Dr. Ephemeron

Never mind... I found it in messages... it is a big mess of stuff.

Mar 29 19:23:52 localhost shutdown: shutting down for system reboot
Mar 29 19:23:52 localhost init: Switching to runlevel: 6
Mar 29 19:23:52 localhost gconfd (dr_ephemeron-2714): Received signal 1, shutting down cleanly
Mar 29 19:23:53 localhost gconfd (dr_ephemeron-2714): Exiting
Mar 29 19:23:54 localhost kernel: bootsplash 3.1.3-2003/11/14: looking for picture.... silentjpeg size 14856 bytes, found (800x600, 14808 bytes, v3).
Mar 29 19:23:54 localhost kernel: bootsplash: status on console 0 changed to on
Mar 29 19:23:54 localhost kernel: Unable to handle kernel paging request at virtual address e1df50f0
Mar 29 19:23:54 localhost kernel:  printing eip:
Mar 29 19:23:54 localhost kernel: c023fe77
Mar 29 19:23:54 localhost kernel: *pde = 1d3b2067
Mar 29 19:23:54 localhost kernel: *pte = 00000000
Mar 29 19:23:54 localhost kernel: Oops: 0000 [#1]
Mar 29 19:23:54 localhost kernel: CPU:    0
Mar 29 19:23:54 localhost kernel: EIP:    0060:[splash_renderc+247/448]    Tainted: PF  VLI
Mar 29 19:23:54 localhost kernel: EIP:    0060:[<c023fe77>]    Tainted: PF  VLI
Mar 29 19:23:54 localhost kernel: EFLAGS: 00010202
Mar 29 19:23:54 localhost kernel: EIP is at splash_renderc+0xf7/0x1c0
Mar 29 19:23:54 localhost kernel: eax: 00000001   ebx: e1df50f0   ecx: e09d86f0   edx: 00000000
Mar 29 19:23:54 localhost kernel: esi: 00000000   edi: 00000000   ebp: db391db4   esp: db391d90
Mar 29 19:23:54 localhost kernel: ds: 007b   es: 007b   ss: 0068
Mar 29 19:23:54 localhost kernel: Process splash (pid: 2822, threadinfo=db390000 task=db3978c0)
Mar 29 19:23:54 localhost kernel: Stack: 00000640 00000010 00000003 00000000 0000ad55 00000001 00000240 e1831000 
Mar 29 19:23:54 localhost kernel:        00000007 db391dec c024052a e1831000 c041da40 00000007 00000000 c041dc34 
Mar 29 19:23:54 localhost kernel:        0000025e 00000158 00000010 00000008 00000002 c1516000 c041da40 db391e78 
Mar 29 19:23:54 localhost kernel: Call Trace:
Mar 29 19:23:54 localhost kernel:  [splash_cursor+362/480] splash_cursor+0x16a/0x1e0
Mar 29 19:23:54 localhost kernel:  [<c024052a>] splash_cursor+0x16a/0x1e0
Mar 29 19:23:54 localhost kernel:  [fbcon_cursor+647/992] fbcon_cursor+0x287/0x3e0
Mar 29 19:23:54 localhost kernel:  [<c0238bb7>] fbcon_cursor+0x287/0x3e0
Mar 29 19:23:54 localhost kernel:  [do_IRQ+229/256] do_IRQ+0xe5/0x100
Mar 29 19:23:54 localhost kernel:  [<c010d845>] do_IRQ+0xe5/0x100
Mar 29 19:23:54 localhost kernel:  [common_interrupt+24/32] common_interrupt+0x18/0x20
Mar 29 19:23:54 localhost kernel:  [<c010bf28>] common_interrupt+0x18/0x20
Mar 29 19:23:54 localhost kernel:  [complement_pos+18/368] complement_pos+0x12/0x170
Mar 29 19:23:54 localhost kernel:  [<c01fc222>] complement_pos+0x12/0x170
Mar 29 19:23:54 localhost kernel:  [hide_cursor+69/96] hide_cursor+0x45/0x60
Mar 29 19:23:54 localhost kernel:  [<c01fc715>] hide_cursor+0x45/0x60
Mar 29 19:23:54 localhost kernel:  [redraw_screen+312/464] redraw_screen+0x138/0x1d0
Mar 29 19:23:54 localhost kernel:  [<c01fc9a8>] redraw_screen+0x138/0x1d0
Mar 29 19:23:54 localhost kernel:  [vc_resize+923/1184] vc_resize+0x39b/0x4a0
Mar 29 19:23:54 localhost kernel:  [<c01fd09b>] vc_resize+0x39b/0x4a0
Mar 29 19:23:54 localhost kernel:  [splash_status+249/368] splash_status+0xf9/0x170
Mar 29 19:23:54 localhost kernel:  [<c023d429>] splash_status+0xf9/0x170
Mar 29 19:23:54 localhost kernel:  [splash_write_proc+478/1056] splash_write_proc+0x1de/0x420
Mar 29 19:23:54 localhost kernel:  [<c023d73e>] splash_write_proc+0x1de/0x420
Mar 29 19:23:54 localhost kernel:  [dentry_open+173/416] dentry_open+0xad/0x1a0
Mar 29 19:23:54 localhost kernel:  [<c015314d>] dentry_open+0xad/0x1a0
Mar 29 19:23:54 localhost kernel:  [proc_file_write+46/64] proc_file_write+0x2e/0x40
Mar 29 19:23:54 localhost kernel:  [<c017fece>] proc_file_write+0x2e/0x40
Mar 29 19:23:54 localhost kernel:  [vfs_write+142/224] vfs_write+0x8e/0xe0
Mar 29 19:23:54 localhost kernel:  [<c0153cde>] vfs_write+0x8e/0xe0
Mar 29 19:23:54 localhost kernel:  [sys_write+46/80] sys_write+0x2e/0x50
Mar 29 19:23:54 localhost kernel:  [<c0153dae>] sys_write+0x2e/0x50
Mar 29 19:23:54 localhost kernel:  [sysenter_past_esp+82/121] sysenter_past_esp+0x52/0x79
Mar 29 19:23:54 localhost kernel:  [<c010afa9>] sysenter_past_esp+0x52/0x79
Mar 29 19:23:54 localhost kernel: 
Mar 29 19:23:54 localhost kernel: Code: c6 07 00 00 00 75 11 8b 45 18 8a 10 40 81 e2 ff 00 00 00 89 45 18 89 d7 89 f8 8b 55 ec 84 c0 78 15 8b 45 f0 8b 55 e8 85 c0 74 0b <66> 8b 03 89 c2 81 e2 ff ff 00 00 83 c3 02 f7 c7 40 00 00 00 75 
Mar 29 19:23:54 localhost kernel:  <6>bootsplash: status on console 0 changed to o

So what log would I look in to see details of the shutdown seg fault of the "bootsplash"... I see nothing in messages?

Err I mean I see nothing that looks like the same message that scrolls up the screen that is... I see stuff that may be it, that I don't understand.

What is the difference between starting Gnome with init 5 and starting it with startx?

I notice that sometimes startx can give you problems with permissions and init 5 doesn't... I am not sure exactly how it works.

For example, /dev/nvidia* is set 0600 by pam_console and if I use startx to start GNOME, glxgears won't run because it needs 0660. If I su and then init 5, it will run.

Also I notice that shutdown from gnome that has been started using startx can give errors or seg faults when a shutdown after being started with init 5 will go smoothly.

For example if I shutdown from GNOME after starx from console. I see a seg fault in the bootsplash routine. Although I have no ill effects from it. I need to look and see if it is a file permission thing causing it... probably is.

I am a big tightwad myself... but I think the time is ripe to kick in some left over lunch money just to help the cause. (The cause being to phuck M$ a good one.)

Looks like a fix is on the way...

 

http://www.mandrakeclub.com/modules.php?op...=21231&forum=11

Here is the fix now:

Edit the file "/usr/share/harddrake/service_harddrake".

row 24, col. 103 should read:

qw(NVdriver nvidia.ko))

The n is missing from "nvidia.ko"

In XF86Config-4, change back "nv" to "nvidia" and make sure Load "glx" is not commented out. :D

10 CE is not 'unstable' and Official Edition is not 'stable' -- it is just 'more' stable :-)

Remember, however unstable CE may be (fine for me) bugzilla is for cooker - CE is no longer cooker...

Most everyone and anyone says report to qa.mandrakesoft.com and in fact many, many people do. 10 is on the list of accepted versions for reports there too... like I said... Mandrake doesn't know how to classify it.

Ebay is full of fraudulent auctions, fradulent bids, fradulent feedbacks and they even have a large staff to clean it up... I don't see how a company giving away software and basically begging for someone to buy it would have the resources to clean up after some monkeys.

Who is a big tightwad and will never pay for Mandrake or the club as long as they can dowload it? :P

I will buy a club membership when 10 is officially released.

Hmm... You went to Bugzilla, but we're (at least I was) talking about Anthill, the bugzilla for 10.0 CE  (Bugzilla deals with Cooker). That's why you didn't find that bug. Anthill is located at http://bugs.mandrakelinux.com/index.php

Enjoy.

I don't know why you are reporting 10 CE to Anthill when it is an unstable release. I am not sure why they are even responding to them there when many people are using Bugzilla. I guess even Mandrake doesn't know how to classify it.

Quote Anthill:

"Anthill is a bug tracking system similar to Bugzilla which is used for tracking bugs in stable releases of MandrakeSoft products including Mandrake Linux and Corporate Server. For bugs in development releases of MandrakeSoft products (aka "Cooker"), use the MandrakeSoft Bugzilla instead."

are you using esd? Or are the event sounds working w/o esd? If not, I'd guess that gnome, not finding esd, hasn't figured out to use or found alsa at the time it's suppose to be playing the startup2.wav. Just a guess.

I have Enable sound server startup in GNOME and esd is a running process in System Monitor. Is there a way to change the sequence of when the service is run or force it to run early or something?

what is it? Do you use/need it? To rm a directory it would have to use -R (recurse)

 

rm -fR /var/run/fastcgi/dynamic

 

Probably just a bug in a script somewhere?

It apparently was created when I installed fastcgi or is created by fastcgi when Apache starts a fastcgi script. I don't think I need it... it's empty.

Edit: How can I tell if someone has submitted this to Mandrake in a bug report? I looked at the FAQ but didn't see anything about submitting bugs, etc...

Click here to see Bug 9300

Does anyone have problems with formatting a floppy in MDK 10. At first I could not read my floppy and I had to completly reinstall from scratch and now it works but if I use the format utility in Gnome, it does it's thing and then the disk is not readable or writeable. I can put in a pre-formated floppy and it works, but I have noticed that copying a file to the floppy takes 4x longer than copying a file from the floppy to the hard drive.

I am using GNOME and ALSA with MDK 10 and all the Event Sounds work except the Login Sound. I have starup2.wav set in Sound preferences and I can highlight it and hit the Play button and it works, but yet no work on Login.

I have an Audigy sound card.

Anyone know why?

I installed some recent updates on Mandrake 10 and now I noticed that on boot I get this right before the swapspace is enabled:

rm: cannot remove '/var/run/fastcgi/dynamic' : Is a directory

I don't know what is trying to remove that directory or why... but I don't think it has anything to do with fastcgi.

Weird.

I stopped harddrake starting up on boot and that stopped it.

I also noticed that I couldn't run glxgears as user because updates made it so reboots was chmod 600 /dev/nvidia* and it should be 666. I had to edit /etc/security/console.perms to fix it.

Thanks! :D

After installing some updates to mandrake 10, when I reboot, the stock "nv" drivers load.

I check XF86Config-4 and Load "glx" is commented out like this: # Load "glx" and Driver "nvidia" is changed back to Driver "nv".

If I edit it and change it back, then logout and type startx again, it works fine, no error messages.

When I reboot, login, startx again, the lines are changed again.

Anyone have an idea what's going wrong? It worked fine for a few weeks and some updates changed something for the bad.

Thanks

I reinstalled and that fixed it. Althouth copying a file from the floppy is about 4x slower than copying a file to it.

I actually removed magicdev with rpmdrake just to see if that would do the trick.

That ps command you gave me gives this error:

# ps -aux |grep magic

Warning: bad syntax, perhaps a bogus '-'? See http://procps.sf.net/faq.html

root 3203 0.0 0.1 1936 644 pts0 S 01:21 0:00 grep magic

Here is ps -A if there is anything in there that would be causing it, I wouldn't know what it is.

# ps -A

PID TTY TIME CMD

1 ? 00:00:02 init

2 ? 00:00:00 ksoftirqd/0

3 ? 00:00:00 events/0

4 ? 00:00:00 kblockd/0

5 ? 00:00:00 kapmd

6 ? 00:00:00 pdflush

7 ? 00:00:00 pdflush

8 ? 00:00:00 kswapd0

9 ? 00:00:00 aio/0

11 ? 00:00:00 kseriod

15 ? 00:00:00 kjournald

145 ? 00:00:00 devfsd

269 ? 00:00:00 khubd

606 ? 00:00:00 kjournald

607 ? 00:00:00 kjournald

762 ? 00:00:00 knodemgrd_0

1099 ? 00:00:00 dhclient

1163 ? 00:00:00 mysqld_safe

1204 ? 00:00:00 mysqld

1205 ? 00:00:00 portmap

1229 ? 00:00:00 syslogd

1237 ? 00:00:00 klogd

1300 ? 00:00:00 rpc.statd

1794 ? 00:00:00 xfs

1866 ? 00:00:00 gdm-binary

1899 ? 00:00:00 atd

1955 ? 00:00:00 ntpd

1985 ? 00:00:00 gdm-binary

1986 ? 00:00:14 X

1987 ? 00:00:00 sshd

2030 ? 00:00:00 xinetd

2072 ? 00:00:00 nfsd

2073 ? 00:00:00 nfsd

2074 ? 00:00:00 nfsd

2075 ? 00:00:00 nfsd

2080 ? 00:00:00 lockd

2081 ? 00:00:00 rpciod

2076 ? 00:00:00 nfsd

2077 ? 00:00:00 nfsd

2078 ? 00:00:00 nfsd

2079 ? 00:00:00 nfsd

2092 ? 00:00:00 rpc.mountd

2253 ? 00:00:00 master

2282 ? 00:00:00 pickup

2283 ? 00:00:00 qmgr

2284 ? 00:00:00 tlsmgr

2319 ? 00:00:00 crond

2351 ? 00:00:00 smbd

2365 ? 00:00:00 nmbd

2599 ? 00:00:00 gnome-session

2635 ? 00:00:00 ssh-agent

2694 ? 00:00:00 httpd2

2702 ? 00:00:00 numlock <defunct>

2708 ? 00:00:00 advxsplitlogfil

2715 ? 00:00:01 gconfd-2

2716 ? 00:00:00 httpd2

2717 ? 00:00:00 httpd2

2718 ? 00:00:00 httpd2

2719 ? 00:00:00 httpd2

2720 ? 00:00:00 httpd2

2782 ? 00:00:00 esd

2795 ? 00:00:00 bonobo-activati

2858 ? 00:00:00 gnome-settings-

2876 tty1 00:00:00 mingetty

2877 tty2 00:00:00 mingetty

2878 tty3 00:00:00 mingetty

2879 tty4 00:00:00 mingetty

2880 tty5 00:00:00 mingetty

2881 tty6 00:00:00 mingetty

2883 ? 00:00:00 fam

2935 ? 00:00:00 xscreensaver

2983 ? 00:00:00 gnome-smproxy

2985 ? 00:00:00 metacity

2989 ? 00:00:00 gnome-panel

2991 ? 00:00:01 nautilus

2993 ? 00:00:00 pam-panel-icon

2996 ? 00:00:00 pam_timestamp_c

3002 ? 00:00:00 mapping-daemon

3006 ? 00:00:00 wnck-applet

3008 ? 00:00:00 notification-ar

3010 ? 00:00:00 mixer_applet2

3061 ? 00:00:00 soundwrapper

3064 ? 00:00:00 gaim

3131 ? 00:00:02 gnome-terminal

3132 ? 00:00:00 gnome-pty-helpe

3133 pts0 00:00:00 bash

3166 pts0 00:00:00 su

3169 pts0 00:00:00 bash

3261 pts0 00:00:00 ps

Any time I try to access the floppy, MDK 10 goes into some sort of read loop trying to read the floppy and it goes on forever and sometimes it stops and sometimes I have to kill the power on the computer.

I tried the exact same fstab line that I used in 9.2 and also tried disabling supermount and no go.

I can boot up in windows and the floppy still works fine.

My MDK 10.0 works fine with cdrecord and cdrdao from the command line with:

dev=ATAPI:0,0,0 for cdrecord

--device ATAPI:0,0,0 for cdrdao

Of course you need to set other options specific for your burner type and the actual bus, target, lun it's on.

I gave up with any GUI for cdrecord and cdrdao a few months ago, even with 9.2 because they were all inheritably flakey.

Never Mind

I think they need to work the bugs out of bugzilla, I submitted a bug and it loaded a page that said wait, submission in process or something of that nature and nothing happened. Checked "My Bugs" later and nothing was there. :unsure:

I have been decrying Mandrake's sloppy releases ever since the 9.2 fiasco.  I swore that I would give them one last chance to redeem themselves by releasing a good, relatively bug-free Mandrake 10...

 

...and they screwed the pooch again!

 

bla bla bla

Cry me a river.

WINEHQ site is not very helpful for a first time installation... you don't need to read an entire developers "bible" to install it. Here are the basics:

Install the RPM then to setup the config file, run

$ winesetup

Winesetup may use the wrong string for you cdrom and it won't work, so you need to manual enter it. For MDK 9.2, the cdrom "device" for the config file is /dev/scd0 and the "path" will be /mnt/cdrom For MDK 10.0 the "device" will probably be /dev/hdx (where x is a letter) and the "path" is /mnt/cdrom

Update 3/13/2004: I installed wine-20040309-mdk.i586.rpm on MDK 10.0 and the winesetup auto detected path for my IDE cdrom works correctly without having to manually change it.

A wine folder (.wine) is created in your home directory with the config file and a fake_windows directory.

You may need to use the option in fstab to show hidden files on the CD if you have problems installing programs. I don't remember the option for this. I have noticed that some programs won't install unless you cd /mnt/cdrom first before you run setup.

Update 3/13/2004: You should definitely use the "unhide" option to show hidden files on the CD.

You mainly just need to make sure all of the config file [Drive X] sections are setup proper for you computer. The sound setup defaults to using OSS and you will have to manual change the config to use ALSA.

Run any windows program like this:

$ wine program_filename

If a Windows program you are installing prompts you to reboot, select to reboot later and then you have to run:

$ wineboot

This simulates rebooting of the computer and finalize the setup.

To remove windows programs run:

$ uninstaller

Update 3/13/2004: Now you can actually install something like Office97 or Office2000 and it should run, but many functions may lockup unless you play dll musical chairs with the config file and copying native dlls into your fake_windows directory to help make some functions work.

FYI - You will find it is more trouble trying to run it than what it is worth since it is very unfinished, unreliable, and generally buggy.