Fedora To Get a New Partition Manager

← Back to Stories (view on slashdot.org)

Fedora To Get a New Partition Manager

Posted by timothy on Sunday September 7, 2014 @11:01AM from the nostalgia-is-a-hard-force-to-overcome dept.

sfcrazy writes Developer Vratislav Podzimek has announced the next-gen partition manager for Fedora, blivet-gui. It is eventually going to replace GParted, the most popular GUI based partition manager, found in all major distros. The new tool is named blivet-gui after the blivet python library (originally Anaconda's storage management and configuration tool). The need of a new partition manager stems from the fact that none of the existing GUI partitioning tools supports all modern storage technologies. Fedora's Anaconda base supports all, though, and is hence chosen as the back-end for this new tool. The application is only a few months old but is already looking nice and useful. Features like RAID and BTRFS support are being worked on. Vojtech Trefny is the other developer working with Vratislav on blivet-gui. Here's the announcement.

28 of 170 comments (clear)

Min score:

Reason:

Sort:

So.... by Anonymous Coward · 2014-09-07 11:03 · Score: 4, Insightful

Fedora is going to replace GParted none of the existing GUI partitioning tools supports all modern storage technologies. Theyre replacing it with blivet-gui which doesnt support features like RAID and BTRFS.
That hat too tight?
1. Re:So.... by Jeremiah+Cornelius · 2014-09-07 11:39 · Score: 2, Funny
  
  Are you the Judean People's Front?
  Listen. The only people we hate more than the Romans are the fucking Judean People's Front.
  P.F.J.: Yeah...
  JUDITH: Splitters.
  P.F.J.: Splitters...
  FRANCIS: And the Judean Popular People's Front.
  P.F.J.: Yeah. Oh, yeah. Splitters. Splitters...
  LORETTA: And the People's Front of Judea.
  P.F.J.: Yeah. Splitters. Splitters...
  
  --
  "Flyin' in just a sweet place,
  Never been known to fail..."
2. Re:So.... by ysth · 2014-09-07 13:00 · Score: 3, Insightful
  
  No, we need such a command line tool or possibly library with a command line tool wrapped around it. The GUI is entirely optional and certainly shouldn't be bundled.
3. Re: So.... by Guy+Smiley · 2014-09-07 15:15 · Score: 5, Interesting
  
  In typical open source fashion, their replacing a tool (GParted) that doesn't support a few features they want with a new one that (at least initially) didn't support _any_ features at all because it was written from scratch.
  Why not just fix GParted to add the few missing features instead of writing a completely new too? The new one will of course itself not support all the features GParted had, but instead be chok full of new bugs that will take years to find and fix...
  Why is it that everyone wants to reinvent the wheel instead of using and improving the tools we already have?
4. Re: So.... by bored · 2014-09-07 16:09 · Score: 5, Insightful
  
  I considered moderating you, but I think this is really a case of <whine> "C++ is haaardddd, learning it enough to understand how to plug in a new module is going to take me months. Instead I'm going to rewrite it" </whine>
  Or similar bullshit by people who think "scripting" languages are appropriate for base system tools. Now you will have python dependency hell every-time you want to do something simple like repartition your disks. Oh, and is that project python 2 or python 3? On and on..
  Frankly, its fsking stupid and its another sign that redhat is jumping the shark.
  Plus, do you really want to depend on the skills of some "leet" hacker that thinks python is an appropriate tool for this?
5. Re: So.... by thegarbz · 2014-09-07 20:37 · Score: 3, Insightful
  
  In typical open source fashion, their replacing a tool (GParted) that doesn't support a few features they want with a new one that (at least initially) didn't support _any_ features at all because it was written from scratch.
  Why not just fix GParted to add the few missing features instead of writing a completely new too? The new one will of course itself not support all the features GParted had, but instead be chok full of new bugs that will take years to find and fix...
  Why is it that everyone wants to reinvent the wheel instead of using and improving the tools we already have?
  In the typical open source fashion people thinking they're experts will blindly criticise someone's decision without understanding it. How about you start with blivet-gui is not a partition manager and then work onwards from there with your understanding.
  Blivet-gui is a standalone implementation of the storage manager used during the install process. Yes it can partition, and in the true open source fashion it uses another program to do so (parted), but that's a small subset of what they want to use it for which is more like be a one stop shop for all disk management, volume management, and RAID management.
  Please put the pitchfork away.
6. Re: So.... by DrXym · 2014-09-07 22:27 · Score: 3, Informative
  
  Or similar bullshit by people who think "scripting" languages are appropriate for base system tools. Now you will have python dependency hell every-time you want to do something simple like repartition your disks. Oh, and is that project python 2 or python 3? On and on..
  gparted is a graphical tool for editing partitions and already has a raft of dependencies. One more won't make a difference especially since python is used increasingly in core distributions for scripting instead of bash.
  Secondly, perhaps the reason that gparted is considered a mess is precisely because it mixes up the graphical parts and the low level stuff in one package, a problem compounded because the installer also has its own partition editor. Fedora appears to have written a layer called blivet to abstract out partitioning from the installer GUI and therefore it makes sense that they use it in the desktop also.
7. Re: So.... by gilboad · 2014-09-08 00:43 · Score: 2
  
  God knows why anyone moderated your comment as insightful.
  1. Fedora / RedHat is targeting, wait for it, Fedora / RedHat. End of story.
  As someone that maintains a fairly large DPI (kernel/user-land) software and a management software that uses PyGI, I can from attest from personal experience that both PyGI and PyGTK under both Fedora and RHEL are top-notch with zero dependencies issues.
  2. PyGI is *far* easier to use than Qt/C++. We wrote the original management code in Qt, but switched to PyGI as it increased that rate of the development by 3/1. We may switch back to Qt due to performance issues and lackluster mutli-threading support (Known python limitation) and problematic Windows support. (Both of which are non-issues when it comes to *Linux* storage management)
  3. I can agree that if you want to scan incoming packets at 150Gbps using software only, or write output at 1GB/s, you'd most likely need C++ or even C/asm combination. But what-exactly prevents python from calling a couple of syscalls to executing /sbin/cryptsetup / /sbin/lvm or /sbin/mkfs.ext4? What exactly makes Qt/C++ better at calling external binaries (such as the one mentioned above)?
  - Gilboa
The origin of the term "blivet" by Nimey · 2014-09-07 11:07 · Score: 3, Informative

Ten pounds of shit in a five-pound bag, i.e. a nasty, dirty situation. It seems to have originated around the 1940s as US military slang.

--
Hail Eris, full of mischief...

E pluribus sanguinem
1. Re:The origin of the term "blivet" by Anonymous Coward · 2014-09-07 11:16 · Score: 3, Funny
  
  Ten pounds of shit in a five-pound bag,
  
  A fantastic deal! if you're a dung beetle.
2. Re:The origin of the term "blivet" by Anonymous Coward · 2014-09-07 11:50 · Score: 4, Informative
  
  http://en.wikipedia.org/wiki/Blivet
  It is a poiuyt, devil's fork or widget, is an undecipherable figure, an optical illusion and an impossible object. It appears to have three cylindrical prongs at one end which then mysteriously transform into two rectangular prongs at the other end.
Damn the GUI! by turbidostato · 2014-09-07 11:08 · Score: 2

"The need of a new partition manager stems from the fact that none of the existing GUI partitioning tools supports all modern storage technologies"
Does the backend -and kickstart, support all those "modern storage technologies"?
That's the important part. For a GUI-based installation, the ability to format -and install into, a single root partition is good enough for me.
1. Re:Damn the GUI! by phantomfive · 2014-09-07 11:20 · Score: 4, Interesting
  
  Yeah, another example of NIH coming from RedHat.
  
  --
  "First they came for the slanderers and i said nothing."
Why not contribute to gparted? by sayfawa · 2014-09-07 11:12 · Score: 4, Insightful

Instead of making another program, I wonder what was wrong with sharing the code with gparted so that they could incorporate support for more filesystems?

TFA didn't say if that option had been explored.

--
Free the Quark 3 from asymptotic confinement! Bring your charm! Don't get down! All colours and flavours welcome!
1. Re:Why not contribute to gparted? by Burdell · 2014-09-07 12:39 · Score: 5, Informative
  
  Because (as usual) the summary got it wrong. This is not a partition manager, it is a disk/filesystem manager. Partitions make up one part of that, but it is also intended to manage LVM, RAID, btrfs filesets, etc. I believe it uses the parted library on the backend for partitions.
  This is based on the years-of-development code used in the backend of anaconda, the Fedora/Red Hat installer. The code has been pulled out, split up into a library, and set up for stand-alone use (after install). I believe the intention is that anaconda keeps using the library, but now there will be the same interface during install and afterwards for managing disks and filesystems.
Re:fuck up your disks with greater ease! by Anonymous Coward · 2014-09-07 11:27 · Score: 2, Informative

And aimed at the same audience.
Idiots by Anonymous Coward · 2014-09-07 11:35 · Score: 2, Insightful

I hate the FOSS mentality sometimes. So much unnecessary reinventing the wheel when all that's need is some enhancement of an existing tool. It seems like it'd make much more sense to take GParted, a very mature, well-supported and proven tool, and extend its feature set to incorporate the "modern technology" they require, rather than create a whole new tool almost from scratch and deal with an unproved base. They can either work with the GParted developers to incorporate these new features, or fork it and do the work themselves.
Reusing existing tech is supposed to be the whole fucking POINT of FOSS - it's freely available code, take it and use what's already been developed and use it as a base to create something new/better. So many people though want to start from scratch because they believe that their implementation will be better. It's one of the reasons I see so many music players that do 90% of what Winamp can do for example, each player though doing a different 90% than the other. No-one seems to be able to collaborate in FOSS.
1. Re:Idiots by Fwipp · 2014-09-07 12:01 · Score: 2
  
  Exactly - a fork, not a rewrite.
We have a winner! by dbc · 2014-09-07 12:00 · Score: 4, Informative

"RedHat is also known for having a bad case of Not-Invented-Here as well as wanting more control over a significant piece of their distro."
What does it support that others don't? by dbc · 2014-09-07 12:04 · Score: 3, Interesting

Not flamage, this is data-seeking. The announcement only vaguely states that existing tools don't support all modern storage technologies. So, what are the technologies where blivit gets a "yes" and gparted gets a "no" in the "supports " column?
1. Re:What does it support that others don't? by eulernet · 2014-09-07 12:10 · Score: 5, Funny
  
  slashdvertisement:
  Blivet: yes
  GParted: no
Command line? by ToasterMonkey · 2014-09-07 12:07 · Score: 3, Interesting

I'm completely fine seeing things move away from the older "GUI driving non-interactive commands in the background" model, to GUIs and CLIs that are built on shared libraries, because that potentially gives us THREE usable interfaces. However, is it normal for a CLI to lag behind the GUI now in Linuxland?
I see that blivet comes from Anaconda, so I expect some integration there.
It seems like a good CLI could be used to avoid the awkward practice of writing out a kickstart partition fragment from the pre section. We could just drive Anaconda's partitioning directly from %pre with shell logic instead of pooping out Anaconda-ese to be parsed later.
So where's my damned anaconda partitioning CLI already, this would affect more [important] people than yet another partition GUI!!
Re:How to know if a component of Linux.. by Nimey · 2014-09-07 12:32 · Score: 5, Funny

Could be worse: it could have been written by Lennart Poettering.

--
Hail Eris, full of mischief...

E pluribus sanguinem
Anaconda's base supports all? by Antique+Geekmeister · 2014-09-07 12:47 · Score: 3, Insightful

The only reason that "anaconda's base supports all" is because anaconda, and kickstart tools, have the ability to support '%pre" scripts that allow manual use of hte command line partitioning tools to tune the partitioning as desired, and completely skip anaconda partition. Anaconda has never, and from all signes will never, be able to support all disk management and partitioning tools.
Since it's a Python based wrapper for the actual system tools used, features can be added. But there will be inevitable mismatches between configurations manageable through anaconda, and configurations manageable through command line tools for new disk and filesystem tools. And anaconda's use in system critical critical tools like kickstart mean that it _must_ be thoroughly tested before updates. This will slow feature addition in a way that gparted, or other tools, need not support.
Re:How to know if a component of Linux.. by Anonymous Coward · 2014-09-07 14:36 · Score: 2, Funny

Could be worse: it could have been written by Lennart Poettering.
Hey, you watch your tone! Lennart's arguably leading Linux development almost single-handedly now. Linus might be focused entirely on the kernel, but Lennart is doing all the work on system elements that are crucial to bring Linux into modern times. After Linus, Lennart's now the most prolific (and famous) Linux contributor of all time.
Of course I'm just an AC, so there's no reason to listen to me. But I'm sure once Bennett Haselton write an article on systemd, you can be Slashdot will finally understand the genius of Lennart!
Lie by Vlijmen+Fileer · 2014-09-07 16:52 · Score: 2

"The need of a new partition manager stems from the fact that none of the existing GUI partitioning tools supports all modern storage technologies."
That would be a lie. If that is all, just contribute to Parted already. As always more will be at stake, probable things like "not invented here" and "I wanna have the power". And as a result we get to have a partition editor that needs Python??
1. Re:Lie by thatkid_2002 · 2014-09-08 01:45 · Score: 2
  
  It's not rewriting parted. Parted can't handle all modern storage technologies as it only deals with partitions which are only one part (pun intended) of the picture. In the [your favourite distro here] installer the UI calls out a *suite* of tools just like Blivet-GUI does. Previously in Fedora this was all piled into Anaconda - but now it is split out into this "Blivet-GUI" thing.
  If you bothered to read the articles or browse the source you'd know that it depended on Blivet and subprocess calls to normal system utilities. Blivet has been around for at least two years already and has matured through its use in Anaconda. Anaconda has been around for many years and has always (AFAIK) depended on Python. Nearly every Linux distro (and other Unixes, OSX and even Haiku) come with Python by default. Most installation environments can use whatever they please without impacting the resulting system so even if you didn't want to install Python (or the more concerning GTK, IMO) you don't have to. This person was not engaging in NIH - they were simply writing a new GUI to an existing tool to allow for better integration, easier modification and fewer dependencies... Shock horror! OMG WHAT AN IDIOT RITE???
  This mindless Red Hat bashing has really gone too far.
This is rumour control, here are the facts by AdamWill · 2014-09-08 07:40 · Score: 2

Unfortunately, Mukt completely mis-reported this and Slashdot picked up their errors for the summary, which is making for a lot of confusion.
tl;dr:
1. blivet-gui isn't supposed to (and in fact cannot) 'replace' gparted in any reasonable sense of that term.
2. blivet-gui is a new application, but its backend is the Fedora installer's storage management code, which is a very old codebase. There is no new storage management backend being written here.
3. Lennart and systemd have nothing at all to do with this.
4. It wouldn't really be practical to 'contribute' this to gparted, as it would involve completely ripping and replacing gparted's backend and then very rapidly proposing significant changes to the GUI, and hence would be a project takeover by any other name.
5. blivet uses standard underlying tools for performing operations, it's just a logic/configuration layer for them.
1: what the original announcement says is that blivet-gui uses a gparted-like UI to make it instantly familiar for gparted users. It doesn't say anything at all about it 'replacing' gparted. That's a pure invention (likely based on a misunderstanding) in the Mukt article. See the original announcement at https://lists.fedoraproject.or... to verify this, if you like. There's no sense in which blivet-gui really *could* "replace" gparted, if you think about it. gparted is an independent project; Red Hat doesn't own or maintain it, so Red Hat can't stop it existing or being maintained. gparted isn't a significant component for either RHEL or Fedora: it's just a leaf package, an app like any other. It's not like anaconda uses gparted as its partitioning tool, or anything like that. So talking about blivet-gui 'replacing' gparted doesn't make any sense, not upstream, not downstream. So long as upstream gparted devs see a need to keep developing gparted, gparted will continue to exist upstream, and so long as a Fedora packager wants gparted to be in Fedora, it'll be in Fedora, whether or not blivet-gui or any *other* storage management GUI app is also in Fedora. We have lots of space in the repos.
2: the backend for blivet-gui is blivet: https://git.fedorahosted.org/g... (packaged in Fedora as python-blivet). This codebase is simply the storage management backend of anaconda (the Fedora installer) split out into its own repository. The split happened back in 2012: http://www.redhat.com/archives... . The intent was to allow for exactly this kind of code re-use. So there really isn't some kind of new NIH effort going on here: the storage management code is not new, all that's new is the light wrapper around blivet to produce a standalone GUI app rather than using it as a part of the anaconda installer. The underlying codebase has existed basically as long as anaconda has existed, which is rather longer than gparted has existed. anaconda dates back to 1999 (https://fedoraproject.org/wiki/History_of_Red_Hat_Linux ), gparted AFAICT dates back to 2004 (http://gparted.org/news.php?item=180 ).
3: Doesn't really need expanding on, but no, there is absolutely zero link to Lennart, systemd, or any other systemd developers.
4: so the reason to do blivet-gui at all, and the reason anaconda doesn't just call gparted for "partitioning" like ubiquity does, is it doesn't cover anywhere near the functionality we actually need for the Fedora (and, more to the point, RHEL) installer. gparted really is a *partitioning* tool, and there's a reason I keep referring to blivet as "storage management". It handles things that aren't just partitions. The most obvious examples are mdraid, LVM, and btrfs (insofar as btrfs acts as a volume management and redundancy system, not just as a simple filesystem like ext), but blivet has all sorts of other interesting capabilities too, primarily of interest t