Re: [gentoo-user] system update
On 2005-07-15 00:46:45 +0100 (Fri, Jul), Edward Catmur wrote: On Thu, 2005-07-14 at 17:58 -0500, Qv6 wrote: I am currently running kde-3.3, and do not want to have both 3.3 and 3.4 on the box. How can I update kde or any other app and not have the old version still installed. Update kde as usual, then # emerge -av --depclean Make sure that portage doesn't offer to remove packages you want to keep installed! Also a good idea to run revdep-rebuild afterwards. As this version of kde is slotted I believe that after your compilation finishes, and you find the new KDE in working state, you should: emerge prune kde emerge -va prune arts and then run emerge -av --depclean After depclean there will be a few leftovers from kde if I recall well. You may find them looking at /usr/kde/3.3/ and finding packages that own files left there. qpkg -f /usr/kde/3.3/some-file emerge -va prune the-package-found-if-it-is-part-of-kde Some files will not belong to any package, you may delete them, some packages will have to be rebuild to use the new KDE version and leave the old - dev-db/knoda in my case... Surely you may cook a nice script that will find and clear the KDE leftovers in more hackish way, but it was very hot the day I was doing this and my brain cooling system was not working properly... :-) Remember to run revdep-rebuild -p after you remove some packages. It may help you to find ones that need to be unmerged, and it is anyway a good habit to run it. HTH. -- No virus found in this outgoing message. Checked by 'grep -i virus $MESSAGE' Trust me. pgpkV0Qwr8Owi.pgp Description: PGP signature
[gentoo-user] RealPlayer-not-fetching-when-wget
hi! i tried to fetch Realplayer new version using emerge tool (#emerge -f realplayer). but it was not possible. then i tried it download it by manually from https://helixcommunity.org/download.php/1343/RealPlayer-10.0.5.756-20050513.i586.rpm this worked for me. THIS IS ONLY F.Y.I-- ...The future lies ahead. ___ Have you mooed today? ---\^__^ \ (oo) \___ (__) \ )\/\||--w||| ||Gentoo Linux 2.6.12-gentoo-r4-Adapt-v1.1#
[gentoo-user] vmware/xorg.conf issues
I am going NUTS trying to figure out how to make full-screen mode work properly in vmware-workstation 5. I can go into full-screen mode, but it looks like only half of the lines get drawn, so it's very ugly, impossible to read. I had beautiful screens in 4.5.2 which turned ugly at some point (don't know causal factor). I emerged 5, got beautiful screens again. I have been trying to get mplayer hdtv working, so I've been playing around with xorg.conf a lot, and I guess I managed to break it. I think I need additional modelines to make vmware work properly, not sure how to get the right ones. At some point in this blur, I used a tool which automatically calculated all valid modelines for my combination of nvidia 6600gt and Dell monitor, using DDC and EDID. I threw them into xorg.conf, they didn't fix mplayer problems, I thought I didn't need them removed them without backup. Now I can't find the tool I used (I thought it was read-edid, but that won't run properly on my platform), I don't see a good way of reproducing the modelines, and I'm not even sure that they were responsible for the good screens I had in vmware. Both my vmware virtual machines (XP), and my real desktop run 1024x768, 24 bit color. Oh, one other oddity. The vmware modules vmmon and vmnet which should be loaded by /etc/init.d/vmware in default mode, don't get loaded. Can't figure out why. We are buying 15 Thinkpads T-43, I'm planning on installing Gentoo w/ vmware-workstation on top to create a mobile classroom. But I may have to use XP plus vmware if I can't sort these full-screen issues out. Any pointers are appreciated. Bill Roberts pgpoCgQ9mdCFc.pgp Description: PGP signature
[gentoo-user] ldconfig breaks my system
Hi, Applogies for the cross-post. I'm unsure if this is and amd64 specific problem or a more general b0rkage. I've posted about my failing glibc emerges before: http://thread.gmane.org/gmane.linux.gentoo.user/114627 I think I've narrowed it down to ldconfig running during the emerge. In fact at the moment any time I run ldconfig it breaks the system so that I cannot spawn any new shells (they segfault straight away). I can restore my system by untaring a rescue glibc binary. Can anyone give me any hints on how to diagnose the problem with ldconfig and why things stop working when its done? I'm currently building a parallel world on a second partition (in a chrooted environment) to see if there are any differences between the two setups (i.e a subtle breakage occured and was never picked up). -- Alex, homepage: http://www.bennee.com/~alex/ It destroys one's nerves to be amiable every day to the same human being. -- Benjamin Disraeli -- gentoo-user@gentoo.org mailing list
[gentoo-user] Konqueror/KDE3.4.1/udev - media:/ won't browse removable usb devices
I have just installed KDE 3.4.1. I've got udev and the latest 2.6.12
kernel on an amd64. I've got my udev rules configured so that my
flash card is seen as a symlink named /dev/usbdrive. Works fine most
of the time.
If I go to media:/ in konq, it displays my usb drive as Removeable
Device (usbdrive), but when I click on it I get a blank window and
the gear at the top right-hand corner keeps going round - as if its
loading something.
I'm able to access my usb drive using a device icon on my desktop -
the drive mounts automatically if I click on this. Also - some of my
usb devices are listed as hard drives and there is no safely
remove option when I right click on them.
Here is the relevant part of /etc/udev/rules.d/10-udev.rules:
BUS=scsi, SYSFS{model}=Flash Disk , SYSFS{vendor}=USB ,
NAME=%k, SYMLINK=usbdrive
.. and here is the relevant part of /etc/fstab:
/dev/usbdrive /mnt/usbvfatrw,noauto,user
0 0
Have I not installed something vital?
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] system update
Thanks On Friday 15 July 2005 03:08 am, Mariusz Pękala wrote: On 2005-07-15 00:46:45 +0100 (Fri, Jul), Edward Catmur wrote: On Thu, 2005-07-14 at 17:58 -0500, Qv6 wrote: I am currently running kde-3.3, and do not want to have both 3.3 and 3.4 on the box. How can I update kde or any other app and not have the old version still installed. Update kde as usual, then # emerge -av --depclean Make sure that portage doesn't offer to remove packages you want to keep installed! Also a good idea to run revdep-rebuild afterwards. As this version of kde is slotted I believe that after your compilation finishes, and you find the new KDE in working state, you should: emerge prune kde emerge -va prune arts and then run emerge -av --depclean After depclean there will be a few leftovers from kde if I recall well. You may find them looking at /usr/kde/3.3/ and finding packages that own files left there. qpkg -f /usr/kde/3.3/some-file emerge -va prune the-package-found-if-it-is-part-of-kde Some files will not belong to any package, you may delete them, some packages will have to be rebuild to use the new KDE version and leave the old - dev-db/knoda in my case... Surely you may cook a nice script that will find and clear the KDE leftovers in more hackish way, but it was very hot the day I was doing this and my brain cooling system was not working properly... :-) Remember to run revdep-rebuild -p after you remove some packages. It may help you to find ones that need to be unmerged, and it is anyway a good habit to run it. HTH. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware/xorg.conf issues
Oh, one other oddity. The vmware modules vmmon and vmnet which should be loaded by /etc/init.d/vmware in default mode, don't get loaded. Can't figure out why. Make sure the kernel version in /usr/src/linux matches the version of the kernel currently running. What does the following command says? # ls /etc/vmware Cheers, AS -- gentoo-user@gentoo.org mailing list
[gentoo-user] Init script dependency creation?
I've got a Sangoma ADSL card that needs wanpipe to run before starting rp-pppoe. Having seen the /etc/init.d files, dependencies can be declared within the scripts themselves to make sure a dependency is satisfied before the script starts. I want to create such a dependency in /etc/init.d/rp-pppoe. The first question I have is whether this is the right way to integrate this dependency. And if it is, how do I go about doing it? Thanks! -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware/xorg.conf issues
On 09:41 Fri 15 Jul , Antonino Sabetta wrote: Oh, one other oddity. The vmware modules vmmon and vmnet which should be loaded by /etc/init.d/vmware in default mode, don't get loaded. Can't figure out why. Make sure the kernel version in /usr/src/linux matches the version of the kernel currently running. What does the following command says? # ls /etc/vmware ls -al /etc/vmware/ total 85 drwxr-xr-x 3 root root 160 Jul 15 10:22 . drwxr-xr-x 83 root root 5496 Jul 15 10:04 .. -rw-r--r-- 1 root root 211 Jul 15 10:22 config drwxr-xr-x 9 root root 240 Jul 15 10:04 init.d -r-xr-xr-x 1 root root 14482 Jul 15 10:04 installer.sh -rw-r--r-- 1 root root 58854 Jul 15 10:22 locations Kernel version matches. I also did an emerge -C', removed all files, and re-emerged. Still blurred full-screen. Thanks Bill Roberts pgpZPj3OGEZVZ.pgp Description: PGP signature
[gentoo-user] CONFIG_PROTECT problem
Hi, Apologies if I've misunderstood the use of CONFIG_PROTECT, but I think I've found a hole in it. As I have lots of stuff under /var/www/localhost/htdocs which contains configuration files mixed in with the code ( phpmyadmin, phpldapadmin, phpwiki, squirrelmail, gallery etc ) I have put the path /var/www/localhost/htdocs into CONFIG_PROTECT in make.conf. When one of these packages is upgraded this seems to work fine. Last night, after upgrading PHP to 4.4.0 my wiki was broken. I thought a good place to start would be to re-emerge phpwiki, so I did. During the emerge it flashed up a message about this being a package that it couldn't upgrade, so it would be unmerged it first. It appears that this bypassed the CONFIG_PROTECT mechanism, as when the new files were installed the original had been removed, so no ._cfg_ files were created for the changed files. Having no recent backup (lesson learned!) I had to recreate the phpwiki config, which is a non-trivial job. So the question is, how can config files be protected in this kind of situation (other than backing them up) - is there another mechanism to protect files from being overwritten, and how many packages are likely to do an unmerge before re-emerging, and is there ay way of knowing? I believe the default behaviour on umnerging a package is to leave its configuration files in place, this doesn't seem to apply to the web apps. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Subversion 1.2
Holly Bostick wrote: The thing is Portage doesn't *remember* ACCEPT_KEYWORDS, beyond the original compile in which it is used. So if you use it, and keep the package, as soon as you do an emerge -u world, Portage will try to downgrade the package to the last stable version, which is the only one that it knows to be allowed (because /etc/make.conf says xarch, not ~arch, and no exception for this particular package and its dependencies has been made in /etc/portage/package.keywords). OK - now using package.keywords make far more sense to me. I'd always assumed (wrongly I guess) that emerge -u would only upgrade and never downgrade... Now I see why I'd need the entry in package.keywords. It's a beautiful system :-) . I'm closer to believing you. :-) The only way in which I'm not yet as convinced as you are is with respect to dependencies. I'm comfortable with the idea that I browse the bugs to verify that none of the issues affect my install directly - then to accept an unstable version of a specific package... but I'd prefer not to have to dig out the package dependencies and explicitly allow the unstable branch for those packages too (as seems to have been indicated earlier in this thread.) Is there a simple way to say, for example, I'm willing to accept the unstable version of Subversion-1.2.1, and (naturally) the unstable version of any package on which Suversion-1.2.1 depends? It was my wish to side-step explicitly dealing with package dependencies which prompted me to use ACCEPT_KEYWORDS with emerge -uD ... Steve -- gentoo-user@gentoo.org mailing list
[gentoo-user] [OT?] Advice on hardware
Greetings list, After my mobo fried, I decided to upgrade and buy a new one, with a new processor, I would get an Asus last generation with an Athlon 64 as processor and the rest inherit from my previous machine (so, compatible as far as I know). I've seen posts regarding imcompatibility issues with A64 and Gentoo, and heard about an specific division of Gentoo Foundation studying and developing for the A64. I'm kinda a stable user, so, I'm afraid of buying hardware that will still be tested and maybe encounter some weird errors. Any advice? PS: Nvidia would be my choice for video, anyone have suggestions? -- Daniel da Veiga Computer Operator - RS - Brazil -- gentoo-user@gentoo.org mailing list
[gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
I was wondering if others have tackled this before me? I only have remote SSH access to my Gentoo box (everything else is fire walled) and I use this to map ports on my client machine to corresponding ports on my server machine - so, for example, I can access IMAP; SMTP, X applications etc. on the Gentoo server remotely. My remote client is WinXP, and I use Cygwin OpenSSH. Owing to low bandwidth. While Xwindows is usable it is very slow - command line interaction without X is acceptable for most tasks but imposes a very restrictive interface. If possible I would like to be able to edit (relatively small) configuration files remotely in a text editor running under windows. What would be the easiest way to somehow map a network drive on my Gentoo server from Windows given that I need to tunnel over SSH? -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] system update
qpkg -f /usr/kde/3.3/some-file emerge -va prune the-package-found-if-it-is-part-of-kde Thanks for this very useful hint! I have managed to free a lot of precious disk space!! I did it in a semi-automated way: # emerge -p prune prune-list.txt then I edited the file, leaving only the rows with packages I really wanted to prune (for instance I wouldn't prune old kernels) # vi prune-list.txt and then: for i in `cat prune-list.txt`; do emerge -v prune $i ; done Cheers, AS -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Subversion 1.2
Steve [Gentoo] wrote: The only way in which I'm not yet as convinced as you are is with respect to dependencies. I'm comfortable with the idea that I browse the bugs to verify that none of the issues affect my install directly - then to accept an unstable version of a specific package... but I'd prefer not to have to dig out the package dependencies and explicitly allow the unstable branch for those packages too (as seems to have been indicated earlier in this thread.) Is there a simple way to say, for example, I'm willing to accept the unstable version of Subversion-1.2.1, and (naturally) the unstable version of any package on which Suversion-1.2.1 depends? It was my wish to side-step explicitly dealing with package dependencies which prompted me to use ACCEPT_KEYWORDS with emerge -uD ... Steve Roy has created a perl script called unmask.pl which automatically unmasks keyword masked dependencies: http://marc.theaimsgroup.com/?l=gentoo-userm=111472741321054 Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
What would be the easiest way to somehow map a network drive on my Gentoo server from Windows given that I need to tunnel over SSH? Could you try something like WebDAV? It should be easy to forward the necessary port over ssh to a web server on your Gentoo box. Plus, XP has built-in support for WebDAV through its Web Folders stuff. Matt -- gentoo-user@gentoo.org mailing list
[gentoo-user] Printing woes; firefox incompetent; cups printer disabled and cannot enable
Woe is me.
First: Firefox 1.0.4 could not print. It seemed to go through all the
motions,The printer command was some oddity involving ${MOZ_PRINTER}
or some such, and I just changed it to 'lpr', and it still wouldn't
work.
I forget why, but I tried 'disable lp0' to see what it did to the
output of 'lpstat -t', and
got the results one would expect. However, 'enable lp0' gives this
odd error message:
-/bin/bash: enable: lp0: not a shell builtin
Now, this makes no sense to me at all.
So, now I have a completely unusable setup. Any clues?
++ kevin
--
Kevin O'Gorman, PhD
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
Steve [Gentoo] wrote: I was wondering if others have tackled this before me? I only have remote SSH access to my Gentoo box (everything else is fire walled) and I use this to map ports on my client machine to corresponding ports on my server machine - so, for example, I can access IMAP; SMTP, X applications etc. on the Gentoo server remotely. My remote client is WinXP, and I use Cygwin OpenSSH. Owing to low bandwidth. While Xwindows is usable it is very slow - command line interaction without X is acceptable for most tasks but imposes a very restrictive interface. If possible I would like to be able to edit (relatively small) configuration files remotely in a text editor running under windows. What would be the easiest way to somehow map a network drive on my Gentoo server from Windows given that I need to tunnel over SSH? Filezilla is an ftp client that supports file transfer over ssh: http://filezilla.sourceforge.net/ For remote X you should use nx (emerge nxserver-freenx) or vnc (emerge tightvnc). To map a network drive, you should probably tunnel samba through ssh. Google for samba tunnel putty. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
On 16:08 Fri 15 Jul , Steve [Gentoo] wrote: I was wondering if others have tackled this before me? I only have remote SSH access to my Gentoo box (everything else is fire walled) and I use this to map ports on my client machine to corresponding ports on my server machine - so, for example, I can access IMAP; SMTP, X applications etc. on the Gentoo server remotely. My remote client is WinXP, and I use Cygwin OpenSSH. Owing to low bandwidth. While Xwindows is usable it is very slow - command line interaction without X is acceptable for most tasks but imposes a very restrictive interface. If possible I would like to be able to edit (relatively small) configuration files remotely in a text editor running under windows. What would be the easiest way to somehow map a network drive on my Gentoo server from Windows given that I need to tunnel over SSH? I use putty as my sshclient whenever I'm on a Windows machine. For file editing, I use vim, but nano would be more user friendly. putty is nice because you can download the executable from the internet and don't even have to install, just run the executable. I have used tightvnc to access a remote box, which gives you the desktop, and it can be tunneled over ssh. Very nice. The fastest desktop product is nx and nxserver. It is not opensource, but I believe there is some provision for personal use. I would guess that it can be tunneled over ssh. Not sure about the mapping. Bill Roberts pgp1riCwg2HQA.pgp Description: PGP signature
Re: [gentoo-user] [OT?] Advice on hardware
On Friday 15 July 2005 17.06, Daniel da Veiga wrote: Any advice? I´ll give my 2 cents. Asus is to begin with a (in my experience) very good vendor. Even though you buy a AMD64 now, you are not forced to compile Gentoo for amd64 arcitecture. You could compile it for AthlonXP (x86). PS: Nvidia would be my choice for video, anyone have suggestions? I have a Nvidia 6800 (AGP) and is extremly pleased with it. X has never crashed one single time during the last 6 months. Best regards, Andreas Karlsson pgp4k7ALpJwKu.pgp Description: PGP signature
Re: [gentoo-user] [OT?] Advice on hardware
Daniel da Veiga wrote: Greetings list, After my mobo fried, I decided to upgrade and buy a new one, with a new processor, I would get an Asus last generation with an Athlon 64 as processor and the rest inherit from my previous machine (so, compatible as far as I know). I've seen posts regarding imcompatibility issues with A64 and Gentoo, and heard about an specific division of Gentoo Foundation studying and developing for the A64. I'm kinda a stable user, so, I'm afraid of buying hardware that will still be tested and maybe encounter some weird errors. Any advice? PS: Nvidia would be my choice for video, anyone have suggestions? Gentoo amd64 works great! The main place you might have difficulties is with IA32 dependent apps such as wine. If you need IA32 compatibility there is some info here: http://www.gentoo.org/proj/en/base/amd64/technotes/index.xml?part=1chap=4 Also, for amd64 specific stuff, you can use the gentoo-amd64 mailing list. Zac -- gentoo-user@gentoo.org mailing list
[gentoo-user] shutdown -h on Power-button...
Hi, I want to set up my computer so that it would do automatically shutdown -h after short pressing of power button. Is the only solution to emerge acpid (found on gentoo-forum)? Would not it be easier with apm? Because I remember once in the past having a comp with some other linux-distro, where I think I did not install acpid, only edited some file in /etc, and yet after pressing the power button it correctly shutdowned my computer. I just want to keep number of running daemons low. It is not a notebook, so I think it is not worth to have one more daemon running all the time just because of one possible event-trigger... Jarry -- gentoo-user@gentoo.org mailing list
[gentoo-user] glsa
How long between when a GLSA is distributed and glsa-check will pick it up? I just got a message from announce that has a GLSA I believe my server will be vulnerable to but glsa-check isn't picking it up. In fact I just did a 'glsa-check -l | grep 200507' and nothing showed up. Does the portage db need to be updated? I figured it would use the rss. -- Noah Roberts Saint Martin's University Information Technology Services 486-8814 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] shutdown -h on Power-button...
Jarry wrote: Hi, I want to set up my computer so that it would do automatically shutdown -h after short pressing of power button. Is the only solution to emerge acpid (found on gentoo-forum)? Would not it be easier with apm? Because I remember once in the past having a comp with some other linux-distro, where I think I did not install acpid, only edited some file in /etc, and yet after pressing the power button it correctly shutdowned my computer. I just want to keep number of running daemons low. It is not a notebook, so I think it is not worth to have one more daemon running all the time just because of one possible event-trigger... Jarry AFAIK your choices are acpid and apmd. Either will do the job as log as it can detect power button events. I wouldn't worry about running the daemon because resource usage is probably quite low. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] glsa
On Saturday 16 July 2005 00:56, Noah Roberts wrote: How long between when a GLSA is distributed and glsa-check will pick it up? I just got a message from announce that has a GLSA I believe my server will be vulnerable to but glsa-check isn't picking it up. In fact I just did a 'glsa-check -l | grep 200507' and nothing showed up. Does the portage db need to be updated? I figured it would use the rss. GLSAs are held in $PORTDIR/metadata/glsa -- Jason Stubbs pgpmKxKDEUGzX.pgp Description: PGP signature
Re: [gentoo-user] strange /init kernel-2.6.12-gentoo-r5 boot problems
solution are simple. there is a new version of genkernel in portage. try: emerge --sync and emerge genkernel and finally rebuild the entire kernel and that should solve. At least I've solve mine. good luckzz On 7/15/05, Zac Medico [EMAIL PROTECTED] wrote: Iain Buchanan wrote: On Thu, 2005-07-14 at 08:36 +0200, Richard Fish wrote: Iain Buchanan wrote: -install: applet not found /init: 41: In: not found /init: 45: cat: not found /init: 150: sed: not found Apparently it is a problem with busybox configuration: From another thread on this list less than 12 hours old: Zac Medico wrote: It's not as bad as it looks. I had the same problem. You simply need to enable CONFIG_FEATURE_INSTALLER in the busybox config. Then you can do busybox --install and it automatically creates hardlinks for all the enabled applets. What busybox config? `equery files busybox` doesn't show any config files for busybox. I found the previous thread here http://marc.theaimsgroup.com/?l=gentoo-userm=112127857419654w=2 and Zac suggested some different options: root (hd0,0) kernel /kernel-genkernel-x86-2.6.12-gentoo-r4n root=/dev/hda11 video=vesafb:[EMAIL PROTECTED] splash=verbose gentoo=nodevfs udev devfs=nomount initrd /initramfs-genkernel-x86-2.6.12-gentoo-r4n but that didn't help either. So it still uses the initramfs even though you didn't specify root=/dev/ram0? Richard said something about that (I wasn't aware of this behavior). Apparently you need to remove the initrd parameter in order to disable it. Genkernel builds busybox with the config file in /usr/share/genkernel/$arch/busy-config. Of course, it will only build busybox if it doesn't find a suitable binary in your bincache (specified in /etc/genkernel.conf). And why do I need to do a `busybox --install` ?? shouldn't this come up in some emerge warning or be done by genkernel or something? I monitor all the output from emerge with enotice, but I didn't see anything about busybox... TIA, You don't do busybox --install, that happens automatically in the init script. It seems that the genkernel devs may have overlooked this minor detail. It's easy to forget about it after you have a working busybox in your bincache. Zac -- gentoo-user@gentoo.org mailing list -- Regards, Chee Ong -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Seeking recommdations for multi-card reader.
On Wed, Jul 13, 2005 at 10:01:17PM -0400, Colin wrote: I've got an AFT PRO-9. It's a 9-in-1 connects via USB 2.0 internally/externally and has Secure Digital, Multimedia Card, SmartMedia, xDigital, Memory Stick, Memory Stick Pro, CompactFlash, MicroDrive and a 4-pin FireWire 400 port (pass-through, you need a FireWire port somewhere on your computer). I've only tested it with an SD card, and it seems to work perfectly. To use this, you need to compile in SCSI support (the same you need for USB Mass Storage) and make sure you select the Probe all LUN's option. Thanks Colin. It seems to get some pretty good reviews elsewhere too. John -- Contrary to the lie machine, the world is not safer. -- gentoo-user@gentoo.org mailing list
[gentoo-user] VIDEO_CARDS setting for 3D Rage Pro
Converting my last FC machine to Gentoo since it didn't survive an FC2 to FC4 upgrade. Questions: 1) Is the VIDEO_CARDS= setting in /etc/make.conf still considered the right way to configure a new machine for a specific card? 2) If the answer to #1 is yes then what's the right setting for a 3D Rage Pro? (I.e. - where's the list of legal names?) 3) If no, then what's the right way to do this these days? Thanks, Mark -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
On Fri, July 15, 2005 11:08 am, Steve [Gentoo] wrote: I was wondering if others have tackled this before me? If possible I would like to be able to edit (relatively small) configuration files remotely in a text editor running under windows. I use UltraEdit for this purpose on a daily basis. It is the best editor for all my Windows needs. It supports remote servers through sftp(ssh2). And has support for non *DOS* files. It's only drawback is that its Shareware(Costs $$$) but has a time limited demo. www.ultraedit.com Charles Andrews -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] VIDEO_CARDS setting for 3D Rage Pro
Mark Knecht wrote: Converting my last FC machine to Gentoo since it didn't survive an FC2 to FC4 upgrade. Questions: 1) Is the VIDEO_CARDS= setting in /etc/make.conf still considered the right way to configure a new machine for a specific card? 2) If the answer to #1 is yes then what's the right setting for a 3D Rage Pro? (I.e. - where's the list of legal names?) 3) If no, then what's the right way to do this these days? Thanks, Mark It looks like the gentoo ati faq might be helpful: http://www.gentoo.org/doc/en/ati-faq.xml It mentions an x11-drm package that can be used to build kernel DRI modules. Inside the ebuild there is a list: IUSE_VIDEO_CARDS=3dfx ffb i810 i830 i915 mach64 matrox rage128 radeon savage sis via I'm not sure if any of those will work for you but I'm sure one of the ati gurus will be along in a moment to help you. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] VIDEO_CARDS setting for 3D Rage Pro
On 7/15/05, Zac Medico [EMAIL PROTECTED] wrote: Mark Knecht wrote: Converting my last FC machine to Gentoo since it didn't survive an FC2 to FC4 upgrade. Questions: 1) Is the VIDEO_CARDS= setting in /etc/make.conf still considered the right way to configure a new machine for a specific card? 2) If the answer to #1 is yes then what's the right setting for a 3D Rage Pro? (I.e. - where's the list of legal names?) 3) If no, then what's the right way to do this these days? Thanks, Mark It looks like the gentoo ati faq might be helpful: http://www.gentoo.org/doc/en/ati-faq.xml It mentions an x11-drm package that can be used to build kernel DRI modules. Inside the ebuild there is a list: IUSE_VIDEO_CARDS=3dfx ffb i810 i830 i915 mach64 matrox rage128 radeon savage sis via I'm not sure if any of those will work for you but I'm sure one of the ati gurus will be along in a moment to help you. Zac -- gentoo-user@gentoo.org mailing list Thanks Zac. On this machine I don't care right now about 3D performance. Everything in my audio recording is pretty much 2D (VU meters and the like) so possibly it won't make much difference at all. I guess I'll choose rage128 and see what happens. Since DRM is modular I guess I don't have to load it if it causes trouble. Thanks for reading the ebuild for me. cheers, Mark -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] VIDEO_CARDS setting for 3D Rage Pro
Hello, I also have this video card. On Fri, 15 Jul 2005, Mark Knecht wrote: Converting my last FC machine to Gentoo since it didn't survive an FC2 to FC4 upgrade. Questions: 1) Is the VIDEO_CARDS= setting in /etc/make.conf still considered the right way to configure a new machine for a specific card? Yes it is the right way but only used by the x11-drm package. If you want dri to work you also have to emerge xorg with the 'insecure-drivers' USE flag. That's right, the dri drivers for this video card are considered insecure... 2) If the answer to #1 is yes then what's the right setting for a 3D Rage Pro? (I.e. - where's the list of legal names?) The 3D Rage Pro has uses a mach64 chip so the right setting is VIDEO_CARDS=mach64. Finally, here is my video card section in xorg.conf: Section Device Identifier Card0 Driver ati VendorName ATI BoardName 3D Rage Pro 215GP ChipSet ati ChipId 0x4750 ChipRev 0x5c Option backingstore EndSection HTH -- T.G. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] VIDEO_CARDS setting for 3D Rage Pro
Thanks Tero. I think htis email has probably saved me hours and hours of research. cheers, Mark On 7/15/05, Tero Grundström [EMAIL PROTECTED] wrote: Hello, I also have this video card. On Fri, 15 Jul 2005, Mark Knecht wrote: Converting my last FC machine to Gentoo since it didn't survive an FC2 to FC4 upgrade. Questions: 1) Is the VIDEO_CARDS= setting in /etc/make.conf still considered the right way to configure a new machine for a specific card? Yes it is the right way but only used by the x11-drm package. If you want dri to work you also have to emerge xorg with the 'insecure-drivers' USE flag. That's right, the dri drivers for this video card are considered insecure... 2) If the answer to #1 is yes then what's the right setting for a 3D Rage Pro? (I.e. - where's the list of legal names?) The 3D Rage Pro has uses a mach64 chip so the right setting is VIDEO_CARDS=mach64. Finally, here is my video card section in xorg.conf: Section Device Identifier Card0 Driver ati VendorName ATI BoardName 3D Rage Pro 215GP ChipSet ati ChipId 0x4750 ChipRev 0x5c Option backingstore EndSection HTH -- T.G. -- gentoo-user@gentoo.org mailing list -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ldconfig breaks my system
Alex Bennee wrote: Hi, Applogies for the cross-post. I'm unsure if this is and amd64 specific problem or a more general b0rkage. I've posted about my failing glibc emerges before: http://thread.gmane.org/gmane.linux.gentoo.user/114627 I think I've narrowed it down to ldconfig running during the emerge. In fact at the moment any time I run ldconfig it breaks the system so that I cannot spawn any new shells (they segfault straight away). I can restore my system by untaring a rescue glibc binary. Hmm, don't have an AMD64 yet (waiting for the first dual-core laptop to become available!), but this certainly sounds like a 32-bit vs 64-bit library issue. Take a look at /etc/ld.so.conf. This file defines the library search path for your system, and ldconfig takes this file and generates /etc/ld.so.cache from it. I suspect removing the entries for the 32-bit library directories, or possibly just moving them to the end of the list, would solve the problem for the majority of the programs. You may run into trouble with some closed source 32-bit stuff though. (VMWare, Adobe Acrobat, etc) -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] CONFIG_PROTECT problem
Peter Campion-Bye wrote: Hi, Apologies if I've misunderstood the use of CONFIG_PROTECT, but I think I've found a hole in it. As I have lots of stuff under /var/www/localhost/htdocs which contains configuration files mixed in with the code ( phpmyadmin, phpldapadmin, phpwiki, squirrelmail, gallery etc ) I have put the path /var/www/localhost/htdocs into CONFIG_PROTECT in make.conf. When one of these packages is upgraded this seems to work fine. Last night, after upgrading PHP to 4.4.0 my wiki was broken. I thought a good place to start would be to re-emerge phpwiki, so I did. During the emerge it flashed up a message about this being a package that it couldn't upgrade, so it would be unmerged it first. It appears that this bypassed the CONFIG_PROTECT mechanism, as when the new files were installed the original had been removed, so no ._cfg_ files were created for the changed files. Having no recent backup (lesson learned!) I had to recreate the phpwiki config, which is a non-trivial job. So the question is, how can config files be protected in this kind of situation (other than backing them up) - is there another mechanism to protect files from being overwritten, and how many packages are likely to do an unmerge before re-emerging, and is there ay way of knowing? I believe the default behaviour on umnerging a package is to leave its configuration files in place, this doesn't seem to apply to the web apps Ouch. Do you by chance have CONFIG_PROTECT_MASK set also? Check emerge --info.. -Richard -- gentoo-user@gentoo.org mailing list
[gentoo-user] QOS
Hi, Are there people on the list who are using QOS, i need it to preserve 4MB for IP telephonie on a 100Mbit line. What are good site's / howto's for QOS. TIA Patrick -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] shutdown -h on Power-button...
Zac Medico wrote: Jarry wrote: Hi, I want to set up my computer so that it would do automatically shutdown -h after short pressing of power button. Is the only solution to emerge acpid (found on gentoo-forum)? Would not it be easier with apm? Because I remember once in the past having a comp with some other linux-distro, where I think I did not install acpid, only edited some file in /etc, and yet after pressing the power button it correctly shutdowned my computer. I just want to keep number of running daemons low. It is not a notebook, so I think it is not worth to have one more daemon running all the time just because of one possible event-trigger... Jarry AFAIK your choices are acpid and apmd. Either will do the job as log as it can detect power button events. I wouldn't worry about running the daemon because resource usage is probably quite low. Zac Acpid on my system is consuming 592K of memory and has accumulated 0 seconds of CPU time out of the 40-or-so hours that my laptop has been up. So it is definitely quite frugal on resources!! -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware/xorg.conf issues
Well, nothing obvious is popping out at me, but see my comments inline below... Bill Roberts wrote: I am going NUTS trying to figure out how to make full-screen mode work properly in vmware-workstation 5. I can go into full-screen mode, but it looks like only half of the lines get drawn, so it's very ugly, impossible to read. A major reason for this can be enabling framebuffer graphics in the kernel. If you are using framebuffer graphics, try compiling a new kernel with only text-mode console support. Full-screen vmware sessions seem most reliable this way. I had beautiful screens in 4.5.2 which turned ugly at some point (don't know causal factor). I emerged 5, got beautiful screens again. I have been trying to get mplayer hdtv working, so I've been playing around with xorg.conf a lot, and I guess I managed to break it. I think I need additional modelines to make vmware work properly, not sure how to get the right ones. My guess is that this has nothing to do with the modelines, because on a modern system with a DDC capable monitor, everything should be autodetected. Specifically, double check your monitor section and try commenting out any HorizSync or VertRefresh settings that you have. Oh, one other oddity. The vmware modules vmmon and vmnet which should be loaded by /etc/init.d/vmware in default mode, don't get loaded. Can't figure out why. Explain a bit more please...are you saying the vmware init script runs as part of your startup, but produces errors? You didn't by chance accidentally replace this with the VMWare version, did you? The second line of that script should say Copyright ... Gentoo Foundation. If this doesn't give you any ideas, I would like to see the device and monitor sections from your xorg.conf file. -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT?] Advice on hardware
Daniel da Veiga wrote: Well, I'm planning on using cedega for gaming, so, your mail kinda scared me (*LOL*), but anyway, there's always the choice to build a compatible mini-system at another partition or disc with the athlon-xp arch... If I ever plan on having a windows partition (I'm a gamer, ratter prefer transgamer, but not all games run at linux) I'll have my first tri-boot system :) Thanks for the replies! It doesn't even have to be on another parition. You can have a 32-bit chroot inside any directory. For some things you can use emul-linux-x86* packages to avoid a chroot entirely. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
I use UltraEdit for this purpose on a daily basis. It is the best editor for all my Windows needs. It supports remote servers through sftp(ssh2). And has support for non *DOS* files. It's only drawback is that its Shareware(Costs $$$) but has a time limited demo. www.ultraedit.com I used to use that one till I discovered jEdit: http://www.jedit.org/. Yes it does SSH2/sftp and it's free, 100's of plugins too. /djb -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] QOS
On Fri, 15 Jul 2005, Patrick wrote: Are there people on the list who are using QOS, i need it to preserve 4MB for IP telephonie on a 100Mbit line. http://gentoo-wiki.com/HOWTO_Packet_Shaping -- -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] glsa
On Fri, 15 Jul 2005, Noah Roberts wrote: How long between when a GLSA is distributed and glsa-check will pick it up? I just got a message from announce that has a GLSA I believe my server will be vulnerable to but glsa-check isn't picking it up. In fact I just did a 'glsa-check -l | grep 200507' and nothing showed up. Does the portage db need to be updated? You prob. need to run emerge --sync I figured it would use the rss. We wish... -- -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
On Fri, 15 Jul 2005, Chaz Andrews wrote: I use UltraEdit for this purpose on a daily basis. It is the best editor for all my Windows needs. There is only ONE editor: vim (it runs on everything incl. Windoze) Let the flames begin... ;-) -- -- gentoo-user@gentoo.org mailing list
[gentoo-user] LDAP everywhere?
After a recent --sync I see now that the update wants to emerge openldap because an ldap flag has been added to openssh and it seems to be the default. Anyone else seen this? What is the thinking behind making it a default flag? Makes no sense to me. I dont want or need openldap on my boxes... -- Aj. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] LDAP everywhere?
A. Khattri wrote: After a recent --sync I see now that the update wants to emerge openldap because an ldap flag has been added to openssh and it seems to be the default. Anyone else seen this? What is the thinking behind making it a default flag? Makes no sense to me. I dont want or need openldap on my boxes... Do you have openldap installed? The ldap flag is listed in /usr/portage/profiles/base/use.defaults so it should default to -ldap unless openldap is installed. You can always add -ldap to make.conf. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] LDAP everywhere?
On Fri, 15 Jul 2005, Zac Medico wrote: Do you have openldap installed? The ldap flag is listed in /usr/portage/profiles/base/use.defaults so it should default to -ldap unless openldap is installed. You can always add -ldap to make.conf. Poking around it looks like ldap support was compiled into php and yes it looks like openldap is installed (I will need to rebuild php and remove openldap). Still, it doesn't explain why its switched on for openssh? -- Aj. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] LDAP everywhere?
A. Khattri wrote: On Fri, 15 Jul 2005, Zac Medico wrote: Do you have openldap installed? The ldap flag is listed in /usr/portage/profiles/base/use.defaults so it should default to -ldap unless openldap is installed. You can always add -ldap to make.conf. Poking around it looks like ldap support was compiled into php and yes it looks like openldap is installed (I will need to rebuild php and remove openldap). Still, it doesn't explain why its switched on for openssh? That's the way that it works with use.defaults. When openldap is installed, it's just like you have +ldap in make.conf. It's enabled for everything unless you add -ldap to make.conf or /etc/portage/package.use. Zac -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware/xorg.conf issues
Richard I tried all of your suggestions, without success. See inline comments: Note device and monitor section of xorg.conf at end: I've had one further idea. I have a 4-port kvm switch (Cybex), it has never interferred with anything in the past four or five years. Wonder if. . . , maybe that could be having an effect. One other thought. In full screen, ever other line is black, giving the awful looking screen. Couldn't that have something to do with interlacing?? Wish I understood all of this better. On 21:05 Fri 15 Jul , Richard Fish wrote: Well, nothing obvious is popping out at me, but see my comments inline below... Bill Roberts wrote: I am going NUTS trying to figure out how to make full-screen mode work properly in vmware-workstation 5. I can go into full-screen mode, but it looks like only half of the lines get drawn, so it's very ugly, impossible to read. A major reason for this can be enabling framebuffer graphics in the kernel. If you are using framebuffer graphics, try compiling a new kernel with only text-mode console support. Full-screen vmware sessions seem most reliable this way. I did have framebuffer graphics enabled, just for the smaller fonts. I diabled them, didn't make any difference. Also, when I had good full screens, I did have framebuffer graphics enabled. I had beautiful screens in 4.5.2 which turned ugly at some point (don't know causal factor). I emerged 5, got beautiful screens again. I have been trying to get mplayer hdtv working, so I've been playing around with xorg.conf a lot, and I guess I managed to break it. I think I need additional modelines to make vmware work properly, not sure how to get the right ones. My guess is that this has nothing to do with the modelines, because on a modern system with a DDC capable monitor, everything should be autodetected. Specifically, double check your monitor section and try commenting out any HorizSync or VertRefresh settings that you have. Modelines are supposed to be generally obsolete, and my Dell monitor is DDC compliant, but I've been having difficulty with mplayer/mythtv. Also, VmWare says I can run into these kind of problems if there isn't an appropriate resolution available. Oh, one other oddity. The vmware modules vmmon and vmnet which should be loaded by /etc/init.d/vmware in default mode, don't get loaded. Can't figure out why. Explain a bit more please...are you saying the vmware init script runs as part of your startup, but produces errors? You didn't by chance accidentally replace this with the VMWare version, did you? The second line of that script should say Copyright ... Gentoo Foundation. I did a rc-update add vmware default and back when, it actually started the vmmon/vmnet modules. No longer. No error messages, no hiccups, nothing. And when I manually do a /etc/init.d/vmware start, it works perfectly. The Copyright ... is there. If this doesn't give you any ideas, I would like to see the device and monitor sections from your xorg.conf file. Here are the device and monitor sections of xorg.conf, plus a couple of extra, maybe pertinent items. Section Monitor DisplaySize 350 255 # mm Identifier Monitor0 VendorName DEL ModelNameDELL D1226H # HorizSync30.0 - 95.0 # VertRefresh 50.0 - 160.0 Option DPMS EndSection Section Device Identifier n6600gt Driver nvidia Option NoLogo 1 EndSection Section Screen Identifier Screen0 Device n6600gt MonitorMonitor0 DefaultDepth 24 SubSection Display Depth 16 Modes 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 24 Modes 1024x768 800x600 640x480 EndSubSection EndSection Maybe some of the following should be disabled??? Section Module Load record Load extmod Load dbe Load glx Load xtrap Load type1 Load freetype EndSection pgpJzYEyhAeru.pgp Description: PGP signature
Re: [gentoo-user] vmware/xorg.conf issues
Bill Roberts wrote: Richard I tried all of your suggestions, without success. See inline comments: Note device and monitor section of xorg.conf at end: I've had one further idea. I have a 4-port kvm switch (Cybex), it has never interferred with anything in the past four or five years. Wonder if. . . , maybe that could be having an effect. One other thought. In full screen, ever other line is black, giving the awful looking screen. Couldn't that have something to do with interlacing?? Wish I understood all of this better. Well, I'm not completely certain, but I don't think so. To the best of my knowledge, PC cards today work on a fairly simply frame-buffer concept...the software renders the image it wants displayed to a memory location, usually on the card itself. It then instructs the GPU that it has updated the image, and it is up to the GPU to read and create the analog signals required for the monitor. Interlacing should have nothing to do with the framebuffer, it should only be considered in instructing the GPU how to communicate with a given monitor. So it is only between the GPU and the monitor that I would expect a disagreement over interlacing to result in a problem like this, but the image would also not be synchronized correctly (rolling, distorting horizontally, basically looking like a 1970's-era American TV!) Note that the above is even true for accelerated 2D or 3D graphics...the main difference there is that the GPU has many additional instructions available for common drawing operations so that the system CPU doesn't have to perform them. The idea is still the same...render first to a memory buffer, then let the hardware figure out how to transmit that. My best guess here is that VMWare and the X server are having a disagreement about the layout of the framebuffer. In addition to width and height, a particular framebuffer is expected to have a particular alignment in memory for how many pixels per line, lines per screen, bits per pixel, and bits per color, and the order of the colors. So, on that front, I have several other suggestions to try: 1. Try using the x.org 'nv' driver instead of the proprietary nvidia driver. 2. Try setting DefaultDepth to 16. (with both drivers) 3. Try setting DefaultDepth to 32 (and create the appropriate subsection for Depth 32). This may only work with the proprietary driver HTH, -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware/xorg.conf issues
Bill Roberts wrote: Richard I tried all of your suggestions, without success. See inline comments: Note device and monitor section of xorg.conf at end: I've had one further idea. I have a 4-port kvm switch (Cybex), it has never interferred with anything in the past four or five years. Wonder if. . . , maybe that could be having an effect. One other thought. In full screen, ever other line is black, giving the awful looking screen. Couldn't that have something to do with interlacing?? Wish I understood all of this better. Oh, one more thing. Modelines can be an issue, but only if VMWare and X are running at different resolutions or color depths. For example, my normal laptop resolution is 1600x1200. If I go fullscreen VMWare at 1600x1200 or 1280x1024, everything works normally. However there is also a 1400x1050 mode in XP, but my hardware doesn't seem to support that resolution, so if I try to go fullscreen, I get a resolution that results in a corrupted display for the right 1/3 of the screen. So it is possible that a particular modeline could fix that issue for me, and may be related to why you had to specify modelines previously. But if the resolution and depth are the same, then the same modeline should apply whether VMWare is fullscreen or not, AFAIK. -Richard -- gentoo-user@gentoo.org mailing list
[gentoo-user] Samba crashes after upgrading system
Hello, I just upgraded my system (emerge -u world) and after restart samba do not start. (Samba was not upgraded.) In the /var/log/messages I see: Jul 15 23:59:25 amit smbd[7829]: [2005/07/15 23:59:25, 0] lib/fault.c:fault_report(36) Jul 15 23:59:25 amit smbd[7829]: === Jul 15 23:59:25 amit smbd[7829]: [2005/07/15 23:59:25, 0] lib/fault.c:fault_report(37) Jul 15 23:59:25 amit smbd[7829]: INTERNAL ERROR: Signal 11 in pid 7829 (3.0.10) Jul 15 23:59:25 amit smbd[7829]: Please read the appendix Bugs of the Samba HOWTO collection Jul 15 23:59:25 amit smbd[7829]: [2005/07/15 23:59:25, 0] lib/fault.c:fault_report(39) Jul 15 23:59:25 amit smbd[7829]: === Jul 15 23:59:25 amit smbd[7829]: [2005/07/15 23:59:25, 0] lib/util.c:smb_panic2(1482) Jul 15 23:59:25 amit smbd[7829]: PANIC: internal error Jul 15 23:59:25 amit smbd[7829]: [2005/07/15 23:59:25, 0] lib/util.c:smb_panic2(1490) Jul 15 23:59:25 amit smbd[7829]: BACKTRACE: 22 stack frames: Jul 15 23:59:25 amit smbd[7829]:#0 /usr/sbin/smbd(smb_panic2+0x64) [0x575794] [snip] Jul 15 23:59:25 amit smbd[7829]: Jul 15 23:59:26 amit nmbd[7831]: [2005/07/15 23:59:26, 0] lib/fault.c:fault_report(36) Jul 15 23:59:26 amit nmbd[7831]: === Jul 15 23:59:26 amit nmbd[7831]: [2005/07/15 23:59:26, 0] lib/fault.c:fault_report(37) Jul 15 23:59:26 amit nmbd[7831]: INTERNAL ERROR: Signal 11 in pid 7831 (3.0.10) Jul 15 23:59:26 amit nmbd[7831]: Please read the appendix Bugs of the Samba HOWTO collection Jul 15 23:59:26 amit nmbd[7831]: [2005/07/15 23:59:26, 0] lib/fault.c:fault_report(39) Jul 15 23:59:26 amit nmbd[7831]: === Jul 15 23:59:26 amit nmbd[7831]: [2005/07/15 23:59:26, 0] lib/util.c:smb_panic2(1482) Jul 15 23:59:26 amit nmbd[7831]: PANIC: internal error Jul 15 23:59:26 amit nmbd[7831]: [2005/07/15 23:59:26, 0] lib/util.c:smb_panic2(1490) Jul 15 23:59:26 amit nmbd[7831]: BACKTRACE: 22 stack frames: Jul 15 23:59:26 amit nmbd[7831]:#0 /usr/sbin/nmbd(smb_panic2+0x64) [0x481844] [snip] Mounting network filesystems (smb shares) and various samba comands (smbtree, smbstatus) also crashes. I've tried re-emerge samba but it did not help. Any idea where could be the problem? I have amd64 system. The important packages that was upgraded are: [ebuild U ] sys-libs/zlib-1.2.2-r1 [1.2.2] -build 0 kB [ebuild U ] sys-devel/gcc-config-1.3.11-r4 [1.3.11-r3] 0 kB [ebuild U ] sys-kernel/linux-headers-2.6.11-r2 [2.6.8.1-r4] 0 kB [ebuild U ] sys-libs/glibc-2.3.5 [2.3.4.20041102-r1] -build -debug -erandom -hardened (-multilib) +nls -nptl -nptlonly -pic (-selinux) +userlocales 0 kB [ebuild U ] sys-devel/libtool-1.5.18-r1 [1.5.16] 0 kB [snip] [ebuild NS ] sys-libs/db-4.2.52_p2 -bootstrap -doc -java -nocxx +tcltk 0 kB [ebuild U ] sys-libs/pam-0.78-r2 [0.77-r6] +berkdb -nis -pam_chroot -pam_console -pam_timestamp -pwdb (-selinux) 0 kB [ebuild U ] sys-apps/shadow-4.0.7-r3 [4.0.5-r3] +nls +pam (-selinux) -skey 0 kB [ebuild U ] sys-apps/pam-login-3.17 [3.14] -livecd +nls (-selinux) 0 kB [snip] [ebuild U ] sys-fs/udev-058 [056] (-selinux) -static 0 kB Thanks Robert -- Robert Cernansky E-mail: [EMAIL PROTECTED] Jabber: [EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
[gentoo-user] how do I change to /etc/conf.d/hostname
I have a day off work and it is too hot to play outside, so I got the bright idea to clean up some minor warnings I have been getting. When I am booting my computer, I get warnings to change from using /etc/hostname to /etc/conf.d/hostname. Also the same with my /etc/domainname. My first though was these topics would be covered in the Gentoo install manual. But after re-reading the manual I find no difference now than when I installed everything a few months ago. Is there something I need to do, or do I live with this for now? -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how do I change to /etc/conf.d/hostname
Robert Crawford wrote: On Friday 15 July 2005 06:31 pm, George Roberts wrote: I have a day off work and it is too hot to play outside, so I got the bright idea to clean up some minor warnings I have been getting. When I am booting my computer, I get warnings to change from using /etc/hostname to /etc/conf.d/hostname. Also the same with my /etc/domainname. My first though was these topics would be covered in the Gentoo install manual. But after re-reading the manual I find no difference now than when I installed everything a few months ago. Is there something I need to do, or do I live with this for now? IIRC, I just copied my /etc/hostname to /etc/conf.d/hostname. Did that remove the warning for you? After I re-did both files, I am still getting the same warning. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how do I change to /etc/conf.d/hostname
I did this: rm /etc/hostname nano -w /etc/conf.d/hostname [ edit properly ] env-update reboot No problems... /djb George Roberts wrote: Robert Crawford wrote: On Friday 15 July 2005 06:31 pm, George Roberts wrote: I have a day off work and it is too hot to play outside, so I got the bright idea to clean up some minor warnings I have been getting. When I am booting my computer, I get warnings to change from using /etc/hostname to /etc/conf.d/hostname. Also the same with my /etc/domainname. My first though was these topics would be covered in the Gentoo install manual. But after re-reading the manual I find no difference now than when I installed everything a few months ago. Is there something I need to do, or do I live with this for now? IIRC, I just copied my /etc/hostname to /etc/conf.d/hostname. Did that remove the warning for you? After I re-did both files, I am still getting the same warning. -- gentoo-user@gentoo.org mailing list
[gentoo-user] No sound or picture on saa7134 TV Tuner card
The module (saa7134) is loaded and has the correct board type and tuner but no matter what application I use (I have tried xawtv, kdetv and tvtime) I cannot get any picture or sound. I've tried the instructions found at http://gentoo-wiki.com/HARDWARE_saa7134 but they haven't helped at all. The card is a Pinnacle PCTV Stereo which uses the Philips saa7134 module, I can get it to work under Mandrake but have not had any luck under Gentoo which is my prefered Linux version. Any help greatly appreciated otherwise my PVR machine may have to end up being Windows based :( -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how do I change to /etc/conf.d/hostname
David Busby wrote: I did this: rm /etc/hostname nano -w /etc/conf.d/hostname [ edit properly ] env-update reboot No problems... /djb Removing the files did the job. Sounds like another evil plot to force me to clean up unused files. :-) Thanks! George Roberts wrote: Robert Crawford wrote: On Friday 15 July 2005 06:31 pm, George Roberts wrote: I have a day off work and it is too hot to play outside, so I got the bright idea to clean up some minor warnings I have been getting. When I am booting my computer, I get warnings to change from using /etc/hostname to /etc/conf.d/hostname. Also the same with my /etc/domainname. My first though was these topics would be covered in the Gentoo install manual. But after re-reading the manual I find no difference now than when I installed everything a few months ago. Is there something I need to do, or do I live with this for now? IIRC, I just copied my /etc/hostname to /etc/conf.d/hostname. Did that remove the warning for you? After I re-did both files, I am still getting the same warning. That did is. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fdisk errors after repartitioning
Hi Benno, 2005/7/15, Benno Schulenberg [EMAIL PROTECTED]: Hi Robert, /dev/hda52212 7971739063024 83 Linux /dev/hda6 1312211 1048792+ 82 Linux swap / Solaris cfdisk doesn't even start but stops with the following message: FATAL ERROR: Bad logical partition 6: enlarged logical partitions overlap Press any key to exit cfdisk Have you tried using parted on this disk? I'm quite curious as to what it will say about these out-of-order partitions, and the unusual boundaries. (Be careful though with executing other commands than print, as they act immediately.) I tried parted, but it doesn't report any errors: (parted) print Disk geometry for /dev/hda: 0.000-57231.562 megabytes Disk label type: msdos MinorStart End Type Filesystem Flags 1 0.031 63.984 primary ext2 2 63.984 39235.710 extended 6 64.015 1088.226 logical linux-swap 5 1088.227 39235.710 logical reiserfs 3 39235.711 47230.158 primary boot 4 47231.789 57231.070 primary fat32 lba Benno Robert. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT?] Advice on hardware
On Fri, 15 Jul 2005 12:06:52 -0300 Daniel da Veiga [EMAIL PROTECTED] wrote: Any advice? Yes - Get a 939 pin processor. 754 pin socket is going away. Also, it's not just the motherboard. If you are a gamer - start with the Gfx card. If you are going for one of the high end cards, be aware they suck down 75+ watts - more than the cpu will. A specifically designed power supply for these cards is called for, otherwise you;ll be back here asking about random compile failures, graphics lockups, corrupt file systems, etc. To support the gfx card and the system, you'll need a power supply that has a decent continous power rating, such as a PC PowerCooling Silencer 470ATX. If you want an SLI config, then the PC PowerCooling Turbo-Cool 510 Express/SLI is called for. PS: Nvidia would be my choice for video, anyone have suggestions? Unless you have a specific need for high end, a 6600 GT works well with a smaller power draw - 350 W power supply. Here is a partial output of the Shuttle SN95G5 that I'm running with an xfx 6600GT - Operating System Information Linux 2.6.12-gentoo-r4 #1 Mon Jul 11 19:02:55 PDT 2005 on x86_64 Memory Information Total RAM 1025124 kB Total Swap 1999864 kB CPU Information CPU 01 AuthenticAMD AMD Athlon(tm) 64 Processor 3000+ Bob -- - -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Filesystem access to Gentoo from XP over SSH - suggestions?
I have tried shfs ( a kernel module allowing the reomote mount of filesystems over ssh) in the past but over less than perfect connections had file system corruption. Lately I am using WebDav (via davfs or cadaver), both running on my own server, and mounting other systems I have access to onto my laptop as I move around with good success. Gentoo's implementation of davfs is clunky and problematic to manage (interfacing with fstab, having to kill a process after unmounting), but it works reliably. BillK On Fri, 2005-07-15 at 16:08 +0100, Steve [Gentoo] wrote: I was wondering if others have tackled this before me? I only have remote SSH access to my Gentoo box (everything else is fire walled) and I use this to map ports on my client machine to corresponding ports on my server machine - so, for example, I can access IMAP; SMTP, X applications etc. on the Gentoo server remotely. My remote client is WinXP, and I use Cygwin OpenSSH. Owing to low bandwidth. While Xwindows is usable it is very slow - command line interaction without X is acceptable for most tasks but imposes a very restrictive interface. If possible I would like to be able to edit (relatively small) configuration files remotely in a text editor running under windows. What would be the easiest way to somehow map a network drive on my Gentoo server from Windows given that I need to tunnel over SSH? -- William Kenworthy [EMAIL PROTECTED] Home! -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how do I change to /etc/conf.d/hostname
On Friday 15 July 2005 06:31 pm, George Roberts wrote: I have a day off work and it is too hot to play outside, so I got the bright idea to clean up some minor warnings I have been getting. When I am booting my computer, I get warnings to change from using /etc/hostname to /etc/conf.d/hostname. Also the same with my /etc/domainname. My first though was these topics would be covered in the Gentoo install manual. But after re-reading the manual I find no difference now than when I installed everything a few months ago. Is there something I need to do, or do I live with this for now? I never had that problem, as I reboot every day, so env-update gets run. -- gentoo-user@gentoo.org mailing list
