(Original: https://developer.apple.com/library/archive/documentation/DeviceDrivers/Conceptual/IOKitFundamentals/art/registryexplorer.jpg)

If you are using the IORegistryExplorer on macOS while working with Dark Mode enabled, then you probably noticed the fact that the explorer doesn't fit the new DarkMode, hence the UI might look like that:

Which makes it hard to read the content.

I found this article which shows you how to disable DarkMode for a specific application.

Running the following commands:

osascript -e 'id of app "IORegistryExplorer"'

defaults write com.apple.IORegistryExplorer NSRequiresAquaSystemAppearance -bool yes

Solves the issue and make it usable again.

Have fun,

Tal Kain

This executable should be:

• Run as a standalone, even if there is no Python installed on the machine

• Have all required packages even if they aren't installed on the machine

• Doesn't have extra packages in it (just what needed)

• Single executable (no shared objects)

In order to achieve this task I used PyInstaller, which according to this StackOverflow post seems like one of the good options to do so.

I also wanted to make my environment as clean as possible, so I installed virtualenv:

virtualenv ipython_env/

source ipython_env/bin/activate

Inside this environment, I ran the following pip command to install PyInstaller and IPython:

pip install pyinstaller

pip install ipython 

Following rgtk's comment: https://stackoverflow.com/a/31412509/132847, I created the following script:

from IPython import embed

embed()

and called it ipython.py

From PyInstaller help manual:

  -F, --onefile         Create a one-file bundled executable.

then, I ran:

pyinstaller -F ipython.py

The output of the script generated a (default named) directory called dist which the output binary was in.

$ file dist/ipython

dist/ipython: Mach-O 64-bit executable x86_64

$ otool -L ./dist/ipython

./dist/ipython:

/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1226.10.1)

/usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.5)

Running this binary:

$ ./dist/ipython

Python 2.7.16 (default, Apr 12 2019, 15:32:40)

Type "copyright", "credits" or "license" for more information.

IPython 5.8.0 -- An enhanced Interactive Python.

?         -> Introduction and overview of IPython's features.

%quickref -> Quick reference.

help      -> Python's own help system.

object?   -> Details about 'object', use 'object??' for extra details.

In [1]:

That's all for today.

Good luck!

Tal Kain

In order to do so, I wanted to use the famous lsof, so I downloaded the package from saurik's packages using Cydia, only to find out that the given lsof is not working on my device:

lsof: PID 40 information error: Cannot allocate memory

lsof: PID 39 information error: Cannot allocate memory

lsof: PID 38 information error: Cannot allocate memory

lsof: PID 37 information error: Cannot allocate memory

lsof: PID 36 information error: Cannot allocate memory

lsof: PID 35 information error: Cannot allocate memory

lsof: PID 33 information error: Cannot allocate memory

lsof: PID 31 information error: Cannot allocate memory After trying to figure it out, I decided that it would be the best to just try compiling my own lsof and use it on the device (I also liked the challenge :-) )

So my journey begins with downloading lsof from Apple's site:

http://opensource.apple.com/tarballs/lsof/lsof-53.tar.gz

(tarballs can be found here: http://opensource.apple.com/tarballs/ while sources can be found here: https://opensource.apple.com/source/lsof/ )

$ wget http://opensource.apple.com/tarballs/lsof/lsof-53.tar.gz

$ tar zvxf lsof-53.tar.gz

Inside the lsof directory, I ran the configure script:

$ ./Configure darwin

I tried several ways to compile the executable for my iPhone, and while doing so I got several errors.

Using Gregory Pakosz post from here, running:

make CC="$(xcrun --sdk iphoneos --find clang) -isysroot $(xcrun --sdk iphoneos --show-sdk-path) -arch armv7 -arch armv7s -arch arm64"

Gave me a list of errors of missing headers, for example:

In file included from usage.c:39:

In file included from ./lsof.h:195:

./dlsof.h:56:10: fatal error: 'netinet/tcp_fsm.h' file not found                          

#include <netinet/tcp_fsm.h>                                                            

         ^                                                                              

1 error generated.

make: *** [usage.o] Error 1

What I did, following this answer was to create my own copy of the SDK headers' folder, adding the missing headers from /usr/include:

• netinet/tcp_fsm.h

• rpc/pmap_prot.h

• libproc.h

• sys/proc_info.h

• sys/kern_control.h

• net/route.h

I also looked at emonti's missing_headers folder from here to make sure I am on the right track.

Trying to build it again gave me the following linking error:

ld: library not found for -lcrt1.3.1.o

clang: error: linker command failed with exit code 1 (use -v to see invocation)

Then, I found this SO post that used the -miphoneos-version-min flag, so I added the flag:

$ make CC="$(xcrun --sdk iphoneos --find clang) -isysroot /Users/talkain/tmp/sdk -arch armv7 -arch armv7s -arch arm64 -miphoneos-version-min=8.1"

Which successfully created the executable.

$ otool -L lsof

lsof (architecture armv7):

        /usr/lib/libncurses.5.4.dylib (compatibility version 5.4.0, current version 5.4.0)

        /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1238.0.0)

lsof (architecture armv7s):

        /usr/lib/libncurses.5.4.dylib (compatibility version 5.4.0, current version 5.4.0)

/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1238.0.0)

lsof (architecture arm64):                                                                                                                            /usr/lib/libncurses.5.4.dylib (compatibility version 5.4.0, current version 5.4.0)                                     /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1238.0.0)

$ file lsof

lsof: Mach-O universal binary with 3 architectures

lsof (for architecture armv7):  Mach-O executable arm

lsof (for architecture armv7s): Mach-O executable arm

lsof (for architecture arm64):  Mach-O 64-bit executable 

Trying to run it on the phone - SUCCESS! The binary was successfully compiled on my OSX for the iOS 8.1 which I was using (Note that the bare minimum I managed to compile it with no errors was to iOS 6.0) and I managed to get the output I was looking for.

That was fun.

Credits goes to:

• webie for his Q&A: http://stackoverflow.com/a/10118021/132847

• Gregory Pakosz for his great post: https://coderwall.com/p/heonhw/compiling-for-ios-outside-of-xcode-with-xcrun

• CRDave for his answer: http://stackoverflow.com/a/21422550/132847

• http://stackoverflow.com/a/6719314/132847

• https://github.com/emonti/iOS_app_re_tools/tree/master/missing_headers

• HaneTV for http://stackoverflow.com/q/15475302/132847

Till next time,

Tal Kain

All credits goes to Yann-R that wrote here the following command:

brew cask install `brew cask list`

Works perfectly.

Note that you can also run this command using the -f flag (force) to force installation.

- Tal Kain

I had an ipython console open inside a tmux split window and I needed to save the history to a file.

Since I can't copy & paste the data from the screen (I was in the middle of a session, so the next time I would configure it correctly using this).

So I found a quicker way, Roberto Z wrote in his comment that in order to save the session's history you can use readline package:

import readline

readline.write_history_file('/home/user/current_history')

This works like a charm in Ubuntu.

- Tal Kain

Let's assume we would like to add some rules:

iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE

I can use the comment match and add a comment to this line:

iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE -m comment --comment "SOME_COMMENT"

Now, cleaning all the relevant rules in a simple command would be:

# iptables-save | grep -v SOME_COMMENT | iptables-restore

 Probably not the best way to do it, but it's simple and fast.

Have fun,

-Tal Kain

While trying to install the machine, I started my research by reading Traun Leyden's great blog post: http://tleyden.github.io/blog/2014/10/25/cuda-6-dot-5-on-aws-gpu-instance-running-ubuntu-14-dot-04/ (you should too)

Amazon offers two types of machines that includes GPUs (https://aws.amazon.com/ec2/instance-types/#g2)

High-performance NVIDIA GPUs, each with 1,536 CUDA cores and 4GB of video memory

While writing this post, I used the g2.2xlarge machine, but you can also use the 8xlarge.

This will be quick and simple:

1. Make sure you are fully up-to-date

sudo apt-get update && sudo apt-get upgrade && sudo apt-get dist-upgrade

 When prompted, choose the "install maintainer package...."

2. Reboot the machine (so it will load the new kernel)

3. Install the kernel's header files

sudo apt-get install -y linux-image-extra-virtual linux-headers-`uname -r`

3. Configure the xorg-edgers PPA

sudo add-apt-repository ppa:xorg-edgers/ppa

sudo apt-get update 4. Install the NVIDIA's CUDA repository package

OUTPUT_PATH=/tmp/cuda.deb

wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1404/x86_64/cuda-repo-ubuntu1404_7.0-28_amd64.deb -O ${OUTPUT_PATH}

dpkg -i ${OUTPUT_PATH}

 5. Update the package manager and install CUDA

sudo apt-get update

sudo apt-get install cuda

6. Make sure everything is installed as expected:

sudo nvidia-smi -a

7. You can also compile and run a sample to make sure everything is ready:

cd /usr/local/cuda/samples/1_Utilities/deviceQuery

make

./deviceQuery

That's all!

- Tal Kain

root@ubuntu:~# modprobe carl9170

modprobe: ERROR: ../libkmod/libkmod.c:556 kmod_search_moddep() could not open moddep file '/lib/modules/3.10.17/modules.dep.bin'

and the following error:

modprobe: ERROR: ../libkmod/libkmod-index.c:821 index_mm_open() magic check fail: d00dfeed instead of b007f457

To solve this issue, run depmod

# depmod

From http://manpages.ubuntu.com/manpages/dapper/man8/depmod.8.html:

"depmod creates a list of module dependencies, by reading each module

under /lib/modules/version and determining what symbols it exports, and

what symbols it needs. By default this list is written to modules.dep

in the same directory. If filenames are given on the command line,

only those modules are examined (which is rarely useful, unless all

modules are listed)."

After running depmod, everything should work as expected.

- Tal Kain

Several weeks ago I had to compile gdb and strace to a target device that runs on ARM (EBI). 

This time, instead of cross compile it, I decided to use my favorite way - schroot environment.

I had the pleasure of using schroot on my old system - an Ubuntu 12.04 desktop, and this was the first time I needed to use it on my current Ubuntu 14.04 system.

The problem started after creating the schroot environment and trying to use it.

As you will see in the following lines, I tried getting into the environment and got an error:

root@ubuntu:/home/talkain# schroot -c ubuntu-armel -u talkain

W: line 11 [ubuntu-i386] union-type: Configuration key name ‘union-type’ is not a permitted name.

W: line 11 [ubuntu-armel] union-type: Configuration key name ‘union-type’ is not a permitted name.

W: Failed to change to directory ‘/home/talkain’: No such file or directoryI: The directory does not exist inside the chroot. Use the --directory option to run the command in a different directory.W: Falling back to directory ‘/’

(ubuntu-armel)talkain@ubuntu:/$

This was weird, since the directory exist, I had permissions to this directory (this is my home directory) and this was a fresh installation of this environment.

After digging a little bit, I found an excellent start point here:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=675189

So I looked at the schroot's config file:

/etc/schroot/chroot.d/sbuild-ubuntu-armel:

[ubuntu-armel]
description=ubuntu-armel
groups=sbuild,root
root-groups=sbuild,root
# Uncomment these lines to allow members of these groups to access
# the -source chroots directly (useful for automated updates, etc).
#source-root-users=sbuild,root
#source-root-groups=sbuild,root
type=file
profile=sbuild
union-type=overlayfs
file=/var/lib/schroot/tarballs/ubuntu-armel.tgz
So in order to solve this error, I changed the schroot configuration file's parameter called "profile" from sbuild to default as follows:

profile=sbuild
to

profile=default
and now I can get into the schroot environment with no issues.

Hope this helps,

- Tal Kain

Bluetooth: Firmware file "ath3k-1.fw" not found

According to http://wireless.kernel.org/en/users/Drivers/ath3k:

ath3k is the Linux Bluetooth driver for Atheros AR3011/AR3012 Bluetooth chipsets.

I saw this blog post: http://hexwave.blogspot.com/2011/04/debian-6-with-atheros-3011-ath3k-usb.html

So I decided to try installing the package linux-firmware which contains the relevant file (I used dlocate to

$ dlocate linux-firmware | grep -i ath3

linux-firmware: /lib/firmware/ath3k-1.fw or you can find it by searching for ath3k:

$ dlocate ath3k-1.fw

linux-firmware: /lib/firmware/ath3k-1.fw I just ran:

# apt-get install linux-firmware

and it fixed it by installing the missing firmware file.

Good luck!

Tal Kain

ImportError: No module named OpenSSL

You can install the relevant package using the following command:

$ sudo apt-get install python-openssl

 The other option would be to install the OpenSSL paackage using Python's pip:

$ sudo pip install pyOpenSSL

Good luck!

- Tal Kain

One of the first things that I do after setting a new Ubuntu system is changing the location of the Close && Minimize && Maximize buttons to be on the right side of the window.

There are several ways to achieve this, but the simplest way is using the command line, running the following command:

gsettings set org.gnome.desktop.wm.preferences button-layout ":minimize,maximize,close"

(The colon should be on the left side of the string)

-Tal

As a first step I ran apt-get update, which (you guessed correctly) failed fetching several index files:

Err https://private-ppa.launchpad.net precise/main amd64 Packages            

  HttpError401

Err https://private-ppa.launchpad.net precise/main i386 Packages              

  HttpError401

Ign https://private-ppa.launchpad.net precise/main Translation-en_US          

Ign http://security.ubuntu.com trusty-security/main Translation-en_US    

.....

... 

W: Failed to fetch https://private-ppa.launchpad.net/commercial-ppa-uploaders/steam/ubuntu/dists/precise/main/binary-amd64/Packages  HttpError401

W: Failed to fetch https://private-ppa.launchpad.net/commercial-ppa-uploaders/steam/ubuntu/dists/precise/main/binary-i386/Packages  HttpError401

E: Some index files failed to download. They have been ignored, or old ones used instead.

Searching for a possible reason led me to Artūras Norkus (xeranas)'s answer: https://bugs.launchpad.net/ubuntu/+source/software-center/+bug/1130945/comments/4

It seems that my /etc/apt/auth.conf pointed to an incorrect username (a user that was used to merge between 2 different accounts that I had), changing the username to my current username fixed the problem and I was able to update the index files with no errors/warnings.

That's all,

Tal Kain

After upgrading my system from Ubuntu 12.04 to Ubuntu 14.04 I decided to check whether the latest package of virt-manager ( http://virt-manager.org/ ) comes with the feature that I needed the most - snapshots.

Unfortunately, The current Ubuntu 14.04's version of virt-manager is still based on the 0.9 branch,

So I decided to clone the latest version from the project's git repository ( https://git.fedorahosted.org/git/virt-manager.git ) and try it out.

While doing so, I had several issues which I will list here so you won't have to waste your time doing the same research.

Summary for busy people

There are several packages that you will need to install:

$ sudo apt-get install gir1.2-libvirt-glib-1.0 python-ipaddr libosinfo-bin libosinfo-1.0-dev libgtk-3-dev gir1.2-gtk-vnc-2.0 python-spice-client-gtk gir1.2-spice-client-gtk-3.0

This should fix all the following issues.

End of summary for busy people

After the git-clone's process was over, I read the README and INSTALL files, which told me that I should just run the virt-manager script ./virt-manager from the root directory. This was the result:

$:~/work/virt-manager$ ./virt-manager

ERROR:root:Could not find any typelib for LibvirtGLib

Traceback (most recent call last):

  File "./virt-manager", line 30, in <module>

    from gi.repository import LibvirtGLib

ImportError: cannot import name LibvirtGLib

According to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=736547 you should install gir1.2-libvirt-glib-1.0:

sudo apt-get install gir1.2-libvirt-glib-1.0

Running the virt-manager again will produce:

Traceback (most recent call last):

  File "./virt-manager", line 32, in <module>

    from virtinst import util as util

  File "/home/tal/work/virt-manager/virtinst/__init__.py", line 40, in <module>

    from virtinst.interface import Interface, InterfaceProtocol

  File "/home/tal/work/virt-manager/virtinst/interface.py", line 26, in <module>

    import ipaddr

ImportError: No module named ipaddr

Searching for the python package using apt-cache:

apt-cache search ipaddr

Revealed the python-ipaddr package:

sudo apt-get install python-ipaddr

Moving forward and running again:

ERROR:root:Could not find any typelib for Libosinfo

Traceback (most recent call last):

  File "./virt-manager", line 32, in <module>

    from virtinst import util as util

  File "/home/tal/work/virt-manager/virtinst/__init__.py", line 70, in <module>

    from virtinst.distroinstaller import DistroInstaller

  File "/home/tal/work/virt-manager/virtinst/distroinstaller.py", line 32, in <module>

    from virtinst import urlfetcher

  File "/home/tal/work/virt-manager/virtinst/urlfetcher.py", line 35, in <module>

    from virtinst import osdict

  File "/home/tal/work/virt-manager/virtinst/osdict.py", line 25, in <module>

    from gi.repository import Libosinfo as libosinfo

ImportError: cannot import name Libosinfo

This was solved by installing libosinfo-1.0-dev and libosinfo-bin:

sudo apt-get install libosinfo-bin libosinfo-1.0-dev

This time the application successfully ran, but pressing the "Open" button generates the following error:

Error launching details: cannot import name GtkVnc

Traceback (most recent call last):

  File "/home/tal/work/virt-manager/virtManager/engine.py", line 753, in _show_vm_helper

    details = self._get_details_dialog(uri, vm.get_connkey())

  File "/home/tal/work/virt-manager/virtManager/engine.py", line 731, in _get_details_dialog

    obj = vmmDetails(conn.get_vm(connkey))

  File "/home/tal/work/virt-manager/virtManager/details.py", line 495, in __init__

    from virtManager.console import vmmConsolePages

  File "/home/tal/work/virt-manager/virtManager/console.py", line 26, in <module>

    from gi.repository import GtkVnc

ImportError: cannot import name GtkVnc

After reading https://www.redhat.com/archives/virt-tools-list/2014-March/msg00128.html and https://www.mail-archive.com/debian-bugs-dist@lists.debian.org/msg1188846.html:

sudo apt-get install libgtk-3-dev gir1.2-gtk-vnc-2.0

Which led to the next issue:

Error launching details: cannot import name SpiceClientGtk

Traceback (most recent call last):

  File "/home/tal/work/virt-manager/virtManager/engine.py", line 753, in _show_vm_helper

    details = self._get_details_dialog(uri, vm.get_connkey())

  File "/home/tal/work/virt-manager/virtManager/engine.py", line 731, in _get_details_dialog

    obj = vmmDetails(conn.get_vm(connkey))

  File "/home/tal/work/virt-manager/virtManager/details.py", line 495, in __init__

    from virtManager.console import vmmConsolePages

  File "/home/tal/work/virt-manager/virtManager/console.py", line 27, in <module>

    from gi.repository import SpiceClientGtk

ImportError: cannot import name SpiceClientGtk

Now, I ran the following command:

sudo apt-get install python-spice-client-gtk gir1.2-spice-client-gtk-3.0

Finally! Now you should be able to easily run your new virt-manager.

NOTE: This post was written based on the master branch with a specific commit (aa823b5b58d689e2a69a9822fd462a1333d8d415) since the current state had a small issue creating new machines (I sent a patch https://www.redhat.com/archives/virt-tools-list/2014-July/msg00051.html).

Have fun!

Tal Kain

Can't start disassembler. Please set correct path in the main settings first

The solution for this error is very simple:

While installing BinDiff, the installation process asks for the installation path of IDA. The default path (on x64 machine) would be C:\Program Files (x86)\IDA\. This is not always correct.

At my case the installation was at C:\Program Files (x86)\IDA 6.3\.

Moving the path inside BinDiff will not be sufficient since part of BinDiff's files are still on the old directory.

Easiest way to fix it (or by doing it in advance): Uninstall BinDiff and install it while pointing to the correct IDA's path.

-Tal

Error: Unable to find vcvarsall.bat

Although I had Visual Studio 2010 installed, and made sure the batch file exists and the batch file creates the compilation environment correctly, I still got this error.

Googling the issue brought this question with this specific solution.

The short answer (credits to fmuecke) is to point the environment variable VS90COMNTOOLS to the value of  VS100COMNTOOLS.

SET VS90COMNTOOLS=%VS100COMNTOOLS%

This solved the issue.

-Tal

1. OCZ Vertex2 60GB

2. Western Digital Caviar Blue (1TB)

The SSD for the operating system and the Caviar Blue for all my data.

I installed my Windows 7 Ultimate edition (with no SP1), Visual Studio, Office and updated them all using Windows Update.

After doing so, I was amazed to find out that I had only 20GB left on my SSD drive – bummer…

In order to find out what is taking me so much space I used WinDirStat (http://windirstat.info) and find out that my C:\WINDOWS\Installer folder took almost 6GB of my expensive SSD space!

After searching the web for a while I found out about Mark Russinovich’s tool called "Junction"

From his site (http://technet.microsoft.com/en-us/sysinternals/bb896768):

Windows 2000 and higher supports directory symbolic links, where a directory serves as a symbolic link to another directory on the computer. For example, if the directory D:\SYMLINK specified C:\WINNT\SYSTEM32 as its target, then an application accessing D:\SYMLINK\DRIVERS would in reality be accessing C:\WINNT\SYSTEM32\DRIVERS. Directory symbolic links are known as NTFS junctions in Windows. Unfortunately, Windows comes with no tools for creating junctions—you have to purchase the Win2K Resource Kit, which comes with the linkd program for creating junctions. I therefore decided to write my own junction-creating tool: Junction.Junction not only allows you to create NTFS junctions, it allows you to see if files or directories are actually reparse points. Reparse points are the mechanism on which NTFS junctions are based, and they are used by Windows’ Remote Storage Service (RSS), as well as volume mount points.

So I have downloaded Junction, created a folder (with the correct credentials of course) on my other hard drive and pointed C:\WINDOWS\Installer to the new folder... success! Everything is working as expected (Windows Update) and I just cleared 6GB from my SSD hard drive…

You can download Junction from here: http://technet.microsoft.com/en-us/sysinternals/bb896768

(Using instructions can be found under the download page)

-Tal