Re: [Cooker] some files disapear in php rpms between 8.2 and9.0/cooker
Sorry for the late reply, I couldn't figure it out at first, then I moved, then there was the holidays, etc. But I finally have the answer: The Log.php file has disappeared from PHP, and is now only on PEAR. You can get it at http://pear.php.net/get/Log There are three types of files on PEAR: - The PFC (PHP Foundation classes), which are really important scripts, are stable, and conform to the coding standards. Those are bundled with the main PHP distribution. Log.php did not conform to this. - The other extensions, which are not bundled with PHP, but are on pear.php.net. - PECL, which are binary extensions to PHP that will not be bundled with the main PHP release. Jean-Michel Le dim 08/12/2002 à 12:32, Yves Duret a écrit : > It is the Log.php file (and the one under log/), wich is required by horde2. > Can we reintroduce back this file (and other related) ? or there is a > real reason to have wiped out them ? > > 08/12 17:26 yves@katu ~% urpmf Log.php > php-devel:/usr/lib/php/Log.php > phpgroupware:/var/www/html/phpgroupware/chora/inc/CVSLib/Log.php > phpgroupware:/var/www/html/phpgroupware/chora/lib/CVSLib/Log.php > 08/12 17:28 yves@katu ~% cat /etc/mandrake-release > Mandrake Linux release 8.2 (Bluebird) for i586 > 08/12 17:28 yves@katu ~% rpm -q php-devel > php-devel-4.1.2-1mdk > > > [yduret@klama rpm]$ urpmf Log.php > phpgroupware:/var/www/html/phpgroupware/chora/inc/CVSLib/Log.php > [yduret@klama rpm]$ cat /etc/mandrake-release > Mandrake Linux release 9.1 (Cooker) for i586 > > thx > -- > Yves Duret > [EMAIL PROTECTED] > piouk toujours et meme apres ! >
[Cooker] [Bug 688] [XFree86-server] XFree86 segfaults with radeon7500 xinerama mode
https://qa.mandrakesoft.com/show_bug.cgi?id=688 [EMAIL PROTECTED] changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED --- Additional Comments From [EMAIL PROTECTED] 2003-01-05 01:21 --- problem fixed by XFree86-xfs-4.2.99.3-1.20021223.3mdk XFree86-75dpi-fonts-4.2.99.3-1.20021223.3mdk XFree86-100dpi-fonts-4.2.99.3-1.20021223.3mdk XFree86-server-4.2.99.3-1.20021223.3mdk XFree86-devel-4.2.99.3-1.20021223.3mdk XFree86-libs-4.2.99.3-1.20021223.3mdk XFree86-4.2.99.3-1.20021223.3mdk --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] description: XFree86 segfaults with radeon7500 when configured in dual head xinerama mode # rpm -qa XFree* XFree86-100dpi-fonts-4.2.1-14mdk XFree86-devel-4.2.1-14mdk XFree86-75dpi-fonts-4.2.1-14mdk XFree86-doc-4.2.1-14mdk XFree86-libs-4.2.1-14mdk XFree86-compat-libs-4.1.0-2mdk XFree86-xfs-4.2.1-14mdk XFree86-server-4.2.1-14mdk XFree86-static-libs-4.2.1-14mdk XFree86-4.2.1-14mdk
Re: [Cooker] Mozilla won't start since last urpmi --auto-select--auto ....
On Thu, 2 Jan 2003, Ibukun Olumuyiwa wrote: > > > On Wed, 2003-01-01 at 21:39, Ibukun Olumuyiwa wrote: > >> Anyways...and I hope the Mozilla maintainer(s) take note of this: The > >> dependencies in that rpm need to be fixed. I'm not sure which it is, > >> whether glibc or GTK, but upgrading both of those packages fixed the > >> problem for me. There's no reason why any upgrade should fail if all > >> dependencies are correctly resolved. > >> > > I don't think you read my post completely. The problem is not with urpmi. > The problem is with the mozilla rpm itself. The rpm needs to be updated > with new dependencies. That's exactly what I'm ranting about. > The dependencies are fine for the distribution it is intended for. It is virtually impossible to keep all packages perfect for all distributions, heck, it's difficult just keeping some packages working well on just the supported distros. Is it really worthwhile (financially) to keep people who are (against recommendations) runnning some unknown mismatch of packages that only semi-resembles any Mandrake release? Rebuild the SRPM on your system, and then post back. Dynamic (ie output from ldd) dependencies are supposed to do this, assuming authors bump the major version o f their libraries when they break something. So go file a bug in Gnome bugzilla (or wherever best) about it if your own build works. Of course, you're on your own building the SRPM ... Buchan -- |Registered Linux User #182071-| Buchan MilneMechanical Engineer, Network Manager Cellphone * Work+27 82 472 2231 * +27 21 8828820x121 Stellenbosch Automotive Engineering http://www.cae.co.za GPG Key http://ranger.dnsalias.com/bgmilne.asc 1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
Re: [Cooker] Printing from MS
On Thu, 2 Jan 2003, Palmer, Hilary wrote: Is there any reason you posted 2 identical messages, except for the subject line? I responded to the first, since there were no replies to it ... to find there was a response to the 2nd ... Please try and make it easier for everyone to help resolve your problems by making it easy to follow email threads (especially using text-based mail apps remotely over a 28kbps connection ...) Regards, Buchan -- |Registered Linux User #182071-| Buchan MilneMechanical Engineer, Network Manager Cellphone * Work+27 82 472 2231 * +27 21 8828820x121 Stellenbosch Automotive Engineering http://www.cae.co.za GPG Key http://ranger.dnsalias.com/bgmilne.asc 1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
Re: [Cooker] Re: [CHRPM] hdparm-5.3-2mdk
On Thu, 2 Jan 2003, [iso-8859-1] Götz Waschk wrote: > Am Donnerstag, 2. Januar 2003, 16:15:17 Uhr MET, schrieb Oden Eriksson: > > > +* Thu Jan 02 2003 Thierry Vignaud <[EMAIL PROTECTED]> 5.3-2mdk > > > +- build release > > > > Excuse a stupid question, but what exactly does "build release" mean? > > Some people are too lazy to write "rebuild for new glibc and rpm", > including myself :-) isn't 'rebuild' just as descriptive, and shorter? ;-) -- |Registered Linux User #182071-| Buchan MilneMechanical Engineer, Network Manager Cellphone * Work+27 82 472 2231 * +27 21 8828820x121 Stellenbosch Automotive Engineering http://www.cae.co.za GPG Key http://ranger.dnsalias.com/bgmilne.asc 1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
Re: [Cooker] Re: New cooker mozilla doesn't work...
On 2 Jan 2003, Pbt wrote: > All right. Excuse me then. > When it's kindly answered, it's very clear ! ;) > The one option would be to rebuild mozilla on your system before making bug reports, since you probably have a glibc issue (although a rebuild would solve most). Of course, problems rebuilding on your system will nt be appreciated on cooker either, so you're on your own ... Regards, Buchan -- |Registered Linux User #182071-| Buchan MilneMechanical Engineer, Network Manager Cellphone * Work+27 82 472 2231 * +27 21 8828820x121 Stellenbosch Automotive Engineering http://www.cae.co.za GPG Key http://ranger.dnsalias.com/bgmilne.asc 1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
Re: [Cooker] SAMBA/CUPS Printing Issues.
On Tue, 31 Dec 2002, Palmer, Hilary wrote:
> Am I missing something?
>
> In order for me to get my Windows 2000 box to print to a HP DeskJet 1220c I
> need to do the following:
>
> 1. Have the printer in cups set to "HP DeskJet 1220C, CUPS+GIMP-print
> v4.2.3"
> 2. Make sure that I use a PostScript printer driver on Windows 2000 (I am
> using HP DJ 1200/PS).
> 3. Change the print command in smb.conf from "lpr-cups -P %p -o raw %s -r"
> to "lpr-cups -P %p -r %s"
Instead of 2. and 3. you could (I guess, don't have a DJ 1200 PS printer
available) use the PCL driver.
>
> Now when I want to print I need to:
> 1. Pause the printer on my Linux system.
> 2. Print my document in Windows.
> 3. Edit the print file with vi and remove the top 2 and bottom 2 lines that
> contain some JPL commands.
> 4. Edit the print file with Kedit to remove the "^M" character from every
> line.
dos2unix? (or is it d2u?)
> 5. Reenable the printer.
What happens when you print directly from linux?
Does your windows PS driver have options such as "optimise for
portability", which is usually not the default, since "optimise for speed"
is. This is the case with the Adobe drivers.
BTW, we have a Canon LBP-660 laser printer, and I have to jump through
hoops to print from linux. Till, if you read this thread, interestingly
enough I could print to the LBP-660 via the network with the printer test
pages in Mandrake 8.{1,2} (IIRC), but nothing else. It seems to do PCL
emulation in software, which is only available if the printer is on a
windows 9x machine (AFAIK, win2k support doesn't look good ...). So I was
using some sort of HP Laserjet (1 or IIP) PCL driver.
>
> Also the status information on the Windows system says "Access denied,
> unable to connect". Is there a way to fix that?
Are you authenticated by samba? ie do you see a share with your username
when you browser the machine from windows? I would guess you aren't, and
you should really just add yourself a windows password (as root):
# smbpasswd -a
should be a valid unix account, which you also log into windows
with, and the password you enter should be the password you log into
windows with.
If you aren't authenticating, you won't be able to do things like delete
print jobs ...
If it were a winxp box, it would probably be that you had the firewall
enabled, which by default firewalls off ports 137-139, which would give
you similar problems to pre-windows2000 servers (including Windows NT
server).
FYI, newer CUPS versions are supposed to be able to auto-detect PS vs
non-PS jobs, so hopefully newer versions of samba should be able to drop
the default -o raw option. Also, CUPS-1.17 had a PS driver available for
windows, which coud make setting up printers easier (so that CUPS could
auto-install windows print drivers on the samba server).
Unfortunately I have been spending most samba time on samba3 thus far, but
if you have time to experiment I could think of some things for you to try
...
Oh, and I just got back from holiday (no connection ...) and have waded
through my inbox, and about half the cooker backlog ...
Regards,
Buchan
--
|Registered Linux User #182071-|
Buchan MilneMechanical Engineer, Network Manager
Cellphone * Work+27 82 472 2231 * +27 21 8828820x121
Stellenbosch Automotive Engineering http://www.cae.co.za
GPG Key http://ranger.dnsalias.com/bgmilne.asc
1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
[Cooker] cooker hard drive install lilo config and options for expert mode
Hi. To complete Buchan Milne contribution in response to Narfi Stefansson, here is a detailed HD install lilo config. please replace my config by YOUR CONFIG for all "hdb5 labels" COOKER-BASE-ABSOLUTE-DIRECTORY => /home/hdb5/devel/cooker/i586/ COOKER-BASE-RELATIVE-DIRECTORY => /devel/cooker/i586/ COOKER-BASE-PARTITION => hdb5 COOKER-BASE-DISK => hdb COOKER-LILO-INSTALL-LABEL => ins_hd_img_hdb5 [root@pc root]# cd /boot [root@pc boot]# mkdir hd_img [root@pc boot]# mkdir -p /var/tmp/tmp_mount [root@pc boot]# mount /home/hdb5/devel/cooker/i586/images/hd.img -o loop /var/tmp/tmp_mount/ [root@pc boot]# cp -p /var/tmp/tmp_mount/* hd_img edit /etc/lilo.conf to obtain that difference [root@pc boot]# diff /etc/lilo.conf /etc/lilo.conf.svg 69,76d68 < image=/boot/hd_img/vmlinuz < label=ins_hd_img_hdb5 < root=/dev/ram3 < initrd=/boot/hd_img/hd.rdz < ramdisk=32000 < vga=791 < append="automatic=method:disk,disk:hdb,par:hdb5,dir:/devel/cooker/i586" < read-only [root@pc boot]# lilo Added linux * Added linux-nonfb Added failsafe Added floppy Added ins_hd_img_hdb5 [root@pc boot]# reboot Thanks again for Buchan Milne to start this contribution. I have try to add expert in append string like append="expert automatic=method:disk,disk:hdb,par:hdb5,dir:/devel/cooker/i586" but "automatic install" feature fail on two newer question "Third party modules" and "Scsi access" does anybody know the string to realize "automatic install" in expert mode. BCL
Re: [Cooker] Re: [CHRPM] XFree86-4.2.99.3-0.20022312.1mdk
On Sun, Jan 05, 2003 at 01:12:41AM +0200, Buchan Milne wrote: > Only if the date on the souce tarball (notice -r flag) had changed, which > should only really happen if the source had changed. It's a bit of a hack, > but it should be useful IMHO. I haven't tested this yet (just got back > from holiday ..).. Ohhh I see what you're doing. I missed that you were getting the date from the tarball... -- Ben Reser <[EMAIL PROTECTED]> http://ben.reser.org "If you're not making any mistakes, you're flat out not trying hard enough." - Jim Nichols
Re: [Cooker] Re: [CHRPM] XFree86-4.2.99.3-0.20022312.1mdk
On Thu, 26 Dec 2002, Ben Reser wrote:
> On Thu, Dec 26, 2002 at 10:22:44PM +0200, Buchan Milne wrote:
> > I think we could do with a few more rpm macros, or some scripts to
> > automate building from cvs a bit more. How about:
> >
> > %define rel 1mdk
> >
> > %if build_cvs
> > Release:%(echo "0.`date +%Y%m%d -r %{SOURCE}`.%{rel}")
> > %else
> > Release:rel
> > %endif
> >
> > I guess I should look at finishing up urpmb, with or without a working
> > 'rpm -q --qf "%{SOURCE}\n" --specfile $PACKGE.spec' ...
>
> Problem that would make the release tag change if the package gets
> rebuilt a few days later.
Only if the date on the souce tarball (notice -r flag) had changed, which
should only really happen if the source had changed. It's a bit of a hack,
but it should be useful IMHO. I haven't tested this yet (just got back
from holiday ..)..
Regards,
Buchan
--
|Registered Linux User #182071-|
Buchan MilneMechanical Engineer, Network Manager
Cellphone * Work+27 82 472 2231 * +27 21 8828820x121
Stellenbosch Automotive Engineering http://www.cae.co.za
GPG Key http://ranger.dnsalias.com/bgmilne.asc
1024D/60D204A7 2919 E232 5610 A038 87B1 72D6 AC92 BA50 60D2 04A7
[Cooker] gtk+2.0/XFree86/pango rebuilding & Xft missing development files
Hi,
I've found a collection of related issues centering around Xft.
Previously, the Xft include files (/usr/X11R6/include/X11/Xft/*) were
provided in the libXft2-devel RPM. This RPM was made obsolete by
XFree86-devel, which says it provides "libXft2-devel". However, the Xft
include files and /usr/X11R6/bin/xft-config script are *not* included.
This causes a rebuild of the pango SRPM to fail on a current cooker
(BuildRequires are fine, but the Xft files are missing).
Also, gtk+2.0-2.2.0-1mdk has a BuildRequire on Xft-devel.
libXft2-devel Provided libXft-devel, libXft2-devel and Xft-devel.
XFree86-devel only Provides libXft2-devel.
I've attached two patches:
(1) Fix gtk+2.0 spec to BuildRequire libXft2-devel instead of Xft-devel
(2) Fix XFree86-devel spec to package the Xft header files and xft-config.
I'm not sure whether (1) should be fixing the gtk+2.0 spec or adding
Provides to XFree86-devel. I decided updating the gtk+2.0 spec was
better than adding more to the XFree86 spec.
Also, I'm not that familiar with Xft/fontconfig/XFree86, so take (2)
carefully. There is a warning of some other unpackaged files at the end
of the XFree86 rebuild - perhaps someone knowledgeable should have a
look at whether they should be included.
In case it isn't clear these are the versions that I'm talking about:
XFree86-4.2.99.3-1.20021223.3mdk
gtk+2.0-2.2.0-1mdk
pango-1.2.0-1mdk
--
Mark Scott
--- gtk+2.0.spec.erroneous 2003-01-02 08:44:44.0 +
+++ gtk+2.0.spec2003-01-03 15:59:05.0 +
@@ -50,7 +50,7 @@
URL: http://www.gtk.org
Requires: common-licenses
-BuildRequires: Xft-devel
+BuildRequires: libXft2-devel
BuildRequires: autoconf2.5
BuildRequires: gettext-devel
BuildRequires: libglib2.0-devel >= %{req_glib_version}
--- XFree86.spec.xft2002-12-31 13:38:28.0 +
+++ XFree86.spec2003-01-04 00:16:12.0 +
@@ -1715,9 +1715,7 @@
/usr/X11R6/include/X11/Xaw
/usr/X11R6/include/X11/Xmu
/usr/X11R6/include/X11/Xcursor
-%if ! %{with_new_fontconfig_Xft}
-%{_x11dir}/include/X11/Xft
-%endif
+/usr/X11R6/include/X11/Xft
/usr/X11R6/include/X11/*.h
/usr/X11R6/include/GL
/usr/X11R6/include/DPS
@@ -1731,6 +1729,7 @@
/usr/X11R6/bin/makedepend
/usr/X11R6/bin/gccmakedep
/usr/X11R6/bin/xmkmf
+/usr/X11R6/bin/xft-config
/usr/X11R6/man/man1/imake.1x*
/usr/X11R6/man/man1/makedepend.1x*
[Cooker] rpm-helper ghost code
del-user seems to miss actual code to remove user: if [ $# != 3 ]; then echo "usage: $0 " 1>&2 exit 1 fi pkg=$1 # name of the package num=$2 # number of packages installed name=$3 # name of the user # del-user ends here [guillaume@baader www]$ rpm -q rpm-helper rpm-helper-0.8-1mdk -- Any cool program always requires more memory than you have. -- Murphy's Computer Laws n°2
[Cooker] Request for IMQ in Mandrake kernel
Hi everybody! I would like to see IMQ (http://trash.net/~kaber/imq) in Mandrake kernel as I think is the only way to do ingress shaping in linux. What do you think? Is this possible? --- Correo enviado usando el servicio de Webmail del SIAIN: http://siain.udc.es
[Cooker] DrakX 1.760 new features?
Noticed a few changes in the way things work and would like any hints about new or removed or changed behaviour regarding the auto install? Couple questions that come to mind: 1. Pixel, did you ever get around to fixing the raidtab problem? eg. If you re-installed on a system that previously had raid set up, and this time around you decided to use standard partitions, the old raidtab file was left in /etc with the result that the first boot of the newly install system always failed. 2. Noticed 'libsafe' in the auto_config.pl file. What is it for? How to use? Dave
Re: [Cooker] xine-dxr3-1-0.beta2.1mdk.i586.rpm
Am Samstag, 4. Januar 2003, 13:43:51 Uhr MET, schrieb Charles A Edwards: > On Sat, 4 Jan 2003 19:12:32 +0100 > Götz Waschk <[EMAIL PROTECTED]> wrote: > If you build xine-lib yourself, it will pick up an installed libfame. > > The official cooker version of this package doesn't have this > > dependancy. > Yes, but it does carry the BuildRequire. > Are you saying it is OK for a pkg in main to have a BuildRequire for a > pkg that is not in the distro as long as all pkgs it Requires are in the > distro? Yes, the buildrequire is only for the plf package, if you build the package with rpm --rebuild --with plf This was already discussed on the list, the common sense was that it's ok to build the plf package from the same spec file as the mdk version. > On another page but still about xine. > Have you been able to find any info anywhere regarding the the > non-building xvid? No > And I am guessing that to be able to use an external libdvdnav you have > to go cvs for the version xine wants. Yes, at the moment I use the included static version of libdvdnav and libdvdread. After the next release of libdvdnav I'll enable the buildrequires and link to the shared version. > 1 more thing xine-ui 0.17 has been released and easily builds with > existing spec modified only for version change done -- Götz Waschk <> master of computer science <> University of Rostock http://wwwtec.informatik.uni-rostock.de/~waschk/waschk.asc for PGP key --> Logout Fascism! <--
Re: [Cooker] DrakX: cdrom modules
No Problem. On Saturday 04 January 2003 10:43, Pixel wrote: > David Eastcott <[EMAIL PROTECTED]> writes: > > On Saturday 04 January 2003 08:59, rcc wrote: > > > drakx fails at install_steps_interactive.pm line 196 with an > > > "undefined" error. Seems that this is on loading firewire mods. > > > However, I'm not that much concerned about that, I can easily skip that > > > and rebuild stage2. > > > > Actually the error is: > > > > Warning: undefined subroutine &modules::interactive::load_category called > > ... > > > > removing the ::interactive at lines 170, 196 and 198 solves the problem. > > Maybe additional issue because the number of parameters at lines 170, 196 > > and 198 are different. > > oops. the pb is a missing "use modules::interactive;" > > fixing! > > > Also, earlier error still present which prevents printer setup: > > > > could not start chmod! at ... /common.pm line 226 ... > > there are quite a few pbs with printerdrake at the moment. I tried to > fix some but there are tvignaud's bugs and don't want him to reproach > me for stealing his bugs ;) > > > > What troubles me is that the cdrom modules aren't loaded with hd > > > install. Both cdrom and ide-cd get a "not found". And I don't know > > > where these modules are. I can't extract them from > > > modules.cz-2.4.19-16mdkBOOT. I took parts of the insmod perl out of > > > commands.pm into a small perl script and used it to extract modules, > > > but those cdrom modules seem not to be in there. Are they in the > > > initial ramdisk? > > kernel 2.4.19-16mdkBOOT has neither ide-cd nor cdrom, those modules > are built-in the kernel. > > > > btw, are there any docs about making the initial boot floppy (stage1 > > > and the .rdz)? > > available documentation is in > http://www.mandrakelinux.com/cgi-bin/cvsweb.cgi/gi/docs and > http://www.mandrakelinux.com/cgi-bin/cvsweb.cgi/gi/mdk-stage1/doc
Re: [Cooker] internet connection on bootup
Le Samedi 4 Janvier 2003 18:30, Sascha Noyes a écrit : > > However, this is not the perfect solution, as you still need loopback for > > most applications. My personal solution is to have a network profile with > > loopback only for when i'm roaming, and to switch between different > > profiles with symlinks. > > Could you give a short synopsis of how you create a netword profile, and > how you switch between different profiles with symlinks? Or maybe point me > in the way of some documentation? Sure. Identify all files involved in a given network configuration, which means basically: /etc/sysconfig/network /etc/sysconfig/network-scripts/ifcfg-eth0 /etc/hosts /etc/resolv.conf And potentially others: /etc/exports /etc/rcx.d etc... Just rename them to name.foo, where name is original file name, and foo is your profile name. Then use attached script to switch from a given profile to another. draknet also had limited support for different network profiles, but i don't know current state. -- No matter how good of a deal you get on computer components, the price will always drop immediately after the purchase. -- Murphy's Computer Laws n°7 switch-network-profile Description: application/shellscript
Re: [Cooker] xine-dxr3-1-0.beta2.1mdk.i586.rpm
On Sat, 4 Jan 2003 19:12:32 +0100 Götz Waschk <[EMAIL PROTECTED]> wrote: > If you build xine-lib yourself, it will pick up an installed libfame. > The official cooker version of this package doesn't have this > dependancy. Yes, but it does carry the BuildRequire. Are you saying it is OK for a pkg in main to have a BuildRequire for a pkg that is not in the distro as long as all pkgs it Requires are in the distro? On another page but still about xine. Have you been able to find any info anywhere regarding the the non-building xvid? And I am guessing that to be able to use an external libdvdnav you have to go cvs for the version xine wants. 1 more thing xine-ui 0.17 has been released and easily builds with existing spec modified only for version change Charles Man is a military animal, glories in gunpowder, and loves parade. -- P.J. Bailey -- Mandrake Linux 9.1 Kernel- 2.4.20-2mdk --
[Cooker] Sane Problem.
I have a ScanJet IIcx C2500A 3332 SCSI scanner. It worked just fine with Mandrake 8.2 if I do a sane-findscaner it says: found SCSI processor "HP C2500A 3332" at /dev/scanner found SCSI processor "HP C2500A 3332" at /dev/sg0 but if I try to run xsane or scanimage -L it says that it can't find any scanner. if I run scannerdrake it says that HP C2500A not found in the database. If i chose to configure it manualy it just dies. I have: those rpm's installed: [pucko@c-571472d5 pucko]$ rpm -qa | grep sane libsane-hpoj0-0.90-4mdk xsane-gimp-0.90-1mdk libsane1-devel-1.0.9-1mdk xsane-0.90-1mdk libsane1-1.0.9-1mdk sane-frontends-1.0.9-1mdk sane-backends-1.0.9-1mdk [pucko@c-571472d5 pucko]$
Re: [Cooker] xine-dxr3-1-0.beta2.1mdk.i586.rpm
Am Samstag, 4. Januar 2003, 07:18:27 Uhr MET, schrieb Charles A Edwards: > xine-dxr3-1-0.beta2.1mdk.i586.rpm has a BuildRequire for > libfame0.8-devel and requires libfame-0.8.so.0. > The libfame rpms are plf. > Should not this rpm also be plf. If you build xine-lib yourself, it will pick up an installed libfame. The official cooker version of this package doesn't have this dependancy. CU -- Götz Waschk <> master of computer science <> University of Rostock http://wwwtec.informatik.uni-rostock.de/~waschk/waschk.asc for PGP key --> Logout Fascism! <--
Re: [Cooker] Saving Mandrake - older distros
Le Dimanche 29 Décembre 2002 06:08, Leon Brooks a écrit : > > Not only can ISOs be found for versions back to 7.2 but also many of the > > above carry the full directory tree for those versions. > > Ta. That's the most helpful, labour-saving response I've seen. I don't think old ISO are really usefuls. However, directory trees are, especially for production servers you don't want to update. Unfortunatly, the very fact that path change as they are moved to mandrake-old obsoletes all urpmi settings... -- Any cool program always requires more memory than you have. -- Murphy's Computer Laws n°2
Re: [Cooker] DrakX: cdrom modules
David Eastcott <[EMAIL PROTECTED]> writes: > On Saturday 04 January 2003 08:59, rcc wrote: > > drakx fails at install_steps_interactive.pm line 196 with an "undefined" > > error. Seems that this is on loading firewire mods. However, I'm not > > that much concerned about that, I can easily skip that and rebuild > > stage2. > > > > Actually the error is: > > Warning: undefined subroutine &modules::interactive::load_category called ... > > removing the ::interactive at lines 170, 196 and 198 solves the problem. > Maybe additional issue because the number of parameters at lines 170, 196 and > 198 are different. oops. the pb is a missing "use modules::interactive;" fixing! > Also, earlier error still present which prevents printer setup: > > could not start chmod! at ... /common.pm line 226 ... there are quite a few pbs with printerdrake at the moment. I tried to fix some but there are tvignaud's bugs and don't want him to reproach me for stealing his bugs ;) > > What troubles me is that the cdrom modules aren't loaded with hd > > install. Both cdrom and ide-cd get a "not found". And I don't know where > > these modules are. I can't extract them from > > modules.cz-2.4.19-16mdkBOOT. I took parts of the insmod perl out of > > commands.pm into a small perl script and used it to extract modules, but > > those cdrom modules seem not to be in there. Are they in the initial > > ramdisk? kernel 2.4.19-16mdkBOOT has neither ide-cd nor cdrom, those modules are built-in the kernel. > > > > btw, are there any docs about making the initial boot floppy (stage1 and > > the .rdz)? available documentation is in http://www.mandrakelinux.com/cgi-bin/cvsweb.cgi/gi/docs and http://www.mandrakelinux.com/cgi-bin/cvsweb.cgi/gi/mdk-stage1/doc
Re: [Cooker] internet connection on bootup
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Saturday 04 January 2003 12:24, Guillaume Rousse wrote: > Le Lundi 30 Décembre 2002 12:36, Sascha Noyes a écrit : > > Hi, > > > > Yes, the box is a laptop. It is sometimes connected to a broadband > > gateway which gives out an IP address via DHCP, but sometimes not. If it > > is disconnected everything (as Brook Humphrey noted) runs extremely > > slowly. I did: "chkconfig --level 2345 network off", but this only turned > > off networking on the next boot. > > You must use chkconfig --del network to permanently prevent network at > boot. Then use service network start/stop manually to launch/shutdown the > network. > > However, this is not the perfect solution, as you still need loopback for > most applications. My personal solution is to have a network profile with > loopback only for when i'm roaming, and to switch between different > profiles with symlinks. Could you give a short synopsis of how you create a netword profile, and how you switch between different profiles with symlinks? Or maybe point me in the way of some documentation? Thank you very much, Sascha - -- Please encrypt all correspondence. PGP key available from: http://individual.utoronto.ca/noyes/snoyes.asc - -- -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQE+FxpHgzJdfX+cTW8RAuFVAJ0eNeYEzNGoM0f0b5fifo4513/mWwCeJ+8X 7xQhfGJQPQ3rWny5C+7yNqo= =UcqP -END PGP SIGNATURE-
Re: [Cooker] DrakX: cdrom modules
On Saturday 04 January 2003 08:59, rcc wrote: > drakx fails at install_steps_interactive.pm line 196 with an "undefined" > error. Seems that this is on loading firewire mods. However, I'm not > that much concerned about that, I can easily skip that and rebuild > stage2. > Actually the error is: Warning: undefined subroutine &modules::interactive::load_category called ... removing the ::interactive at lines 170, 196 and 198 solves the problem. Maybe additional issue because the number of parameters at lines 170, 196 and 198 are different. [...] Also, earlier error still present which prevents printer setup: could not start chmod! at ... /common.pm line 226 ... Dave
Re: [Cooker] internet connection on bootup
Le Lundi 30 Décembre 2002 12:36, Sascha Noyes a écrit : > Hi, > > Yes, the box is a laptop. It is sometimes connected to a broadband gateway > which gives out an IP address via DHCP, but sometimes not. If it is > disconnected everything (as Brook Humphrey noted) runs extremely slowly. I > did: "chkconfig --level 2345 network off", but this only turned off > networking on the next boot. You must use chkconfig --del network to permanently prevent network at boot. Then use service network start/stop manually to launch/shutdown the network. However, this is not the perfect solution, as you still need loopback for most applications. My personal solution is to have a network profile with loopback only for when i'm roaming, and to switch between different profiles with symlinks. -- When you finally buy enough memory, you will not have enough disk space. -- Murphy's Computer Laws n°3
Re: [Cooker] PATCH: 2.4.20-2mdk: scsi_error timeout resubmitted
On Saturday 04 January 2003 13:53, Andrey Borzenkov wrote: > I have some old patches still lingering around. so you do read cooker:) > Danny, I understand you maintain some unofficial kernel. You are adviced to > add this one as well, current SCSI error handling too easily goes into > endless loop on innocent media errors. It is already in my kernel on mandrakeclub.com Danny
[Cooker] DrakX: cdrom modules
drakx fails at install_steps_interactive.pm line 196 with an "undefined" error. Seems that this is on loading firewire mods. However, I'm not that much concerned about that, I can easily skip that and rebuild stage2. What troubles me is that the cdrom modules aren't loaded with hd install. Both cdrom and ide-cd get a "not found". And I don't know where these modules are. I can't extract them from modules.cz-2.4.19-16mdkBOOT. I took parts of the insmod perl out of commands.pm into a small perl script and used it to extract modules, but those cdrom modules seem not to be in there. Are they in the initial ramdisk? btw, are there any docs about making the initial boot floppy (stage1 and the .rdz)? - Mark
[Cooker] Please, enable GRKERNSEC_CHROOT in secure kernel
This probably has been disabled due to problems with initrd, am I right? Current implementation of initrd in kernel has been changed so it is safe to enable CHROOT features again. I would advice to enable all of them because you always can disable them (using sysctl) on case by case basis. This is safe to enable in 2.4.20 and above. It is NOT safe to enable in 2.4.19 and before. In case somebody does not remember it - linuxrc from initrd is (and has been) run as chrooted process. It means, grsecurity denied mknod, mount and pivot_root. Implementation of linuxrc in current kernel has been changed in such way that does not interfere with grsecurity (I do not know if this was intentional, is not, it is a pleasant side effect :)) cheers -andrey P.S. I am not on list so I appreciate Cc on replies if any.
[Cooker] Is secure kernel really secure?
grsecurity has one "feature" - if you enable sysctl support it starts up with most security features disabled. You must manually enable them using sysctl interface. Mandrake 9.0 and above now enables sysctl support. Which means that most grsecurity features are actually off on default installation: [bor@iap-pxy-mow1 bor]$ sudo sysctl -a | grep grsec kernel.grsecurity.grsec_lock = 0 kernel.grsecurity.rand_bind = 0 kernel.grsecurity.cap_prot = 0 kernel.grsecurity.rand_rpc = 0 kernel.grsecurity.dmesg = 0 kernel.grsecurity.audit_mount = 0 kernel.grsecurity.altered_pings = 0 kernel.grsecurity.rand_tcp_src_ports = 0 kernel.grsecurity.rand_ip_ids = 0 kernel.grsecurity.rand_pids = 0 kernel.grsecurity.chroot_caps = 0 kernel.grsecurity.timechange_logging = 0 kernel.grsecurity.forkfail_logging = 0 kernel.grsecurity.signal_logging = 0 kernel.grsecurity.execve_limiting = 0 kernel.grsecurity.fifo_restrictions = 0 kernel.grsecurity.linking_restrictions = 0 this is BIG INCOMPATIBLE change comparing with previous versions. How many people installed secure kernel just to be fooled by its "security"? I suggest adding a two line patch that removes check for sysctl and always makes grsecurity to come up with features enabled during compilation. This is much better than leaving users to do it manually (besides, in this case you always can disable them if needed). In case it is agreed patch will follow (do not have handy). -andrey P.S. I am not currently on cooker so I appreciate Cc in replies if any.
[Cooker] PATCH: 2.4.20-2mdk: ide-proc && ide_scan cleanup
Again from very old backlog.
This fixes my old error. The patch is against 2.4.20-2mdk, but it reverts DI92, so
Juan, if you intend to ever apply it :) just remove DI92 and skip drivers/ide/ide.c
chunk.
The patch fixes driver changing by writing into /proc/ide/hdX/driver. It does work to
some extent, but current code does not null-terminate written driver name so
ide_scan_device does not (always) find it.
DI92 must be removed in any case (it is buggy); this patch is optional as probably
nobody ever tried to play with IDE drivers this way :)) I repost it so it is not lost.
-andrey
--- linux-2.4.20-2mdk/drivers/ide/ide-proc.c.ide-proc-write-driver 2002-12-09
14:38:12.0 +0300
+++ linux-2.4.20-2mdk/drivers/ide/ide-proc.c2003-01-01 15:43:06.0 +0300
@@ -640,14 +640,23 @@
PROC_IDE_READ_RETURN(page,start,off,count,eof,len);
}
+#define DRIVER_REQ_SIZE (sizeof(((ide_drive_t *)0)->driver_req))
static int proc_ide_write_driver
(struct file *file, const char *buffer, unsigned long count, void *data)
{
ide_drive_t *drive = (ide_drive_t *) data;
+ chartmpbuf[DRIVER_REQ_SIZE];
+ size_t s = count < DRIVER_REQ_SIZE ? count : DRIVER_REQ_SIZE-1;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
- if (ide_replace_subdriver(drive, buffer))
+ /*
+* Driver name may be less than 9 characters
+* Make sure it is properly NULL-terminated
+*/
+ strncpy(tmpbuf, buffer, s);
+ tmpbuf[s] = '\0';
+ if (ide_replace_subdriver(drive, tmpbuf))
return -EINVAL;
return count;
}
--- linux-2.4.20-2mdk/drivers/ide/ide.c.ide-proc-write-driver 2002-12-09
14:38:17.0 +0300
+++ linux-2.4.20-2mdk/drivers/ide/ide.c 2003-01-01 15:25:53.0 +0300
@@ -3843,7 +3843,7 @@
for (unit = 0; unit < MAX_DRIVES; ++unit) {
ide_drive_t *drive = &hwif->drives[unit];
char *req = drive->driver_req;
- if (*req && !strstr(req, name))
+ if (*req && !strstr(name, req))
continue;
if (drive->present && drive->media == media && drive->driver
== driver && ++i > n)
return drive;
[Cooker] PATCH: 2.4.20-2mdk: scsi_error timeout resubmitted
I have some old patches still lingering around.
- 2.4.19-q18.
...
* scsi_error timeout patch
Juan, this is still unapplied (or have been lost during 2.4.19 -> 2.4.20 move). I
attach it against 2.4.20-2mdk.
Danny, I understand you maintain some unofficial kernel. You are adviced to add this
one as well, current SCSI error handling too easily goes into endless loop on innocent
media errors.
cheers
-andrey
--- linux-2.4.20-2mdk/drivers/scsi/scsi_error.c.scsi-error-timeout 2002-11-29
02:53:14.0 +0300
+++ linux-2.4.20-2mdk/drivers/scsi/scsi_error.c 2003-01-01 15:44:49.0 +0300
@@ -1100,6 +1100,8 @@
*/
STATIC int scsi_eh_completed_normally(Scsi_Cmnd * SCpnt)
{
+ int rtn;
+
/*
* First check the host byte, to see if there is anything in there
* that would indicate what we need to do.
@@ -1113,14 +1115,18 @@
* otherwise we just flag it as success.
*/
SCpnt->flags &= ~IS_RESETTING;
- return NEEDS_RETRY;
+ goto maybe_retry;
}
/*
* Rats. We are already in the error handler, so we now get to try
* and figure out what to do next. If the sense is valid, we have
* a pretty good idea of what to do. If not, we mark it as failed.
*/
- return scsi_check_sense(SCpnt);
+ rtn = scsi_check_sense(SCpnt);
+ if (rtn == NEEDS_RETRY) {
+ goto maybe_retry;
+ }
+ return rtn;
}
if (host_byte(SCpnt->result) != DID_OK) {
return FAILED;
@@ -1139,7 +1145,11 @@
case COMMAND_TERMINATED:
return SUCCESS;
case CHECK_CONDITION:
- return scsi_check_sense(SCpnt);
+ rtn = scsi_check_sense(SCpnt);
+ if (rtn == NEEDS_RETRY) {
+ goto maybe_retry;
+ }
+ return rtn;
case CONDITION_GOOD:
case INTERMEDIATE_GOOD:
case INTERMEDIATE_C_GOOD:
@@ -1154,6 +1164,17 @@
return FAILED;
}
return FAILED;
+
+ maybe_retry:
+
+ if ((++SCpnt->retries) < SCpnt->allowed) {
+ return NEEDS_RETRY;
+ } else {
+/*
+ * No more retries - report this one back to upper level.
+ */
+ return SUCCESS;
+ }
}
/*
[Cooker] xine-dxr3-1-0.beta2.1mdk.i586.rpm
xine-dxr3-1-0.beta2.1mdk.i586.rpm has a BuildRequire for libfame0.8-devel and requires libfame-0.8.so.0. The libfame rpms are plf. Should not this rpm also be plf. Charles While you don't greatly need the outside world, it's still very reassuring to know that it's still there. -- Mandrake Linux 9.1 Kernel- 2.4.20-2mdk --
[Cooker] Problems with harddrake
When I try to run harddrake It dies.
Here is the output:
Name "harddrake::data::tree" used only once: possible typo at
/usr/sbin/harddrake2 line 201 (#1)
(W once) Typographical errors often show up as unique variable names.
If you had a good reason for having a unique name, then just mention it
again somehow to suppress the message. The our declaration is
provided for this purpose.
sbus_probing::main::prom_walk() called too early to check prototype at
/usr/lib/libDrakX/sbus_probing/main.pm line 126 (#2)
(W prototype) You've called a function that has a prototype before the
parser saw a definition or declaration for it, and Perl could not check
that the call conforms to the prototype. You need to either add an
early prototype declaration for the subroutine in question, or move the
subroutine definition ahead of the call to get proper prototype
checking. Alternatively, if you are certain that you're calling the
function correctly, you may put an ampersand before the name to avoid
the warning. See perlsub.
sbus_probing::main::prom_walk() called too early to check prototype at
/usr/lib/libDrakX/sbus_probing/main.pm line 127 (#2)
Use of uninitialized value in hash element at
/usr/lib/libDrakX/sbus_probing/main.pm line 91 (#3)
(W uninitialized) An undefined value was used as if it were already
defined. It was interpreted as a "" or a 0, but maybe it was a mistake.
To suppress this warning assign a defined value to your variables.
To help you figure out what was undefined, perl tells you what operation
you used the undefined value in. Note, however, that perl optimizes your
program and the operation displayed in the warning may not necessarily
appear literally in your program. For example, "that $foo" is
usually optimized into "that " . $foo, and the warning will refer to
the concatenation (.) operator, even though there is no . in your
program.
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 100 (#3)
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 124 (#3)
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 125 (#3)
Use of uninitialized value in split at /usr/lib/libDrakX/detect_devices.pm
line
111 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/detect_devices.pm line 514 (#3)
Use of uninitialized value in pattern match (m//) at
/usr/lib/libDrakX/detect_devices.pm line 182 (#3)
TODO: XSetInputFocus if force_focus
TODO: ensure focus stuff
Use of uninitialized value in string eq at /usr/lib/libDrakX/ugtk2.pm line 887
(#3)
TODO: XSetInputFocus if force_focus
TODO: ensure focus stuff
Use of uninitialized value in string eq at /usr/sbin/harddrake2 line 216 (#3)
Use of uninitialized value in string eq at /usr/sbin/harddrake2 line 229 (#3)
Use of uninitialized value in pattern match (m//) at /usr/sbin/harddrake2 line
229 (#3)
Use of uninitialized value in sprintf at /usr/sbin/harddrake2 line 235 (#3)
Use of uninitialized value in index at /usr/lib/libDrakX/detect_devices.pm
line
199 (#3)
Use of uninitialized value in index at /usr/lib/libDrakX/detect_devices.pm
line
203 (#3)
Use of uninitialized value in pattern match (m//) at
/usr/lib/libDrakX/detect_devices.pm line 190 (#3)
Argument "SCSI" isn't numeric in sprintf at /usr/sbin/harddrake2 line 235 (#4)
(W numeric) The indicated string was fed as an argument to an operator
that expected a numeric value instead. If you're fortunate the message
will identify which operator was so unfortunate.
Use of uninitialized value in string eq at /usr/lib/libDrakX/harddrake/data.pm
line 46 (#3)
Use of uninitialized value in string ne at /usr/lib/libDrakX/harddrake/data.pm
line 15 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/scanner.pm line 33 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/scanner.pm line 34 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/mouse.pm line 264 (#3)
Uncaught exception from user code:
Is gtk2-perl missing Gtk2::GObject::set_active ?
Call trace:
Gtk2::_Object::AUTOLOAD() called from /usr/sbin/harddrake2:262
Gtk2::_Object::AUTOLOAD('Gtk2::GObject=SCALAR(0x8fd2178)',0) called at
/usr/sbin/harddrake2 line 262
Use of uninitialized value in subroutine entry at /usr/lib/libDrakX/c.pm line
10.
[Cooker] ScannerDrake does not work...
Scanner drake says thtat it can not find my scanner in the database and ask me
if I want to configure it manualy. I click yes and it just dies: Here is the
output:
Name "harddrake::data::tree" used only once: possible typo at
/usr/sbin/harddrake2 line 201 (#1)
(W once) Typographical errors often show up as unique variable names.
If you had a good reason for having a unique name, then just mention it
again somehow to suppress the message. The our declaration is
provided for this purpose.
sbus_probing::main::prom_walk() called too early to check prototype at
/usr/lib/libDrakX/sbus_probing/main.pm line 126 (#2)
(W prototype) You've called a function that has a prototype before the
parser saw a definition or declaration for it, and Perl could not check
that the call conforms to the prototype. You need to either add an
early prototype declaration for the subroutine in question, or move the
subroutine definition ahead of the call to get proper prototype
checking. Alternatively, if you are certain that you're calling the
function correctly, you may put an ampersand before the name to avoid
the warning. See perlsub.
sbus_probing::main::prom_walk() called too early to check prototype at
/usr/lib/libDrakX/sbus_probing/main.pm line 127 (#2)
Use of uninitialized value in hash element at
/usr/lib/libDrakX/sbus_probing/main.pm line 91 (#3)
(W uninitialized) An undefined value was used as if it were already
defined. It was interpreted as a "" or a 0, but maybe it was a mistake.
To suppress this warning assign a defined value to your variables.
To help you figure out what was undefined, perl tells you what operation
you used the undefined value in. Note, however, that perl optimizes your
program and the operation displayed in the warning may not necessarily
appear literally in your program. For example, "that $foo" is
usually optimized into "that " . $foo, and the warning will refer to
the concatenation (.) operator, even though there is no . in your
program.
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 100 (#3)
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 124 (#3)
Use of uninitialized value in string eq at
/usr/lib/libDrakX/sbus_probing/main.pm line 125 (#3)
Use of uninitialized value in split at /usr/lib/libDrakX/detect_devices.pm
line
111 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/detect_devices.pm line 514 (#3)
Use of uninitialized value in pattern match (m//) at
/usr/lib/libDrakX/detect_devices.pm line 182 (#3)
TODO: XSetInputFocus if force_focus
TODO: ensure focus stuff
TODO: XSetInputFocus if force_focus
TODO: ensure focus stuff
Use of uninitialized value in string eq at /usr/sbin/harddrake2 line 216 (#3)
Use of uninitialized value in string eq at /usr/sbin/harddrake2 line 229 (#3)
Use of uninitialized value in pattern match (m//) at /usr/sbin/harddrake2 line
229 (#3)
Use of uninitialized value in sprintf at /usr/sbin/harddrake2 line 235 (#3)
Use of uninitialized value in index at /usr/lib/libDrakX/detect_devices.pm
line
199 (#3)
Use of uninitialized value in index at /usr/lib/libDrakX/detect_devices.pm
line
203 (#3)
Use of uninitialized value in pattern match (m//) at
/usr/lib/libDrakX/detect_devices.pm line 190 (#3)
Argument "SCSI" isn't numeric in sprintf at /usr/sbin/harddrake2 line 235 (#4)
(W numeric) The indicated string was fed as an argument to an operator
that expected a numeric value instead. If you're fortunate the message
will identify which operator was so unfortunate.
Use of uninitialized value in string eq at /usr/lib/libDrakX/harddrake/data.pm
line 46 (#3)
Use of uninitialized value in string ne at /usr/lib/libDrakX/harddrake/data.pm
line 15 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/scanner.pm line 33 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/scanner.pm line 34 (#3)
Use of uninitialized value in concatenation (.) or string at
/usr/lib/libDrakX/mouse.pm line 264 (#3)
Uncaught exception from user code:
Is gtk2-perl missing Gtk2::GObject::set_active ?
Call trace:
Gtk2::_Object::AUTOLOAD() called from /usr/sbin/harddrake2:262
Gtk2::_Object::AUTOLOAD('Gtk2::GObject=SCALAR(0x8fafd7c)',0) called at
/usr/sbin/harddrake2 line 262
Use of uninitialized value in subroutine entry at /usr/lib/libDrakX/c.pm line
10.
Re: [Cooker] Fwd: PATCH: 2.4.20-2mdk: supermount combined patch
Le Vendredi 3 Janvier 2003 21:27, Danny Tholen a écrit : > Finally, Andrey send me a fix for supermount. > I will probably also merge this with my kernel on the club, unless an > official update is released? I've patched my current cooker kernel, and it works fine! new year is starting fine! > > > > Danny, your last problem - "invalid argument" after cd /mnt/cdrom, > > ejecting and reinserting CD, ls . is not a bug - it is a feature :-) > > After CD has been ejected all file information is invalidated. So your > > current directory does not exist anymore. The same as in case of > > somebody removing directory on NFS server. So the patch changes > > error to ESTALE in this case (it is more appropriate). Just do cd /; > > cd /mnt/cdrom and you can use CD again. With the patch, I don't get a ESTALE, but the normal content of the CD :) Emmanuel
[Cooker] [Bug 642] [kernel-2.4.20.2mdk] supermount doesn't work after some time...
https://qa.mandrakesoft.com/show_bug.cgi?id=642
[EMAIL PROTECTED] changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution||FIXED
--- Additional Comments From [EMAIL PROTECTED] 2003-01-04 11:59 ---
patch from: "??" fixes the problem.
--- You are receiving this mail because: ---
You are on the CC list for the bug, or are watching someone who is.
--- Reminder: ---
assigned_to: [EMAIL PROTECTED]
description:
With latest packaged kernel, the supermount problem hasn't been solved.
It's the same bug as reported before, by other people.
Closing the cdrom, I can use files. After some time (5mn perhaps), doing an ls on
the cdrom directory give me the names of the subdirectory, with 'No such file or
directory'.
but, not for all!
The latest directory who was used, didn't produce a 'no such file or directory', but
was a normal directory.
Here a sample from a strace from "ls /mnt/cdrom2"
22:00:36.102211 stat64("/mnt/cdrom2/", {st_mode=S_IFDIR|0777, st_size=0, ...}) = 0
22:00:36.103734 open("/mnt/cdrom2/",
O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_DIRECTORY) = 4
22:00:36.105731 fstat64(4, {st_mode=S_IFDIR|0777, st_size=0, ...}) = 0
22:00:36.106350 fcntl64(4, F_SETFD, FD_CLOEXEC) = 0
22:00:36.106631 brk(0x8061000) = 0x8061000
22:00:36.106875 getdents64(0x4, 0x805fcd0, 0x400, 0x40011a4c) = 504
22:00:36.107135 lstat64("/mnt/cdrom2/.", {st_mode=S_IFDIR|0777, st_size=0, ...}) =
0
22:00:36.108623 lstat64("/mnt/cdrom2/..", {st_mode=S_IFDIR|0755, st_size=4096,
...}) = 0
22:00:36.109319 lstat64("/mnt/cdrom2/Candy Dulfer - Sax-A-Go-Go", 0x805cff4) =
-1 ENOENT (No such file or directory)
22:00:36.110110 write(2, "ls: ", 4) = 4
22:00:36.110349 write(2, "/mnt/cdrom2/Candy Dulfer - Sax-A"..., 38) = 38
[Cooker] kmix in kde3.1rc5 is krazy !!!
Hello, I have talk about this problem some months ago, Laurent Montel said that he'll take a look about it ( http://www.mail-archive.com/cooker@linux-mandrake.com/msg80545.html ) but the problem is here, in rc5. Here, as part, a screenshot of kmix and here my lspcidrake : [root@cosmic cosmicflo]# lspcidrake unknown : VIA Technologies|VT8363/8365 [KT133/KM133] unknown : VIA Technologies|VT8363/8365 [KT133/KM133 AGP] unknown : VIA Technologies|VT82C686 [Apollo Super] unknown : VIA Technologies|VT82C586 IDE [Apollo] usb-uhci: VIA Technologies|VT82C586B USB usb-uhci: VIA Technologies|VT82C586B USB unknown : VIA Technologies|VT82C686 [Apollo Super ACPI] via82cxxx_audio : VIA Technologies|VT82C686 [Apollo Super AC97/Audio] 8139too : Realtek|RTL-8139 snd-emu10k1 : Creative Labs|SB Live! (audio) emu10k1-gp : Creative Labs|SB Live! (joystick) Card:ATI Rage 128 TVout: ATI|Rage 128 GL AGP 1x/2x unknown : Virtual|Hub [] unknown : Virtual|Hub [] ov518_decomp: OmniVision Technologies, Inc.|OV518 WebCam [] unknown : Alcatel|USB ADSL Modem (Speed Touch) [] [root@cosmic cosmicflo]# Thanks to fix it (before kde3.1 please), Florent <>
