ntfs-3g segmentation fault

My dev/sda6, which is a ntfs partition, can not mounted automatically anymore

Syndrome:
-Used to work ok, mounted under my home directory (/home/DangVinh/Documents) every boot time by a line in fstab
-Now: mount failed:

kernel: mount.ntfs[5279]: segfault at 9ddc238 ip 00b974e7 sp bfb52b9c error 4 in libntfs-3g.so.71.0.0[b8f000+58000]

-But I CAN mount /dev/sda6 /mnt, or any other directory, as long as it is not the old directory under my home dir

Solution:

umount /dev/sda6
ls -al /home/DangVinh

–>Spotted that /home/DangVinh/Documents directory is corrupted, listed by ????????. Tux doesn’t understand what kind of dir it is.

rm -r /home/DangVinh/Documents
–>Transport endpoint not connected

Can’t delete this folder.

After several tries, this is how I manage to do the job:

$umount -all
–>That’s sucks. Don’t panic

$rm /home/DangVinh/Documents
–>No complaints

$ls -al /home/DangVinh
–>Documents folder is gone. Fantastic

$reboot now
–>Oops… Have to push reset button

After reboot, recreate /home/DangVinh/Documents

$mount /dev/sda6
–>Works

A happy ending.

ktorrent not running anymore

Somehow ktorrent refused to run. I don’t know if this happen after or before kernel update

This is the message:

Warning: Calling appendChild() on a null node does nothing.

$tail /var/log/messages shows several complaints about corrupt file system. I’ll check if this is the reason

Update: fixed. Fixed errors with filesystem and ktorrent runs

vmware and FC11 kernel update

I have just updated my machine with the latest kernel (Jan 06 2010)

[DangVinh@localhost ~]$ uname -r
2.6.30.10-105.fc11.i686.PAE

Ran vmware, took 2 minutes to rebuild modules.

And so far so good.

Merry Christmas everyone!

I hate tinyurl

Many people use tinyurl, but I don’t. And I don’t like posts that contain tinyurl link

Why? Because with tinyurl, I could not see the real url. Everytime I click a tiny url, I feel like I am ambushed.

“This could be heaven or this could be hell”

What will happen if I click an url, and something like www.childporn.com/age-12.html appear?

Absolutely not good for me.

So, I’ll install greasemonkey, install urldecoder in order to get rid of these stupid tiny urls

Why we love *nix?

Because of these commands

unzip, strip, touch, finger, grep, mount, fsck, more, yes, fsck, fsck, fsck, umount, sleep

Well, maybe just one fsck

Wow Factor: Compiz on Fedora 11

Okay, this stuff is only for fun. But sometime you need to refesh your /dev/brain, isn’t it?

I’ve been using fedora for nearly 2 years but I have never used Desktop Effects. Every time I click that button, my desktop is not manageable:windows missing title bar, flashing screen,… but I didn’t bother to check why.

And today, I’ve found out the answer: I need to run compiz-manager.

How stupid I am!

Hey, desktop cube is so cool !!!

vmware and excessive long guest shutdown time

Another issue with vmware. When you shutdown your guest OS, it take too long, vmware freeze, screen not refresh. After that, you have to kill vmware-vmx in order to run vmware again.

Solution:

Just kill pcscd. Disable it if you don’t have PCMCIA cards

vmware 6.5.0, fedora 11, no /dev/dsp

I ran into this issue, and the solution is:

modprobe snd-pcm-oss

VMWare module compile nightmare with kernel 2.6.30

After update running system to Fedora 11, kernel 2.6.30, vmware does not run any more.

I’ve been struggling with this issue for weeks…and finally, I’ve made it

This is my stuff:
-vmware 6.5.0
-fedora 11 with kernel 2.6.30.9-90.fc11.i686.PAE

This has been discussed in vmware community. Here I just rewrite my steps:

1. Install kernel source, gcc and some library header. I don’t remember exactly what library is needed, but this step is easy, so do it yourself

2. Download vmware-6.5.2-newkernmods.patch and vmware-6.5.2-newkernmods.sh

3. This is the important step. Because this update is intended for vmware workstation 6.5.2, the patching script is coded with fixed md5 hash of 6.5.2 module files. I use 6.5.0, so I have to change it. Simple enough, just run the shell script. It will complain about md5 hash incorrect. Then copy your md5 hash, vi , paste your md5 hash to the script, run it again, edit again…

After several edit times, everything will run.
And vmware will run without a glitch.

This is md5 hash for 6.5.0:

[root@localhost vmware-any-any-update117c]# head -n 100 vmware-6.5.2-newkernmods.sh
#!/bin/sh
# VMWare 6.5.2 New Kernel Modules
# The purpose of this script is to help people install the accompanying
# patch, which corrects for bit rot in some VMware kernel modules,
# allowing them to compile and run on newer versions of the Linux kernel.
# It is only intended for VMware Workstation version 6.5.2 (but should
# be adaptable to any similar version, just change the MD5SUM variables
# below).
# Josh Lehan (krellan) 7/2009
### TODO handle VMware Server
### TODO handle VMware Player

# Strict error handling
set -e

# Change these if desired
VMWAREVER=”6.5.2″
MODSRCDIR=”/usr/lib/vmware/modules/source”
PATCHFILE=”vmware-${VMWAREVER}-newkernmods.patch”
BACKEXT=”.backup-${VMWAREVER}”

# Known good MD5SUM values for VMware Workstation 6.5.2 64-bit
# If you change these, be sure to also change VMWAREVER above
KG64FILE=`mktemp -t`
cat <”$KG64FILE”
e2a82d15ca08e4dc59c7ef911f0ee91f vmblock.tar
ce33586383ef575107fdbf198267db4b vmci.tar
fa9889eaaf62ea2517d9c2b57e7fa0b6 vmmon.tar
213c0b027db9fc872b0fe9e66323ffc2 vmnet.tar
d70497dbb937e47b978b7af0d86ab1a7 vmppuser.tar
61d5d0b57299f99d7452a7cc7890ec9b vsock.tar
EOF

I’m too lazy to change VMWAREVER to “6.5.0″