAIL PROTECTED] writes: 1) I have a DSL account through T-Online here in Germany. I have a Linksys 4 port router which runs DHCP. During install I am never asked about hostname - and it aparently gets a domainname from the T-Online host (result of command hostname: localhost.BS1.srv.t-online.de) - but an entry in the /etc/hosts file is not there - and therefore Squid fails! This is directly after a fresh install! gc, That stage 1 dhcp patch I sent you a while ago may be of interest for this. Argh, i've not looked at it yet :-(.. -- Guillaume Cottenceau - http://www.frozen-bubble.org/
[Cooker] diskdrake and jfs
Hi! When creating a new partition, diskdrake allows to create the partition as JFS, even if the jfsprogs are not installed. IMO, if jfsprogs are not installed and a JFS partition is to be created, jfsprogs should be installed automatically (or even better: ask the user if he wants to install the necessary tools. If he says no, ask if he really wants to creat a partition of this type. If yes, go ahead and let it silently fail. If no, cancel the action.) Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.iso-top.de | Jabber: [EMAIL PROTECTED] iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 3 hours 45 minutes
Re: [Cooker] bug: supermount, iomega zip
serge == Serge Ryabchun [EMAIL PROTECTED] writes: serge kernel-source-2.4.17-17mdk.i586.rpm serge iomega ZIP100 serge I cant eject ZIP disc when use ide-floppy.o (IDE/ATAPI FLOPPY) - door is serge locked and I need to unmount manually, but if I use sd_mod.o (SCSI emulation) serge then supermount work fine. problem is in ide-floppy land, if you have mounted it, it don't allow you to eject :((( Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
RE: [Cooker] bug: supermount, iomega zip
Juan Quintela [EMAIL PROTECTED] writes: serge == Serge Ryabchun [EMAIL PROTECTED] writes: serge kernel-source-2.4.17-17mdk.i586.rpm serge iomega ZIP100 serge I cant eject ZIP disc when use ide-floppy.o (IDE/ATAPI FLOPPY) - door is serge locked and I need to unmount manually, but if I use sd_mod.o (SCSI emulation) serge then supermount work fine. problem is in ide-floppy land, if you have mounted it, it don't allow you to eject :((( i switch to ide-scsi for zips... Please, folks, supermount does NOT work fine for SCSI as well!!! Read-write mount locks up the system (available info has been sent to kernel@mandrake). Read-only (super-)mount works I agree. -andrej
Re: [Cooker] bug: supermount, iomega zip
Juan Quintela [EMAIL PROTECTED] writes: serge == Serge Ryabchun [EMAIL PROTECTED] writes: serge kernel-source-2.4.17-17mdk.i586.rpm serge iomega ZIP100 serge I cant eject ZIP disc when use ide-floppy.o (IDE/ATAPI FLOPPY) - door is serge locked and I need to unmount manually, but if I use sd_mod.o (SCSI emulation) serge then supermount work fine. problem is in ide-floppy land, if you have mounted it, it don't allow you to eject :((( i switch to ide-scsi for zips...
[Cooker] draksync (rsync) mem leak
It turns out that at least one of the problems I originally associated with the new kernel may actually be an rsync problem. I was able to fill up all ram and swap by using draksync to sync two directories. The only noticable difference between this one and others was that the directory name contained a space in the one that created the memory leak. It was an rsync process that grabbed all available memory, however. = SI Reasoning [EMAIL PROTECTED] A requirement of creativity is that it contributes to change. Creativity keeps the creator alive. -FRANK HERBERT, unpublished notes __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com
[Cooker] mindi/mondo
Lenny have you tested them? Last time I tested there were several issues. My patch has been included but I never got time to test them thereafter. The main problem was that mindi could not use cdrom with stock kernel because inflate_fs. Mindi kernel worked fine IIRC. -andrej
[Cooker] diskdrake doesn't handle devfs style partitions correctly
Hi. When creating a new partition with diskdrake, it rewrites /etc/fstab (which is bad, btw). However, I just now ended up with something like this: /dev/hda6 swap swap defaults 0 0 /dev/ide/host0/bus0/target0/lun0/part6 swap swap defaults 0 0 Well, /dev/hda6 == /dev/ide This only happens for swap partitions. Normal partitions work fine. drakxtools-newt-1.1.7-76mdk Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.iso-top.de | Jabber: [EMAIL PROTECTED] iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 3 hours 47 minutes
Re: [Cooker] Quota, still unperfect...
Claudio [EMAIL PROTECTED] writes: In /etc/fstab the default option is missing, is it normal? yes it is. default is only needed if you don't have any options so that you don't have an empty string :)
[Cooker] urpmi.update much smoother - uromi or kernel?
On PPro 180 with 32MB. For a long time I could not use urpmi.update then it has been fixed but I always had several cannot read hdlist when updating media. Now several times already urpmi.update runs surprisingly fast and no warnings ... I wonder if VM became really better or urpmi (rpmtools) has been changed? -andrej
Re: [Cooker] atd core dumps
»Martin Ma?ok« sagte am 2002-02-20 um 13:35:59 +0100 :
On Wed, Feb 20, 2002 at 12:57:13PM +0100, Alexander Skwar wrote:
atd from at-3.1.8-5mdk core dumps at startup.
Can you gdb the core? Can you strace atd?
gdb atd
GNU gdb 5.1.1
Copyright 2002 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type show copying to see the conditions.
There is absolutely no warranty for GDB. Type show warranty for details.
This GDB was configured as i386-mandrake-linux...
(no debugging symbols found)...
(gdb) run
Starting program: /usr/sbin/atd
(no debugging symbols found)...
Program received signal SIGSEGV, Segmentation fault.
0x0804a005 in strcpy ()
(gdb) where
#0 0x0804a005 in strcpy ()
(gdb)
execve(/usr/sbin/atd, [atd], [/* 42 vars */]) = 0
uname({sys=Linux, node=cpo, ...}) = 0
brk(0) = 0x804c478
open(/etc/ld.so.preload, O_RDONLY)= -1 ENOENT (No such file or directory)
open(/etc/ld.so.cache, O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=62606, ...}) = 0
old_mmap(NULL, 62606, PROT_READ, MAP_PRIVATE, 3, 0) = 0x40016000
close(3)= 0
open(/lib/libc.so.6, O_RDONLY)= 3
read(3, \177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0p\204\1..., 1024) = 1024
fstat64(3, {st_mode=S_IFREG|0755, st_size=1275588, ...}) = 0
old_mmap(NULL, 1292288, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x40026000
mprotect(0x40158000, 38912, PROT_NONE) = 0
old_mmap(0x40158000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x131000)
= 0x40158000
old_mmap(0x4015e000, 14336, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS,
-1, 0) = 0x4015e000
close(3)= 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x40162000
munmap(0x40016000, 62606) = 0
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++
Do you have everything synced to cooker? Files are okay?
Yes, it is synced. Hmm:
rpm -V at
.M.. c /etc/rc.d/init.d/atd
..5. /usr/sbin/atd
After erasing and reinstalling the rpm, at works. Thanks for the hints!
Alexander Skwar
--
How to quote: http://learn.to/quote (german) http://quote.6x.to (english)
Homepage: http://www.iso-top.de | Jabber: [EMAIL PROTECTED]
iso-top.de - Die günstige Art an Linux Distributionen zu kommen
Uptime: 4 hours 22 minutes
[Cooker] we want xine in main distribution :-)
Jerome just sent me two new skins for xine (one of which with official mdk colors) and asked me why xine was not in main part of the distribution. With current witch-hunt for napster client, i think potential legal problems are currently main Mandrake concern, even if i never saw any official position on this topic. So, which is the best way to have at least an explanation on this subject, or much better, make this position evolves ? -- Guillaume Rousse [EMAIL PROTECTED] GPG key http://lis.snv.jussieu.fr/~rousse/gpgkey.html
[Cooker] Quota, still unperfect...
hi! now I found usrquota/grpquota between options during installation. But there are some things to adjust: 1) In /etc/fstab the default option is missing, is it normal? 2) quotacheck -avug should be automatically run at first startup, so that aquota.user and aquota.group will be created 3) initscript is still bad. It happens that quotas work right, but during the boot, after interactive startup, between backspace send and Activating swap partitions, there's a boring error message with quota: it says that the device is busy... Claudio -- Claudio Panichi System/Network Admin at Dept. of Physics Tor Vergata University and INFN - Sec. Roma II Remote System is: LINUX Mandrake
Re: [Cooker] accelerated X4.2 with r128 on Inspiron 8000
On Wed, 20 Feb 2002, Jason Straight wrote: BTW - Where can one find the 2.2 patch for the kernel dri for this? The 2.4.18-rc2 only has 2.1.6 still. If you find one let me know. I haven't come across one, and have been tasked to try to port the 2.4 version over. I looked at it briefly, and it doesn't appear to be trivial. Stew Benedict -- MandrakeSoftOH/TN, USA http://people.mandrakesoft.com/~sbenedict/ PPC FAQ: http://www.linux-mandrake.com/en/demos/PPC/FAQ/
Re: [Cooker] Kernel and Lilo appends
robert == Robert Fox [EMAIL PROTECTED] writes: robert On Mon, 2002-02-11 at 16:50, Borsenkow Andrej wrote: Should an LS-120 drive have an append for ide-floppy? In theory kernel should default to ide-floppy for it. -andrej robert And in practice? if you want to use supermount, ide-scsi :) Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] r128 still not good with -19mdk kernel
On Thursday, 21. February 2002 10:31, you wrote: I just tried kernel -19mdk and XFree -6mdk: the infamous error message [drm:r128_cce_indirect] has vanished fromlog messages, kdm initial display is OK, but the box still freeze when switching virtual console while X display is owned by uid O (display manager). Once logged as a user, it's OK. I can't confirm this error with kernel -19mdk and last X. But, unfortunately, I've found a reproducable bug, if glx/dri is enabled (fb plays no role): xmms-open dir and then scroll the list with mouse: freeze, SAK necessary. Could you confirm this?
Re: [Cooker] Hardware sensors.
borsenkow == Borsenkow Andrej [EMAIL PROTECTED] writes: While on this subject: is there any other way for loading sensor-related modules other than /etc/modules ? borsenkow Not that I know of. Halt! Modify rc.local! :-) I normally put the modules in /etc/init.d/sensors in the start option. But extra bonus for the first person that got: sensors-detect to output its things to /etc/modules (here alias options) /etc/sysconfig/sensors (here the modules to load). And source them from /etc/sensors: aka, it will still work after upgrades. That is my idea to fix it, but I am short of time, if somebody beat me doing that, please sent me a patch :) Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] module problems
dave == Dave Seff [EMAIL PROTECTED] writes: dave I have a Philips web cam. when I go to load the pwc modules I get modules dave resolve errors. This is not a recompiled kernel, rather straight from cooker. Hi arghh, you are right, I read badly the first mail, and read modprobe instead of insmod. Please, except if you know that you need insmod, allways use modprobe :) Later, Juan. dave -Dave dave [root@mercury root]# insmod pwc dave Using /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz dave /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz: unresolved symbol dave video_unregister_device_R6adf1a49 dave /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz: unresolved symbol dave video_register_device_R19d759f6 dave [root@mercury root]# -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] via-rhine broken in latest kernels
Juan thanks for the effort, but I had given up and bought a new network card, and now I don't have a problem, as it uses differen't drivers. A couple of other people told me I didn't have a problem, so apparently I didn't have a problem. Anyway, you say it's fixed, I believe you. :) -Bill === Juan Quintela wrote: "bill" == Bill Lea Greenwood [EMAIL PROTECTED] writes: bill Thanks for all the great work on 8.2 guys... you are all very muchbill appreciated! :)bill via-rhine is still broken though:bill # insmod via-rhinebill Using /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gzbill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz:bill unresolved symbol mii_link_ok_Rsmp_ba4049b4bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz:bill unresolved symbol mii_nway_restart_Rsmp_dceb1f73bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz:bill unresolved symbol mii_ethtool_sset_Rsmp_6d966584bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz:bill unresolved symbol mii_ethtool_gset_Rsmp_a986f194bill Has been for at least the last 8 kernels. 2.4.13-12 works perfectly.I am unable to reproduce tha t, this is with kernel-secure kernel?Please try latest.Here depmod -ae has not ouput, i.e. everything is ok. I changedthings around 2.4.17.17mdk than now if there are unresolved symbolscompilation breaks :)Later, Juan.
Re: [Cooker] module problems
dave == Dave Seff [EMAIL PROTECTED] writes: dave I have a Philips web cam. when I go to load the pwc modules I get modules dave resolve errors. This is not a recompiled kernel, rather straight from cooker. Could try latest kernel, it works for me (tm) root$ modprobe pwc root$ lsmod | grep pwc pwc38656 0 (unused) videodev4896 0 [pwc] usbcore51232 1 [pwc usb-uhci] root$ uname -a Linux localhost.mandrakesoft.com 2.4.17-18mdk #1 Thu Feb 14 21:42:16 CET 2002 i686 unknown (same with kernel-2.4.17-20mdk). dave -Dave dave [root@mercury root]# insmod pwc dave Using /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz dave /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz: unresolved symbol dave video_unregister_device_R6adf1a49 dave /lib/modules/2.4.17-16mdk/kernel/drivers/usb/pwc.o.gz: unresolved symbol dave video_register_device_R19d759f6 dave [root@mercury root]# -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] who is APIC decision maker in standard kernel?
si == SI Reasoning [EMAIL PROTECTED] writes: Hi si I was wondering who builds the kernel and made the si decision to have APIC enabled in the standard kernel. I did the decision. Reasons: - you need it in some SMP chipsets (i.e. PIC emulation is wrong on them) - you need it in some UP chipsets (machines with Promise RAID on board come to mind). - new GX chipsets for Intels also need APIC :((( - APIC mode is _way_ nicer for everything Problems: - some machines have a broken APIC implementation (see Dell Inspiron threads here). si I am curious as to why it is still in there. This is si not a flame, I just want to know, and to see if there si was something I could help test so that we can get si this problem resolved. It would be terrible to have a si kernel that so badly breaks many laptops out of the box!! solution, to appear in -21/22mdk: - APIC is always enabled - noapic kernel option will mean _same_ behaviour as before - if you need noapic, send me the output of dmidecode for your BIOS and I will try to incorporate that information to the kernel (i.e. you don't need to put noapic anymore). Is everybody happy with that plan? Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] XFree86-4.20-6
le mer 20-02-2002 à 02:26, John Cavan a écrit : Fabrice FACORAT wrote: http://gatos.sourceforge.net/ati.2.php do you use the right kernel version with the patched drm ? Seems to me that this makes Mandrake less likeable as a kernel development platform. Generally speaking, up until now, major software components (such as XFree86) did not break if you were working with the main kernel trees from Marcelo or Linus. Is this a wise idea to fall off the path like that? yes and no. gatos ati driver need a special drm for ATI cards in order to work correctly. gatos ATI drivers are the best implementation of drivers for ATI cards ( DRI + mpeg hard decode, ... ) because they use the SDK provide by ATI. So have them is better. It might make more sense to create some add-on packages that provide the selection of which XFree86 server and kernel modules are used. so 2 kernel et 2 XFree server just for this drivers ? To my mind you'd better critize incoherence in linux development. There's not enough coherence. For example : why does gatos drivers is still not merged with XFree and dri official drivers ? on top of that why don't they in official XFree team and DRI team ? etc ... On top of that because of DRI we have to sync kernel code with XFree code ! It seems to be the price for Direct rendering and better performance but ... Maybe UTAH-GLX project resurection will provide a better alternative ( acceptable 3D perf with no need to be sync with kernel code ). -- http://perso.wanadoo.fr/linux_wizard/index.html - Les ecologistes reclament une voiture propre. ...Les imbeciles frottent. -- P. Geluck
Re: [Cooker] gcc3.0-doc-3.0.4-0.3mdk
On Thu, 21 Feb 2002, Rainer Hermanns wrote: gcc3.0-doc ## /var/tmp/rpm-tmp.83366: line 3: syntax error near unexpected token `;' /var/tmp/rpm-tmp.83366: line 3: `; # FIXME: that semi-colon should not be needed, nor the blank line' error: execution of %preun scriptlet from gcc3.0-doc-3.0.4-0.3mdk failed, exit status 2 Fixed in -3.0.4-0.4mdk but since it was in a %preun script, effect will be visible in release + 1. So, normal updating to -3.0.4-1mdk will be fine. Or, you can already check now by removing gcc3.0-doc-3.0.4-0.4mdk and realize that it's fixed. Bye, Gwenole.
Re: [Cooker] DiskDrake mount option problem
le mer 20-02-2002 à 10:36, Pixel a écrit : you have to watch if you can see a cursor... sorry, no better way Is it a limitation of gtk_mdk ? gtk 1.2 ? does this pb is solved in gtk 2 ? Even in Motif we can disable input fields and greyed them ! -- http://perso.wanadoo.fr/linux_wizard/index.html - Amants, heureux amants, voulez-vous voyager ? -- Jean de La Fontaine, Les Deux Pigeons
Re: [Cooker] Move RPM_CHECK to msec 4?
On Thu, 21 Feb 2002, Mattias Dahlberg wrote: How insecure is your system without RPM_CHECK=yes? Would it be dangerous to move it to msec level 4? I think about all the discussions about rpmv. I mean as Mandrake Linux default, of course. Regards, Mattias
Re: [Cooker] beta3
le mer 20-02-2002 à 17:20, Yves Duret a écrit : Charles A Edwards [EMAIL PROTECTED] writes: On Wed, 20 Feb 2002 13:24:20 +0100 Warly [EMAIL PROTECTED] wrote: Likely to be the last beta before rc. Tolerance zero (hum) So as not to report on already known is the installer the same as that used in b1 and 2 or has it been updated? it has been of course updated and fixed. why are we doing many beta if we do not fix anything ? lol maybe he should have a look at DrakX changelog -- http://perso.wanadoo.fr/linux_wizard/index.html - L'idée que la science peut et doit être organisée selon des règles à la fois fixes et universelles est utopique et pernicieuse. Paul Feyerabend
Re: [Cooker] via-rhine and drivers
On Thursday 21 February 2002 20:26, Bill Greenwood wrote: thanks for the effort, but I had given up and bought a new network card, and now I don't have a problem, as it uses differen't drivers. Oddly enough, I'm trialling a (Mandrake 8.1) Linux workstation at a local business on a machine with a built-in network card (Intel EtherExpress Pro 100, not via-rhine, which arrive here labelled d-Link). Under Windows, the network would crap out after a day or so's mucking around, so they disabled the built-in LAN interface and installed a separate eepro100 card, which worked more or less fine. Mdk8.1 (with the built-in LAN still disabled!) finds and uses the internal LAN and hasn't wavered yet. I removed the separate LAN card and they're trying out Linux + StarOffice 5.2 + Mozilla + Konsole to replace MS-Whingedows 98 + MS-Orifice 97 + MS-Exploder + MS-Lookout (here comes another virus) + a proprietary terminal emulator on that box. So far, so good. Cheers; Leon
Re: [Cooker] make xconfig still broken in kernel-2.4.17-18mdk
si == SI Reasoning [EMAIL PROTECTED] writes: si [root@sidell linux]# make xconfig si rm -f include/asm si ( cd include ; ln -sf asm-i386 asm) si make -C scripts kconfig.tk si make[1]: Entering directory si `/usr/src/linux-2.4.17-18mdk/scripts' si cat header.tk ./kconfig.tk si ./tkparse ../arch/i386/config.in kconfig.tk si 3rdparty/qla/Config.in: 5: Too many variables defined. si make[1]: *** [kconfig.tk] Error 1 si make[1]: Leaving directory si `/usr/src/linux-2.4.17-18mdk/scripts'make: *** si [xconfig] Error 2 Fixed in -20mdk: * Wed Feb 20 2002 Juan Quintela [EMAIL PROTECTED] 1-1mdk - enable rpmhelper module in all the kernels for now on. - 2.4.18-rc2. - now xconfig has the correct values and works again. - fix dependencies for boot-splash-screen font (jgarzik). - fix in pcmcia modules names links (Andre Duclos).x - fix to kernel-hearders and new naming scheme (Andre Duclos). - update Zaurus driver for ethernet over usb (Dave Seff). - updated ppc configs (stew). - 2.4.17.20mdk. executive sumary: - default array is 2048 elements big, and we have more than 2048 config options in the kernel. Thanks for the report. Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
Re: [Cooker] network.img tokenring
trax [EMAIL PROTECTED] writes: Thanks for the link, the olympic module is included in other.img, however it does not support dhcp. Is there any possibility of a build Sorry? procedure.I would like to include pump. There is already a DHCP client (which is btw the preliminary version of what would eventually become pump later on), I don't see your problem? -- Guillaume Cottenceau - http://www.frozen-bubble.org/
Re: [Cooker] Plse add a Multi_key in the US kbd.
Quel Qun [EMAIL PROTECTED] writes: I use Shift left window key: xmodmap -e 'keycode 116 = Super_R Multi_key' That really helps when I need accentuated letters. This is not a bad idea but it would not be easily findable by people so putting it by default... -- Guillaume Cottenceau - http://www.frozen-bubble.org/
[Cooker] [Fwd: assistance kernel version 2.4.18-2rc]
Hi all, Whos bright idea was it to ban the word HELP from the subject line !!¬!!! -- Best regards Richard Bown JVM RF EMC Solutions Ltd. Tel +44 1952 502550 Fax. +44 1952 275167 JVM RF EMC Solutions Ltd. Registered office, Park View, Chapel St, Telford TF4 3DD. Registered in the United Kingdom 3538979 ---BeginMessage--- Hi gurus is there a mdk version of 2.4.18-rc2, as someones fixed a bug thats been giving me probs with a patch to this kernel version. I would prefer to use mdk kernels TIA -- Best regards Richard Bown ---End Message---
[Cooker] cannot boot from older SCSI device (cannot insert aic7xxx.o)
Hi, I found out that mdk 8.1/8.2 is unable to boot from an older SCSI device. The problem is missing aic7xxx_old.o driver on the bootdisk. With newer aic7xxx.o inserting the kernel module fails. Could you please include it? Red Hat has both of them and older adaptek 7770 works with it. Thanks and have a nice day -- Martin Maok http://underground.cz/ [EMAIL PROTECTED]http://Xtrmntr.org/ORBman/
[Cooker] 2.4.17-20mdk kernel corrupt?
bor@cooker% rpm --checksig /mnt/hd/cooker/i586/Mandrake/RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm /mnt/hd/cooker/i586/Mandrake/RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm: MD5 (GPG) NOT OK (MISSING KEYS: GPG#9B4A4024) Synced off sunsite.uio.no with rsync.
[Cooker] More on cron
30 7 * * 1-5 xmms -p I have that in my user crontab, and it didn't get run this morning. The cron service is running, and this cron job works under Mandrake 7.2 Any ideas?
[Cooker] REPOST: 8.2 request: multiple routing tables option in default kernels
Please compile the kernels with multiple routing tables enabled. It allows multi-homing out of the box using iproute2. Yea or nay? Cheers; Leon
Re: [Cooker] DiskDrake mount option problem
Fabrice FACORAT [EMAIL PROTECTED] writes: le mer 20-02-2002 à 10:36, Pixel a écrit : you have to watch if you can see a cursor... sorry, no better way Is it a limitation of gtk_mdk ? gtk 1.2 ? does this pb is solved in gtk 2 ? Even in Motif we can disable input fields and greyed them ! greying the combo would not do it. there is - grayed combo = not pertinent - not editable entry in the combo - editable entry in the combo
Re: [Cooker] cannot boot from older SCSI device (cannot insert aic7xxx.o)
Martin Maèok [EMAIL PROTECTED] writes: I found out that mdk 8.1/8.2 is unable to boot from an older SCSI device. The problem is missing aic7xxx_old.o driver on the bootdisk. With newer aic7xxx.o inserting the kernel module fails. Could you please include it? done
RE: [Cooker] More on cron
30 7 * * 1-5 xmms -p
I have that in my user crontab, and it didn't get run this morning.
The
cron
service is running, and this cron job works under Mandrake 7.2
Any ideas?
1. path. Why do you think it includes xmms when cron is running?
2. /etc/cron.{allow,deny}?
Re: [Cooker] we want xine in main distribution :-)
On Thu Feb 21 11:59 +0100, Guillaume Rousse wrote: Jerome just sent me two new skins for xine (one of which with official mdk colors) and asked me why xine was not in main part of the distribution. With current witch-hunt for napster client, i think potential legal problems are currently main Mandrake concern, even if i never saw any official position on this topic. So, which is the best way to have at least an explanation on this subject, or much better, make this position evolves ? I believe xine is in Cooker (not Contrib). -- Levi Ramsey [EMAIL PROTECTED] [EMAIL PROTECTED] Boy, that crayon sure did hurt! Linux 2.4.17-16mdk 8:01am up 4 days, 19:01, 8 users, load average: 0.13, 0.05, 0.01
[Cooker] Re: [CHRPM] kdebase-2.2.2-58mdk
Es schrieb Laurent MONTEL: [kdebase] * Thu Feb 21 2002 Laurent MONTEL [EMAIL PROTECTED] 2.2.2-58mdk - Sync with kdebase 2.2.2 branch a lot of fixes in konsole - Remove all my patchs which are commited in this branch Will this allow the usage of the Euro sign out of the box in Konsole? -- _ Tschüss und bis demnächst/à bientôt, _|_|_ () * Stefan /v\ / »( )« Penguin Powered! +(m-m)--+
Re: [Cooker] Plse add a Multi_key in the US kbd.
Le Jeudi 21 Février 2002 14:29, Guillaume Cottenceau a écrit : Stéphane Teletchéa [EMAIL PROTECTED] writes: Le Jeudi 21 Février 2002 13:01, Guillaume Cottenceau a écrit : Quel Qun [EMAIL PROTECTED] writes: I use Shift left window key: xmodmap -e 'keycode 116 = Super_R Multi_key' That really helps when I need accentuated letters. This is not a bad idea but it would not be easily findable by people so putting it by default... In the same idea, how to address a paste function to a key/combinations of keys (example paste with F4). One could select characters somewhere, and paste them with F4 as on SGI ... Shift+Insert already does this... I love U, now i will just map this to another key of mine. thanks a lot, Stef (just in case : where can i find the list of shortcuts you showed me ?)
Re: [Cooker] Plse add a Multi_key in the US kbd.
Stéphane Teletchéa [EMAIL PROTECTED] writes: Shift+Insert already does this... I love U, now i will just map this to another key of mine. thanks a lot, Stef (just in case : where can i find the list of shortcuts you showed me ?) For things like Shift+Insert, I think it's not a keysym, but it's handled by applications. -- Guillaume Cottenceau - http://www.frozen-bubble.org/
[Cooker] tcpwrappers - why this line does not work?
sshd: X.Y.Z.0/255.255.255.0 EXCEPT X.Y.Z.17/255.255.255.255: ALLOW That still allows X.Y.Z.17 even if the only remaining rule denies everything: ALL:ALL EXCEPT localhost:DENY -andrej
Re: [Cooker] beta3
On 21 Feb 2002 13:56:59 +0100 Fabrice FACORAT [EMAIL PROTECTED] wrote: le mer 20-02-2002 à 17:20, Yves Duret a écrit : Charles A Edwards [EMAIL PROTECTED] writes: On Wed, 20 Feb 2002 13:24:20 +0100 Warly [EMAIL PROTECTED] wrote: Likely to be the last beta before rc. Tolerance zero (hum) So as not to report on already known is the installer the same as that used in b1 and 2 or has it been updated? it has been of course updated and fixed. why are we doing many beta if we do not fix anything ? lol maybe he should have a look at DrakX changelog I get the DrakX changelog and I Do 'look' at it. Perhaps my question would have been better phrased if I had asked if All known problems seem in the previous installers had been 'fixed' or if there is need to report that pkg selection can not be saved to floppy if no disk is in USB Zip or that printer can not be set-up during installation. I see no point in continually reporting as problems situations that do not have a resolution (floppy/Zip) or other known problems to which a satisfactory solution has yet been reached. Charles
[Cooker] what's wrong with xcdroast?
When I try to rip audio track in xcdroast of today's cooker here you are what I obtain: recording 4336.06933 seconds stereo with 16 bits @ 44100.0 Hz -'/home/master/track-01'... 0%Fatal error: did not drop group privilege. child reader sem request failed W Child exited with 1 AFAIK it happen since beta2, at least. NB-1) Xcdroast won't work as root too! NB-2) Gcombust work fine (as normal user too) NB-34) cdda2wav and cdparanoia work both fine (as root and as user) Bye, Claudio -- Claudio Panichi System/Network Admin at Dept. of Physics Tor Vergata University and INFN - Sec. Roma II Remote System is: LINUX Mandrake
[Cooker] smbsh tries to load libarary from /usr/bin
Hi! smbsh Username: skwar Password: smbsh: error while loading shared libraries: /usr/bin/smbwrapper.so: cannot open shared object file: No such file or directory urpmf smbwrapper.so samba-common:/usr/lib/smbwrapper.so rpm -q samba-common samba-common-2.2.3a-4mdk Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.iso-top.de | Jabber: [EMAIL PROTECTED] iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 6 hours 50 minutes
Re: [Cooker] cannot boot from older SCSI device (cannot insert aic7xxx.o)
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Maybe it's time you update your firmware? No more development is happening with the _old driver, so you won't see any improvements ever if you keep the old firmware. I don't think Adaptec supports the old firmware anymore. Alternatively, you could probably use the blank.img to make a floppy with the aic7xxx_old.o driver. Martin Ma ok wrote: | Hi, | I found out that mdk 8.1/8.2 is unable to boot from an older SCSI | device. The problem is missing aic7xxx_old.o driver on the bootdisk. | With newer aic7xxx.o inserting the kernel module fails. Could you | please include it? | | Red Hat has both of them and older adaptek 7770 works with it. | | Thanks and have a nice day | | - -- |Registered Linux User #182071-| Buchan MilneMechanical Engineer, Network Manager Cellphone * Work+27 82 472 2231 * +27 21 8828820x202 Stellenbosch Automotive Engineering http://www.cae.co.za GPG Key http://ranger.dnsalias.com/gpg.key -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE8dQI4rJK6UGDSBKcRAlZQAKDAiupjbuTJKuo5uJE6PyTkxXSdnwCeIO4G /FwxRJR32BCHif4Q808CYDM= =OcsB -END PGP SIGNATURE-
[Cooker] IRDA problem with my laptop .
The IRDA server did not work at all on my Omnibook 6000 laptop I would like to syncronize stuff ;-(
Re: [Cooker] we want xine in main distribution :-)
Am Donnerstag, 21. Februar 2002, 08:47:25 Uhr MET, schrieb Levi Ramsey: On Thu Feb 21 11:59 +0100, Guillaume Rousse wrote: Jerome just sent me two new skins for xine (one of which with official mdk colors) and asked me why xine was not in main part of the distribution. With current witch-hunt for napster client, i think potential legal problems are currently main Mandrake concern, even if i never saw any official position on this topic. So, which is the best way to have at least an explanation on this subject, or much better, make this position evolves ? I believe xine is in Cooker (not Contrib). Well, I've just updated the Contrib package with the new mdk skin. -- Götz Waschk student of computer science university Rostock http://www.informatik.uni-rostock.de/~waschk/waschk.asc for PGP key -- Logout Fascism! --
Re: [Cooker] Plse add a Multi_key in the US kbd.
Stéphane Teletchéa [EMAIL PROTECTED] writes: Le Jeudi 21 Février 2002 13:01, Guillaume Cottenceau a écrit : Quel Qun [EMAIL PROTECTED] writes: I use Shift left window key: xmodmap -e 'keycode 116 = Super_R Multi_key' That really helps when I need accentuated letters. This is not a bad idea but it would not be easily findable by people so putting it by default... In the same idea, how to address a paste function to a key/combinations of keys (example paste with F4). One could select characters somewhere, and paste them with F4 as on SGI ... Shift+Insert already does this... -- Guillaume Cottenceau - http://www.frozen-bubble.org/
Re: [Cooker] tcpwrappers - why this line does not work?
On Thu, 21 Feb 2002 17:05:08 +0300 Borsenkow Andrej [EMAIL PROTECTED] wrote: sshd: X.Y.Z.0/255.255.255.0 EXCEPT X.Y.Z.17/255.255.255.255: ALLOW That still allows X.Y.Z.17 even if the only remaining rule denies everything: ALL:ALL EXCEPT localhost:DENY -andrej This is an old bug that has been ignored... here's the notes I keep in my /etc/hosts.allow: # WARNING: Never use mask=255.255.255.255!! While this is perfectly # legal, tcpwrappers treats it as a bad mask. Yet, it # treats the following as valid: # 1.1.1.1/255.255.255.254 -- no host space in subnet; # impossible match # 1.1.1.1/255.255.255.254 -- impossible match (last octet) # Go figure!! Just list the address *without* a mask if you want 255.255.255.255 or /32 I reported this to the author; but got blown off... he has since GPL'ed the code... Fixing it may break some systems; but it's plain wrong and *should* be fixed IMNSHO... HTH, Pierre
[Cooker] Permission denied to lp-errs
Hi, After installing lpd I get a boot warning saying 'Cannot open /var/log/lp-errs: Permission denied'. $ls -l lp-errs -rw-r-1 rootroot 0 feb 20 16:36 lp-errs JFYI. Regards, Mattias
[Cooker] Please, add localhost.localdomain to exception list in hosts.deny
Line looks like ALL: ALL except localhost: DENY Unfortunately, installer puts by default 127.0.0.1 localhost.localdomain localhost which means this exception does not match. Please add full name i.e. ALL: ALL except localhost localhost.localdomain: DENY -andrej
Re: [Cooker] urpmi.update much smoother - uromi or kernel?
Borsenkow Andrej [EMAIL PROTECTED] writes: On PPro 180 with 32MB. For a long time I could not use urpmi.update then it has been fixed but I always had several cannot read hdlist when updating media. Now several times already urpmi.update runs surprisingly fast and no warnings ... I wonder if VM became really better or urpmi (rpmtools) has been changed? rpmtools has been changed :-) some abusive memory leaks where inside thus causing a lot of memory to be wasted (so a 32MB machine will use swap much faster) according to number of media defined. Another bug in rpmtools (I will call it a workaround of rpmlib) was fixed some time ago (a month ?) which wait for data to be available on a stream before calling rpmlib parsing library (instead of letting rpmlib doing it which seems not correctly done, a timeout is used (set to 1 second but almost never above 0.05 second). François.
[Cooker] mdk8.2b2: /var/log errors
There are some error in /var/log/messages: irattach: Failed to open /dev/ttyS2: No such file or directory modprobe: modprobe: Can't locate module eth1 modprobe: modprobe: Can't locate module eth2 modprobe: modprobe: Can't locate module eth3 modprobe: modprobe: Can't locate module eth4 modprobe: modprobe: Can't locate module eth5 modprobe: modprobe: Can't locate module eth6 modprobe: modprobe: Can't locate module eth7 modprobe: modprobe: Can't locate module eth8 modprobe: modprobe: Can't locate module eth9 missing kernel or user mode driver usbcore Can't locate module char-major-226
[Cooker] mdk8.2b2 bug: artswrapper fails
artswrapper must be the super bit set, otherwise artsd will not have realtime scheduler support.
Re: [Cooker] Plse add a Multi_key in the US kbd.
Le Jeudi 21 Février 2002 13:01, Guillaume Cottenceau a écrit : Quel Qun [EMAIL PROTECTED] writes: I use Shift left window key: xmodmap -e 'keycode 116 = Super_R Multi_key' That really helps when I need accentuated letters. This is not a bad idea but it would not be easily findable by people so putting it by default... In the same idea, how to address a paste function to a key/combinations of keys (example paste with F4). One could select characters somewhere, and paste them with F4 as on SGI ... Stef
Re: [Cooker] we want xine in main distribution :-)
Guillaume Rousse [EMAIL PROTECTED] writes: Jerome just sent me two new skins for xine (one of which with official mdk colors) and asked me why xine was not in main part of the distribution. With current witch-hunt for napster client, i think potential legal problems are currently main Mandrake concern, even if i never saw any official position on this topic. So, which is the best way to have at least an explanation on this subject, or much better, make this position evolves ? moved to main -- Warly
Re: [Cooker] via-rhine broken in latest kernels
bill == Bill Lea Greenwood [EMAIL PROTECTED] writes: bill Thanks for all the great work on 8.2 guys... you are all very much bill appreciated! :) bill via-rhine is still broken though: bill # insmod via-rhine bill Using /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_link_ok_Rsmp_ba4049b4 bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_nway_restart_Rsmp_dceb1f73 bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_ethtool_sset_Rsmp_6d966584 bill /lib/modules/2.4.17-16mdksecure/kernel/drivers/net/via-rhine.o.gz: bill unresolved symbol mii_ethtool_gset_Rsmp_a986f194 bill Has been for at least the last 8 kernels. 2.4.13-12 works perfectly. I am unable to reproduce that, this is with kernel-secure kernel? Please try latest. Here depmod -ae has not ouput, i.e. everything is ok. I changed things around 2.4.17.17mdk than now if there are unresolved symbols compilation breaks :) Later, Juan. -- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy
[Cooker] mdk8.2b2: Missing Euro Support in KDE
The Euro Symbol is not available in kde. Some applications like gvim will show the euro sign when pressing alt-gr E, but I found no way to do the same in a kde application. This is a very importand feature for the mandrake 8.2 release or mandrake is not very usefull in office desktop enviroments.
Re: [Cooker] what's wrong with xcdroast?
Claudio [EMAIL PROTECTED] writes: When I try to rip audio track in xcdroast of today's cooker here you are what I obtain: recording 4336.06933 seconds stereo with 16 bits @ 44100.0 Hz -'/home/master/track-01'... 0%Fatal error: did not drop group privilege. child reader sem request failed W Child exited with 1 AFAIK it happen since beta2, at least. NB-1) Xcdroast won't work as root too! NB-2) Gcombust work fine (as normal user too) NB-34) cdda2wav and cdparanoia work both fine (as root and as user) does cdparanoia works in standalone ? -- Warly
[Cooker] rpmdb: var/lib/rpm/Basenames: No such file or directory
Hi. I'm right now doing a network install from ftp.sunet.se of the current cooker. On console 1, I noticed a lot of these messages: rpmdb: var/lib/rpm/Basenames: No such file or directory Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.iso-top.de | Jabber: [EMAIL PROTECTED] iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 8 hours 26 minutes
Re: [Cooker] 2.4.17-20mdk kernel corrupt?
Borsenkow Andrej [EMAIL PROTECTED] writes: bor@cooker% rpm --checksig /mnt/hd/cooker/i586/Mandrake/RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm /mnt/hd/cooker/i586/Mandrake/RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm: MD5 (GPG) NOT OK (MISSING KEYS: GPG#9B4A4024) Synced off sunsite.uio.no with rsync. mirror pb: bash-2.05# rpm -K /RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm /RPMS/kernel-2.4.17.20mdk-1-1mdk.i586.rpm: md5 gpg OK -- Warly
[Cooker] mdk8.2b2 bug: root 18n,nls and locale problems
Translations may be wrong and are sometimes hard to understand. Also many translated man pages are out of date. This is very dangerous for system adminstration. Disable i18n, nls and locale support for user root. Simply do an echo -n /root/.i18n to disable it.
[Cooker] mdk8.2b2 bug: artsd
atrsd will crash when noatum is invoked. Try: # artswrapper artsd -F 4 -S 2048 -m artsmessage -l 3 -f [1] 9692 running as realtime process now (priority 50) There are already artsd objects registered, looking if they are active... ... cleaned 5 unused mcop global references. # noatun [2] 9741 unix_connect: can't connect to server (unix:/tmp/mcop-stefani/deepthought-25dc-3c74e0c4) Launched ok, pid = 9745 [1]- Segmentation fault artswrapper artsd -F 4 -S 2048 -m artsmessage -l 3 -f
[Fwd: Re: [Cooker] Beta 3 Install report]
On Wednesday 20 February 2002 07:42 am, you wrote: Gateway 9150Xl notebook - Xircom Realport (RBEM-56G) card 1) I have a DSL account through T-Online here in Germany. I have a Linksys 4 port router which runs DHCP. During install I am never asked about hostname I have a Linksys 4 port router running DHCP and I am asked, when doing a fresh install, the hostname, IP address, etc. But I am not with T-Online in Germany and my Linux server is given a static IP address. The DHCP is only for my laptop and other desktops. Luc -- Luc Roseberry Consultant Tivoli certifié / Certified Tivoli Consultant Facilité Informatique Canada -- Luc Roseberry Consultant Tivoli certifié / Certified Tivoli Consultant Facilité Informatique Canada
Re: [Cooker] Tonights Install - DrakX 1.658
Pixel [EMAIL PROTECTED] writes: David Eastcott [EMAIL PROTECTED] writes: [...] b) During installation, msec is changing the the permissions of the lilio.conf file from 600 to 644 when I have the keyword 'password' is in it: DrakX sets it correctly. The result is a Warning message from lilo that lilio.conf should be readable by root only. fredl, can you make sure msec doesn't *add* any permission to lilo.conf? ok in next upload of msec -- Fred - May the source be with you
Re: [Cooker] mdk8.2b2: /var/log errors
On Thursday 21 February 2002 10:09 am, you wrote: Can't locate module char-major-226 This is fixed by adding alias char-major-226 agpgart to /etc/modules.conf A bug in 8.1, shouldn't this be fixed in 8.2? Hoyt
[Cooker] mdk8.2b2: dvd support
Some programs like videolan expect a dvd entry under /dev. Harddrake should make a dvd link entry to the dvd-device.
[Cooker] mdk8.2b2 bug: konsole looks ugly
The kde konsole looks very ugly when font antialiasing is enabled. Adding the parameter --noxft and it will have a much nicer look.
[Cooker] mdk8.2b2 bug: PATH Bug
The search order of the PATH enviroment variable is different under root and non-root user. This can make a different behaivor for shell scripts under user root and non-root user. PATH root is set as: /sbin:/usr/sbin:/bin:/usr/bin:/usr/X11R6/bin:/usr/local/bin:/usr/local/sbin PATH non-root is set as: /usr/X11R6/bin:/usr/local/bin:/bin:/usr/bin:/usr/games The best soultion would be: PATH for user root: /sbin:/usr/sbin:/usr/local/sbin:/bin:/usr/bin/:/usr/X11R6/bin:/usr/games:/usr/local/bin PATH for non-root user: /bin:/usr/bin/:/usr/X11R6/bin:/usr/games:/usr/local/bin
