Linux | Windows Development

ASUS Windows_8 Recovery soft/hard OEM

2013-06-04T21:51:00.000-07:00

The below post shows you how to create a soft recovery and a hard recovery on windows 8 using an ASUS machine.
(NOTE:a1recovery has been discontinued)

1) Create a Soft Recovery USB/CD in case you need to do a soft recovery.

with Metro UI
right click>choose "all apps">>choose "control panel"
search "recovery"
choose>>"create a recovery drive"
NOTE: if you want to use CD/DVD then unclick the box stating "Copy the recov......."
after burn complete, you should be able to boot by adjusting your BIOS settings to point to the specific device.

2) From ASUS Download the application so you can create your recovery drive.
NOTE: unfortunately ASUS has taken out the CD/DVD option so you will need a USB thumbdrive around 16GB or a SATA device etc.. micro USB is not supported.

using your browser navigate to:
http://support.asus.com/Download.aspx?SLanguage=en&p=3&s=480&m=ASUS%20Backtracker&os=36&ft=14&f_name=AsusBacktracker_Win8_64_VER200.zip#AsusBacktracker_Win8_64_VER200.zip

or if the link gets corrupt just search "backtracker" on their site:
http://support.asus.com/ServiceHome.aspx?SLanguage=en

once installed open the app and choose your options, once burned to the drive, reboot and adjust the BIOS to point to the device, then once loaded follow the onscreen questions.

NOTE: before reformatting, make sure you write down your product ID and KEY so incase windows activation fails you can supply the info.

that's it!

NFS simple server/client connect

2013-05-15T20:29:00.003-07:00

The below post gives a simple tutorial for setting up a simple server/client session with NFS using debian and opensuse 11.3

1) install the proper packages for the server, then edit the configuration file for the server. restart the server, validate with netstat

apt-get install nfs-common nfs-kernel-server *optionally nfsv4-acl-tools

cd /etc/

gvim exports (add entries to point to your location/mount point)

save and exit

/etc/init.d/nfs-kernel-server stop/start

netstat -atu | grep nfs

2) disable firewall for server, then on client side connect to the NFS server, then verify with netstat

server side as root:

ufw disable (add port entry to iptables in order to leave enabled)

client side as root:

mount -t nfs 10.0.0.44:/mnt /mnt

verify:

3) verify you are connected then check the directory permissions in order to send/receive as regular user

netstat -atu | grep nfs

cd /mnt

ls -l (check ACL's) if wrong change them with chown

chown username:username directories

NOTE: if you have the right permissions, and you don't see anything(like mount on mount) then add the crossmnt to /etc/exports config this way you can use external storage devices and NFS

and that's it! you should have a simple server/client connect.

OpenLDAP GPG KeyServer Private

2013-03-31T13:25:00.000-07:00

The below post shows you how you can create your own Private gpg keyserver through LDAP.
tested on Fedora 17 (other distros may be different)
NOTE: all configs are at bottom of blog

1) install the proper packages. then change into the LDAP config directory. move slapd.d because we are using the older API then modify ldap.conf to add the servers IP

sudo -s

yum install openldap openldap-servers openldap-clients

optionally: yum install gq (LDAP browser)

service slapd stop (stop LDAP)

cd /etc/openldap

ls -l

mv slapd.d{,-old}

gvim ldap.conf (add ip address of server) (config below)

nm-tool

2) create slapd.conf, create db directory download and install gpg-keyserver schemas, then start the server

gvim slapd.conf (config at bottom)

mkdir /var/db/openldap

chown ldap:ldap /var/db/openldap

download and install the schemas

wget http://members.kstp.at/wh/pgp/openldap_pgp_keyserver.tar.gz

tar xvfz openldap_pgp_keyserver.tar.gz | cp -Rv openldap/schema/* /etc/openldap/schema

slapd -d -1 -u ldap -g ldap (use -d 128 for ACL's)

3) create the LDAP directory structure and add the extra users

gvim directory_base_user.ldif (config below)

ldapadd -H ldap://10.0.0.11 -x -D "cn=Justin,dc=Manager,dc=Private" -f directory_base_user.ldif -w dirtysecret

gvim remote_user.ldif (config below)

ldapadd -H ldap://10.0.0.11 -x -D "cn=Justin,dc=Manager,dc=Private" -f remote_user.ldif -w dirtysecret

exit as root

4) create and exchange keys with host and remote host:

cd .gnupg

gpg --gen-keys (answer questions)

gpg --list-keys

gvim gpg.conf (add the entries for gpg keyserver, above picture

send and recv keys:

gpg --keyserver ldap://10.0.0.11 --send-key C9AD4D9B (send key)

gpg --keyserver ldap://10.0.0.11 --recv-key C9AD4D9B (recvkey)

on the remote hot perform the same steps to create the new key and send and recv

5) use gq to view the entries

and there you have it, your own Private GPG Keyserver!!

below are the configuration files for the above:

ldap.conf

# LDAP Defaults

# See ldap.conf(5) for details

# This file should be world readable but not world writable.

BASE dc=Manager,dc=Private

#URI ldap://ldap.example.com ldap://ldap-master.example.com:666

URI ldap://10.0.0.11

SIZELIMIT 0

TIMELIMIT 900

#DEREF never

#TLS_REQCERT try

#TLS_CACERT /etc/openldap/ssl/certs/ca.pem

#TLS_CERT /etc/openldap/ssl/certs/ldap.pem

#TLS_KEY /etc/openldap/ssl/keys/ldap.key

slapd.conf

include /etc/openldap/schema/core.schema

include /etc/openldap/schema/cosine.schema

include /etc/openldap/schema/inetorgperson.schema

include /etc/openldap/schema/pgp-keyserver.schema

include /etc/openldap/schema/pgp-recon.schema

include /etc/openldap/schema/pgp-remte-prefs.schema

loglevel -1

pidfile /var/run/openldap/slapd.pid

argsfile /var/run/openldap/slapd.args

modulepath /usr/lib/openldap

moduleload back_bdb.la

moduleload back_ldap.la

moduleload back_monitor.la

allow bind_anon_dn

allow update_anon

allow bind_anon_cred

access to dn.base=""

by * read

access to dn="cn=Subschema"

by * read

access to attrs="userPassword"

by self write

by * auth

# OPTIONAL: gq for viewing directories etc..

access to dn="dc=Manager,dc=Private"

by * read

access to dn="ou=PGP Users,dc=Manager,dc=Private"

by * read

access to dn.subtree="uid=Justin,ou=PGP Users,dc=Manager,dc=Private"

by * read

access to dn.subtree="uid=Justin_2,ou=PGP Users,dc=Manager,dc=Private"

by * read

# read server pgpKeyInfo

access to dn="cn=pgpServerInfo,dc=Manager,dc=Private"

by * read

# remote access.. change for better ACL's

access to dn.subtree="ou=PGP Keys,dc=Manager,dc=Private"

by * write

by * read

access to * by * none

database bdb

directory /var/db/openldap/

suffix "dc=Manager,dc=Private"

rootdn "cn=justin,dc=Manager,dc=Private"

rootpw dirtysecret

cachesize 10000

checkpoint 128 15

index objectClass eq

index pgpCertID,pgpKeyID,pgpKeyType,pgpUserID,pgpKeyCreateTime sub,eq

index pgpSignerID,pgpSubKeyID,pgpKeySize,pgpKeyExpireTime sub,eq

index pgpDisabled,pgpRevoked eq

database monitor

directory_base_user.ldif

# domain

dn: dc=Manager,dc=Private

dc: Manager

objectClass: domain

objectClass: top

# person part of domain

dn: cn=Manager,dc=Manager,dc=Private

cn: Manager

sn: Manager

objectClass: person

objectClass: top

# applicationProcess and/or organizationalUnit works

dn: ou=PGP Keys,dc=Manager,dc=Private

objectclass: applicationProcess

cn: PGP KEYSERVER PRIVATE "KEYS"

ou: PGP Keys

description: gpg keyserver for private company use.

# Main area for gpg to do stuff

dn: cn=pgpServerInfo,dc=Manager,dc=Private

cn: pgpServerInfo

objectclass: pgpserverinfo

pgpSoftware: OpenLDAP 2.4.33

pgpVersion: gpg (GnuPG) 1.4.13

pgpBaseKeyspaceDN: ou=PGP Keys,dc=Manager,dc=Private

# applicationProcess and/or organizationalUnit works

dn: ou=PGP Users,dc=Manager,dc=Private

objectclass: applicationProcess

cn: PGP KEYSERVER PRIVATE "Users"

ou: PGP Users

description: gpg keyserver for private company use.

# company employee(s)

dn: uid=Justin,ou=PGP Users,dc=Manager,dc=Private

objectClass: inetOrgPerson

objectClass: uidObject

objectClass: userSecurityInformation

uid: Justin

# password is test56

userPassword: {SSHA}PGJjmqxXpro95gho76Gz27qiqQp59LwM

sn: Mattock

cn: Private Company Employee(s).

# photo stuff never got it working maybe u can!

# adds using URL format

#add: jpegphoto

#jpegphoto: < file://path/to/jpeg/file.jpg

# adding a binary cert, use gpgsm to convert format(DER,PEM,etc..)

#changetype: modify

#add: userCertificate;binary

#userCertificate;binary:< file:///path/to/file/binary.asc

#objectClass: stongAuthenticationUser

#UserCertificate: file:///path/to/regular.asc

remote_user.ldif

# company employee(s)

dn: uid=Justin_2,ou=PGP Users,dc=Manager,dc=Private

objectClass: inetOrgPerson

objectClass: uidObject

objectClass: userSecurityInformation

uid: Justin_2

# password is test56

userPassword: {SSHA}PGJjmqxXpro95gho76Gz27qiqQp59LwM

sn: Mattock_2

cn: Private Company Employee(s).

# photo stuff never got it working maybe u can!

# adds using URL format

#add: jpegphoto

#jpegphoto: < file://path/to/jpeg/file.jpg

# adding a binary cert, use gpgsm to convert format(DER,PEM,etc..)

#changetype: modify

#add: userCertificate;binary

#userCertificate;binary:< file:///path/to/file/binary.asc

#objectClass: stongAuthenticationUser

#UserCertificate: file:///path/to/regular.asc

XFS: xorg font server simple server/client connect

2013-03-17T21:23:00.000-07:00

a little outdated due to systems having TTF fonts and such, but was more curious to get this thing working.. anyways the below post shows you how to do a simple server/client connect with xorg font server

1) install the xfs server and utilities. stop the server and prepare the config for the server to service the protocol

su
yum install xorg-x11-xfs xorg-x11-xfs-utils
chkconfig xfs on (if you want xfs starting at boot)
service xfs stop
cd /etc/X11/fs
gvim config

<--- my config if you want to use --->
#no-listen = tcp
client-limit = 10
clone-self = on
default-point-size = 120
default-resolutions = 75,75,100,100
deferglyphs = 16
use-syslog = no
error-file = /var/log/xfs.log
port = 7100
#seems to only read symlinks
catalogue = catalogue:/home/fonts2

save..

2) prepare to the server to serve the fonts. download from the url and prepare the files/directories(catalogue:) so the server/client can communicate with each other.

cd /home (or wherever you want to prepare the fonts)
mkdir /home/fonts
mkdir /home/fonts2
cd /home/fonts
wget http://hea-www.harvard.edu/~fine/Tech/cursive.bdf
mkfontdir
(prepare the catalogue:)
ln -sv /home/fonts /home/fonts2
cd ../fonts2
ls -l
<start xfs manually>
/usr/bin/xfs
netstat -atu (to verify *xfs is on)

NOTE: url where to find fonts and where I got the info to install the *.bdf fonts http://hea-www.harvard.edu/~fine/Tech/x11fonts.html

3) on the client side test the font server to see if we have signs of life..

/usr/bin/fslsfonts -server 10.0.0.9:7100

4) connect to the xorg font server. use xset to view what entries are there, then use xset to add the new font server, verify with xset and netstat. then use fslsfonts to pick a font to use through xterm.

xset -q

xset fp+ tcp/10.0.0.9:7100

xset -q

netstat -atu

fslsfonts -server 10.0.0.9:7100

xterm -fn xterm -fn -fine-cursive-medium-i-normal--0-0-72-72-c-0-iso8859-1

view cursive fonts in the terminal!! isnt it cute!

5) OPTIONAL: add a script to gdm to auto connect the client to the server.

cd /etc/gdm/PostLogin

gvim Default

init 3 / reboot

<--copy/paste or make your own -->

#add your ip and port number

X_CLIENT_XFS_START= /usr/bin/xset fp+ tcp/10.0.0.9:7100

if [ -z "$GDMSESSION" ]; then

exec $X_CLIENT_XFS_START

NOTE: for wanting to connect before login(clients that depend on all fonts from the server) either disable gnome and use startx with xinitrc(add a startup in there), or (if you know gnome terms) create a script in:

/usr/share/gdm/greeter/applications/*.desktop

CRON: simple kernel pull/build with email notification

2013-03-13T23:09:00.001-07:00

The post below shows you how to setup up a simple crontab to do simple tasks with an email notification..

1) depending on your distro you might already have cronwrap intsalled or in the repo(s)
if not download and install

cd Downloads

wget https://pypi.python.org/packages/source/c/cronwrap/cronwrap-1.4.tar.gz

tar xvfz cronwrap*.tar.gz

cd cron*

python setup.py install

which cronwrap

exit ; cd

2) /etc/anacrontab is the main config file as well as /etc/crontab for "root" jobs in the system.

this exercise does not deal with these files, but they can be modified for added root tasks for the system for regular tasks.

3) create/start your own crontab under your username located in /var/spool/cron. the simple task of resetting the kernel and pulling.

crontab -e

enter the entries for the times you want etc.. then save and exit.

MAILTO:justinmattock@gmail.com

30 21 * * * /usr/bin/cronwrap -c "/home/kernel./pull_script" -e justinmattock@gmail.com -v

NOTE: below is my simple script I used for this example

<----- copy/paste------->

#/bin/sh

# my cheesy script to pull the kernel

for dir in /home/kernel/3.0*

cd /home/kernel/3.0 ; git reset --hard origin ;

git checkout master ; git pull ;

git checkout Mattock ; git reset --hard origin ; make

done

Mail info:

I used cronwrap since it makes sending info to myself much easier, and supplies good info in the email on how everything went(not just when things crap out)

crond does send emails but only if the scripts fails. to add this option add:

MAILTO= yourname@.com at the top of your crontab -e

4) create another task as "root" to complete the job

crontab -e

add the root commands that need to fire off after the pull and build go through.

30 22 * * * /usr/bin/cronwrap -c "cd /home/kernel/3.0 ; make modules_install ; make install" -e justinmattock@gmail.com -v

probably should of created another script, but wanted to show you don't need to make scripts to do this..

5) check the logs,(/var/log/cron) and wait for the email.. hopefully everything goes good for whatever commands you use!

the first script for pulling went good, but then when it came for the root crontab I hit a strange error:

ERROR OUTPUT:

/sbin/installkernel: line 82: new-kernel-pkg: command not found

/sbin/installkernel: line 86: new-kernel-pkg: command not found

make[1]: *** [install] Error 127

make: *** [install] Error 2

quick fix is to edit this file at 82 and 86 by adding the full path to the programs

and there you have it.. a simple cron example of pulling and resetting the kernel, then building and installing everyday between 8-9:30pm without having to do anything..because we all know how hard entering commands in a terminal can be!

FSCK: clean/fix BADSECTORS on a harddrive | READ/WRITE

2013-03-11T00:20:00.001-07:00

This post is to show you how I was able to recovery a hard Drive that was reporting to have BADSECTORS on it.
NOTE: determining the BADSECTORS can be tough either its alignment or it actually has physical damage

1) obtain a hard drive that is reporting BADSECTORS on it. over here I was able to buy one off of ebay
(keep in mind this is a BADSECTOR issue)

2) have access to the device so you can work on it

I use an external plugin to access the device through a terminal. if your device is internal then use a liveCD to perform the clean/recovery.

3) determine the state of the device. can you copy the storage device with dd or is it totally thrashed(mine was set at unknown device).

4) since I had an unknown partition I used fdisk to create a partition so you I can clean the disk e2fsck

(the layer between hardware and software)

fdisk /dev/sdb

(create a new partition with fdisk save and exit)

create ext4 on the device:

/sbin/mkefs.ext4 /dev/sdb1

5) once the fresh filesystem is created one would think a simple fsck /dev/sdb1 would of done the trick!!!!

guess this is what the guy meant by having BADSECTORS!!

(thousands of these messages)

using:

fdisk -y /dev/sdb*

above makes it easier to say "yes" to all the BADSECTORS on this thing and hopefully will fix them.

what I was not wanting to happen..: an error out with fsck.

"Error storing directory block information (inode=8209852, block=0, num=4181895) : Memory allocation failed"

NOTE: reading through seems this was a problem with bigfiles and x86_32. seems most of the people just used a x86_64 livecd system with fsck to fix this issue. but not in my case

6)using fdisk create a partition scheme with small multiple partitions so e2fsck is not dealing with largefiles(bigfiles). then fill them up(WRITE) and use fsck to clean(READ)

fdisk /dev/sd*

(plenty of tutorials for this)

mkefs.ext4 /dev/sd*

NOTE: reading through I found that with disks to repair them you need to READ/WRITE to the device to realign the sectors. hopefully this fixes my BADSECTOR issue.

7) after creating the small partitions, READ/WRITE to the partition(s).

dmesg | grep sd*

mount /dev/sd* /mountpount(s)

cd mountpoint(s)

WRITE(s) to each partition(s) either using "yes" or "dd"(yes is faster)

yes abcdefghijklmnopqrstuvwxyz0123456789 > largefile

repeat this for all partitions until all WRITE(s)

8) once full to capacity clean them with fsck(READ)

as you can see WRITE(s) to the disk seems to make things work better, as well as having a smaller partition for fsck(READ) to deal with.

all clean!

9) remove the "largefile" so you have disk space to use.

from here its whatever you want.. I ended up re-partitioning so I can run windows/linux on this disk. using fsck with 100Gigs went fine without any issues after doing the above realignment READ/WRITE(s)

Automatic Login with SSH: authorized_keys(2&) | publickey(s)

2013-03-06T00:00:00.001-08:00

For whatever reason ssh seems so buggy, even though its not. in anycase these are my findings to having a session without having to login each time or having a public Login type scenario..:

1) adjust your firewall to allow ssh port 22

menu>administration>firewall

2) open a terminal change into .ssh then create a new private/public key(save the artwork as well).

cd .ssh

ls -l

ssh-keygen -t rsa -f /home/justin/.ssh/id_rsa -C ' '

gvim artwork <--copy/paste your key artwork from the screen then save -->

optional artwork from the default:

ssh-keygen -lv -f /etc/ssh/ssh_host_rsa_key >> artwork

3) add the key to the database then create authorized_keys. After creating the file(s) then create a compressed image of everything to be transferred over to the other host system either by disk drive or scp whatever you prefer.

ssh-add id_rsa

ssh-add -l

ls -l

cp id_rsa.pub authorized_keys

tar -cf ssh_key_pub.tar artwork authorized_keys id_rsa id_rsa.pub

ls -l

4) using the other machine, use scp to copy the image of files the uncompress and add the key to the database.

scp justin@10.0.0.17:/home/justin/.ssh/ssh_key_pub.tar .

ls -l

tar -xf ssh_key_pub.tar

ls -l

ssh-add id_rsa

NOTE: for whatever reason I hit some connection issue with scp, strange thing is everything worked as it should.

(BUGGY)

5) adjust the config file(s) so you can be using RSA authentication

gvim /etc/ssh/sshd_config

RSAAuthentication yes

PubkeyAuthentication yes

AuthorizedKeysFile /home/justin/.ssh/authorized_keys

PasswordAuthentication no

/etc/ssh/ssh_config

RhostRSAAuthentication yes

RSAAuthentication yes

PasswordAuthentication no

VisualHostKey yes

save and exit.

6) restart sshd and login

service sshd restart

ssh justin@10.0.0.17|3 etc..

NOTE: the first login seems to have /usr/libexec/gcr-prompter in of which requires the password you used for the RSA keyfile, after giving the password click on the box on the bottom to remember. after the initial login all other login(s) are without a password. still need to figure out why gcr-prompter fires off.

hope this helps anybody having issues with this.

Calculate CIDR Netmask Manually

2013-02-27T10:40:00.000-08:00

This is a quick tutorial on how to calculate the CIDR netmask number if you find yourself without a CIDR calculator(like taking a test or something).

1) understand the setup for CIDR i.g. 4 groups of 8 bits equaling 32 (example)

11111111.11111111.11111111.11111111 = 32 bits (4 groups of 8 separated by a "." )

2) total number in each group = 255

11111111 = 255

WTF? each group is easily calculated with these numbers(blocks)!

128+64+32+16+8+4+2+1 = 255 (each bit(block) has a number, add all of them to get 255)

-----------------------------------------------------------------------------------------------------------------------------------------------------

3) Calculate the CIDR netmask of /30

a) count out 30 bits of 1 one(s) and the rest are zero(s) equaling 32 bit(s)

11111111.11111111.11111111.11111100

30 one(s) and 2 zero(s) which equal = 32 bits

4) seperate and count each group:

11111111.11111111.11111111.11111100

group 1 = 11111111 = 255
128+64+32+16+8+4+2+1 = 255

group 2 = 11111111 = 255
128+64+32+16+8+4+2+1 = 255

group 3 = 11111111 = 255
128+64+32+16+8+4+2+1 = 255

group 4 = 11111100 = 252
128+64+32+16+8+4 = 252

NOTE: for each zero in a group remove a bit(block) starting from 1(right to left)
128+64+32+16+8+4+2+1 = 255

the calculated CIDR netmask of /30 = 255.255.255.252

------------------------------------------------------------------------------------------------------------------------------------------------------

diagram of CIDR:

Netmask              Netmask (binary)                 CIDR     Notes    
_____________________________________________________________________________
255.255.255.255  11111111.11111111.11111111.11111111  /32  Host (single addr)
255.255.255.254  11111111.11111111.11111111.11111110  /31  Unuseable
255.255.255.252  11111111.11111111.11111111.11111100  /30    2  useable
255.255.255.248  11111111.11111111.11111111.11111000  /29    6  useable
255.255.255.240  11111111.11111111.11111111.11110000  /28   14  useable
255.255.255.224  11111111.11111111.11111111.11100000  /27   30  useable
255.255.255.192  11111111.11111111.11111111.11000000  /26   62  useable
255.255.255.128  11111111.11111111.11111111.10000000  /25  126  useable
255.255.255.0    11111111.11111111.11111111.00000000  /24 "Class C" 254 useable

255.255.254.0    11111111.11111111.11111110.00000000  /23    2  Class C's
255.255.252.0    11111111.11111111.11111100.00000000  /22    4  Class C's
255.255.248.0    11111111.11111111.11111000.00000000  /21    8  Class C's
255.255.240.0    11111111.11111111.11110000.00000000  /20   16  Class C's
255.255.224.0    11111111.11111111.11100000.00000000  /19   32  Class C's
255.255.192.0    11111111.11111111.11000000.00000000  /18   64  Class C's
255.255.128.0    11111111.11111111.10000000.00000000  /17  128  Class C's
255.255.0.0      11111111.11111111.00000000.00000000  /16  "Class B"
     
255.254.0.0      11111111.11111110.00000000.00000000  /15    2  Class B's
255.252.0.0      11111111.11111100.00000000.00000000  /14    4  Class B's
255.248.0.0      11111111.11111000.00000000.00000000  /13    8  Class B's
255.240.0.0      11111111.11110000.00000000.00000000  /12   16  Class B's
255.224.0.0      11111111.11100000.00000000.00000000  /11   32  Class B's
255.192.0.0      11111111.11000000.00000000.00000000  /10   64  Class B's
255.128.0.0      11111111.10000000.00000000.00000000  /9   128  Class B's
255.0.0.0        11111111.00000000.00000000.00000000  /8   "Class A"
  
254.0.0.0        11111110.00000000.00000000.00000000  /7
252.0.0.0        11111100.00000000.00000000.00000000  /6
248.0.0.0        11111000.00000000.00000000.00000000  /5
240.0.0.0        11110000.00000000.00000000.00000000  /4
224.0.0.0        11100000.00000000.00000000.00000000  /3
192.0.0.0        11000000.00000000.00000000.00000000  /2
128.0.0.0        10000000.00000000.00000000.00000000  /1
0.0.0.0          00000000.00000000.00000000.00000000  /0   IP space

Note: I dont want to plagerise so i have to give most of the credit to the people in this forum who discussed on
how to get this calculated. I just(hopefully) made sense of the confusion!
http://www.techexams.net/forums/ccna-ccent/38980-help-figuring-out-cidr-notation-calculations.html

IBM Thinkpad G40 BIOS password bypass hack

2013-02-21T23:57:00.000-08:00

The below post will show you how to bypass the BIOS password on the IBM Thinkpad G40.

1) obtain an IBM Thinkpad G40 with a locked BIOS

2) disconnect the battery and the A/C

3) start disassembling the machine with a screwdriver

4) find the CMOS battery and disconnect it. then turn the machine on(A/C only) to see if that's all you need to do to reset the password. if not then continue reading.(leave the CMOS battery out of the machine)

5) after disassembling the machine look for the ATMEL chip!

Oops.. wrong chip! this is the processor(PENTIUM M)

Ahh.. there it is!! you cant miss it, says ATMEL all over it.

6) now that you found the ATMEL chip, now you need to short the chip while turning on the machine.

spec of the chip:

you want to short the PROT and GRND out!

as you can see I simply used tweezers to do the job. now that you are shorting the machine hold and turn the machine on, and wait until you are in the BIOS..:

NOTE: dont let go of the tweezers or the screen goes grey.

after a few seconds you should have access to the BIOS. there disable/change the password.

save and exit!

hopefully this helps anybody who has this issue.

MySQL: simple database session

2013-02-19T22:43:00.000-08:00

The previous post was to turn on the MySQL server, now its time to actually do something with it!

1) start a MySQL session as root to create the user and GRANT permissions then verify and flush

sudo mysql -u root -hlocalhost

mysql>

CREATE USER 'justin'@'localhost' IDENTIFIED BY 'test';
GRANT ALL ON db1.* TO 'justin'@'localhost';
GRANT USAGE ON *.* TO 'justin'@'localhost' WITH MAX_QUERIES_PER_HOUR 90;
GRANT ALL ON mydb.* TO 'justin'@'localhost';
GRANT SELECT, INSERT ON mydb.* TO 'justin'@'localhost';
GRANT ALL ON mydb.mytbl TO 'justin'@'localhost';

verify and restart the privileges:

SHOW GRANTS FOR 'justin'@'localhost';
FLUSH PRIVILEGES;

2) exit from the root mysql session, start mysql as normal user, then create a database and use it, verify at the end

mysql>

create database test;
use test;
\s

3) create a table in the database and add names in the tables

mysql>

CREATE TABLE test (name VARCHAR(20), last_name VARCHAR(20));

show tables;

INSERT INTO test VALUES ('justin','mattock');

DESCRIBE test;

SELECT * FROM test;

4) delete a table entry, delete a table, delete a database and delete a user

mysql>

ALTER TABLE test DROP name;

SELECT * FROM test;

drop table test;

show tables;

DROP database test;

NOTE: to delete user exit normal users session, then start a root mysql session then the below command to DROP the user:

mysql>

DROP USER 'justin'@'localhost';

5) have fun with MySQl.. as you can see once you get the hang of it, MySQL is really easy to use..:

Starting Mysqld on fedora 17

2013-02-18T22:37:00.002-08:00

below is a post about how I was able to start mysqld on fedora 17.

1) yum install mysql mysql-server

2) service mysqld start
NOTE: if you get a: Job failed. see system journal and 'systemctl status' for details
then hopefully the below helps.

3) cat /var/log/messages|mysqld.log should give you a better idea of whats going on

after a bit of searching my findings found the best way to deal with the above ERROR's is to:

4) change to the variable directory and rename mysql and create a whole new set of files/dir(s)

cd /var/lib ; ls -l mysql

mv mysql{,-old}

mysql_install_db

5) now correct the ownership and SELinux contexts

chown -R mysql: mysql

restorecon -R /var/lib/mysql

NOTE: should we have mysql UID|GID for all? or just some files?

7) start the server:

service mysqld start

check the logs to verify we have mysqld started!

cat /var/log/mysqld.log

7) check/upgrade tables for Mysqld

mysql_upgrade

8) connect to Mysql database (localhost)

mysql -u username -p -hlocalhost

and that's it you are now using a Mysql database.

NOTE: press 'ENTER' for the password.. or setup password(s) logins for Mysql!

Copy NAND Partition From One SAMSUNG SGH-T499 to Another to recover it!

2013-02-17T18:47:00.001-08:00

This is a quick tutorial on how to copy the partition from one SAMSUNG SGH-T499 to another so it can be recovered from an unworkable(hard bricked) state.

1) Solder the JTAG points on both devices a) working device b) bricked device

NOTE: some devices may not need to be soldered(simply USB)

2) plug the _good_ device in the JTAG box so you can copy all the partitions.

3) open JTAG MANAGER and choose the resurrector or manually choose the chipset for qualcomm(msm7627)

4) open qualcomm partitioning tool, then load "get partitions from device"

5) once partitions are loaded double-click each one to select it, then press "read"

6) once finished reading right click and save/download full/spare all partitions to a certain directory.

7) disconnect the good device and plugin the bricked device

(NOTE: the resurrector that JTAG comes with is for another SGH-T499 so it will not fix the TMOBILE version; just a flashing SAMSUNG and reboot; in this case with the image below I tried firmware for another device

and only had flashing LEDS)

8) Now open DCC read/write plugin

9) choose the first partition(MIBIB) firmware that you copied and downloaded.

10) enter the first block of memory address and the last block of memory address for this partition.

11) make sure settings by code is clicked with "write to NAND"

12) start writting to the first partition block of memory on the device.

13) repeat until all partitions are written to the bricked device.

14) disconnect and hope it works! in my case I finally had a solid SAMSUNG logo, then booted into android

15) sit back and realize next time you decide to zero out all partitions make sure you have the proper firmware so you dont have to go out and buy another of the same phone to get things working!

HTC MyTouch 4G(GLACIER) Qualcomm HS-USB QDLoader 9008 USB JTAG ByPass Signature"!"

2013-02-10T00:00:00.000-08:00

So instead of going and buying a new shinny phone for $500 bucks, I've been fascinated with buying completely HARD bricked boards/phones in hopes to restore them! kind of like Frankenstein or something..:

Anyways the below post is a work in progress on how to get Qualcomm HS-USB QDLoader 9008
loaded and running with triggering the TP(test point) to initiate the USB JTAG.

1) start device manager so you can see the hardware activity, then start JTAG Manager

2)Disable the eMMC to bring up Qualcomm HS-USB QDLoader 9008
NOTE: if your phone is still in good condition you can plugin the usb and Qualcomm HS-USB QDLoader 9008
will appear in the Port COM & LPT section. if not then you need to use a needle to ground one of the chips to disable the eMMC to allow Qualcomm HS-USB QDLoader 9008. The below image is what worked for me:

Once you ground the right capacitor/chip then you should see Qualcomm HS-USB QDLoader 9008

in Device Manager.

as you can see the module is loaded, but we have a warning.

NOTE: with windows 7 there is a signature verification mechanism that will prevent drives such as the above from engaging.

3) google search showed lots of hacks for this, but the best and easiest was to reboot and press F8

once in the bootloader option choose: Disable Driver Signature Enforcement so

Qualcomm HS-USB QDLoader 9008 can run.

4) once rebooted load JTAG Manager and preform the above TP to bring up Qualcomm HS-USB QDLoader 9008.

and JTAG Manager running the resurrection for USB:

A simple reboot option makes life that much easier.

7) This is yet to be written..

NOTE: as you can see in the above picture I have another warning with USB either RIFF is unable to read the eMMC on this device or the chp is damaged(which will need to be replaced) Also a direct JTAG I will try but from what I read as well the fuse has been disabled on these devices not having the capability.

(which gives me nothing but a cool looking paper weight on my desk).

Better Battery Life on Android: build.prop

2013-02-03T23:07:00.000-08:00

The below post is to share my findings on how to improve battery with jellybean and ICS, and possibly older versions of android.

1) root your android so that you can remount /system to make changes to build.prop
(google is your friend with rooting)

2) remount /system with read/write access to modify /system/build.prop

mount -t ext3 -o rw,remount /dev/block/devicename /system

NOTE: you are remounting the partition with /system

3) make the modifications to build.prop so you can have better battery life.

vi /system/build.prop

add the lines below:

#add pm
pm.sleep_mode = 1
ro.ril.disable.power.collapse = 0
wifi.supplicant_scan_interval = 180

4) reboot into recovery mode and wipe cache/delvick then reboot so the settings take affect.

5) once booted up add some changes to the kernel: copy/paste to a file on the sdcard:

copy/paste

<------- cut --------->

fs.nr_open = 1053696
fs.inotify.max_queued_events = 32000
fs.inotify.max_user_instances = 256
fs.inotify.max_user_watches = 10240
fs.lease-break-time = 10
fs.file-max = 165164
kernel.threads-max = 525810
kernel.random.write_wakeup_threshold = 256
kernel.random.read_wakeup_threshold = 128
kernel.panic = 5
kernel.panic = 0
kernel.panic_on_oops = 1
kernel.sched_child_runs_first = 0
fs.lease-break-time = 10
fs.file-max = 65536
vm.dirty_ratio = 90
vm.dirty_background_ratio = 80
vm.oom_kill_allocating_task = 1
vm.overcommit_memory = 1
vm.page-cluster = 3
vm.drop_caches = 3
vm.min_free_kbytes = 4096
vm.panic_on_oom = 0
vm.dirty_expire_centisecs = 1000
vm.dirty_writeback_centisecs = 2000
vm.oom_kill_allocating_task = 0
vm.vfs_cache_pressure = 10
vm.min_free_order_shift = 4
vm.laptop_mode = 0
vm.block_dump = 0

add the entries into the kernel
NOTE: to make permanent add a file in /system/etc/init.d/10sysctl so every reboot the entries are automatically entered if not you need to perform the below command every time.

sysctl -pa somefilewiththeentries

NOTE: the above changes are only what I found, there probably is more that can be added

6) disable or uninstall power hungry apps!
gps,servers, etc.. that you don't need running all the time.

you should have better battery life than before. I noticed on my table and phone the life went from crappy 5hrs to 14hrs.

Custom Home Directory with QUOTA enabled

2013-01-28T23:55:00.002-08:00

This post will show you how to create a new user with a custom directory, then enable QUOTA on the individual.

First add the new user to the account:

with fedora 17 search for "users" and you will see the listed program. there create the new account.
then in a terminal give the user a password: sudo passwd username
(in my case my new user is 'linus')

Choose a custom location for the home directory. in my case I have a seperate AIX partition with JFS2 so might as well use it up and test QUOTA:

now that I have my location specified now to modify the new user 'linus' to have their home directory pointing to my AIX partition at: /dev/sdb2 /srv/AIX

I used usermod to make the changes:

usermod -m -d /srv/AIX/home/linus linus

verify that 'linus' is part of the workgroup:

cat /etc/passwd | grep linus

Now logout and login as the new user 'linus'

NOTE: for whatever reason(maybe missed something) I could not login due to some issue with fedora:

Could Not Update ICEauthority file /srv/AIX/home/linus/.ICEauthority

Clearly a permission issue with creating a new directory in an unknown location.
(SELinux was even in permissive mode but still!).

So the quick solution was to create the directory for .ICE since it cant connect the dots with ACL
and give it the proper group permissions:

as root: cd /srv/AIX ; mkdir home ; chown linus: home

after the quick fix, I was able to login as 'linus' with my home directory located at /srv/AIX/* on a separate partition.

Now to enable quota on the system and the user.
so logout as 'linus' and login as you the administrator.

enable quota on the filesystem:

my fstab file shows my external drive mounted every boot for the users home directory.
you can mount without rebooting with:

mount -t auto -o rw,userquota,grpquota /device/dir /mntpoint

once the filesystem and the kernel have been enabled you can start configuring the user to use QUOTA open a terminal and issue these commands:

using quotacheck -v /srv/AIX
gives you info on what is wrong.

once you see whats wrong use quotacheck to fix the issue:

quotacheck -cugv /srv/AIX

verify:

quotacheck -v
should results proper results.

now to set the soft and hard limits for the individual:

edquota linus

make the changes to the user 'linus' with how much space you want to allow them to use
(kind of like GMAIL giving 10 gig's of free storage for emails)

NOTE: the size of each block in quota is 1024k. the above numbers ended up being to much
So I ended up using:
soft = 2202010 = 2.*GiG / hard = 2203010 - 2.1GiG (1000 bites more)

then enable quota:

quotaon -a

Now that we have the user's QUOTA defined lets go test it out.. logout and login as 'linus'

then start filling a file with Zero's to make a file to exceed the soft and hard limit:

dd if=/dev/zero of=/srv/AIX/home/linus/quota_test.txt

depending on how big of a file, you should see a warning, then "Disk quota exceeded" telling you
that you have no more space.

And thats it! QUOTA support on users so you have control of the amount of data.
hope this wasn't to tough to understand.

easy CIFS(SAMBA) howto server/client connect

2013-01-02T22:42:00.000-08:00

This post is to explain how I started and connected CIFS(SAMBA) server/client

First install samba and cifs-utils(most major distros have it already preinstalled)

start and stop the server:

/etc/init.d/smb stop/start

NOTE:for more debug info in a terminal:(you may need to killall smbd before starting again)
/usr/sbin/smbd -S -F

now lets create a special location for our CIFS(SAMBA) server:

cat /etc/samba/smb.conf

[global]
security = share
log level = 3

[srv]
comment = your own personal directory for sharing.
path = /srv/samba
read only = no
public = yes

then: (copy/paste the commands, or not)

install -dv /srv/samba ; touch /srv/samba/samba.README ; echo THIS IS A SAMBA DIRECTORY YOU MOUNTED... > /srv/samba/samba.README

this will give you the directories, and a file so you can view once mounted.

there you can see the device location [srv] and the directory we are mounting /srv/samba
NOTE: [srv] and path = /srv/samba are important for the custom location.

now to mount our custom location..:

sudo /sbin/mount.cifs //10.*.*.*/srv /mnt -o sec=lanman

NOTE:replace 10.*.*.* with your IP address and also if you see

mount error913): Permissions denied
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

then you need to experiment with sec=*

lanman worked for me as well as ntlm2 may be different for you.

and the final results

df
//10.*.*.*/srv 213515520 15944524 196486404 8% /mnt

now you should have a mounted CIFS(SAMBA) directory..:

Quick start for OpenLDAP and LDAP Browser

2012-12-30T23:27:00.003-08:00

@(#) $OpenLDAP: slapd 2.4.31 $
NOTE: this is for opensuse12.2 may not work on other distros
So after doing some research I found setting up LDAP to be a kind of a beast to setup. With that in mind I decided to simplify the process so others don't have to go through the hell I did.
So to start:

using any distro make sure you have LDAP installed..:

from that point go to /etc/openldap/*
there you will find all the configuration files for LDAP. with this tutorial you will need to only look at slapd.conf

database bdb
suffix "dc=my-domain,dc=com"
checkpoint 1024 5
cachesize 10000
rootdn "cn=Manager,dc=my-domain,dc=com"

under suffix and rootdn we find our my-domain category for LDAP
(changing this requires changing the two entries and changing the *ldif entry in order for this to work)

once deciding the changes starting ldap is as easy as:
/etc/init.d/ldap start
or
sudo /usr/lib/openldap/slapd -4 -d 1

now that ldap is started you can now add entries in your LDAP
create a new *.ldif entry file:
vi newentry.ldif

with the image to the left you can see what I did in order for LDAP to work properly.
after you have tweaked your *.ldif you can go ahead and add it into your DB with this:

ldapadd -x -W -D "cn=Manager,dc=my-domain,dc=com" -f newentry.ldif

NOTE: keep in mind "my-domain" with all three entries(slapd.conf, ldapadd, and *.ldif)if one is off then things go wrong.

from there you should be able to connect to the localhost to test your server. using ldap browser you can see:

configure ldap browser:

connect to the server:

and there you have it.. a simple test of starting LDAP.
(for ssl/tls and all other stuff stay tuned..)

about:debug not about:plugins for the UAString

2012-11-01T00:45:00.002-07:00

for those who are a bit confused with about:debug working on the android ICS web browser here is a tutorial for that..(said to myself this doesnt work; but then learnt quickly).

in the _stock_ android ICS web browser enter about:debug in the url address bar

yeah I know it says about:plugins should be about:debug(figured you would figure that out once there)

now after clicking the right arrow in the top right corner click on the navigation button next to the home

button or the top right _three_ bars for the web browser settings(below)

as you can see scrolling down after settings we have more options once the about:debug is activated

now go into settings

then into debug remember its about:debug not about:plugins thats for firefox!

then if you want to change the UAString for whatever then change it!

supposedly changing this for tether was supposed to get me free tethering but it did not work.

so just figured it would be a nice little tutorial since it seemed a bit confusing from the start.

remember its about:debug not about:plugins

MacbookPro reflow ati radeon solder joints.

2012-09-18T15:16:00.001-07:00

So I was watching magnum PI and my machine started to show garble on the screen, then would crash all of a sudden. upon reboot I could see EFI but was garbled. after that the machine was useless.

Looking at the image to the left of this, you can see what I mean. After doing a bit of googling I had noticed that this was not an uncommon thing to have happen with apple products as well as other brands.
two procedures can fix this issue:
1) re-balling the solder joints and/or
2) stick the board in the oven at 200`F for 8min both will reflow the solder either way you put it.

ultimately I wanted to leave the reballing for later on, since that requires a heat gun capable of 300`F or more and some other tools I dont have, nor care to go out and buy at this time(would rather spend that money on gas since gas prices are so high in Los Angeles!).

so here is what I ended up doing.

1) open the macbook pro and remove the mother board.. yes the motherboard. underneath you will see four squares one of those should say ATI or NVIDIA depending on the graphics card you have.

2) either you have access to an oven, if so google "graphics oven reflow" and you will have your info.
in my case I dont have an oven but I do have a soldering iron; so I ended up improvising.. the image below should say enough on how to do it:

as you can see I just took a fuse cover from the fuse package(because its metal and can act as a heat sink, and covers the chip)which was in the glovebox of my vehicle. put the heat sink on top of the chip, then rested the solder iron tip on top, then waited about 10-15min. probably should of counted to 8min like the oven process, but figured might as well heat it up as much as possible.

After heating up the chip and hoping the chip solder joints reflowed(they just need to turn liquid a bit then back to solid form, so that any disconnected solder joints, rejoin themselves(in my case it was over 100`F in Los Angeles, as well as watching movies, a combination heated up the joints to causing them to break)). it was time to put the board back and turn the machine on.. as you can see to the right ubuntu is up and running, and no more garbled screen..

hope this helps anybody who is experiencing the same with their machine.

zeepad: Rename Phone.apk/TelephonyProvider.apk without rooting your android..:

2012-04-24T21:24:00.001-07:00

So I went and bought one of those Zeepad's 7.1 for $60 bucks from amazon.. first and foremost its no ipad but it does the job.. secondly after playing around with the machine, I was unable to install google play on it(or I did, but to have a functioning app market a no go..) so how do you install the apps? just find the *.apk package and email it to yourself and install within the email client..(for those who have no idea..)
thirdly:

as for rooting the machine I looked into all the rooting type apps, and _none_ of them worked.. something tells me they plugged up the holes in the kernel with patches.. So how to access the machine to rename those files to extend the battery life? simple! just open up the machine take the microSD out and mount it in your laptop running any linux distro.. then locate system/app/*apk and rename what you want(as well as any other file in the file system of the great and all mighty _android_ system..)
no need for adb, no need for some root app.. and hopefully this reduces the power consumption from standby cell on a tablet that does not send and receive phone calls(but can do 3g data!!)

Dual boot windows 8 and Linux on a Lenovo Z series laptop

2012-02-07T08:55:00.000-08:00

This is a quick tutorial on how to dual boot windows 8 and Linux(any distro) on a Lenovo laptop.

step 1: boot up windows 8 and load OneKey Rescue app; there you should see something about making a DEFAULT copy of windows 8(thats your live reinstall CD, rather than getting a copy when you buy the product). after burning(2 dvd's) then you are ready for the partitioning and installing of both systems.

step 2: after spending hours on end creating a partition for linux and windows(logic volume etc... with fdisk), I realized that windows Onekey rescue will always _delete and use the entire disk for a reinstall_ removing any partition scheme I created(is this right? guess windows is a control freak, but is also a nice control freak in this case).
So how to dual boot? simple! first watch this video on how to resize your partition..:
http://www.youtube.com/watch?v=utUvkgKAtKU

once you figure that _easy_ step resize and allocate what you need(remember to create a swap! even though its not really necessary according to some reports). then after re-sizing, load the live Linux distro and go through the installer.

step 3: reboot after installing and you should see grub/lilo with both options to boot either windows or linux..

step 4: enjoy the world of windows 8 and the world of linux without having to go through brain surgery to do so...:

Rebootless System with kexec

2011-11-19T22:00:00.000-08:00

Well after messing around with kexec tools on ubuntu/opensuse I finally realized how to get things functioning properly(or at least get working).

Below is the command that gets kexec working on opensuse and ubuntu over here..:

sudo /sbin/kexec -d -f --reset-vga /boot/vmlinuz-3.2.0-rc2-00141-gd633180 --append="root=UUID=c46f8c15-0a19-4694-998e-4c64eb836cab ro debug splash=off security=selinux selinux=1 vt.handoff=7" --initrd=/boot/initrd.img-3.2.0-rc2-00141-gd633180

seems the "-f" switch is what makes all the difference, basically disables the whole /etc/init.d/* reboot script (which seems to be the problem with getting this working properly).

a bug report in of which seems to be hitting the same that I was hitting; with a similar resolution..:

https://bugs.launchpad.net/linuxmint/+bug/741083

Hopefully the above command works for you as it does over here!!

Note: make sure the .config has this:

KEXEC [=y]

I guess optional Kconfig's

CRASH_DUMP [=y]

KEXEC_JUMP [=y]

LFS on QEMU

2011-09-13T08:17:00.000-07:00

well, I wanted to keep my blog free of any technology related stories, but given things change I might as well..
Anyways I have gone through various tutorials on how to get qemu running with a fresh build of LFS, but most of the tutorials are either toooo difficult for the normal person, and/are just wrong(wrong possibly due to different hardware types etc..). Hopefully this helps with anybody wanting to try this!

20gig's to work with(change to what you want in size)
dd if=/dev/zero of=lfs.img bs=20480000 count=1000

create a partition on the main device /dev/loop0 (sda):
losetup /dev/loop0 lfs.img
fdisk -C1000 -S63 -H16 lfs.img

Device Boot      Start         End      Blocks   Id System
lfs.img1   *        2048    39999999    19998976   83 Linux

now multiply for the offset:
2048 * 512 = 1048576   (use fdisk to find the number:(hint) under "start")

mount /dev/loop1 (sda1) and ready it for booting:
losetup -o 1048576 /dev/loop1 lfs.img
mkfs.ext4 /dev/loop1
tune2fs -c 0 /dev/loop1 2>&1 > /dev/null
mount /dev/loop1 /mnt

unpackage and/or create an OS for /mnt (in my case I use a tar ball of an old LFS I built a while ago).

adjust lilo.conf to point to the devices.
my lilo.conf:

menu-scheme=Gb:Gk:Gb:Yb
menu-title= "LFS with qemu"
prompt
timeout=30
install = menu

#/dev/loop0 is the main disk(sda) /dev/loop1 is the partition(sda1)

boot=/dev/loop0
disk=/dev/loop0
bios=0x80
sectors=40
heads=226
cylinders=4424
partition=/dev/loop1
start=     2048    #use fdisk to find this.

map=/boot/map
vga=normal
image=/boot/vmlinuz-2.6.35-rc6-00191-ga2dccdb
root=/dev/sda1
label=Linux
read-only
append="selinux=0 enforcing=0"

run lilo, and unmount everything, then run qemu with the new system:

lilo -r /mnt -C /etc/lilo.conf -v 3 -L
umount /mnt
losetup -d /dev/loop1
losetup -d /dev/loop0
/usr/bin/qemu-system-x86_64 -hda lfs.img

you should see the lilo menu, and you should boot all the way to login.
(if you see L and a bunch of numbers, the start= 2048 is wrong in lilo.conf. or if you boot all the way, and some panic occurs, make sure the disk is pointing to the right partition)

references:(had to adjust things here and there for my system to work but theirs provided usefule info)
http://www.linuxfromscratch.org/pipermail/lfs-dev/2008-April/061257.html
http://download.ehas.org/misc/image_generation.txt

anyways hope this helps anybody wanting to try this!

token ID: JFTZVNF3GT9M