 DistroWatch Weekly |
| DistroWatch Weekly, Issue 878, 10 August 2020 |
|
Welcome to this year's 32nd issue of DistroWatch Weekly!
On this website we mostly talk about Linux-powered desktop operating systems. However, while there are millions of desktop computers running Linux distributions, where Linux really shines is in the server market. This week we kick off with a quick look at Zentyal Server, an Ubuntu-based distribution that ships with a convenient, user-friendly interface. Read on to hear how the latest release of Zentyal performs. In our News section we share user survey results from the CentOS project and a similar overview of how Linux Mint is utilised. Plus we are pleased to share network security tips from the IPFire team and link to a new list of devices which work with the mobile UBports operating system. How to package software and how to make the many pieces of an operating system work together is an ongoing debate among developers. In our Questions and Answers column we look at one aspect of this debate: dynamic versus static linking where we talk about some of the benefits and drawbacks to these approaches of linking applications to their dependencies. Last week we discussed a GRUB security patch which caused some systems to no longer boot and we would like to hear if you were affected by this issue in our Opinion Poll. Plus we are pleased to list last week's releases and share the torrents we are seeding. We wish you all an incredible week and happy reading!
Content:
- Review: Zentyal Server 6.2
- News: CentOS shares survey results, Linux Mint gives version usage overview, IPFire shares DNS security tips, UBports updates list of working devices
- Questions and answers: Pros and cons of dynamic linking versus static linking
- Released last week: BSD Router Project 1.97, Ubuntu 20.04.1
- Torrent corner: BSD Router Project, KDE neon, Kubuntu, Lubuntu, Redo Rescue, Robolinux, Ubuntu, Ubuntu Budgie, Ubuntu Kylin, Ubuntu MATE, Ubuntu Studio, Xubuntu
- Opinion poll: Were you affected by the GRUB Boothole security patch?
- Reader comments
|
| Feature Story (by Jesse Smith) |
Zentyal Server 6.2
Zentyal is an Ubuntu-based server distribution which is designed to be easy to set up and then manage using a friendly, web-based interface. The distribution targets small and medium office and business environments. The Zentyal distribution is intended to take on such tasks a as a storage server, Internet gateway, or to provide other office IT infrastructure - all through a convenient, point-n-click web portal.
The latest version of Zentyal is based on Ubuntu 18.04.4 and mostly features minor updates. There are new anti-virus packages, improved DNS management, easier management of hard drives, and the AppArmor security software is enabled by default.
The download for Zentyal is 1GB in size and is available for 64-bit (x86_64) machines only. Booting from the install media brings up a menu asking us to select our preferred language from a list. Then we are given the choice of wiping the hard drive and installing Zentyal or launching an expert installer. Both menu options launch a text-based installer which should be familiar to people who have set up Ubuntu Server or used Debian's text installer.
Installing
Setting up Zentyal through the expert install mode brings up a series of text-based menus where we are asked to supply our preferred language (again), choose our keyboard layout, time zone, and to make up a username and password for ourselves. The installer then asks if we want to use guided or manual partitioning. The former takes over the entire disk, creating an ext4 formatted LVM volume, while the manual option walks us through an unusually long series of partitioning and filesystem options screens.
The installer then asks if we would like to install a graphical desktop environment (LXDE in this case). Packages are copied to our hard drive and then we are presented with a few more configuration steps. We are asked if we wish to enable a web proxy and if we want to install the GRUB boot loader. Then we confirm our system clock settings and the system offers to restart itself.
The installer uses an unusual combination of white text on a bright green background for text entry boxes which I found difficult to read. The rest of the interface uses a high contrast, but entering text into fields was hard on the eyes.
After the installer finished its work and offered to reboot the computer I ran into my first serious issue. The shutdown process started off well, showing systemd status messages about services being shutdown. However, the process took a long time. In fact, once the system reached a point where it was reportedly cleaning up temporary directories and had been stuck on that task for 15 minutes (longer than the installer had taken to set up the operating system) I finally forced a hard reset.
Early impressions
Zentyal booted and, after performing a quick filesystem check, brought me to a text console. (I had opted not to install the desktop environment.) I could sign in with the credentials I had supplied to the installer and was able to confirm my Internet connection worked. I was able to shutdown the system (reboots happened quickly and were error-free at this time), and some basic command line programs worked.
Once I got logged in I noticed a few oddities with Zentyal. The first was that the operating system identified itself as Ubuntu 18.04.4 LTS rather than Zentyal Server 6.2. Which makes some sense since Ubuntu is this project's parent. The message which is displayed when the user logs in indicates that the web interface can be accessed by pointing a web browser to the server's IP address and using port 8443 over the HTTPS protocol.
Attempts to connect to port 8443 were refused. As were connection attempts to communicate over port 443. Attempts to connect with port 80 (the default HTTP port) worked, but only showed a generic home page for the nginx web server.
At this point I tried to install the telnet client to help me troubleshoot the connection issues and found my package database was a mess. I had to run "dpkg --configure -a" to fix it. This seemed odd as the APT package tools had not been used yet since the installation had completed. After letting dpkg run I was able to install telnet and confirm only ports 22 (for OpenSSH) and 80 (for the generic nginx install) were available. There were no services listening on port 8443. This remained true after a reboot and so I decided to start over from scratch.
Second impressions
Things had gone poorly enough the first time trying Zentyal that I decided to be extra careful the next time around. I double-checked the distribution's ISO checksum, decided to try running it in a virtual machine to avoid any potential hardware incompatibilities, and then opted to take the automated installation option.
As it turns out, the automated install option walks us through almost all of the same steps as the expert installer. We are asked the same questions about language, keyboard layout, time zone, and login credentials. There appear to be just three differences in the automated version of the system installer. The first is we are warned partitions on the main hard disk are already mounted and asked if they should be unmounted before the disk is partitioned. Since the plan is to wipe the disk, umounting partitions seemed wise. Though I'm not sure why any partitions would be mounted by the installer to begin with.
The second difference is that the installer handles disk partitioning, setting up an LVM volume and formatting it with ext4 rather than have us set up partitions and mount points. The third difference is the installer does not ask if we would like to install a graphical desktop, it just assumes we want the desktop. This causes LXDE to be installed, along with a handful of X.Org and Wayland libraries.
Zentyal Server 6.2 -- Exploring the LXDE menu
(full image size: 52kB, resolution; 1920x1200 pixels)
The installation process finished, indicated it was successful and offered to restart the system. Once again Zentyal got stuck shutting down. Background services stopped gradually for a few minutes, then got stuck cleaning out temporary directories. The system at this point appeared to be idle (hard drive and CPU activity were minimal) and remained stuck for the next twenty minutes. Eventually I forced a restart.
This time my fresh copy of Zentyal booted to a graphical environment. I could sign into the LXDE desktop where the Firefox browser opened and automatically attempted to connect to the localhost. This connection failed and, again, all connections (either over HTTP or HTTPS) failed to reach port 8443 and succeeded on port 80 where the generic web server was running. I tried to take a screenshot of this error and discovered that the keyboard shortcut for screenshots was mapped, but to a GNOME screenshot application that was not available. This broken shortcut feels all the more out of place given Zentyal uses the LXDE desktop rather than GNOME.
Zentyal Server 6.2 -- Trying to take a screenshot of Firefox connecting to the Zentyal server
(full image size: 96kB, resolution: 1920x1200 pixels)
I wanted to use the nmap tool to scan local ports and see which services were available. I soon found that nmap was installed, but one of its dependencies (libblas) was missing, causing the tool to not run. I could reinstall nmap through the APT package manager and a scan turned up no new services, other than OpenSSH and nginx showing its generic home page.
Conclusions
After my second failed attempt at using Zentyal, and some troubleshooting, I came to the realization the distribution was not going to work as expected and put it aside. According to the documentation, I should be able to simply install the distribution and connect to it using a web browser, but this did not work, either locally or over the LAN. This was disappointing as I have used Zentyal in the past and generally had positive experiences with it. I've even recommended the distribution to a few people who wanted to run a light office server with an easy, point-n-click interface.
I have three theories as to why Zentyal did not work for me this time around. One is that the documentation is out of date (or updated in places I'm not looking) and additional steps are now required to set up the web portal service. The second is that there is a bug in the web portal software that is preventing it from running.
Personally, I suspect neither of these are true and, instead, something (or multiple somethings) are going wrong during the setup phase. While the installer appears to finish copying its files to my hard drive and reports it is done, the fact the system does not shut down cleanly afterwards suggests something is not finished in the background. The shutdown services never conclude and, while disk and CPU activity was virtually non-existent all twenty minutes I waited, I suspect additional configuration steps were supposed to be happening during that time. It is hard to say for certain though since no status messages are displayed and the installer claims to be finished. I would also consider it odd for services to be enabled during the shutdown phase of the live media, but stranger things have happened.
Whatever the case, Zentyal did not work for me and, unfortunately, did not display any errors or status messages which would help explain why. The documentation, while normally helpful, did not offer any tips to help me get going. In the past Zentyal has proven to be easy for me to use, but this version has left me with a server-sized void to fill.
* * * * *
Visitor supplied rating
Zentyal Server has a visitor supplied average rating of: 8.7/10 from 3 review(s).
Have you used Zentyal Server? You can leave your own review of the project on our ratings page.
|
| Miscellaneous News (by Jesse Smith and Ladislav Bodnar) |
CentOS shares survey results, Linux Mint gives version usage overview, IPFire shares DNS security tips, UBports updates list of working devices
The CentOS team has published their monthly community newsletter and it shares some important information on the Boothole issue we reported last week. The newsletter also links to survey data the CentOS team has collected about how their distribution is used, how often it is updated, and how people would like to be involved in the project. "Over the past 3 or 4 months, we have been running a survey about how you use CentOS. Many thanks to those who participated in this, to help us better understand how we can give you what you need. I've written up the results of the survey here. While some of the results were expected - most of you use CentOS in small to medium shops, running services either for work or personal use - there were some eye-opening things in this. To me personally, the volunteerism question shows that a lot of you are looking for places to get involved, and that we haven't done a great job telling you where and how. We'll be working to fix that."
* * * * *
The Linux Mint team has been looking at which versions and editions of their distribution get the most attention. The project looked at which versions are most commonly used and added some observations to explain why this is likely the case: "The newest release attracts many users and grows faster than any other but it does not yet represent the largest percentage. As we can see on the chart, more than half of Linux Mint users use Linux Mint 19.x. The older Mint 18.x and the brand new Mint 20 both represent about 20% of the user base. Linux Mint 17.x despite reaching EOL still represents 6% and LMDE represents about 1%." In the project's monthly newsletter the team also acknowledged some common concerns users had with Linux Mint 20, including the lack of a Chromium web browser package, and reported there are plans to address these issues.
* * * * *
The IPFire team is continuing its series of blog posts detailing recommendations on how to improve computer and network security. The latest post concerns DNS configurations and how to make adjustments to this key component of virtually every network-connected device. "After having roamed around infosec in general last week, this post gives some advice on how to gain additional privacy by changing your IPFire's DNS configuration. DNS happens to be a very basic thus quite important protocol of today's internet, but is still being considered a low-risk one when it comes to security and privacy. If you are familiar with IPFire, you might have noticed DNSSEC validation is mandatory, since it defeats entire classes of attacks...."
* * * * *
The UBports team has published information about the project's work, development progress, and updated ports on various devices. The PinePhone and Nexus 6P received special attention: "Florian has been working on a port for the Nexus 6P from Huawei for about a year and it is in a pretty good state. The only issue is that it is big and heavy. It has a good camera , still minus some functions. The GPS is the fastest on any UT device. It takes about 30 seconds to get a fix. Moving the old core devices to Halium 7.1 build is in the works and progressing well. We now have about 40 devices listed on the devices page. That is way more options than in the past." The list of devices which work with UBports, along with their porting status, can be found on the project's new devices page.
* * * * *
These and other news stories can be found on our Headlines page.
|
| Questions and Answers (by Jesse Smith) |
Pros and cons of dynamic linking versus static linking
Linking-it-all-together asks: I came across this article about the benefits of static linking over dynamic linking. If dynamic linking is slower and doesn't offer practical benefits then why do most distros still dynamic link? Is this a hold over from the past or is there a reason I'm missing that make distros still use dynamic linking?
DistroWatch answers: Before we get to the article, I'd like to share a brief, overly generalized view of what statically linked and dynamically linked programs are. A statically linked program is self-contained. Any libraries or functions that are required for the program to run are baked into the program itself. This reduces the program's external dependencies and makes the program bigger. On the positive side of things this means statically linked programs do not rely on other libraries and can be copied to other systems more easily. It also means if libraries are updated on the operating system, the statically linked program will not break due to the changing libraries in other parts of the system.
A dynamically linked program is smaller and does not keep all of its parts internally. Instead a dynamically linked program relies on other libraries on the system to provide key functions. This approach offers a few benefits. In particular it means programs that share common functionality can share one copy of a library rather than have each program carry its own copy. This also means when security updates become available we only need to update each library once and every program on the system which uses pieces of that library automatically benefit, even often without being updated themselves. In theory dynamic linking means smaller programs, smaller updates, and less duplication of functionality. However, dynamically linked programs can break if a library they depend on changes too much or a library they depend on is removed from the system.
I read through the article provided and it does share some interesting statistics on dynamically linked programs versus statically linked programs. The author appears to be making a case against dynamic linking and in favour of static linking, or at least presenting facts which would support such a case. For the sake of this discussion I am going to assume the observations the article's author makes are accurate and factually correct, at least for their own distribution.
The author addresses some interesting questions, such as how often are dynamically libraries used on the system, which indicates how many resources avoid duplication by sharing libraries. They also explore how quickly dynamic and static programs load and how much larger statically linked programs are compared to their dynamically linked counterparts. The author points out that many libraries on their distribution are not shared by many programs, that statically linked programs can load faster, and that not a lot of bandwidth is saved by using dynamically linked programs.
Reading through the page of observations the author shares, it's understandable we might wonder why developers continue to favour dynamically linked applications in most situations. Let's look at some of the specific arguments from the article.
Do your installed programs share dynamic libraries? Findings: not really. Over half of your libraries are used by fewer than 0.1% of your executables.
This is an interesting way of presenting the statistic. Over half of libraries are linked to by fewer than 0.1% of programs. Which is all well and good, but it seems to be focusing on where dynamically linked libraries are not being used, or are used rarely. It is not looking at how often dynamically linked libraries are being used. Imagine if I told you 95% of the items I own do not use wheels, therefore wheels are not important. That may be true, for most items. Many of the objects I own probably do not need wheels. My table, fridge, and laptop do not need wheels. However, the automobile and bicycle really do need wheels to be effective.
I feel the author's argument is similar to the above example using wheels. They are correct that in some instances, maybe even in many instances, libraries are not being shared much, which reduces their value. However, the low level libraries that are used a lot, such as the C library, GTK, and Qt, are likely to be used by hundreds, maybe even thousands, of applications on your distribution. A small number of libraries are used a lot, so it does not matter much (in my opinion) whether other libraries are used rarely. The benefits we get from sharing code in a handful of low-level libraries are enough to make dynamic linking worth while in many situations.
Is loading dynamically linked programs faster?
Findings: definitely not.
This is an interesting point. The author points out that their system loads static applications in about half the time as dynamic ones. This seems like a lot, until we consider the load times are being displayed in nanoseconds. A nanosecond is a billionth of a second, meaning the slowly loading dynamically linked application is taking a whole 0.000137263 seconds to start. It is not physically possible for a person to notice the difference between the load times of a dynamically linked program versus a statically linked one. In fact, you could load a thousand programs of each type before seeing a 0.1 second difference between the two types.
Wouldn't statically linked executables be huge?
Findings: not really. On average, dynamically linked executables use only 4.6% of the symbols on offer from their dependencies. A good linker will remove unused symbols.
I think this point is entirely valid and I do not have much to add to it. A good build system will make individual statically linked programs and dynamically linked ones of similar size. The statically linked ones should not be much larger. However, I believe if we look at how many programs most operating systems run these days, we can see how small benefits will add up over time. Having one program be 4% larger is not a big deal. Having 2,000 programs be 4% larger begins to accumulate additional storage space. Still, the difference, on a modern system, will not be significant so we do not need to worry much about application size. I feel the point here is well made.
Will security vulnerabilities in libraries that have been statically linked cause large or unmanageable updates? Findings: not really.
Not including libc, the only libraries which had "critical" or "high" severity vulnerabilities in 2019 which affected over 100 binaries on my system were dbus, gnutls, cairo, libssh2, and curl. 265 binaries were affected by the rest.
The total download cost to upgrade all binaries on my system which were affected by CVEs in 2019 is 3.8GiB. This is reduced to 1.0GiB if you eliminate glibc.
I think there are two key points to consider carefully here. The author is correct that the difference of a few gigabytes when performing updates over the span of a year is not a lot to most modern computer users. However, I feel it worth noting that, given the choice between larger and smaller updates, most people will want the smaller updates as they use less bandwidth. It might be a relatively small savings, but it is still savings in the range of gigabytes of data and one which will be noticeable on capped or slow Internet connections. In other words, if you can download a 1GB of data in 10 minutes, the size of package updates will not matter to you. However, if you're on a connection where a 1GB download can take most of the day, then you will quickly see the benefit of using dynamic linking to reduce the size of your downloads.
We might also wish to look at this situation from the distribution's point of view. Let's say, just for example, Canonical has 50 million Ubuntu users. And let us assume each user can save around 1GB per year of bandwidth by using dynamic linking to reduce the number packages we need to download. A single extra gigabyte of bandwidth is not much for the end user to absorb, but for Canonical that is a difference of 50 million gigabytes (50 petabytes) per year.
The linked article also skips over the main reason most people want to use dynamically linked programs, which is having the ability to update one library with a security fix and have it correct the issue for every program on the system which uses that library. If one library is vulnerable on a dynamically linked system, we only need to update one package to fix it. When running a statically linked system, we need to update every single program which used the vulnerable library. This can mean a vulnerability in the C library (for example) means updating hundreds, sometimes thousands, of packages. That means in a statically linked scenario we need the maintainer of every package on our system to notice the vulnerability, update their package, and push out a fix. When we use dynamic linking we only need one maintainer, not dozens, to update just one package.
In my opinion, dynamical linking might offer only small, or questionable, benefits a lot of the time, but it offers a great deal of benefit in some key areas with very few drawbacks. There are certainly pros and cons for each approach, but the ease of package management and security updates, along with the reduction of work required by package build servers, means dynamic linking tends to benefit developers and users more than it causes problems. Which is why I suspect dynamic linking will continue to be the default approach for most distributions for the foreseeable future.
* * * * *
Additional answers can be found in our Questions and Answers archive.
|
| Released Last Week |
BSD Router Project 1.97
Olivier Cochard-Labbé has announced the release of BSD Router Project (BSDRP) 1.97, the latest stable build of the project's free and open-source software router distribution based on embedded FreeBSD. This release is upgrades the underlying operating system to FreeBSD 12.1: "BSDRP 1.97 is online. Based on a FreeBSD 12.1-STABLE (r363822), it fixes some missing Ethernet NIC modules (Chelsio Ethernet VF driver and Ethernet QLogic 3200 series). This version adds some new packages - Mellanox Firmware tools (lite version), WireGuard, vim-tiny, mrtparse (MRT format data parser), nrpe3 (Nagios client including nagios-plugins), frr7-pythontools (helper script to help reload frr). New features: load of Intel microcodes by default; update to 12.1-STABLE. Bug fixes: add Chelsio Ethernet VF driver (if_cxgbev); add missing if_qlxgb.ko for Ethernet QLogic 3200 series; correctly disabling ICMP redirect by default. New packages: Mellanox firmware tools (lite version); WireGuard; vim-tiny...." See the release announcement and release notes for more details.
Ubuntu 20.04.1
Łukasz Zemczak has announced the release of Ubuntu 20.04.1, along with the distribution's community editions. These new builds mostly contain bug fixes for packages, reducing the amount of time required to update new installations. "The Ubuntu team is pleased to announce the release of Ubuntu 20.04.1 LTS (Long-Term Support) for its Desktop, Server, and Cloud products, as well as other flavours of Ubuntu with long-term support. As usual, this point release includes many updates, and updated installation media has been provided so that fewer updates will need to be downloaded after installation. These include security updates and corrections for other high-impact bugs, with a focus on maintaining stability and compatibility with Ubuntu 20.04 LTS." Further details can be found in the release announcement and in the release notes.
* * * * *
Development, unannounced and minor bug-fix releases
|
| Torrent Corner |
Weekly Torrents
The table below provides a list of torrents DistroWatch is currently seeding. If you do not have a bittorrent client capable of handling the linked files, we suggest installing either the Transmission or KTorrent bittorrent clients.
Archives of our previously seeded torrents may be found in our Torrent Archive. We also maintain a Torrents RSS feed for people who wish to have open source torrents delivered to them. To share your own open source torrents of Linux and BSD projects, please visit our Upload Torrents page.
Torrent Corner statistics:
- Total torrents seeded: 2,090
- Total data uploaded: 33.0TB
|
| Upcoming Releases and Announcements |
|
Summary of expected upcoming releases
|
| Opinion Poll (by Jesse Smith) |
Were you affected by the GRUB Boot Hole security patch?
Last week we talked about a security bug in the GRUB boot loader which, once the issue was patched, the fix could cause some Red Hat and CentOS systems to no longer boot. The issue was quickly resolved and new patches pushed out which worked around the problem. Were you one of the CentOS or Red Hat users who were hit by the GRUB patch problem? Let us know how you worked around it in the comments.
You can see the results of our previous poll on using a Linux-powered mobile phone in last week's edition. All previous poll results can be found in our poll archives.
|
Were you affected by the Boothole patch?
| I am a CentOS/RHEL user who was affected: | 18 (2%) |
| I am a CentOS/RHEL user who was not affected: | 91 (8%) |
| I am not a CentOS/RHEL user who was affected: | 42 (4%) |
| I am not a CentOS/RHEL user and not affected: | 999 (87%) |
|
|
| Website News (by Ladislav Bodnar) |
DistroWatch database summary
* * * * *
This concludes this week's issue of DistroWatch Weekly. The next instalment will be published on Monday, 17 August 2020. Past articles and reviews can be found through our Article Search page. To contact the authors please send e-mail to:
- Jesse Smith (feedback, questions and suggestions: distribution reviews/submissions, questions and answers, tips and tricks)
- Ladislav Bodnar (feedback, questions, donations, comments)
- Bruce Patterson (podcast)
|
|
| Tip Jar |
If you've enjoyed this week's issue of DistroWatch Weekly, please consider sending us a tip.
(Tips this week: 0, value: US$0.00) |
|
|
|
 bc1qxes3k2wq3uqzr074tkwwjmwfe63z70gwzfu4lx  lnurl1dp68gurn8ghj7ampd3kx2ar0veekzar0wd5xjtnrdakj7tnhv4kxctttdehhwm30d3h82unvwqhhxarpw3jkc7tzw4ex6cfexyfua2nr  86fA3qPTeQtNb2k1vLwEQaAp3XxkvvvXt69gSG5LGunXXikK9koPWZaRQgfFPBPWhMgXjPjccy9LA9xRFchPWQAnPvxh5Le paypal.me/distrowatchweekly • patreon.com/distrowatch |
|
| Extended Lifecycle Support by TuxCare |
|
|
| Reader Comments • Jump to last comment |
1 • Static vs dynamic (by Technosaurus on 2020-08-10 01:55:52 GMT from United States)
First, some libraries are just plain bad at static builds (gnu-libc, x11, glib, gtk, most c++ libs) but many have alternatives such as musl-libc, tinyx11, etc...
If you use link time optimization (or similar), the size difference in the binaries is often negligible (and sometimes smaller than the shared binary)
Since the system doesn't need to load up additional shared libraries, the initial start time of static builds is often much quicker.
Static builds can work on different distros regardless of what packages are installed.
Dynamic library updates don't break static binaries (important for system repair tools).
Updates to static binaries are only necessary iff it uses the library code that required a CVE.
The updates for static builds could take advantage of binary diff tools to minimize network traffic - it's usually pretty minimal if the build systems don't change.
Anecdote:
Back when I was working on Puppy Linux (4.x timeframe), we used a hybrid method in the main iso where any library that was only used once was built in statically, but that was based on iso size constraints - static builds of single use libs reduced the iso size by 10-20%.
2 • Dynamic linking & static linking (by Friar Tux on 2020-08-10 01:58:22 GMT from Canada)
Isn't that the whole idea behind Flatpak, Snaps, and/or AppImages? And don't developers benefit in that it reduces the amount of work needed to satisfy the various distros? I've come to the point where I'm really not worried about duplication of libraries on my system, thanks to the amount of memory most laptops come with now-a-days. (This may be from my Windows years as Windows installs each app/program into its own folder WITH all the libraries needed. There was plenty of duplicating there.)
3 • Grub update issue (by Bin on 2020-08-10 04:22:22 GMT from United Kingdom)
Running Debian SparkyLinux 5.12 with unattended upgrades.
Normally UA handles grub updates - this time it didn't.
Dumped out at grub rescue prompt
Used rescatux to correct grub and all well. grub is now on the exception list along with xserver.
4 • Static vs dynamic linking (by Alexandru on 2020-08-10 07:37:14 GMT from Austria)
First of all, congrats for insightful article.
Static linking and dynamic linking both have their use cases.
1. Size. Static linking only gets those bits of a library which are actually used, dynamic linking always provides whole library. On the other hand, dynamic linking takes the storage space only once, while static linking takes space as many times as many different programs use that library.
2. Loading time. Statically linked programs usually load faster than dynamically linking ones except the case when the dynamic library is already loaded into memory by some other application that links against it. In this case new dynamically linked application (whose size is smaller than that of statically linked one) is loaded faster.
3. Uniformity. The same application which is statically linked to different versions of the same library, or different libraries that export the same symbols, may look or behave differently. And this is counter-intuitive. Additionally static linking creates problems with the size of updates and shortness of time they are available.
In conclusion:
- Static linking use cases:
1. The library is small
2. The library is used by few programs
- Dynamic linking use case are all the rest.
5 • Static vs dynamic linking (by Robert_M on 2020-08-10 09:17:35 GMT from Canada)
Historically, dynamic linking was used over static linking in order to reduce the memory and disk space footprint on a system.
When I started my career in computers, a 10 megabyte Winchester disk drive was several thousand dollars and memory was $1 per word. This was for unix based systems. I had a university professor who said that times were great as he remembered memory being $1 per bit. People of my generation actually remember real "core" memory.
It was not unusual to have systems with 1-2 K of memory and 20-40 mbytes of storage. We were memory constrained at the time. A system with 4gig of storage was a large system.
Dynamic linking reduced the need for expensive upgrades but came at a cost. When I worked for an R&D firm that had several thousand unix boxes, it would take about 6 months to a year to ensure that all applications were converted over to work with a new release. A very painful process.
This was at the time when tcp/ip networks were fairly new and the internet was a collection of statically connected sites. We initially had two diskless workstations configured to a single disk based system to reduce deployment costs.
Times were different back then. Today, with current memory and storage prices , it makes sense to come full circle and look static linking to take advantage of its simplicity. As an end user, I am constantly upgrading and I think of my systems as a collection of packages and not libraries. I feel that anything that makes a package self-contained is a good thing.
I am having a blast working with raspberry pi s that have the same power and storage of $30k workstation in the past for sub $100.
My apologies for the long rant, but to understand where we are, we need to understand where we've come from. Context is everything.
Now back to lurking.....
6 • Zentyal (by Bernard Fuller on 2020-08-10 09:28:17 GMT from United Kingdom)
I tried Zentyal in the past and never felt comfortable with it. I seem to remember that one server distribution I tried didn't include a Mail Server, was it this one??
I've been using ClearOS now for several years. If you get the chance, or have the energy, I'd be interested to hear what you make of that. Apart from their forum, I sometimes feel I'm working in a vacuum with it.
Some other independent view would be interesting.
7 • Static vs Dynamic Linking (by Newby on 2020-08-10 10:12:50 GMT from Canada)
@5 Enjoyed reading about your experience with "period era" hardware. Brought back painful memories of trying to input data on a Data General computer with 4K of core memory, via 8 toggle switches on the front panel.
For a more "excruciating" experience, there was inputting data into an IBM mainframe via punch cards. It could be a while before you got the results and found out if you were successful or not, and then try and remember what you were trying to do so you could "troubleshoot". And that was assuming the mechanics were aligned properly so the cards didn't "jam up".
And who could forget the joys of waiting ages for a cassette tape program to finish loading a simple game on a Commodore or Atari system (as well as Apple, Sinclair/Timex, BBC "Micro", etc.)
At the very end of your history lesson about "linking", you intrigued as with a reference about "lurking". Would be rather interesting if you could expound upon that. So many nefarious possibilities inferred by that last line..... :)
8 • LMDE (by OstroL on 2020-08-10 10:23:44 GMT from Poland)
>> LMDE is small but that’s not particularly relevant. It doesn’t get point-releases, it’s not promoted or given the same exposure, and its purpose isn’t to compete with Linux Mint or to attract new users. It’s developed as a plan B in case we need to switch package base one day. It could be seen as a costly investment but it’s strategically important to our project. It tells us exactly how much we rely on Ubuntu, how well we can do without it, and how much work would be involved if we had to stop using it.
9 • grub update issue (by CatInAHat on 2020-08-10 11:14:55 GMT from Australia)
on 'extra' OS, MX with grub on root Partition: update used grub-install (as a tui) to ask which location to update/install grub - eh? chose my current location of course... next boot gave me grub rescue prompt :P had to use SystemRescueCd to boot OS, repaired grub with MX Boot Repair tool
primary OS, MX with grub on mbr: same process, reboot = fine...
secondary OS, Artix, no issue...
10 • Static linking (by bison on 2020-08-10 14:14:33 GMT from United States)
One class of programs that would benefit from static linking are games.
I like to play older versions of SuperTuxKart (before they changed the game physics), but these will not run on modern Linux systems because they depend on obsolete shared libraries, which cannot be installed because they conflict with other libraries on the system.
I keep a copy of Ubuntu 14.04 around just for games.
11 • Static vs Dynamic (by Nathan on 2020-08-10 14:15:11 GMT from United States)
Static vs dynamic to me is a question of how I want abandonware to fail. As a prolific author of abandonware myself, it always has bugged me how android apps that I released never to update again would benefit from security patches if only I re-compiled them against updated libraries (of course, fixing breakages along the way). Sure they continue working (mostly), but they quickly become vulnerable to attacks because of the vulnerable libraries that are baked in. Same applies to my flatpak apps.
On the other hand, whenever I attempt to launch my dynamically-linked abandonware after a year or so of not touching it, I find that it errors out and I have to fix a few broken library calls first.
In the end, for me it boils down to security. If I want to be forced to have good security, dynamic linking is better. If I just want to run the program, never mind that it exposes me to an ever-expanding list of vulnerabilities, then static linking allows me that luxury. Unfortunately for me, I'm both lazy and a security minded person, so I live in constant guilt and worry about my statically-linked programs betraying me. ;)
12 • Linking, and file hash algorithms (by RJA on 2020-08-10 14:31:09 GMT from United States)
Well, over at the PaleMoon community, using the system libraries instead of libraries that are bundled with the package is a big no-no, because of outdated/wrong library versions.
Well, looks like Microsoft made the move to SHA-2 and considers even SHA-1 a risk.
SHA-1 has been considered a risk for a while, but has been tolerable.
So, I was horrified to see MD5! (Including Zentyal)
-RJA
13 • linking in programs (by Myrtle on 2020-08-10 15:07:03 GMT from United States)
https://stackoverflow.com/questions/1993390/static-linking-vs-dynamic-linking
That page said it well and ended it well with a lot of "it depends" in answer to questions about performance.
Test test test is the takeaway, and in an environment as close to the real world working conditions as possible.
14 • Dynamically linked programs (by Tim on 2020-08-10 15:13:51 GMT from United States)
Jesse, that is an excellent paragraph about the benefits of dynamically linked programs. I have known that for many years (I started using UNIX in the early 80s), but it is great to get such a clear summary. Thank you.
FWIW, for the same reasons, I will avoid Flatpaks and Snaps for as long as possible. If they back me into a corner, I will switch to BSD.
15 • Linux Mint LMDE (by barnabyh on 2020-08-10 16:55:44 GMT from Germany)
Shame only 1% of Mint users run LMDE. It's exceptionally polished and runs really well without as much overhead, it could easily be the main edition. No idea what it is supposedly missing when compared to the Ubuntu Mint. Ok, I only used that one briefly about 11 years ago. Perhaps the hardware driver assistant I read about but how many people actually need that? Most is well supported ootb and adding the nVidia driver is not that hard.
Synaptic shows me 60010 packages available so it can't be availability of packages either.
Hope they don't discontinue it, it's the only Mint I would even consider. Running well for 16 months here incl. the upgrade.
16 • Linux Mint (by narly on 2020-08-10 18:48:53 GMT from United States)
I recently updated from Linux Mint 19 to 20. I wanted to wipe my hard drives and start over again from scratch. I also wanted to switch to the Xfce desktop. I gave up on MATE because it started to get buggy. Cinnamon is OK, but I like the speed and simplicity of Xfce better now. I did try LMDE4, but had intermittant wifi problems with my laptop. Now I'm happily running Linux Mint 20 Xfce on both my laptop and desktop.
17 • Linux Mint (by yrotadnam on 2020-08-10 20:59:06 GMT from New Zealand)
@8,@15 - I see LMDE as no more than an insurance policy by Mint. Rather unpolished in its current state, but important as Canonical could be bought or become unpredictable.
@17 - Cinnamon is good though I see some crashes and hangups. MATE is again in my experience more stable, has wobbly windows, but is a bit dated. I have tried Xfce but find it lacking after Cinnamon's powerhouse of ease-of-use, particularly things like file managers are where many DE's fall down. KDE have improved much over time, but Mint dropped KDE. You can add it, but its 4.1 which is ancient and rubbish compared to the new 5.8 or whatever you would get with Manjaro.
my 10c, YMMV
18 • Mint KDE, @17 (by WhatMeWorry on 2020-08-11 01:03:39 GMT from United States)
"You can add it, but its 4.1"
You can easily add KDE to Mint 20 by enabling Kubuntu backports, and it runs well. Currently it's 5.18.5. (Manjaro is on 5.19.3) Question is, why? If staying in the Ubuntu family, there's Kubuntu; and there's neon, which has the latest. I ran Mint for many years. Had many advantages for me, but in the last few years, the differences between Mint and others have become minor. I just choose a distro that offers the desktop I want.
19 • @18-- Mint. (by R. Cain on 2020-08-11 23:30:04 GMT from United States)
"...Question is, why?...in the last few years, the differences between Mint and others have become minor..."
It's been a very long time since there was any compelling reason to consider using Mint.
20 • LMDE (by other Tim on 2020-08-12 00:20:36 GMT from United States)
I need to push back a little on this idea that LMDE is just a backup plan for Mint and that it's unpolished.
I happily used LMDE 2 for years. There was nothing unpolished about it. The only reason newer LMDEs don't interest me is that I use MATE.
LMDE is a nice option because it's simple to set up a system with it, and by enabling Debian backports you can have relatively fresh packages on a very stable base for many years. There have been many times I've been wishing for a "Ubuntu Backports" but that role is taken by PPAs and sometimes its hard to see who you can trust.
Debian and LMDE are not unpolished. If you take the time to gain even basic knowledge of APT, it's more intuitive and much faster than any of these software stores that people think they need. I use all members of the Debian family (including Ubuntu and Mint) but I think it's been more than six years since I bothered opening a software store. My copies of Ubuntu or Mint are managed the same way my copies of Debian are.
21 • LMDE, @20 (by WhatMeWorry on 2020-08-12 09:32:41 GMT from United States)
"I need to push back a little on this idea that LMDE is just a backup plan for Mint" You'd be pushing back against the Mint devs, since they are the ones saying that. As for being unpolished, it seems OK to me, but I'm not a regular user.
"relatively fresh packages" Debian backports are usually taken from "testing", so "fresh" is definitely relative. Ubuntu does have backports, usually enabled by default.
I don't use the app stores much either, but "basic knowledge of APT" is no substitute for Synaptic. (And vice versa.)
22 • In the last few years... (by Friar Tux on 2020-08-12 13:36:10 GMT from Canada)
@19 (R Cain) "...in the last few years, the differences between Mint and others have become minor..." Some of us see that as a good thing. So long as those others work as well and as problem-free as Mint has worked (for me). Also, it's to be expected. With the number of distros out there, and each one 'improving' with each version, it stands to reason they will soon all meld together. While I like having the choice of hundreds of distros, I can foresee a time when they will all meld into one 'super-distro'. The other route is that we will end up with five or six different versions of Linux and/or BSD, with hundreds of distros in each version. But then, that's freedom for you. (It reminds me of the auto industry. While it only takes four wheels and a motor to get you from point A to point B, still, look how many, many models of vehicles we have come up with.)
23 • Grub Boot(black)hole (by cykodrone on 2020-08-12 19:16:31 GMT from France)
No, my humble dual distro lappy was not affected (Devuan grub), but I must say, that's embarrassing, I bet some faces were as red as RH's logo, eek. :/
24 • @20--"Backup Plan" (by R. Cain on 2020-08-13 13:02:15 GMT from United States)
"I need to push back a little on this idea that LMDE is just a backup plan for Mint..."; "...I happily used LMDE 2 for years..." [this last lends *no* credibility to the commenter's objection; it is NOT ABOUT you, folks]
One needs to be aware of history, to not fall victim to the "Facebook Syndrome" of 'putting it out there' without doing any thinking OR research, before rushing to "...push back..."; to wit--
Clement Lefebvre *HIMSELF* said a few years ago, in one of Mint's blogs or forums, that yes, indeed, the entire, SOLE purpose of developing LMDE was as an insurance policy" against any untoward circumstances which might occur if Mint were absolutely, only tied and committed to Ubuntu.
Push back all you like, folks. Know what you're talking about when you do.
25 • Lmde (by Tim on 2020-08-14 02:07:27 GMT from United States)
I know exactly what I'm talking about: that LMDE is a good distro that gave me a problem free two years.
Clem and Mint are super clear why they make LMDE. My argument is not with them- it's with those voices here dismissing LMDE as unpolished and not worth anything other than this backup plan. If you like Cinnamon and know how to use apt, this might be the distro for you.
26 • LMDE & Mint (by M.Z. on 2020-08-14 20:12:36 GMT from United States)
I've had a lot of good years with LMDE, and it has seemed fairly solid all around to me. It's got plenty of good polish & is very stable, while offering the friendly & easy sort of desktop computing users expect out of Mint. If you use any sort of general PC desktop like KDE/Windows then LMDE could be an excellent distro for you.
The only thing LMDE sort of lacks for me now is the level BTRFS + Timeshift integration in the update manager in LMDE 4 verses LMDE 3. In the previous release I could use Calamares to more easily setup partitions as BTRFS & get the whole thing up & running & doing daily snapshots. There was no need to ever use my daily or weekly BTRFS snaps shots in LMDE 3 give how rock solid stable it was, but that level of stability + a back up plan in a dirt simple desktop distro was very compelling to me. I gave up on additional trouble shooting of getting LMDE4 on BTRFS to load via my Grub customizer in Mageia, though it may well work as well in the main edition as it did for me in LMDE 3.
I could wax poetic about how quickly Mint integrated flatpak into their nice & easy software manager, how long they offered BTRFS + Timeshift before Ubuntu started playing with experimental ZFS support, or how they continue to consistently improve their easy to use set of desktop options. Regardless of all that or how many more details could be given, Mint is one of if not the biggest community based Linux distros & one of the most widely used distros as well, and given how good an alternative it is for windows I can see many compelling reasons why.
27 • Mint (by Friar Tux on 2020-08-14 22:32:28 GMT from Canada)
@26 (MZ) I agree 100%. In fact, Mint is so trouble-free (four years for me) that the joke in my family is that I have to play with other distros to keep my Linux skills up to par. Clem and gang have got one rock solid OS there (LMDE included). Plus, if you don't care for the default polish, you can easily change that. Just hope over to pling.org have have fun. (I'm interpreting polish as the visuals. I could be wrong here.)
28 • SOHO server distros (by Ankleface Wroughlandmire on 2020-08-14 23:16:37 GMT from Ecuador)
Zentyal looks quite underwhelming. If you have hardware that supports it I'd always recommend TrueNAS as the first option. Or if you prefer something Linux-based and/or have less-common or even extremely uncommon hardware, then OpenMediaVault is your solution. It's Debian-based, and can be installed directly as a distro on most normal hardware. But if you have something more arcane you can install it on top of a regular Debian install. Proof in point, I found a site that let me install Debian on an old Lacie NAS with an armv5 processor, and now it runs OpenMediaVault just fine.
29 • "...trouble free.." (by Otis on 2020-08-15 12:16:00 GMT from United States)
@27 I agree but about MX, not Mint. Same thing except it's been only two years (or there abouts) for me. And minus the systemd.
Number of Comments: 29
Display mode: DWW Only • Comments Only • Both DWW and Comments
| | |
| TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite
Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
| *NEW* NovaCustom |
NovaCustom PrivacyGuard Laptops - Escape from Big Tech
The NovaCustom PrivacyGuard Laptop is ideal for anyone who prioritizes privacy. Comes with Dasharo coreboot open source firmware and Zorin OS Pro, free from influence of Big Tech.
|
| Archives |
| • Issue 1169 (2026-04-20): Lakka 6.1, free software and source-based distributions, FreeBSD Foundation publishes compatible laptop list, Debian holds Project Leader election, Haiku progresses ARM64 port, Mint to extend development cycle, Linux 7.0 released |
| • Issue 1168 (2026-04-13): pearOS 2026.03, EndeavourOS 2026.03.06, which distros are adopting age verification, Arch adjusts its firewall packages, Linux dropping i486 support, Red Hat extends its release cycle, Debian's APT introduces rollbacks, Redox improves its scheduler |
| • Issue 1167 (2026-04-06): Origami Linux 2026.03, answering questions for Linux newcomers, Ubuntu MATE seeking new contributors, Ubuntu software centre is expanding Deb support, FreeBSD fixes forum exploit, openSUSE 15 Leap nears its end of life |
| • Issue 1166 (2026-03-30): NetBSD jails, publishing software for Linux, Ubuntu joins Rust Foundation, Canonical plans to trim GRUB features, Peppermint works on new utilities, PINE64 shows off open hardware capabilities |
| • Issue 1165 (2026-03-23): Argent Linux 1.5.3, disk space required by Linux, Manjaro team goes on strike, AlmaLinux improves NVIDIA driver support and builds RISC-V packages, systemd introduces age tracking |
| • Issue 1164 (2026-03-16): d77void, age verification laws and Linux, SUSE may be for sale, TrueNAS takes its build system private, Debian publishes updated Trixie media, MidnightBSD and System76 respond to age verification laws |
| • Issue 1163 (2026-03-09): KaOS 2026.02, TinyCore 17.0, NuTyX 26.02.2, Would one big collection of packages help?, Guix offers 64-bit Hurd options, Linux communities discuss age delcaration laws, Mint unveils new screensaver for Cinnamon, Redox ports new COSMIC features |
| • Issue 1162 (2026-03-02): AerynOS 2026.01, anti-virus and firewall tools, Manjaro fixes website certificate, Ubuntu splits firmware package, jails for NetBSD, extended support for some Linux kernel releases, Murena creating a map app |
| • Issue 1161 (2026-02-23): The Guix package manager, quick Q&As, Gentoo migrating its mirrors, Fedora considers more informative kernel panic screens, GhostBSD testing alternative X11 implementation, Asahi makes progress with Apple M3, NetBSD userland ported, FreeBSD improves web-based system management |
| • Issue 1160 (2026-02-16): Noid and AgarimOS, command line tips, KDE Linux introduces delta updates, Redox OS hits development milestone, Linux Mint develops a desktop-neutral account manager, sudo developer seeks sponsorship |
| • Issue 1159 (2026-02-09): Sharing files on a network, isolating processes on Linux, LFS to focus on systemd, openSUSE polishes atomic updates, NetBSD not likely to adopt Rust code, COSMIC roadmap |
| • Issue 1158 (2026-02-02): Manjaro 26.0, fastest filesystem, postmarketOS progress report, Xfce begins developing its own Wayland window manager, Bazzite founder interviewed |
| • Issue 1157 (2026-01-26): Setting up a home server, what happened to convergence, malicious software entering the Snap store, postmarketOS automates hardware tests, KDE's login manager works with systemd only |
| • Issue 1156 (2026-01-19): Chimera Linux's new installer, using the DistroWatch Torrent Corner, new package tools for Arch, Haiku improves EFI support, Redcore streamlines branches, Synex introduces install-time ZFS options |
| • Issue 1155 (2026-01-12): MenuetOS, CDE on Sparky, iDeal OS 2025.12.07, recommended flavour of BSD, Debian seeks new Data Protection Team, Ubuntu 25.04 nears its end of life, Google limits Android source code releases, Fedora plans to replace SDDM, Budgie migrates to Wayland |
| • Issue 1154 (2026-01-05): postmarketOS 25.06/25.12, switching to Linux and educational resources, FreeBSD improving laptop support, Unix v4 available for download, new X11 server in development, CachyOS team plans server edtion |
| • Issue 1153 (2025-12-22): Best projects of 2025, is software ever truly finished?, Firefox to adopt AI components, Asahi works on improving the install experience, Mageia presents plans for version 10 |
| • Issue 1152 (2025-12-15): OpenBSD 7.8, filtering websites, Jolla working on a Linux phone, Germany saves money with Linux, Ubuntu to package AMD tools, Fedora demonstrates AI troubleshooting, Haiku packages Go language |
| • Issue 1151 (2025-12-08): FreeBSD 15.0, fun command line tricks, Canonical presents plans for Ubutnu 26.04, SparkyLinux updates CDE packages, Redox OS gets modesetting driver |
| • Issue 1150 (2025-12-01): Gnoppix 25_10, exploring if distributions matter, openSUSE updates tumbleweed's boot loader, Fedora plans better handling of broken packages, Plasma to become Wayland-only, FreeBSD publishes status report |
| • Issue 1149 (2025-11-24): MX Linux 25, why are video drivers special, systemd experiments with musl, Debian Libre Live publishes new media, Xubuntu reviews website hack |
| • Issue 1148 (2025-11-17): Zorin OS 18, deleting a file with an unusual name, NetBSD experiments with sandboxing, postmarketOS unifies its documentation, OpenBSD refines upgrades, Canonical offers 15 years of support for Ubuntu |
| • Issue 1147 (2025-11-10): Fedora 43, the size and stability of the Linux kernel, Debian introducing Rust to APT, Redox ports web engine, Kubuntu website off-line, Mint creates new troubleshooting tools, FreeBSD improves reproducible builds, Flatpak development resumes |
| • Issue 1146 (2025-11-03): StartOS 0.4.0, testing piped commands, Ubuntu Unity seeks help, Canonical offers Ubuntu credentials, Red Hat partners with NVIDIA, SUSE to bundle AI agent with SLE 16 |
| • Issue 1145 (2025-10-27): Linux Mint 7 "LMDE", advice for new Linux users, AlmaLinux to offer Btrfs, KDE launches Plasma 6.5, Fedora accepts contributions written by AI, Ubuntu 25.10 fails to install automatic updates |
| • Issue 1144 (2025-10-20): Kubuntu 25.10, creating and restoring encrypted backups, Fedora team debates AI, FSF plans free software for phones, ReactOS addresses newer drivers, Xubuntu reacts to website attack |
| • Issue 1143 (2025-10-13): openSUSE 16.0 Leap, safest source for new applications, Redox introduces performance improvements, TrueNAS Connect available for testing, Flatpaks do not work on Ubuntu 25.10, Kamarada plans to switch its base, Solus enters new epoch, Frugalware discontinued |
| • Issue 1142 (2025-10-06): Linux Kamarada 15.6, managing ZIP files with SQLite, F-Droid warns of impact of Android lockdown, Alpine moves ahead with merged /usr, Cinnamon gets a redesigned application menu |
| • Issue 1141 (2025-09-29): KDE Linux and GNOME OS, finding mobile flavours of Linux, Murena to offer phones with kill switches, Redox OS running on a smartphone, Artix drops GNOME |
| • Issue 1140 (2025-09-22): NetBSD 10.1, avoiding AI services, AlmaLinux enables CRB repository, Haiku improves disk access performance, Mageia addresses service outage, GNOME 49 released, Linux introduces multikernel support |
| • Issue 1139 (2025-09-15): EasyOS 7.0, Linux and central authority, FreeBSD running Plasma 6 on Wayland, GNOME restores X11 support temporarily, openSUSE dropping BCacheFS in new kernels |
| • Issue 1138 (2025-09-08): Shebang 25.8, LibreELEC 12.2.0, Debian GNU/Hurd 2025, the importance of software updates, AerynOS introduces package sets, postmarketOS encourages patching upstream, openSUSE extends Leap support, Debian refreshes Trixie media |
| • Issue 1137 (2025-09-01): Tribblix 0m37, malware scanners flagging Linux ISO files, KDE introduces first-run setup wizard, CalyxOS plans update prior to infrastructure overhaul, FreeBSD publishes status report |
| • Issue 1136 (2025-08-25): CalyxOS 6.8.20, distros for running containers, Arch Linux website under attack,illumos Cafe launched, CachyOS creates web dashboard for repositories |
| • Issue 1135 (2025-08-18): Debian 13, Proton, WINE, Wayland, and Wayback, Debian GNU/Hurd 2025, KDE gets advanced Liquid Glass, Haiku improves authentication tools |
| • Issue 1134 (2025-08-11): Rhino Linux 2025.3, thoughts on malware in the AUR, Fedora brings hammered websites back on-line, NetBSD reveals features for version 11, Ubuntu swaps some command line tools for 25.10, AlmaLinux improves NVIDIA support |
| • Issue 1133 (2025-08-04): Expirion Linux 6.0, running Plasma on Linux Mint, finding distros which support X11, Debian addresses 22 year old bug, FreeBSD discusses potential issues with pkgbase, CDE ported to OpenBSD, Btrfs corruption bug hitting Fedora users, more malware found in Arch User Repository |
| • Issue 1132 (2025-07-28): deepin 25, wars in the open source community, proposal to have Fedora enable Flathub repository, FreeBSD plans desktop install option, Wayback gets its first release |
| • Issue 1131 (2025-07-21): HeliumOS 10.0, settling on one distro, Mint plans new releases, Arch discovers malware in AUR, Plasma Bigscreen returns, Clear Linux discontinued |
| • Issue 1130 (2025-07-14): openSUSE MicroOS and RefreshOS, sharing aliases between computers, Bazzite makes Bazaar its default Flatpak store, Alpine plans Wayback release, Wayland and X11 benchmarked, Red Hat offers additional developer licenses, openSUSE seeks feedback from ARM users, Ubuntu 24.10 reaches the end of its life |
| • Issue 1129 (2025-07-07): GLF OS Omnislash, the worst Linux distro, Alpine introduces Wayback, Fedora drops plans to stop i686 support, AlmaLinux builds EPEL repository for older CPUs, Ubuntu dropping existing RISC-V device support, Rhino partners with UBports, PCLinuxOS recovering from website outage |
| • Issue 1128 (2025-06-30): AxOS 25.06, AlmaLinux OS 10.0, transferring Flaptak bundles to off-line computers, Ubuntu to boost Intel graphics performance, Fedora considers dropping i686 packages, SDesk switches from SELinux to AppArmor |
| • Issue 1127 (2025-06-23): LastOSLinux 2025-05-25, most unique Linux distro, Haiku stabilises, KDE publishes Plasma 6.4, Arch splits Plasma packages, Slackware infrastructure migrating |
| • Issue 1126 (2025-06-16): SDesk 2025.05.06, renewed interest in Ubuntu Touch, a BASIC device running NetBSD, Ubuntu dropping X11 GNOME session, GNOME increases dependency on systemd, Google holding back Pixel source code, Nitrux changing its desktop, EFF turns 35 |
| • Issue 1125 (2025-06-09): RHEL 10, distributions likely to survive a decade, Murena partners with more hardware makers, GNOME tests its own distro on real hardware, Redox ports GTK and X11, Mint provides fingerprint authentication |
| • Issue 1124 (2025-06-02): Picking up a Pico, tips for protecting privacy, Rhino tests Plasma desktop, Arch installer supports snapshots, new features from UBports, Ubuntu tests monthly snapshots |
| • Issue 1123 (2025-05-26): CRUX 3.8, preventing a laptop from sleeping, FreeBSD improves laptop support, Fedora confirms GNOME X11 session being dropped, HardenedBSD introduces Rust in userland build, KDE developing a virtual machine manager |
| • Issue 1122 (2025-05-19): GoboLinux 017.01, RHEL 10.0 and Debian 12 updates, openSUSE retires YaST, running X11 apps on Wayland |
| • Issue 1121 (2025-05-12): Bluefin 41, custom file manager actions, openSUSE joins End of 10 while dropping Deepin desktop, Fedora offers tips for building atomic distros, Ubuntu considers replacing sudo with sudo-rs |
| • Issue 1120 (2025-05-05): CachyOS 250330, what it means when a distro breaks, Kali updates repository key, Trinity receives an update, UBports tests directory encryption, Gentoo faces losing key infrastructure |
| • Issue 1119 (2025-04-28): Ubuntu MATE 25.04, what is missing from Linux, CachyOS ships OCCT, Debian enters soft freeze, Fedora discusses removing X11 session from GNOME, Murena plans business services, NetBSD on a Wii |
| • Issue 1118 (2025-04-21): Fedora 42, strange characters in Vim, Nitrux introduces new package tools, Fedora extends reproducibility efforts, PINE64 updates multiple devices running Debian |
| • Full list of all issues |
| Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
| Random Distribution | 
iBox
iBox was a highly customised and flexible live CD based on Gentoo Linux. Thanks to glc (the Chinese branch of gentoo portage), iBox provides an all-round Chinese (simplified) desktop environment using GNOME with almost all pre-configured popular software. The main feature of iBox was the auto-detection and auto-configuration of hardware, especially with the mkxorgconf script to help create the configuration file for Xorg. Last but not least, iBox can rebuild itself through ibox-builder from a Gentoo box.
Status: Discontinued
|
| TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite
Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
| Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
|
• 
• 
• 
• 
• 
• 
• 
