Slashdot Mirror
Debian GNU/Solaris
An anonymous reader writes "Today "Nexenta" announced an initial pilot program of GNU/Solaris. Initial trials are limited to "Ubuntu developers and the entire Debian community". From the announcement: "As you might know, Sun Microsystems just opened Solaris kernel under CDDL license, which allows one to build custom Operating Systems. Which we did...created a new Debian based GNU/Solaris distribution with (the latest bits of) Solaris kernel & core userland inside. We'll open Nexenta web developer portal completely for the general public by mid-November. Today we are launching a Pilot Program. Ubuntu developers and the entire Debian community - you are welcome to participate in the Pilot!"
--AC
Could someone translate this into English?
Much thanks.
The opposite of progress is congress
GNU/Solaris sounds like only the kernel is not linux.
Uhh, linux is a kernel, and nothing more.
Don't get me wrong, but I don't think this is worthwhile. This is almost as stupid as the debian/freebsd thing. If you want to get the full power from freebsd, use freebsd. If you want to get the full power from solaris, use solaris. If you want to get the full power from linux, use....well....ubuntu? redhat? suse? :P
What is the point of having 3k text editors? What is the point of having 20 window managers? Why don't you stick with only one operating system then? I look forward to see a first version of this OS and experiment it as soon as possible. Choice is one of the most valuable things Open Source Software has.
Besides the kernel it looks like it also uses the Solaris userland. That would include common commands (ps, ftp, m4, etc.) as well as Solaris-specific commands (psrinfo, prtconf, etc.). Also Solaris libc, as well as some other libraries. It will be interesting to see how much of the "original" Solaris userland will need to be replaced with, or supplemented with GNU tools. Some of Solaris' default tools are pretty bare in functionality.
Apt and Solaris should be pretty cool, though Solaris has had BlastWave for a while, which works pretty similar to apt.
hello dear sirs my name is jamesh i are india (bihar) can u guide me install red had linux 9?
Before people comment on why Debian is doing this i suggest reading Debains core statement of what they are all about http://www.debian.org/intro/about
Esspecially this bit:
Debian systems currently use the Linux kernel. Linux is a piece of software started by Linus Torvalds and supported by thousands of programmers worldwide.
However, work is in progress to provide Debian for other kernels, primarily for the Hurd. The Hurd is a collection of servers that run on top of a microkernel (such as Mach) to implement different features. The Hurd is free software produced by the GNU project.
This is an interesting project: developers will have the power of opensolaris tools, like DTrace (a very powerful tool to study the behaviour of programs - and optimize them) available for all debian packages...
a hint that they may just be playing games and chasing publicity as usual rather than showing a serious commitment to f/oss.
Yes, giving us Open Office and paying many full time developers to work on it, and porting DTrace to BSD is obviously not a commitment. They ought to be ashamed of themselves!
Being bitter is drinking poison and hoping someone else will die
It's just what I've always wanted... the kludginess of Solaris combined with the user-friendliness of Debian.
Is this the end of recursive humor?
****
"I'd never want to join a club that would have me as a member" - G. Marx
Actually, the naming story is more complicated than that.
The SunOS 4.x series was just "SunOS", and it was more BSD-based.
The SunOS 5.x series has always also been called Solaris. In the SunOS 5.x series, SunOS is the name of either just the kernel, or the kernel plus the very basics of the operating system, depending on who you listen to, whereas the Solaris brand name refers to the entire "Operating Environment", including SunOS 5.x and a lot of other add-ons like NIS+, LDAP services, CDE, etc, etc. I don't think they have ever sold a stripped down SunOS 5.x, only a Solaris Operating Environment 2.x which happens to contain SunOS 5.x.
And then with the release of Solaris 2.7 (SunOS 5.7) they began marketing it as "Solaris 7", presumably for some dumb marketing reason. Since then that's been the case, so for any given release from 2.7 onwards, the possible alternate namings are:
Solaris 2.X = SunOS 5.X = "Solaris X"
2.7 of course turned out to be an absymal failure of an operating environment, so it's quite fitting that it was the launching point of a marketing-driven renaming policy. Whereas 2.6 was a solid stable platform for many years, building on the successes of 2.4 and 2.5.1, the 2.7 release is where they introduced transitional 64-bit support, and poorly. It wasn't until 2.8 that things settled down again. Many customers unfortunately still have "Solaris 7" installed here and there in production, and it's a nightmare to support. What we can learn from this is never buy/install the first release of an OS right after a major overhaul of its guts to support a new architecture.
11*43+456^2
If they can keep virtually everything outside the kernel consistent with Debian, and replace only the kernel and drivers with Solaris versions, this would prove to be an interesting system for benchmarking and contrasting the two kernels.
As it is right now, you can benchmark entire systems, but it is difficult to do any meaningful kernel benchmarks.
If there is a significant difference in any particular area in either direction, I can forsee future server administrators choosing their kernel based on how well it performs in certain tasks. Perhaps the Linux kernels memory management is better, but the Solaris kernel's SMP scales better. Now you don't have to worry about changing your user or administrative environment, package management tools, or applications -- you can run the same on both, and just change the kernel architecture underneath in order to gain a benefit in an area important to you.
Heck, I can forsee a day when what kernel you want to use is queried as part of the install process, and where you might have a mix of machines that all appear to be running Debian, but which may be using different kernels based on the needs of the system.
We have competing web browsers that (generally) conform to the same interface standards (in terms of HTML rendering) -- why not have a choice in kernels, without having to sacrifice your user and administrative environment (or Open Source ideals)?
Yaz.
It's probably not the best choice for a notebook or a router. It might be a good choice for a server, depending on what you're doing with it.
I've done some Solaris administration, but I have never been even remotely expert. I did get an idea of just how much I don't know, however.
In many ways, Solaris makes Linux and the BSDs look like tinkertoys. There is a vast array of functionality inside. Solaris feels like it was designed 'down', by people who are used to working with mainframes; Linux and the BSDs are more designed 'up' by people inventing (and often re-inventing) things for the first time.
There's a fundamental expectation in Solaris that the server will stay up under virtually all circumstances. If the hardware you're running it on is expensive enough, it's apparently pretty easy to hotswap almost anything... even RAM and CPUs. I believe it supports multiple running instances of Solaris on the same hardware at the same time, even different versions. And I'm fairly sure that a recent feature will let you upgrade OSes 'on the fly', though I think there would have to be at least a brief interruption of service. Pay attention to replies, if any. My knowledge in this area is very sketchy.
It's highly optimized to scale to amazing degrees, given more CPUs to work with. But that means it's not very well optimized to work with only one, which was why it was called 'Slowlaris' in the early days of Linux. Running multiprocessor requires a lot of locking to make sure that different CPUs don't step on each others' toes. This locking takes time. So the first versions of Solaris/Intel were dogmeat slow in comparison to Linux. There was no chance of a conflict, because multi-CPU Intel boxes almost didn't exist, but Sun was and is more interested in having it run WELL than run FAST. Removing all that locking would have introduced bugs. So they left it slow. And most folks went with SPARC boxen or Linux instead, for better performance.
At the time, Linux screamed on the same hardware, because it didn't worry about any of that. Up through 2.2, Linux had just a Big Kernel Lock... only one CPU could be in kernel space at any given time, and the rest of the CPUs either ran user code or sat around idle. Most user code makes fairly frequent kernel calls, so the extra CPUs blocked a lot. Running on one CPU was very fast, but there wasn't much benefit to adding more. A second was a moderate plus, adding maybe 50% overall throughput. Adding further CPUs did very little for most workloads.
Solaris does exactly the opposite. It's slow on one CPU, though 'slow' is pretty relative on a multi-gigahertz processor, but as you add more, it scales almost linearly. 64-way Solaris boxes run very nicely. And they do it without crashing, too. That's an area where Linux, for instance, has had a huge amount of trouble... as they add in new locks and try to rearchitect to let more and more CPUs into kernel code at once, they introduce bugs, often at a furious rate. You don't see much of that in released versions of Solaris. A lot of what you're paying for with the expensive Sun equipment is their QA department, which must be just incredibly good. (if any of you are reading this, thanks!)
Basically, this is enterprise-grade software. It's designed to run things like banks and air traffic control and medical equipment...stuff that just can't ever break or go down. It's not actually USED for air traffic control, as far as I know, but I'm sure Sun would be happy to sell systems into that market. If your hardware is good, Solaris can take an unbelievable beating... you can have loads in the thousands and still be able to connect to the box with SSH (eventually) and rescue it. It'll be slow, but you'll get in. Linux, in contrast, will often still die from dumb stuff like fork bombs. Yes, ulimits can prevent that problem, but Solaris will survive without the extra help.
Basically, Solaris is the kind of OS that you can bet your job on, and remain employed. Linux remai
GNU's Now UNIX
Did you read yesterdays piece about Redhat wanting Xen in the kernel?
Maybe you'll actually run several kernels, managing different tasks - virtualisation can do this. Are we going to see an explosion of different kernels (similar to distributions now), as it becomes easier to try different ones out?
Why shouldn't Debian run on various kernels - Linux, Hurd, Solaris, Interix... the list goes on. I'd personally like to see Debian running on Microsoft's Subsystem for UNIX-based Applications (Interix), which will be included with Windows 2003 Server R2 and apparently also future operating systems like Vista. Here's my post to debian-win32 from last week:
) .
To: debian-win32@lists.debian.org
Subject: Debian GNU/Interix (Windows 2003 Server R2)?
From: Sam Johnston
Date: Mon, 31 Oct 2005 14:55:31 +0000
Afternoon all,
Windows 2003 Server R2 is to ship shortly with a POSIX subsystem
(Interix 5.2 of Services for Unix fame) which "includes more than 300
UNIX utilities and tools that behave as they would on UNIX systems,
plus a software development kit (SDK) that supports more than 1,900
UNIX APIs and migration tools, including make, rcs, yacc, lex, cc,
c89, nm, strip, gbd, as well as the gcc, g++, and g77 compilers.".
Apparently future versions of Windows (eg Vista) will follow suit.
With a tweak of config.guess I have had no problems compiling
bash-3.0, wget, etc. on Windows 2003 and am now interested in tacking
some more interesting packages, like dpkg and apt, with a view to
getting a full port of Debian running on it. The problem is that I
have limited time and porting experience, and the fact that this was
previously attempted under Cygwin is disconcerting; the debian-win32
mailing list has been around for years and yet there there has only
been one non-spam post in the last 18 months according to the
archives.
According to http://www.interix.com/ many interesting packages have
already been ported over and are currently available for download for
$30 as 'Interop Toolworks 2.2'
(http://www.interix.com/InteropToolworks.htm). Presumably source is
provided/available - I have posted the question in their Unix Tools
forum as I figure this would be a good starting point.
While there's no X server included, the X libraries are and the new
release opens up the Win32 API which should pave the way for someone
to build one. In the mean time Starnet Software do ship a free 'LX'
version which will accept localhost connections only
(http://www.starnet.com/xwin32LX/get_xwin32LX.htm
At this stage I'm looking for feedback about the viability of such a
project, information about why the last one appears to have failed and
any suggestions about what the procedure would be (eg build dpkg, then
debootstrap etc.), how best to ensure its success, and so on. I would
like to think it will be possible to bootstrap a base Debian
installation (Debian GNU/Interix?) from an installer executable, or
possibly even deliver it via ActiveX, eventually getting to the point
where one can log into Windows and get a full Debian desktop complete
with your favourite window manager.
For the time being I'll be happy with bash, OpenSSH, etc. but it's
interesting to consider what is possible... SFU/SUA was meant as a
migration path *to* Windows, but there's nothing stopping it from
being a two way street.
Sam
A few corrections / notes:
/etc and everything else (or you can opt to share parts of the install with the 'global' zone), however there is still a single kernel running all of the zones on a given server. Applying a kernel patch, for example, still requires a reboot of the whole enchilada to take effect.
There is no provision for running multiple kernels on the same hardware, aside from 'domaining', which is more like IBM's LPARing (prior to P5 and 5.3, that is - splitting a single server into two or more hardware partitions). But, when you do this, each partition really is its own machine, with distinct and non-shared CPUs, memory, and IO buses. You can move hardware from one domain to another (even without shutting down applications), but a given kernel is only ever running on one machine at a time.
More recently, is the concept of 'Zones'. Here, you can install seperate copies of userland onto the same box, and when you are 'inside' of any one zone, you appear to have your very own box, complete with your own
Actually, I believe 'Slowaris' was coined in 93 or 94 or so, when admins started switching from SunOS 4.1.3/4.1.4 to Solaris on SMP boxes, and ran into bug where no matter how many CPUs, every thread was being scheduled on the same CPU. I forget what version this was - maybe 2.3? I remember logging onto my University's student shell machine (back then, SLIP and PPP were still not being used by home users - you'd dial up to terminal into a UNIX machine to check email, usenet, or IRC). Anyway, I remember logging in and seeing a load in the thousands, just after the upgrade.
You can "upgrade on the fly", in a way. The concept is you mirror your root disk, split the mirror, perform the upgrade against the inactive mirror, and finally reboot, this time off the upgraded side. It more or less works, but you do need to use your brain. Really what it gets you is the ability to immediately revert back to the pre-upgrade image with just a single reboot, and also shortens the outage window to one reboot. They've had this since Solaris 8, I believe, and you could use it to upgrade from say 2.6 to 8.
Over time, 'Slowaris' did come to refer to the performance disparity between Linux and Solaris on x86. Until Solaris 10, Linux pretty clearly outperformed Solaris on like-hardware (x86). I think in part, as you say, due to Sun's lack of effort in optimizing for x86 and small servers, but also because x86-based benchmarks tended to focus on things like web-serving, and Sun's networking stack was long overdue for an overhaul (the first phase of which was completed for Solaris 10, and returned enormous improvements).
And when will Debian start a Debian GNU/MINIX project? There is already a Debian GNU/Hurd project and MINIX is alive, well and ready for production. Now that would be cool.
Slashdot, fix your code or at least hire someone who is competent at it to do it for you.