What’s new in openSUSE Leap 15 – installation experience

A big release

On the 25th May 2018, openSUSE Leap 15 was released for download. Over the last few days I have upgraded both of my systems to this new release. Although this was a big release for openSUSE, the media attention for this release was surprisingly low. The reason why this is a big release, is that the underlying software packages are all new.

openSUSE Leap 42 has a shared core with SUSE Linux Enterprise 12 (SLE 12). For instance Leap 42.2 shares a lot of software packages with Service Pack 2 (SLE 12 SP2). And Leap 42.3 shares a lot of packages with SLE 12 SP3. The shared core was on average 20% of the total number of packages. Because of that shared core, some of the packages were starting to show their age.

openSUSE Leap 15 shares a lot of software packages with SUSE Linux Enterprise 15, which in itself is based on a 2017 fork of openSUSE Tumbleweed. That means that all of the underlying packages in SLE 15 have been updated to a more current version in comparison to SLE 12 SP3. The shared core for openSUSE Leap 15 is (according to a FOSDEM 2018 presentation) about 27% of the total number of packages. And the remaining packages are originating from (an even more recent fork from) openSUSE Tumbleweed. Which means that we get a lot of improvements in openSUSE Leap 15.

A good example (to get an idea about the progress that has been made) is the underlying Linux kernel, which has been updated from version 4.4 to 4.12. Linux kernel 4.4 was released in January 2016 and Linux kernel 4.12 was released in July 2017. You cannot simply assume that the SLE kernel is identical to the upstream Linux kernel, because SUSE includes a lot of back-ports of security fixes and of hardware drivers in their kernels. However, you can assume that most of the newly introduced features in more recent Linux kernels are not being back-ported. So the upgrade from SLE 12 to SLE 15 means that we get 1,5 years of new features from the Linux kernel community.

So openSUSE Leap 15 is a big release. But is it any good? In this article I will focus on the installation experience.

Installation

I have installed openSUSE Leap 15 on 2 machines. The first one is a netbook: the Acer Aspire One 725. This machine has an AMD C70 chip set (CPU + GPU), 4 GB of DDR3 memory and a 120 GB SSD hard drive. The second one is a bare-bones desktop: the Zotac ZBOX Sphere OI520. This machine has an Intel Core i5-4200U CPU, Intel HD Graphics 4400, 16 GB of DDR3L memory and a 256 GB SSD hard drive. For both machines I have opted for a fresh installation instead of doing an upgrade.

Every time I install a new openSUSE release, I evaluate whether to do an update or a fresh install. The ‘problem’ with any Linux is that over time, when you install new software packages, a lot of dependencies (additional software packages that are required for the software to function) are automatically installed. But when you remove that same software package, some of these dependencies don’t get automatically removed. Which means that over time, you can collect a lot of software packages that are not used anymore. The advantage of updating is that all of the old software packages remain installed. The advantage of a fresh install is that all of the old unused packages get removed. So in general, I update between minor releases (Leap 15 to Leap 15 SP1) and I do a fresh install between big releases (Leap 15 SP3 to Leap 16).

There are 2 bottlenecks in the installation experience. The first one is the WiFi setup. The second one is the hard drive partitioning. The installation starts easy enough by asking you if you want to boot from hard disk, perform a new installation, perform an upgrade or more (which includes booting into a rescue system). You select Installation (in this example). You agree to the license.

And then you meet the network settings. This screen is similar to the networking setup for the YaST Wicket Service. You need to click on your network card. Then click Edit. Then select Dynamic Address. Then select Next. Then select Scan Network. Pick your network from the Network Name (ESSID) drop-down menu. Then select WPA-PSK (WPA version 1 or 2) from the Authentication Mode drop-down menu. Then fill in your WiFi password in the Encryption Key field. Click Next and then Click OK. Now you can continue to the next part of the installer.

The advantage of configuring your WiFi settings is that the openSUSE installer can install the latest versions from the online repositories. Which means that your openSUSE installation is fully up to date after the installation has ended. The disadvantage is that the configuration is rather complicated. My advice would be to plug-in an Ethernet cable before starting the installation. This means that the installer will automatically detect the internet connection and will simply skip this step.

Now you can select the Desktop environment. I have used KDE (Plasma) for the last 10 years and therefore always select that one. You can configure the online repositories, but I would advice to simply leave that as-is and configure the repositories after the installation is complete.

The next step is the Partitioning. This is the part where most users will encounter the biggest difficulties. First lets discuss some advantages of the openSUSE approach. The default partitioning setup is great. openSUSE creates a separate /home partition by default. This means that whenever something happens to openSUSE, you can simply blow away the root (“/”) partition and reinstall openSUSE. This is a worst case scenario, because there is a better way to rescue your system: BtrFS. This is a modern file system based on the “Copy on Write” principle. Which means that whenever you install something new, it also keeps the old version around. And you can move between “snapshots”. A snapshot is created every time you install something new. Which means that if you screw up, you can simply undo your mistake from the boot screen. But if even that fails, you can simply blow your operating system partition away, while keeping your personal data safe. This is exactly what I did. I have formatted the root partition and have installed openSUSE Leap 15 in that same partition. I kept my /home partition unchanged.

But say you don’t understand anything about partitioning. Then you can simply go with the default proposal. Nothing wrong with that. As stated before, the default setup is great. Or you can take the Guided setup. This means that you answer some easier questions and the tool automatically determines your setup. This allows you to Encrypt your partitions. Meaning that when your machine is powered off, people cannot read your data by sticking that hard drive in another computer. I would always advice you to create a separate Home partition. XFS is a very stable file system for such a Home partition. For your Root partition, you should select either BtrFS or XFS. There are heated discussions online with arguments against and for BtrFS. After having a personal question/answer with the openSUSE chair on Reddit and after reading about the careful openSUSE implementation, I think selecting BtrFS for the Root partition is now the best way to go.

If you want to determine your partition setup yourself, you need to use the Expert Partitioner. This gives you the option to start with Existing Partitions or with the Current Proposal. It really doesn’t matter too much, if you want to determine everything by yourself. Both selections bring you into the YaST Expert Partitioner tool. Some headlines indicate that it is renewed. But if you read carefully, only the back-end was renewed. It looks exactly the same as I remembered it. Go into your Hard Disks (sda, sdb, et cetera) and select the one that you want to partition. Create at least a Boot partition, a Swap partition, a Root partition and a Home partition. This is a nice tool if you have both an SSD and a spinning HDD. You can install the operating system on the faster SSD and install your Home partition on the spinning HDD which typically can contain much more data.

The remaining steps are easier. You select your timezone. Then you create a new user, or you import the current user. Because I kept my Home partition (by not formatting it and mounting it to /home) I could choose to import my current user. I would advice you to deselect Automatic Login. Finally you see an overview of the installation. Click twice on Install and now wait for the process to finish.

Things to do after the installation

One of the first things that I like to do after installing the base operating system is to add the applications that I like to use. In previous openSUSE releases, a lot of KDE applications got installed by default. This is still the case, but the Amarok music player was missing. And not having a music player installed by default is a strange decision. You can install Amarok from the YaST Software Management application, something I would recommend.

In previous releases, it could take 1 or 2 weeks before all applications were available in the software repositories. This was especially true for open source games. For example: I was used that Warsow and Hedgewars were not immediately available. This time I was pleasantly surprised to see that almost every application in my “Linux essentials list” was immediately available. There is still 1 application that I am still missing: PDF Chain, a graphical PDF editor that makes use of the PDFtk toolkit. I also discovered that a couple of Calligra applications (Author, Braindump and Flow) were depreciated, which explains why I can’t find them in the repositories. I don’t care about Author, but Braindump and Flow were actually very good applications. So I hope that the people of the Calligra team bring them back in the future.

The list of applications that I like to install might be different from yours. But if you want a wider selection of (more current) applications, it is worth the effort to add some online repositories. You can do so by going into YaST and selecting Software Repositories. The picture below shows the repositories that I have configured. This is the first time (that I am aware) that I can specify to make a HTTPS connection to the openSUSE repositories. A nice security improvement.

Another thing that you most likely want to do, is to install Codecs needed to play common audio and video files. You can do this in 2 ways. The first way is to go to: https://opensuse-community.org/ and use the 1-click install option. The second way is to install these software packages with YaST Software Management. Search and install the following software packages:

  • ffmpeg
  • lame
  • gstreamer-plugins-bad
  • gstreamer-plugins-libav
  • gstreamer-plugins-ugly
  • gstreamer-plugins-ugly-orig-addon
  • libdvdcss2
  • xine-ui
  • libxine2
  • libxine2-codecs
  • vlc-codecs
  • vlc

The last part is the system configuration. You might want to configure a printer / scanner. You can do this via YaST. Or if you have a HP (all-in-one) printer, you can follow the instructions on the HP Linux Imaging and Printers site. The last option is the more difficult way to install your printer. But this is the way to get the HP Device Manager installed, something that you woudn’t have otherwise. Before starting the process described by HP, you need to install 1 missing software package: libgphoto2-devel. Also you need to open your Firewall to be able to auto-discover your printer over WiFi. This is done in Yast. See the screenshot below.  After that, the setup of a HP printer should go without a hitch.

In conclusion

I have covered most of the basics of the openSUSE Leap 15 installation. My first impression is very positive. For me the installation went flawless on both machines. And the software availability was great this time.

But you might have different hardware and a different experience. Know that you can ask for help on the openSUSE forums, on the sub-forum for technical help on installations. Another channel to ask for help is the openSUSE Reddit forum. Or you can ask your question directly via IRC.

Enjoy your new openSUSE Leap 15 install and remember to Have a lot of Fun!