Archive

Posts Tagged ‘Debian’

Big distributions, little RAM 9

November 28th, 2015 2 comments

It’s been a while but once again here is the latest instalment of the series of posts where I install the major, full desktop, distributions into a limited hardware machine and report on how they perform. Once again, and like before, I’ve decided to re-run my previous tests this time using the following distributions:

  • Debian 8.2 (Cinnamon)
  • Debian 8.2 (GNOME)
  • Debian 8.2 (KDE)
  • Debian 8.2 (MATE)
  • Debian 8.2 (Xfce)
  • Elementary OS 0.3.1 (Freya)
  • Kubuntu 15.10 (KDE)
  • Linux Mint 17.2 (Cinnamon)
  • Linux Mint 17.2 (MATE)
  • Linux Mint 17.2 (Xfce)
  • Mageia 5 (GNOME)
  • Mageia 5 (KDE)
  • Ubuntu 15.10 (Unity)
  • Xubuntu 15.10 (Xfce)

I also attempted to try and install Fedora 23, Linux Mint 17.2 (KDE) and OpenSUSE 42.1 but none of them were able to complete installation.

All of the tests were done within VirtualBox on ‘machines’ with the following specifications:

  • Total RAM: 512MB
  • Hard drive: 10GB
  • CPU type: x86 with PAE/NX
  • Graphics: 3D Acceleration enabled

The tests were all done using VirtualBox 5, and I did not install VirtualBox tools (although some distributions may have shipped with them). I also left the screen resolution at the default (whatever the distribution chose) and accepted the installation defaults. All tests were run prior to December 2015 so your results may not be identical.

Results

Just as before I have compiled a series of bar graphs to show you how each installation stacks up against one another. Measurements were taken using the free -m command for memory and the df -h command for disk usage.

Like before I have provided the results file as a download so you can see exactly what the numbers were or create your own custom comparisons (see below for link).

Things to know before looking at the graphs

First off if your distribution of choice didn’t appear in the list above its probably not reasonably possible to be installed (i.e. I don’t have hours to compile Gentoo) or I didn’t feel it was mainstream enough (pretty much anything with LXDE). As always feel free to run your own tests and link them in the comments for everyone to see.

Quick Info

  • Out of the Cinnamon desktops tested Debian 8.2 had the lowest memory footprint
  • Out of the GNOME desktops tested Mageia 5 had the lowest memory footprint
  • Out of the KDE desktops tested Mageia 5 had the lowest memory footprint
  • Out of the Xfce desktops tested Debian 8.2 had the lowest memory footprint
  • Out of the MATE desktops tested Debian 8.2 had the lowest memory footprint
  • Elementary OS 0.3.1 had the highest memory footprint of those tested
  • Debian 8.2 Xfce and MATE tied for the lowest memory footprint of those tested
  • Debian 8.2 Xfce had the lowest install size of those tested
  • Kubuntu 15.10 had the largest install size of those tested
  • Elementary OS 0.3.1 had the lowest change after updates (+2MiB)
  • Mageia 5 KDE had the largest change after updates (-265MiB)

First boot memory (RAM) usage

This test was measured on the first startup after finishing a fresh install.

big_distro_little_ram_9_first_boot_memory_usageMemory (RAM) usage after updates

This test was performed after all updates were installed and a reboot was performed.

big_distro_little_ram_9_memory_usage_after_updatesMemory (RAM) usage change after updates

The net growth or decline in RAM usage after applying all of the updates.

big_distro_little_ram_9_memory_usage_change_after_updatesInstall size after updates

The hard drive space used by the distribution after applying all of the updates.

big_distro_little_ram_9_install_sizeConclusion

Once again I will leave the conclusions to you. Source data provided below.

Source Data




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Big distributions, little RAM 8

July 11th, 2015 2 comments

It’s been a while but once again here is the latest instalment of the series of posts where I install the major, full desktop, distributions into a limited hardware machine and report on how they perform. Once again, and like before, I’ve decided to re-run my previous tests this time using the following distributions:

  • Debian 8 (Cinnamon)
  • Debian 8 (GNOME)
  • Debian 8 (KDE)
  • Debian 8 (MATE)
  • Debian 8 (Xfce)
  • Elementary OS 0.3 (Freya)
  • Kubuntu 15.04 (KDE)
  • Linux Mint 17.1 (Cinnamon)
  • Linux Mint 17.1 (KDE)
  • Linux Mint 17.1 (MATE)
  • Linux Mint 17.1 (Xfce)
  • Mageia 4.1 (GNOME)
  • Mageia 4.1 (KDE)
  • OpenSUSE 13.2 (GNOME)
  • OpenSUSE 13.2 (KDE)
  • Ubuntu 15.04 (Unity)
  • Ubuntu Mate (MATE)
  • Xubuntu 15.04 (Xfce)

I also attempted to try and install Fedora 21 and Linux Mint 17.2 (KDE) but it just wouldn’t go.

All of the tests were done within VirtualBox on ‘machines’ with the following specifications:

  • Total RAM: 512MB
  • Hard drive: 8GB
  • CPU type: x86 with PAE/NX
  • Graphics: 3D Acceleration enabled

The tests were all done using VirtualBox 4.3.30, and I did not install VirtualBox tools (although some distributions may have shipped with them). I also left the screen resolution at the default (whatever the distribution chose) and accepted the installation defaults. All tests were run prior to June 2015 so your results may not be identical.

Results

Just as before I have compiled a series of bar graphs to show you how each installation stacks up against one another. Measurements were taken using the free -m command for memory and the df -h command for disk usage.

Like before I have provided the results file as a download so you can see exactly what the numbers were or create your own custom comparisons (see below for link).

Things to know before looking at the graphs

First off if your distribution of choice didn’t appear in the list above its probably not reasonably possible to be installed (i.e. I don’t have hours to compile Gentoo) or I didn’t feel it was mainstream enough (pretty much anything with LXDE). As always feel free to run your own tests and link them in the comments for everyone to see.

First boot memory (RAM) usage

This test was measured on the first startup after finishing a fresh install.

First_Boot_AllMemory (RAM) usage after updates

This test was performed after all updates were installed and a reboot was performed.

After_Updates_AllMemory (RAM) usage change after updates

The net growth or decline in RAM usage after applying all of the updates.

Usage_Changes_AllInstall size after updates

The hard drive space used by the distribution after applying all of the updates.

Install_SizeConclusion

Once again I will leave the conclusions to you. Source data provided below.

Source Data




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Big distributions, little RAM 7

October 13th, 2014 4 comments

It’s been a while but once again here is the latest instalment of the series of posts where I install the major, full desktop, distributions into a limited hardware machine and report on how they perform. Once again, and like before, I’ve decided to re-run my previous tests this time using the following distributions:

  • Debian 7.6 (GNOME)
  • Elementary OS 0.2 (Luna)
  • Fedora 20 (GNOME)
  • Kubuntu 14.04 (KDE)
  • Linux Mint 17 (Cinnamon)
  • Linux Mint 17 (MATE)
  • Mageia 4.1 (GNOME)
  • Mageia 4.1 (KDE)
  • OpenSUSE 13.1 (GNOME)
  • OpenSUSE 13.1 (KDE)
  • Ubuntu 14.04 (Unity)
  • Xubuntu 14.04 (Xfce)

I also attempted to try and install Fedora 20 (KDE) but it just wouldn’t go.

All of the tests were done within VirtualBox on ‘machines’ with the following specifications:

  • Total RAM: 512MB
  • Hard drive: 8GB
  • CPU type: x86 with PAE/NX
  • Graphics: 3D Acceleration enabled

The tests were all done using VirtualBox 4.3.12, and I did not install VirtualBox tools (although some distributions may have shipped with them). I also left the screen resolution at the default (whatever the distribution chose) and accepted the installation defaults. All tests were run between October 6th, 2014 and October 13th, 2014 so your results may not be identical.

Results

Just as before I have compiled a series of bar graphs to show you how each installation stacks up against one another. Measurements were taken using the free -m command for memory and the df -h command for disk usage.

Like before I have provided the results file as a download so you can see exactly what the numbers were or create your own custom comparisons (see below for link).

Things to know before looking at the graphs

First off if your distribution of choice didn’t appear in the list above its probably not reasonably possible to be installed (i.e. I don’t have hours to compile Gentoo) or I didn’t feel it was mainstream enough (pretty much anything with LXDE). As always feel free to run your own tests and link them in the comments for everyone to see.

First boot memory (RAM) usage

This test was measured on the first startup after finishing a fresh install.

 

All Data Points

All Data Points

RAM

RAM

Buffers/Cache

Buffers/Cache

RAM - Buffers/Cache

RAM – Buffers/Cache

Swap Usage

Swap Usage

RAM - Buffers/Cache + Swap

RAM – Buffers/Cache + Swap

Memory (RAM) usage after updates

This test was performed after all updates were installed and a reboot was performed.

All Data Points

All Data Points

RAM

RAM

Buffers/Cache

Buffers/Cache

RAM - Buffers/Cache

RAM – Buffers/Cache

Swap Usage

Swap Usage

RAM - Buffers/Cache + Swap

RAM – Buffers/Cache + Swap

Memory (RAM) usage change after updates

The net growth or decline in RAM usage after applying all of the updates.

All Data Points

All Data Points

RAM

RAM

Buffers/Cache

Buffers/Cache

RAM - Buffers/Cache

RAM – Buffers/Cache

Swap Usage

Swap Usage

RAM - Buffers/Cache + Swap

RAM – Buffers/Cache + Swap

Install size after updates

The hard drive space used by the distribution after applying all of the updates.

Install Size

Install Size

Conclusion

Once again I will leave the conclusions to you. Source data provided below.

Source Data




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

A tale of a gillion installs

January 21st, 2014 1 comment

Install number one: LMDE 201303.  I was hoping for the best of both worlds, but I got driver issues instead.  LMDE has known ATI proprietary driver install issues.  I followed the Mint instructions and got it working, then got a blank screen after too much tinkering.  I was surprised that LMDE had this problem since Debian doesn’t, and LMDE should be a more polished version of LMDE.  This wasn’t a big deal, but I decided to give Debian a chance.

Install number two: debian stable (7.3).  The debian website has a convoluted maze of installation links, but it’s still fairly easy to find an ISO for the stable version you need.  I installed from the live ISO using a USB key.  The installation and ATI driver update went smoothly, and I thought all was well at first.  I soon realized that about 50% of reboots failed; the audio driver was the culprit.  I installed the latest driver from Realtec/ALSA and it sort of worked, but I was still getting some crap from # dmesg and the audio would crackle with some files.

LMDE.  I live booted LMDE to see if the same issue existed there and it did.

Time for Mint 16.  As expected everything worked.  Man I really wish Ubuntu hadn’t chosen the dark side – their OS is really good.  All of these distros use ALSA audio drivers, so why is Ubuntu the only one that works?   Kernel versions:

debian stable (7.3):
cat /proc/asound/version
Advanced Linux Sound Architecture Driver Version 1.0.24.
Mint 16:
cat /proc/asound/version
Advanced Linux Sound Architecture Driver Version k3.11.0-12-generic.

One more thing to check.  What kernel version is the real debian testing “jessie” using:

http://packages.debian.org/testing/kernel/linux-image-3.12-1-amd64

LMDE 201303 = 3.2
debian stable 7.3 = 3.2
Mint 16 = 3.11
debian testing “jessie - Jan 2014” = 3.12!

I determined to try debian testing before settling for Mint.  I tried a netinstall from USB key which killed my PC and grub bootloader.  The debian stable live iso usb key decided to stop working as well.   I finally got a real DVD debian stable install to work, changed the repositories to point to “jessie” and upgraded.  I was very surprised to see this worked!   I’m having some problems with bash, but all of my day to day software is up and running.  Nice.

TL;DR: LMDE was using an old kernel so I needed the real debian testing (jessie) to solve my driver problems.

So many flavours – with bonus privacy rant!

January 21st, 2014 1 comment

It’s interesting reading the old Linux Experiment first posts when people were contemplating which distro to install.  It’s been 4.5 years since then and the linux world has evolved.  Most noticeable, was no one talking about Mint!

I was considering three distros for my home PC dual boot:

  1. Debian
  2. LMDE
  3. Mint

I wanted something in the debian family since it seems to be receiving, by far, the most attention.  I expect this also means it gets the most activity and updates.  Ubuntu would probably work the best out of the box, but as you probably already know:

https://en.wikipedia.org/wiki/Unity_%28user_interface%29#Privacy_controversy

Ubuntu’s privacy issues are a deal breaker of course, but they also made me question Mint.  I don’t want to support Ubuntu and I think using Mint would indirectly do that.  Also, Mint does have some minor default search engine sketchyness going on.   I realize that these developers need funding, but I don’t think selling their users’ stats or useage is the way to do it.  I think donations are the way to go and they seem to be working for Wikimedia.  Developing non-essential non-related commercial software in parallel with the OS might be another alternative… hmm, sounds like a slippery slope.

The plan was: Try LMDE first, Debian stable if more stability is needed, and Mint if I got to the point that I just wanted things to work.  Results to follow!

TL;DR:  I planned to install LMDE or Debian, since Ubuntu wants to track me.

Limit Bandwitdth Used by apt-get

October 22nd, 2012 No comments

It’s easy. Simply throw “-o Acquire::http::Dl-Limit=X” in your apt-get command where X is the kb/s you wish to limit it to. So for example let’s say that you want to limit an apt-get upgrade command to roughly 50kb/s of bandwidth. Simply issue the following command:

sudo apt-get -o Acquire::http::Dl-Limit=50 upgrade

Simple right?




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Big distributions, little RAM 4

April 9th, 2012 No comments

It’s that time again. Like before I’ve decided to re-run my previous tests this time using the following distributions:

  • Debian 6.0 (GNOME)
  • Kubuntu 11.10 (KDE)
  • Linux Mint 12 (GNOME)
  • Linux Mint 201109 LXDE (GNOME)
  • Mandriva 2011 (KDE)
  • OpenSUSE 12.1 (GNOME)
  • OpenSUSE 12.1 (KDE)
  • Sabayon 8 (GNOME)
  • Sabayon 8 (KDE)
  • Sabayon 8 (Xfce)
  • Ubuntu 11.10 (Unity)
  • Ubuntu 12.04 Beta 2 (Unity)
  • Xubuntu 11.10 (Xfce)

I will be testing all of this within VirtualBox on ‘machines’ with the following specifications:

  • Total RAM: 512MB
  • Hard drive: 8GB
  • CPU type: x86 with PAE/NX
  • Graphics: 3D Acceleration enabled

The tests were all done using VirtualBox 4.1.0 on Windows 7, and I did not install VirtualBox tools (although some distributions may have shipped with them). I also left the screen resolution at the default (whatever the distribution chose) and accepted the installation defaults. All tests were run between April 2nd, 2012 and April 9th, 2012 so your results may not be identical.

Results

Following in the tradition of my previous posts I have once again gone through the effort to bring you nothing but the most state of the art in picture graphs for your enjoyment.

Things to know before looking at the graphs

First off if your distribution of choice didn’t appear in the list above its probably not reasonably possible to installed (i.e. Fedora 16 which requires 768MB of RAM) or I didn’t feel it was mainstream enough (pretty much anything with LXDE). Secondly there may be some distributions that don’t appear on all of the graphs, for example Mandriva. In the case of Mandriva the distribution would not allow me to successfully install the updates and so I only have its first boot RAM usage available. Finally when I tested Debian I was unable to test before / after applying updates because it seemed to have applied the updates during install. As always feel free to run your own tests.

First boot memory (RAM) usage

This test was measured on the first startup after finishing a fresh install.

Memory (RAM) usage after updates

This test was performed after all updates were installed and a reboot was performed.

Memory (RAM) usage change after updates

The net growth or decline in RAM usage after applying all of the updates.

Install size after updates

The hard drive space used by the distribution after applying all of the updates.

Conclusion

As before I’m going to leave you to drawing your own conclusions.




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Big distributions, little RAM 3

August 14th, 2011 2 comments

Once again I’ve decided to re-run my previous tests this time using the following distributions:

  • Debian 6.0.2 (GNOME)
  • Fedora 15 (GNOME 3 Fallback Mode)
  • Fedora 15 (KDE)
  • Kubuntu 11.04 (KDE)
  • Linux Mint 11 (GNOME)
  • Linux Mint 10 (KDE)
  • Linux Mint 10 (LXDE)
  • Linux Mint 11 (Xfce)
  • Lubuntu 11.04 (LXDE)
  • Mandriva One (GNOME)
  • Mandriva One (KDE)
  • OpenSUSE 11.4 (GNOME)
  • OpenSUSE 11.4 (KDE)
  • Ubuntu 11.04 (GNOME Unity Fallback Mode)
  • Xubuntu 11.04 (Xfce)

I will be testing all of this within VirtualBox on ‘machines’ with the following specifications:

  • Total RAM: 512MB
  • Hard drive: 8GB
  • CPU type: x86

The tests were all done using VirtualBox 4.0.6 on Linux Mint 11, and I did not install VirtualBox tools (although some distributions may have shipped with them). I also left the screen resolution at the default 800×600 and accepted the installation defaults. All tests were run on August 14th, 2011 so your results may not be identical.

Results

Following in the tradition of my previous posts I have once again gone through the effort to bring you nothing but the most state of the art in picture graphs for your enjoyment.

Things to know before looking at the graphs

First off none of the Fedora 15 versions would install in 512MB of RAM. They both required a minimum of 640MB and therefore are disqualified from this little experiment. I did however run them in VirtualBox with 640MB of RAM just for comparison purposes. Secondly the Linux Mint 10 KDE distro would not even install in either 512MB or 640MB of RAM, the installer just kept crashing. I was unable to actually get it to work so it was not included in these tests. Finally when I tested Debian I was unable to test before / after applying updates because it seemed to have applied the updates during install.

First boot memory (RAM) usage

This test was measured on the first startup after finishing a fresh install.

Memory (RAM) usage after updates

This test was performed after all updates were installed and a reboot was performed.

Memory (RAM) usage change after updates

The net growth or decline in RAM usage after applying all of the updates.

Install size after updates

The hard drive space used by the distribution after applying all of the updates.

Conclusion

As before I’m going to leave you to drawing your own conclusions.




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

One week, three distributions (Day 7: Conclusions)

October 24th, 2010 No comments

Well it’s been an interesting week. I’ve gotten to try out three new distributions and share my thoughts with everyone here. My original goal was to see which one of these distributions offered the best first impression and declare that one as the ‘winner’. However in actually working my way through these great releases I have changed my mind somewhat.

What makes a great distribution great?

This is a very interesting question that I’m sure would generate a wide array of unique and passionate responses. Some prefer ease of use, while others demand nothing less than complete control over what they can tweak. There are people who swear by using nothing but open source solutions, while others are happy to add proprietary code into the mix as well. This is the great thing about Linux, we get so many choices which means we get to decided what we want.

Unfortunately this has also resulted in a bit of distribution zealotry; like choosing Ubuntu over Fedora, or Arch over OpenSUSE is somehow taking a side in some giant war. Instead of all of the infighting we should be celebrating the fact that when Ubuntu comes out with a new piece of user-friendly software, or Fedora introduces a new awesome technology, we can share and integrate it right into all distributions.

So what makes a great distribution great? A distribution is great because it works for you, it suites your needs, fits your personality and lets you do what you want to do. At the end of the day isn’t that what open source is about?

Final thoughts

OK enough of the preachy writing. I think that all of the distributions I have tested this week were very good. They each embody the spirit of open source in their own little ways.

Kubuntu 10.10

Awards: The most improved release. Most likely to recover lost KDE fans.

I was extremely impressed with this release. The folks over at the Kubuntu project deserve a huge round of applause for their continued work on this often forgotten Ubuntu sibling. This release is unlike any other that I’ve tried from Kubuntu, and I hope it marks a turning point in the distribution’s history. If the next release sees anywhere close to the improvement that this release did it may even unseat Ubuntu as the go to Linux release. If you haven’t tried out this release I urge you to give it a shot.

Ubuntu 10.0

Awards: The most refined. Most likely to be installed on a new Linux user’s computer.

Ubuntu makes a return from its last long-term support (LTS) release with this stellar offering. For a release that is meant to experiment with changes, which might eventually be incorporated into a future LTS release, this version feels as polished as ever. The new theme, font, store and integration features make this an absolutely solid release. If you’re an Ubuntu user I’m sure you have already upgraded. If you develop for a different distribution, this might still be worth looking into if only to steal the good parts for your release of choice. Either way I think this release of Ubuntu marks a whole new level of application integration on the Linux desktop and I am excited to see where they go next with it.

Linux Mint Debian Edition

Awards: The most advanced. Most likely to see the fastest improvement.

For people who have been using Debian for a while now this release will feel right at home. It combines the best parts of Debian testing, modern software, stability and thousands of packages, with the Linux Mint team’s renown ability to iron out the kinks in any Linux distribution. I think that this release will see so much improvement in the next couple of months that it has the potential to steal users away from other rolling release distributions with its easy to use desktop. While this current iteration does have some issues I hardly think that they are anything to run away from. For technical users looking for the newest stuff, while hoping avoiding the vast majority of headaches other distributions can cause, this one is for you.

My Choice

For me personally I have been very happy with Linux Mint 9 and look forward to version 10 when it ships later this year. Until then however I think I will be sticking with the one that most closely resembles my current set up. No not Linux Mint Debian Edition, but Ubuntu 10.10. That being said I do look forward to giving Fedora 14 and Linux Mint 10 a ride soon.




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

One week, three distributions (Day 6: Linux Mint Debian Edition)

October 23rd, 2010 7 comments

To round out the week I installed the newly released Linux Mint Debian Edition (LMDE). Like the other posts I have made I will walk you through my first impressions as well as the general overall feel of the distribution.

Install

LMDE’s install is something quite different from the experiences I had with Ubuntu 10.10 and Kubuntu 10.10. While it is still a very polished installer (complete with a single slide slide-show even!) it does lack a couple of features that make it far less user friendly. Where the other installers basically held your hand in every way, LMDE requires at least some technical understanding in order to complete. For instance there is no friendly auto-partition step, instead LMDE leaves the user to do it manually with GParted. While hardly the end of the world it is enough of a challenge that you could no longer just hand this disc to your non-technical friend and let them have at it. However once that step is complete the installer is very straight forward and rips through the installation in a matter of minutes.

First boot and drivers (oh my!)

I have to say that my first impression of LMDE was a mixed one. On one hand it spewed text everywhere as it booted, which I assume came from its Debian heritage. On the other hand the boot was ridiculously fast. I know it’s been one of Canonical’s goals to make Ubuntu the fastest booting Linux distribution but I have to say that a stock install of LMDE (and maybe even Debian Testing) will easily give that claim a run for its money.

Once at my desktop I was presented with a very familiar Linux Mint set up. If you were to place this desktop next to Linux Mint’s Ubuntu derivative (Linux Mint 9 for instance) I would be very hard pressed to spot any differences.

Unfortunately one thing that was glaringly missing was the lack of the Ubuntu automatic driver detection and install system jockey. Without jockey I had to resort to Fedora-esque measures in order to install the correct driver which is necessary for correct display and power management on my laptop. For reference here are the steps I took in order to install the proprietary ATI driver and setup my X configuration:

1) Run the following command in order to make sure you have the most recent package list

$ sudo apt-get update

2) You may want to now install all available updates so that we start with the most recent kernel

$ sudo apt-get dist-upgrade

3) Install the kernel headers so that we can configure it to work with the ATI driver

$ sudo apt-get install linux-headers-2.6-$(uname -r | sed ‘s,.*-,,’)

4) Install the ATI driver and control panel

$ sudo apt-get install fglrx-control fglrx-driver

5) From there just follow the instructions on my post here in order to generate the new X configuration file and maybe even fix your vsync issues at the same time.

I have read that jockey is currently being ported and will appear in the LMDE repositories so hopefully this small issue will be solved quickly.

Rolling release = LOTS OF UPDATES

One of LMDE’s big selling points is that it is a rolling release, which means that you will continue to get new packages and updates on your system without having to reinstall at 6 month intervals like some other distributions. Strangely though LMDE does not ship with update checking enabled which I found kind of weird. A quick forced check later and I discovered why the team may have made that choice

That’s right, almost 500 updates… Almost every package on my newly installed machine had to be replaced updated. This is definitely a release for people looking for the newest software but could easily lead to update fatigue for everyone else.

32-bit working system vs PAE kernel broken system

One of the unfortunate things about this release is that it only comes in a 32-bit version. As I run a 64-bit processor with 4GB of RAM it irks me to know that I am not using the full potential of my system. I starting looking into Physical Address Extension (PAE) kernels as a solution to this problem. PAE kernels, for those who don’t know, use a system of memory indirection in order to allow a 32-bit processor access to more than 4GB of mappable memory. In the case of Linux the PAE kernel can map up to 64GB of RAM.

After a bit of googling I stumbled upon instructions to install a PAE kernel by simply installing the linux-image-686-bigmem meta-package. Unfortunately this quick fix, as most often is the case, didn’t exactly turn out well and actually broke my GDM system. Without GDM I was unable to log into my desktop and this experiment came to an end. In the interest of time I decided to just reinstall instead of trying to troubleshoot how to fix what I had inadvertently broken.

Software selection

The software selection in LMDE is impressive and in many ways is what Ubuntu’s used to be. Here you will find (pre-installed) Flash, the Java and Mono runtimes, an MP3 codec and even the Gimp. Everything, like the other Linux Mint releases, is designed to make it so that the user does not have to search for solutions to missing functionality.

Because this is a rolling release it doesn’t really make sense for me to review the included software as much as it does to just mention it. For web browsing LMDE, like Ubuntu, ships with Firefox. To send and receive e-mail it calls upon Firefox’s cousin Thunderbird. Instant messaging is handled by Pidgin and your music collection is controlled by Rhythmbox. F-Spot remains as the photo manager, unlike Ubuntu which replaced it with Shotwell, while Gwibber and Totem round out the release.

Conclusion

Currently Linux Mint Debian Edition is somewhat of an enthusiast’s release. It has the potential to be a great rolling release but it’s pretty obvious that right now it needs some work to get there. For instance, why when I updated my software, did my GDM background change to a Debian one? The team over at Linux Mint knows how to polish a distribution and so I’m confident that they will do the same for Debian Edition.

Pros:

  • Rolling release which means you always have the most recent software
  • Still has that Linux Mint charm to it

Cons:

  • The distribution still needs a bit of polish before I could see myself recommending it to all but seasoned Linux users
  • Being a rolling release might result in update fatigue



I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Going Linux, Once and for All

December 23rd, 2009 7 comments

With the linux experiment coming to an end, and my Vista PC requiring a reinstall, I decided to take the leap and go all linux all the time. To that end, I’ve installed Kubuntu on my desktop PC.

I would like to be able to report that the Kubuntu install experience was better than the Debian one, or even on par with a Windows install. Unfortunately, that just isn’t the case.

My machine contains three 500GB hard drives. One is used as the system drive, while an integrated hardware RAID controller binds the other two together as a RAID1 array. Under Windows, this setup worked perfectly. Under Kubuntu, it crashed the graphical installer, and threw the text-based installer into fits of rage.

With plenty of help from the #kubuntu IRC channel on freenode, I managed to complete the Kubuntu install by running it with the two RAID drives disconnected from the motherboard. After finishing the install, I shut down, reconnected the RAID drives, and booted back up. At this point, the RAID drives were visible from Dolphin, but appeared as two discrete drives.

It was explained to me via this article that the hardware RAID support that I had always enjoyed under windows was in fact a ‘fake RAID,’ and is not supported on Linux. Instead, I need to reformat the two drives, and then link them together with a software RAID. More on that process in a later post, once I figure out how to actually do it.

At this point, I have my desktop back up and running, reasonably customized, and looking good. After trying KDE’s default Amarok media player and failing to figure out how to properly import an m3u playlist, I opted to use Gnome’s Banshee player for the time being instead. It is a predictable yet stable iTunes clone that has proved more than capable of handling my library for the time being. I will probably look into Amarok and a few other media players in the future. On that note, if you’re having trouble playing your MP3 files on Linux, check out this post on the ubuntu forums for information about a few of the necessary GStreamer plugins.

For now, my main tasks include setting up my RAID array, getting my ergonomic bluetooth wireless mouse working, and working out folder and printer sharing on our local Windows network. In addition, I would like to set up a Windows XP image inside of Sun’s Virtual Box so that I can continue to use Microsoft Visual Studio, the only Windows application that I’ve yet to find a Linux replacement for.

This is just the beginning of the next chapter of my own personal Linux experiment; stay tuned for more excitement.

This post first appeared at Index out of Bounds.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

Debian Device Driver Check

December 19th, 2009 No comments

Just wanted to bring this website that I found today to everybody’s attention. It’s an online tool that checks if your hardware is supported by Debian. All you need to do is boot the target system from a live CD, type lspci -n at the command line, and paste the output into the text field on the site.

The system then checks a database to see if each of your devices is supported, and gives you a handy readout that shows which drivers you should use for each device. Because Debian is so strict about free software (as in speech, not as in beer), if your hardware passes this test, you should be able to find open sourced drivers that will allow any distribution to run on it.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

How is it doing that?

December 15th, 2009 13 comments

Just about everything that I’ve ever read about media playback on Linux has been negative. As I understand the situation, the general consensus of the internet is that Linux should not be relied on to play media of any kind. Further, I know that the other guys have had troubles with video playback in the past.

All of which added up to me being extremely confused when I accidentally discovered that my system takes video playback like a champ. Now from the outset, you should know that my system is extremely underpowered where high definition video playback is concerned. I’m running Debian testing on a laptop with a 1.73 GHz single-core processor, 758MB shared video RAM, and a 128MB Intel GMA 900 integrated graphics card.

Incredibly enough, it turns out that this humble setup is capable of playing almost every video file that I can find, even with compiz effects fully enabled and just a base install of vlc media player.

Most impressively, the machine can flawlessly stream a 1280x528px 1536kb/s *.mkv file over my wireless network.

As a comparison, I have a Windows Vista machine with a 2.3GHz processor, 4GB of RAM, and a 512MB video card upstairs that can’t play the same file without special codecs and the help of a program called CoreAVC. Even with these, it plays the file imperfectly.

I can’t explain how this is possible, but needless to say, I am astounded at the ability of Linux.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

This… looks… awesome!

December 8th, 2009 No comments

Looks being the key word there because I haven’t actually been able to successfully run either of  these seemingly awesome pieces of software.

Amahi is the name of an open source software collection, for lack of a better term, that resembles what Windows Home Server has to offer. I first came across this while listening to an episode of Going Linux (I think it was episode #85 but I can’t remember anymore!) and instantly looked it up. Here is a quick rundown of what Amahi offers for you:

  • Currently built on top of Fedora 10, but they are hoping to move it to the most recent version of Ubuntu
  • Audio streaming to various apps like iTunes and Rhythmbox over your home network
  • Media streaming to other networked appliances including the Xbox 360
  • Acts as a NAS and can even act as a professional grade DHCP server (taking over the job from your router) making things even easier
  • Built in VPN so that you can securely connect to your home network from remote locations
  • SMB and NFS file sharing for your whole network
  • Provides smart feedback of your drives and system, including things like disk space and temperature
  • Built-in Wiki so that you can easily organize yourself with your fellow co-workers, roommates or family members
  • Allows you to use the server as a place to automate backups to
  • Windows, Mac & Linux calendar integration, letting you share a single calendar with everyone on the network
  • Implements the OpenSearch protocol so that you can add the server as a search location in your favorite browser. This lets you search your server files from right within your web browser!
  • Includes an always-on BitTorrent client that lets you drop torrent files onto the server and have it download them for you
  • Supports all Linux file systems and can also read/write to FAT32 and read from NTFS.
  • Sports a plugin architecture that lets developers expand the platform in new and exciting ways
  • Inherits all of the features from Fedora 10
  • Finally Amahi offers a free DNS service so you only have to remember a web address, not your changing home IP address

FreeNAS is a similar product, although I use that term semi-loosely seeing as it is also open source, except instead of being based on Linux, FreeNAS is currently based on FreeBSD 7.2. Plans are currently in the works to fork the project and build a parallel Linux based version. Unlike Amahi, FreeNAS sticks closer to the true definition of a NAS and only includes a few additional features in the base install, letting the user truly customize it to their needs. Installed it can take up less than 64MB of disk space. It can (through extensions) include the following features:

  • SMB and NFS as well as TFTP, FTP, SSH, rsync, AFP, and UPnP
  • Media streaming support for iTunes and Xbox 360
  • BitTorrent support allowing you to centralize your torrenting
  • Built-in support for Dynamic DNS through major players like DynDNS, etc.
  • Includes full support for ZFS, UFS, ext2, ext3. Can also fully use FAT32 (just not install to), and can read from NTFS formatted drives.
  • Small enough footprint to boot from a USB drive
  • Many supported hardware and software RAID levels
  • Full disk encryption via geli

Both of these can be fully operated via a web browser interface and seem very powerful. Unfortunately I was unable to get either up and running inside of a VirtualBox environment. While I recognize that I could just install a regular Linux machine and then add most of these features myself, it is nice to see projects like that package them in for ease of use.

This is definitely something that I will be looking more closely at in the future; you know once these pesky exams are finished. In the mean time if anyone has any experience with either of these I would love to hear about it.

[UPDATE]

While publishing this, the folks over at Amahi sent out an e-mail detailing many new improvements. Turns out they released a new version now based on Fedora 12. Here are their notable improvements:

  • Amahi in the cloud! This release has support for VPS servers (Virtual Private Servers).
  • Major performance and memory improvements, providing a much faster web interface and a 30% smaller memory footprint.
  • Based on Fedora 12, with optimizations for Atom processors built-in, preliminary support in SAMBA for PDC (Primary Domain Controller) with Windows 7 clients and much more.
  • Completely revamped web-based installer.
  • Users are more easily and securely setup now, the with password-protected pages and admin users.
  • Brand new architecture, with future growth in mind, supporting more types of apps, and more importantly, bring us closer to supporting Ubuntu and other platforms. Over 100+ apps are working in this release out of the gates!

It all sounds great. I will be looking into this new version as soon as I have a moment to do so.




I am currently running a variety of distributions, primarily Linux Mint 18.
Previously I was running KDE 4.3.3 on top of Fedora 11 (for the first experiment) and KDE 4.6.5 on top of Gentoo (for the second experiment).

Coming to Grips with Reality

December 8th, 2009 No comments

The following is a cautionary tale about putting more trust in the software installed on your system than in your own knowledge.

Recently, while preparing for a big presentation that relied on me running a Java applet in Iceweasel, I discovered that I needed to install an additional package to make it work. This being nothing out of the ordinary, I opened up a terminal, and used apt-cache search to locate the package in question. Upon doing so, my system notified me that I had well over 50 ‘unnecessary’ packages installed. It recommended that I take care of the issue with the apt-get autoremove command.

Bad idea.

On restart, I found that my system was virtually destroyed. It seemed to start X11, but refused to give me either a terminal or a gdm login prompt. After booting into Debian’s rescue mode and messing about in the terminal for some time trying to fix a few circular dependencies and get my system back, I decided that it wasn’t worth my time, backed up my files with an Ubuntu live disk, and reinstalled from a netinst nightly build disk of the testing repositories. (Whew, that was a long sentence)

Unfortunately, just as soon as I rebooted from the install, I found that my system lacked a graphical display manager, and that I could only log in to my terminal, even though I had explicitly told the installer to add GNOME to my system. I headed over to #debian for some help, and found out that the testing repositories were broken, and that my system lacked gdm for some unknown reason. After following their instructions to work around the problem, I got my desktop back, and once more have a fully functioning system.

The moral of the story is a hard one for me to swallow. You see, I have come to the revelation that I don’t know what I’m doing. Over the course of the last 3 months, I have learned an awful lot about running and maintaining a Linux system, but I still lack the ability to fix even the simplest of problems without running for help. Sure, I can install and configure a Debian box like nobody’s business, having done it about 5 times since this experiment started; but I still lack the ability to diagnose a catastrophic failure and to recover from it without a good dose of help. I have also realized something that as a software developer, I know and should have been paying attention to when I used that fatal autoremove command – when something seems wrong, trust your instincts over your software, because they’re usually correct.

This entire experiment has been a huge learning experience for me. I installed an operating system that I had never used before, and eschewed the user-friendly Ubuntu for Debian, a distribution that adheres strictly to free software ideals and isn’t nearly as easy for beginners to use. That done, after a month of experience, I switched over from the stable version of Debian to the testing repositories, figuring that it would net me some newer software that occasionally worked better (especially in the case of Open Office and Gnome Network Manager), and some experience with running a somewhat less stable system. I certainly got what I wished for.

Overall, I don’t regret a thing, and I intend to keep the testing repositories installed on my laptop. I don’t usually use it for anything but note taking in class, so as long as I back it up regularly, I don’t mind if it breaks on occasion; I enjoy learning new things, and Debian keeps me on my toes. In addition, I think that I’ll install Kubuntu on my desktop machine when this whole thing is over.  I like Debian a lot, but I’ve heard good things about Ubuntu and its variants, and feel that I should give them a try now that I’ve had my taste of what a distribution that isn’t written with beginners in mind is like. I have been very impressed by Linux, and have no doubts that it will become a major part of my computing experience, if not replacing Windows entirely – but I recognize that I still have a long way to go before I’ve really accomplished my goals.

As an afterthought: If anybody is familiar with some good tutorials for somebody who has basic knowledge but needs to learn more about what’s going on below the surface of a Linux install, please recommend them to me.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

Eclipse… Again

November 21st, 2009 No comments

Man I’m beginning to sound like a broken record. Last night I reinstalled my Debian system. Somewhere along the line, I made a mess with my repositories, and as Wayne suggested in the comments on one of my previous posts, a good way to avoid many of these issues is to install your Testing system directly from a netinst daily build cd image instead of installing Lenny and then upgrading.

So I did. Upon inserting the install disc and attempting to use the graphical installer, I was confronted with a terminal spewing error messages about missing drivers or something. Figuring that this was just an error related to the daily installer build, I backed out of the graphical installer and took a shot at the expert install. Now that I know my way around Linux, the expert installer isn’t so daunting, and the rest of the process went smoothly, although it took awhile.

This morning, I figured I’d be productive and write some Java on my freshly installed system. So I went over to synaptic, and searched out Eclipse… only to find that it didn’t exist in the Testing repositories. How strange. A google and a half later and I had found that eclipse is available in Lenny, as well as Sid, but is conspicuously absent from Testing. What to do?

I hit the #debian IRC channel and asked for a bit of help, which i promptly got, in the form of these instructions:

  1. Add the line deb-src http://ftp.us.debian.org/debian sid main non-free contrib to your sources list.
  2. From a root terminal, run apt-get update
  3. From a root terminal, run apt-get install build-essential
  4. Navigate to an empty directory somewhere on your system
  5. Run apt-get build-dep eclipse. This will download almost 200MB of source code to your system. Don’t do it over a wireless connection like I did.
  6. Run apt-get -b source eclipse. Don’t worry if this step takes forever – it took almost an hour on my system.
  7. install the resultant debs. This step is painful, because while all of the dependencies will have been created for you, there is a certain order to installing them that requires a bit of trial and error to figure out.

So after a little over an hour of messing about, I have a working Eclipse install on my system, and can get some real work done. It was frustrating, but hey, thanks to the guys over at #debian, it wasn’t the end of the world.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

I Cannot Has Eclipse?

November 20th, 2009 No comments
Screenshot-Add-Remove Applications

You know, that's interesting, as I'm quite certain that I've installed Eclipse on x86 hardware before...




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

Configuring BlueZync and Failing at Barry

November 6th, 2009 1 comment

After successfully compiling and installing the BlueZync for Thunderbird plugin last night, I decided to take a shot at actually synchronizing my Blackberry with Thunderbird. The first step was a little bit of configuration. For that, I followed this guide on the BlueZync website.

Everything was going fine until I got to the section entitled “Mozilla plugin for OpenSync.” In this section, you are instructed to execute the command ldconfig -p | grep libxpcom.so, which checks if the file libxpcom.so is registered as a symlink on your system. After finding out that it was not, I entered the command locate libxpcom.so from a root terminal, and found three locations for the file in question on my system. I then used the line export LD_LIBRARY_PATH=/usr/lib/icedove:/usr/lib/iceowl:/usr/lib/xulrunner-1.9 to register the symlink. Unfortunately, even after running the export command, ldconfig failed to find the link. Although this one will probably bite me in the ass later on, I’ll skip it for now.

At this point in the install process, I could access the BlueZync settings panel from within Thunderbird, and run the command line osynctool –listplugins and see the mozilla-sync plugin listed, which is the part of the BlueZync suite that really interests me. mozilla-sync is a plugin for OpenSync that should allow me to interface my Blackberry with Thunderbird (with the help of the Barry libraries, which provide another OpenSync plugin that communicates with the phone).

To continue, it was necessary to install all of the elements of the Barry libraries in order to get their OpenSync plugin that would complete the chain. This is where I may have committed my second cardinal sin – dpkg notified me that in order to install the opensync-plugin-barry package, I had to install a version of the libopensync0 package that was between v0.22 and v0.3. As I understand it, Bluezync already installed some version of OpenSync onto my machine, and I have a feeling that reinstalling a different version may ruin all of the progress that I’ve made thus far.

Indeed, after finishing the Barry install and running osynctool –listplugins again, mozilla-sync was still listed, but opensync-plugin-barry was not. This is strange, as in my last three attempts at this process, getting Barry to show up was the easy part. Now the tables have turned, and I have what I assume to be a properly working BlueZync install, but without the Barry component that would make it all work with my phone.

Back to the proverbial drawing board with me…




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

(Finally) Installing Bluezync for Thunderbird

November 5th, 2009 1 comment

After some constructive comments from Henrik, the developer of the BlueZync plugin for Thunderbird, I decided to take another shot at getting Blackberry sync working on Linux. This time, instead of making up my own instructions, I actually followed his (which have been updated somewhat since my last visit).

Surprisingly, when I followed the instructions to the letter, the plugin built correctly the first time without any problems. When I launched Icedove (the Debian rebranding of Mozilla Thunderbird), the plugin even loaded correctly! If you’ve read my past posts detailing this process, you’ll feel as incredulous as I did.

The only trouble that I ran into along the way was actually with version 0.9 of the Lightning plugin for Icedove (Thunderbird). Upon installation of the plugin, I was not able to create a calendar, an event, or a task. Turns out that this Ubuntu bug applies to Debian as well, and that the problem can be easily fixed by uninstalling Lightning, downloading and installing the libstdc++5 package, and reinstalling the Lightning plugin. For whatever reason, I could not find this package in the Debian Testing repositories, and instead downloaded and installed it from the Lenny repositories.

With that issue solved, I tried running the ./test-bluezync.sh script, and was met yet again with a slew of failed tests:

21% tests passed, 15 tests failed out of 19

The following tests FAILED:
5 – thunderbird (Failed)
6 – tbird_empty (Failed)
7 – tbird_slow (Failed)
8 – tbird_slow_3 (Failed)
9 – tbird_fast (Failed)
10 – tbird_add (Failed)
11 – tbird_delete (Failed)
12 – tbird_modify (Failed)
13 – light_empty (Failed)
14 – light_slow (Failed)
15 – light_slow_3 (Failed)
16 – light_fast (Failed)
17 – light_add (Failed)
18 – light_delete (Failed)
19 – light_modify (Failed)

However, unlike in past attempts at this install, this time the Bluezync plugin is visible from within Thunderbird… Now all I have to figure out is how to use it. More on that later.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.

Back at Square 1

November 2nd, 2009 2 comments

This morning I reinstalled my Debian system. I began by downloading an ISO for the current Debian Stable build (called Lenny), and installing it with the graphical installer. That done, I used a couple of my old posts to get my wireless firmware installed and to upgrade my system to the Testing repositories.

Unfortunately, I have realized that a clean install of Debian Linux is a pretty plain place to be in. Even though I have the benefit of my old writings to help me get up to speed, some, like the ones dealing with how to get Compiz working properly, are somewhat lacking in detail.

Naturally, I’ve replaced all of the problems that running multiple desktop environments was causing with all of the problems that an entirely unconfigured system can cause. I’ve already mentioned that I haven’t gotten Compiz working yet (whenever I turn it on, all of my window decorations disappear), and there is some error with Postgre that causes Synaptic and Aptitude to complain whenever I make changes to my system:

E: postgresql-8.4: subprocess installed post-installation script returned error exit status 1
E: postgresql: dependency problems – leaving unconfigured
E: postgresql-contrib-8.4: dependency problems – leaving unconfigured
E: postgresql-contrib: dependency problems – leaving unconfigured

Most stressing is the fact that I cannot get into the preferences for the Nautilus file system browser. Whenever I try to open the preferences dialog from the edit menu, it (and most of GNOME) crash. Running Nautilus from the terminal yields me this output:

(nautilus:4213): Gtk-CRITICAL **: gtk_combo_box_append_text: assertion `GTK_IS_COMBO_BOX (combo_box)’ failed

(nautilus:4213): GLib-GObject-CRITICAL **: g_object_set_data_full: assertion `G_IS_OBJECT (object)’ failed

(nautilus:4213): Gtk-CRITICAL **: gtk_widget_set_sensitive: assertion `GTK_IS_WIDGET (widget)’ failed

(nautilus:4213): GLib-GObject-WARNING **: invalid (NULL) pointer instance

(nautilus:4213): GLib-GObject-CRITICAL **: g_signal_connect_data: assertion `G_TYPE_CHECK_INSTANCE (instance)’ failed

(nautilus:4213): GLib-GObject-WARNING **: invalid (NULL) pointer instance

(nautilus:4213): GLib-GObject-CRITICAL **: g_signal_handlers_block_matched: assertion `G_TYPE_CHECK_INSTANCE (instance)’ failed

(nautilus:4213): GLib-GObject-CRITICAL **: g_object_get_data: assertion `G_IS_OBJECT (object)’ failed
Segmentation fault

Actually, the terminal prints output similar to the above, but so much of it that this post would take up most of the front page of the site were I to post it all. I have no idea what the hell any of that means, or how it got into my system, or why I cannot get into the preferences panel of Nautilus as a result.

Until I do figure it out, I’ll be spending a lot of time on the #debian channel. Along with these major problems come a number of small tasks, like adding myself to the sudo keyring, adding the Testing repository keys to my sources list so that it stops yelling that all of my software is unverifiable.

Fucking Linux.




On my Laptop, I am running Linux Mint 12.
On my home media server, I am running Ubuntu 12.04
Check out my profile for more information.