Blank display on installing Kubuntu


I had a computer with Ubuntu 16.04 running the Unity desktop. I wanted to switch to Kubuntu, so I installed it using this command:

$ sudo apt install kubuntu-desktop

Everything installed fine, except this error at the end:

kde-telepathy-minimal : Depends: kde-config-telepathy-accounts (>= 0.9.0) but it is not installed

After this, there is an error about kaccounts-provider whenever I login into Unity. If I try to login into the KDE desktop, I get a blank display!


This error seems to be hitting lots of people as seen here. This seems to be caused by a conflict between packages required by Unity and KDE. Apparently, Unity (Ubuntu) and KDE (Kubuntu) cannot be installed on the same computer!

Following some clues given by other users, I forcefully removed Telepathy since I did not need it anyway:

$ sudo dpkg --remove kde-telepathy
$ sudo apt update
$ sudo apt dist-upgrade

After this the Unity error on login and the black display in KDE went away. I was able to switch to using KDE as my primary desktop.

So, if you intend to use Kubuntu, you are better off installing a Kubuntu ISO directly instead of installing it from inside Ubuntu.


ACPI_BIOS_ERROR on installing Windows 8


I had a new notebook computer with no operating system installed on it. I had a Windows 8 license and a Windows 8 ISO which I wanted to use to install on this notebook. I prepared a Windows installer USB thumbdrive using the ISO and booted the notebook from it.

The computer would throw this error and keep restarting:

Your PC ran into a problem and needs to restart.
If you'd like to know more you can search online later for this error: ACPI_BIOS_ERROR


Searching online for this error, I found that this occurred to most people when they were updating or upgrading. Mine was neither, I was installing on a fresh computer. In any case, their solutions were either inapplicable or did not work.

While looking up solutions for this error, I discovered that I could use my Windows 8 license to install the newer Windows 10. I decided to give this a try, since Windows 10 ISO is easily available for download here. I made a Windows installer USB thumbdrive and surprise, it booted up normally and I was able to install Windows 10 Pro without any further problems!

Pip uninstall error


I had installed a package using Pip. However, when I tried to uninstall it, I got this error:

$ sudo pip uninstall h5py
Can't uninstall 'h5py'. No files were found to uninstall.


You first want to make sure that you are using the pip version that matches the package. That is, if you installed a Python 2.x package, that can only be uninstalled using pip. A Python 3.x package can only be uninstalled using pip3.

In my case, I could not discover the real reason for this error, I resorted to finding the deleting the installed directory for this package. I found the directory h5py-2.6.0-py2.7.egg-info under /usr/local/lib/python2.7/dist-packages and deleted that. If this was a Python 3.x package, I would similarly have looked under the /usr/local/lib/python3.x/ directory.

Tried with: Ubuntu 14.04

Invalid version number error with Python


I tried to import a Python package that I had installed from source. The import failed with this error:

File "/usr/lib/python2.7/distutils/", line 40, in __init__
File "/usr/lib/python2.7/distutils/", line 107, in parse
  raise ValueError, "invalid version number '%s'" % vstring
ValueError: invalid version number '2.7.0rc3'


It turns out that package version number has to be in the x.y.z format. Else Python throws this error.

Since I had the source code of this package, I found all instances of 2.7.0rc3 and changed it to 2.7.0. Typically, this will be in the and files. I removed the previously installed package and reinstalled this changed source code. I was able to import after this successfully.

Tried with: Ubuntu 14.04

CMake error building with Python libraries


I got this error from CMake when building a project that needs to link with Python 3.4 libraries:

-- Found PythonInterp: /usr/bin/python3 (found suitable version "3.4.3", minimum required is "3.0")
-- Could NOT find PythonLibs (missing:  PYTHON_LIBRARIES PYTHON_INCLUDE_DIRS) (Required is at least version "3.0")


Turns out that the CMake available on my system only supported finding Python 3 packages upto version 3.3. To change it to support Python 3.4 was possible by editing two files:

  • In file /usr/share/cmake-3.4/Modules/FindPythonInterp.cmake find the line containing _PYTHON3_VERSIONS and prepend 3.4 to the versions already listed there.

  • In file /usr/share/cmake-3.4/Modules/FindPythonLibs.cmake find the line containing _PYTHON3_VERSIONS and prepend 3.4 to the versions already listed there.

I was able to build with Python 3.x libraries after that.

Error building Caffe with Python 3 support

Caffe can be built with support for Python 2.x. This allows you to invoke Caffe easily from Python code. However, I wanted to call Caffe from Python 3.x code.

  • I built Boost with Python 3.x support. I could see that libboost_python3 library files were generated.

  • I added this to the normal CMake command that I use to build Caffe: -Dpython_version=3

Sadly, this popped up errors of this type: undefined reference to `PyClass_Type'

This type of error indicates that the Python 2.x Boost library was being used to compile with Python 3.x libraries.

fish_pwd error


I updated the Fish shell using the apt command:

$ sudo apt install fish

After the update, Fish threw up this error:

fish: __fish_pwd
in command substitution
    called on standard input

in command substitution
    called on standard input


This turned out to be because the /usr/share/fish/functions/ function file it was looking for could not be found! The apt update had screwed up and had deleted some essential configuration files of Fish.

I removed and reinstalled Fish shell:

$ sudo apt purge fish
$ sudo apt install fish

I noticed that the above function file was correctly installed and available now. Fish worked fine after this 🙂

Tried with: Fish 2.4.0 and Ubuntu 14.04

Selenium error on geckodriver


I updated Ubuntu packages, which included Firefox and I updated Selenium using pip3 cause that also depends on the Firefox version. But running my existing Python scripts that use Selenium popped this error:

FileNotFoundError: [Errno 2] No such file or directory: 'geckodriver'


Firefox now provides the geckodriver as a separate binary. You can download the version matching your OS and CPU here. Unzip the file and place the binary anywhere that is in your PATH. Your Python scripts should work now.


Tried with: Firefox 49, Selenium 3.0.1 and Ubuntu 16.04

Out of memory error with py-faster-rcnn


When training a Faster-RCNN model using py-faster-rcnn I noticed that it would randomly crash with this error:

out of memory
invalid argument
an illegal memory access was encountered
F0919 18:01:51.657281 21310] Check failed: error == cudaSuccess (77 vs. 0)  an illegal memory access was encountered
*** Check failure stack trace: ***


On closer investigation, I found that a training process would crash when I ran a second or third training on the same GPU. How can running a second training kill the first training? The only scenario I could think of, based on the above error message, was that a cudaMalloc to get more memory was failing. But why was Caffe trying to get more memory in the middle of training. Should not the reshape all be done and finished in the beginning of training?

Anyway, the first problem was to reliably reproduce the error, since it did not always crash on running a second training. It only crashed once in a few times in these scenarios. Since I suspected cudaMalloc, I wrote a small adversary CUDA program that would try to grab as much GPU memory as possible. I ran this program a while after training had started and it reliably crashed the training everytime!

A core dump file was being generated on crash, but it was at first useless since I was running Caffe compiled in release mode with no debugging symbols. I recompiled Caffe in debug mode with debugging symbols and used that to open in GDB:

$ gdb /usr/bin/python core

After the core dump was loaded in GDB, I got its backtrace using bt. It was interesting, but did not point to anything suspicious.

I next monitored GPU memory occupied by the training continuously, using this watch and nvidia-smi:

$ watch -n 0.1 nvidia-smi

I noticed that the GPU memory used by training incremented and decremented by around 18MB consistently all the time. If my adversary CUDA program went and grabbed the 18MB that was released, then the training would crash when it tried to alloc that same memory next time.

So, who is allocating and releasing memory all the time in py-faster-rcnn? Since I had ported the proposal layer recently from Python to C++, I remembered NMS. There is both a CPU and GPU version of NMS in py-faster-rcnn. The GPU NMS is used by default, though this can be changed in By switching it to CPU, I found that the crash no longer happened.

But the problem is that CPU NMS at 0.2s was 10 times slower than the GPU NMS at 0.02s for my setup.


Once I saw that the GPU NMS code in lib/nms/ was doing the cudaMalloc and cudaFree continuously, fixing it was easy. I changed the allocated memory pointers to static and changed the code to hold on to the memory allocated last time. Only if more memory was required then the old one would be freed and new larger one allocated. I basically used the same strategy used by std:vector, that is doubling of memory. A better solution would be to allocate the maximum required memory, based on the box numbers set in, and use it during training.

Tried with: CUDA 7.5 and Ubuntu 14.04

Kubuntu install stuck with unmet dependencies


I tried to install Kubuntu on an existing Ubuntu system using this command:

$ sudo apt install kubuntu-desktop

And I got this package dependency error:

You might want to run 'apt-get -f install' to correct these:
The following packages have unmet dependencies:
 kde-telepathy-minimal : Depends: kde-config-telepathy-accounts (>= 15.04.0) but it is not going to be installed
 unity-scope-gdrive : Depends: account-plugin-google but it is not going to be installed
E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution).

However, running sudo apt-get -f install would stop with the same dependency problem.


The key here is to realize that apt itself cannot resolve this cyclic dependency. So, to fix it we need to use a lower-level tool to explicitly take out the offending package. We can do that by using dpkg:

$ sudo dpkg --purge unity-scope-gdrive
$ sudo dpkg --purge account-plugin-google
$ sudo apt-get -f install

Tried with: Ubuntu 15.10