Debian 7.0 ('Wheezy') Release Planned For 1st Weekend in May
An anonymous reader writes with this good news from the Debian developers who have been working hard to release the next version of the distro: "We now have a target date of the weekend of 4th/5th May for the release. We have checked with core teams, and this seems to be acceptable for everyone. This means we are able to begin the final preparations for a release of Debian 7.0 — 'Wheezy'. The intention is only to lift the date if something really critical pops up that is not possible to handle as an errata, or if we end up technically unable to release that weekend (e.g. a required machine crashes or d-i explodes in a giant ball of fire). Every other RC fix that does not make it in time will be r1 material. Please be sure to contact us about the RC fixes you would like included in the point release!" Of particular interest to casual users, from the list of changes in 7.0: "Debian wheezy comes with full-featured libav (formerly ffmpeg) libraries and frontends, including e.g. mplayer, mencoder, vlc and transcode. Additional codec support is provided e.g. through lame for MP3 audio encoding, xvidcore for MPEG-4 ASP video encoding, x264 for H.264/MPEG-4 AVC video encoding, vo-aacenc for AAC audio encoding and opencore-amr and vo-amrwbenc for Adaptive Multi-Rate Narrowband and Wideband encoding and decoding, respectively. For most use cases, installation of packages from third-party repositories should not be necessary anymore. The times of crippled multimedia support in Debian are finally over!"
xfce 4.8! finally!
Wheezy Waiter?
Why, why, why does it have to be associated with some sick or even dying sounds? How is it ever a good idea to call your product: wheezy?
Sneezy? Leaky? Deathly? Spooky? Sickly? Stinky? Slimy?
I like Debian, why does it want me to think about such things when I think about that distro?
You can't handle the truth.
Maybe it was named for Louise Jefferson, from the TV show. Movin' on up!
I hope they can fix http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=368297 before they ship.
It seems horrible in this day and age that a Linux would exist that couldn't get basic functionality with ldap working.
Why would you want a wizard?
Just edit config files like a big boy.
Pick your smtp server and edit the files. Not like there is a lot to it.
Debian will run Linux kernel 3.2.39
Debian can also run on the FreeBSD kernel. It looks like Wheezy will support both 8.3 and 9.0 kernels.
The giant gap between Woody (the 2.4 release you're thinking of) and Sarge was too much even by Debian standards and so they've sped things up a bit to approximately one stable release every two years, so it's not too terribly antiquated. The only complaint I have about Wheezy is that it comes with Gnome 3 instead of either Gnome 2.32 or Mate. The installer is fine now (better than the new Fedora one in many respects) and the environment is nice and stable.
The Linux flavour will use Linux 3.2 with a couple of backported features.
http://womble.decadent.org.uk/blog/whats-in-the-linux-kernel-for-debian-70-wheezy-part-1.html
http://womble.decadent.org.uk/blog/whats-in-the-linux-kernel-for-debian-70-wheezy-part-2.html
http://womble.decadent.org.uk/blog/whats-in-the-linux-kernel-for-debian-70-wheezy-part-3.html
I pick tested and works well any day over bleeding edge.
And what do you mean by broken media support?
I employ Debian as an email/web server platform. What I've always missed is a command line tool I can use to set these services up without any graphical interface. A wizard to handle the hops would be awesome.
Now, before you label me a troll, let me mention that I loathe the GUI. And as a server described above, I surely don't need it.
Bytemark Symbiosis may be of interest to you.
No, it is/was caused by long release cycles (2 years or longer between stable releases) and the Debian group's dedication to truly Free software. The finally found something they find to be morally acceptable, and it is in the next release after they found it.
Don't blame me, I voted for Kodos
Sure - until the next codec comes along. Then, it's third party or install from test, as is tradition with desktop Debian.
Finding God in a Dog
This is true if you're using Debian on the desktop. As a production server, I want something that Just Works and Doesn't Change except for the occasional security fix.
Finding God in a Dog
So I'm reluctant to upgrade my laptop, although my work machine is already running wheezy (and KDE: flakey but usable).
What's new in Debian 7.0
2.2.3. Hardened security
Many Debian packages have now been built with gcc compiler hardening flags enabled. These flags enable various protections against security issues such as stack smashing, predictable locations of values in memory, etc. An effort has been made to ensure that as many packages as possible include these flags, especially focusing on those in the 'base'-installation, network-accessible daemons and packages which have had security issues in recent years.
Now there are no reasons for using Ubuntu anymore. I do not remember being so excited ever!
The one known as 'limpy'.
You can't handle the truth.
The times of crippled multimedia support in Debian are finally over!
Bleeding edge can work just fine, just be careful what you install.
Let's say that you want the same user accounts on all your machines. Then you often use some kind of directory service in larger installations, and LDAP serves that purpose very well.
I'm actually just quoting the post:
The times of crippled multimedia support in Debian are finally over!
It says crippled, which isn't the same as broken. All they sya there is that the multimedia support has been improved.
Bleeding edge can work just fine, just be careful what you install.
When I install something it is installed in parallell on hundreds or even thousands of machines. It is kind of important that we can rely on that the software we use continue to not only work well but work well consistently.
But that doesn't matter, just install the package in a test VM and if it works fine then deploy it. I've actually ran into more issues with "stable" software then anything else, I've had cases where software was to old to support features and protocols I've needed. I don't have the time to wait 2 years in between updates just to get a feature working, I need it working yesterday.
If you would like to help out, check out the following link. Anything you donate will be matched by a generous donor. Note, this money goes towards DebConf13, the Debian developers conference (free admission to all). Lets get Jessie off to a good start! http://nylug.org/pipermail/nylug-announce/2013-April/001231.html
What, exactly, is wrong with
If I have been able to see further than others, it is because I bought a pair of binoculars.
Will we be getting a multiarch build (armhf+armel) of this one? I found some info that said there would be multiarch for x86 (64-bit+32-bit) but I can't find anything for arm.
Are you willing to post some instructions as to how you did this?
I'm running 6.0 now. I could use a newer kernel but I don't want to give up gnome 2.
friends don't let friends teleport drunk
Find me a "casual user" who can comprehend the above paragraph.
Just change your environment variable, troll.
It uses the EDITOR and VISUAL environment variables, and falls back to vi if they're unset.
If you've been around any enterprises, then yes, you've seen it. It's one of the components of Active Directory (indeed, it's the "Directory" part of Active Directory.)
LDAP is used to centrally store and manage information about all the entities involved in a network - users, computers, etc. With users that includes everything from their email address to their network password - which brings us to the GP's point.
There are numerous implementations of LDAP out there. Unix systems have been administered using LDAP for decades. So it's a big thing if Debian still doesn't support it properly. It should. And actually, with SAMBA4 now out, we should be moving LDAP (and Kerberos and the other major network administration tools) out of "the enterprise" and start being able to use them in smaller networks, like at home. There's no reason why your D-Link router shouldn't have it built in.
You are not alone. This is not normal. None of this is normal.
It's not all that hard, but you do know you can run Webmin on top of Debian?
Don't leave webmin hanging out on the internet and you should be just fine. Firewall it and use SSH tunneling to access.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
I wish it was that simple. We've tested it and it just doesn't work. We want to be able to install a system today, tomorrow or even a few years from now and reliably get the same result. You need stable packages to do that.
Hell you can even use it to authenticate against a Windows domain! Very helpful if you have to cooperate.
For example, my SVN server authenticates against the Windows domain, so everyone can continue to use their domain credentials seamlessly.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
More than that, will Wheezy be Linux only, or will it also come in kFreeBSD and HURD flavors? Will we see a kFreeBSD 9.x? What's the status of Debian HURD?
Redhat/Centos works just fine on a desktop (just like Debian.)
Hell Redhat sells licenses specifically intended for workstations!
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
Have they finally come up with a free solution for media playback? Or have they compromised on their principles?
SURELY NOT!!!!!
I would go as far as saying that Red Hat actually works better on desktops. It's often tricky to get Debian to run well on brand new hardware if stable has been out for two-three years, since it will often require new drivers that aren't available in the Debian kernel. Red Hat on the other hand continuously backports a ton of stuff into their kernel and releases it as part of their point updates (about every 6-12 months). This means that you can often install Red Hat on hardware which Debian won't run on, even if the Debian distribution is newer. I have desktop machines which can run even RHEL 5 (kernel 2.6.18) just fine, while Debian 6 (kernel 2.6.32) don't even recognize half the hardware.
It's 2013 and your finally advertising non broken media support. Even Gentoo has had working media support built in for years, I think if this is one of the selling points of Debian then it's time to move on, your trying to get me to take the Honda N360 off your hands instead of the race car.
I think you're laboring under the mistaken assumption that this reflects on the general state of Debian. The truth is that media support is a very specific issue, which, IIRC, was caused by intellectual property issues. For most other things, Debian has had a very complete and high-quality selection of packages. For proprietary media formats, they basically had none - although getting support for those was as simple as adding a repository that provided them (change one line and run a command, or do a couple of clicks - whichever you prefer).
In particular, the "broken" media support was not an issue with Debian generally being broken (it hasn't been) and it also wasn't an issue with Debian being behind other distros (Debian stable tends to have old software, but that is by design - if you want newer software, you can use backports, unstable, experimental, or third-party repositories).
This is one of the major reasons I could never stick with Debian, I need stuff to work, be up to date and ready to go out of the box, Debian is built off legacy packages in an attempt to claim stability, when in reality it's just outdated in it's release mode.
By all means use the distro that works best for you. For me, that's Debian stable, because I want to minimize the amount of time I spend on maintenance. There is a trade-off between having newer software and having more testing performed on that software, and a trade-off between minimizing system maintenance effort and running up-to-date software, and I'm happy with how Debian stable makes these trade-offs. Every other OS I've used has had a higher maintenance burden.
Please correct me if I got my facts wrong.
Which is exactly the kind of behavior you want from a stable distribution. The previous version included support for ext4 as well so that users could test it. It just wasn't the default.
Define "casual user". If you mean a user who cares about this stuff, they probably understand it. If they don't care about this stuff, they may or may not understand it, but I don't see why it matters.
Please correct me if I got my facts wrong.
Debian's multimedia has always been a bit weird... anyway it really wasn't that hard to add in the deb-multimedia repository.
Also, I hate to say it because I am a huge Debian fan, but even the newest offerings are a bit lacking if you have underpowered equipment - the deb-multimedia versions for wheezy work better on my ancient laptop than the official packages.
That said, for many of the file formats that require the special repository, you need to download a codec pack for Windows - same amount of effort, and the millions of Windows users are fine with it. Not a big deal.
What *is* a big deal is that newer Linux distros seem to be dropping "out of the box" support on things that are far more important than multimedia. I was all set to use Mint, but was forced to go with Debian Testing because both the Ubuntu and Debian-based versions are lacking in a feature that is far more important to work "out of the box": the ability to set up full disk encryption on installation. RedHat based distros, Debian, and Ubuntu have done this for ages. Setting up full disk encryption after the fact is far more difficult than getting a few multimedia packages from a third party repository.
Does anyone know what changed to allow Debian to add MP3 and other libs? There has never been a technical problem with including them, but Debian has always tried to avoid violating patents by distributing patented (or claimed-to-be-patented) software.
I'm glad they've been able to take this step, just wondering what happened.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
LibAV's a badly forked version that's several revisions behind FFmpeg. Plus, this is Debian -- non-free codecs like H.264 are stripped out and are probably really supported by a seperate non-free repository.
I'd rather strip LibAV out and compile my own version of FFmpeg for faster encodes.
--
# Canmephians for a better Linux Kernel
$Stalag99{"URL"}="http://stalag99.net";
You don't have to install them if you don't want. At least they don't include "non-truly-free-patent-encumbered" codecs by default.
It's the Stay-Puft Marshmallow Man.
Hurd didn't make the cut for the wheezy release, apparently the hurd porters plan some sort of unofficial release alongside wheezy but what form this will take (hurd is not and never has been in testing for wheezy) has not been made clear.
AIUI kfreebsd will be included but is still considered to be a "techology preview" rather than a full production release. They seem to be shipping both 8.x and 9.x kernels.
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
I run servers for people who pay me money. I run debian stable exclusively and watch the cash roll in.
"Piter, too, is dead."
You are talking about debian stable, and calling it debian. Are you aware of those other flavours, testing, unstable, experimental? What about a chroot for some packages? what about running ubuntu packages depending on a different libc by using LD_LIBRARY_PATH and not having the rest of the ubuntu stuff shoved down your throat? What about linux mint debian edition?
But let's stick to stable. Who in his right mind uses a system like debian stable, whose software and data formats tends to stay the same, and upgrades do not involve changing habits. One cannot reliably WORK with those systems right? :)
---- MISSING MISCELLANEOUS DATA SEGMENT --- [sigdash] trolololol
Enable squeeze-backports.
Look for a new kernel.
OR
get vanilla kernel source, make localmodconfig | oldconfig, make menuconfig | nconfig | ... to adjust, make deb-pkg. dpkg -i ../linux**.deb
You probably want to enable "build firmware".
(I'm currently on a Lucid system that's still on 2.6.32, but my main system is Squeeze with 3.4 and 3.8)
Its stable its been out over a year, have any distros picked it up yet?
I made an installation a week ago (a basic one from network boot) and apt-get install avidemux got me.. nothing. So, they're now boasting about multimedia but it's missing quite an obvious piece of software that is simple enough to use even if just for grabbing an extract of a video.
I also suffered a few silly things :
- The netinstall iso doesn't work if put on a USB drive with unetbootin. Had to install a tftp and dhcp server on another box. I didn't try the big CD and DVD images. This is a bit of a problem as I've always used and instructed to use unetbootin (I don't want to dd a USB drive, destroying its filesystem and content)
- There was no lxdm. I put lightdm instead, but there was no wallpaper, very ugly boxy look and every time after the first one, it has displayed graphical corruption as a wallpaper. I'll have to see if it was because of --no--install-recommends
- Something I forgot.
- Ah yes, ALSA did not work on this laptop so I had to install OSSv4 instead. Mint 12's ALSA worked.
So, I was ready to hype up every body but sadly I will still recommend Linux Mint 13 Xfce for a ready-made desktop (though it pisses off the user with the duck duck go search, and occasionnally the outdated flash plugin), and for a base system with no graphical environment on the first time you boot Ubuntu 12.04 is a bit easier and with a bit more software.
On the plus side there's a debian squeeze desktop I installed and gave to someone that I will upgrade (though it can't access wireless networks because of it's kernel with free firmware only). The installer also asked me about non-free firmware, it first confused me (sort of implying I was going to get fucked) but the next step worked fine. I could choose a kernel with non-free firmware. So, a badly message peeved me but in the end I can appreciate that I had the option to get a working system.
We have to wait until both the hurd guys come back from holiday to find out :)
The difference in philosophy between linux (thanks for the patch, looks good, I'll merge it in) and hurd (you want to contribute, where's your doctorate? You call that a doctorate? It's not from the right supervisor at MIT at all!) is IMHO why hurd has been a very slow moving project.
The reason not to is the same way the craze of a database for everything (even formerly very simple print server software) doesn't always work. LDAP is perfect for very long and or complex nested lists of information but there is a threshold below where it's just too annoying and confusing when a small flat file will do. I have to admit with the features on even low end routers now it's approaching that threshold, and at the high end with enough NAT, port forwarding and multiple networks it could already be there (dunno how many firewall rules I have on an old snapgear router that's still keeping up with network speeds).
What, GNU Hurd won't make it to stable again?
By the way, I'm already migrating (I'm writting at /. while my server downloads packages). New asterisk and postgres. What's not to like?
Rethinking email
You want to use NIS, be my guest
Climate Progress - Hell and High Water
Casual users can stick with their playskool products written by accountants and marketers. In all fairness, I suppose it's not inconceivable that you can have a manager with no technical experience, but one who is still able to create a well-engineered product - I'd think this is very rare. To counter, look at how MS is doing with Balmer at the helm. I'll just stick with engineering products written and managed by actual engineers and other experts in the field, thanks.
And you know what - when Windows slows down after a few months of use, or Steve Jobs' ghost decides to rip even more user privileges from their consumers, we'll enjoy the technical superiority and enhanced freedom of Linux. Since it's not going to disappear any time soon, I don't mind that much at all.
I have left slashdot and am now on Soylent News. FUCK YOU DICE.
I'll tell you what's wrong - you have to know to put exim4-config in there.
Just try and figure out how to reconfigure your keyboard.
why dpkg-reconfigure doesn't take an option to let you know what the hell it is you _can_ reconfigure is another great mystery.
Absolute statements are never true
But that's me
Here's what I've noticed after two weeks doing an expert install:
* base system looks awful. the console's font rendering is so old and ugly. I've seen some pretty looking screenshots around with sans and fixed fonts that look so nice and even on low res. Why can't debian use something like that ?
Agree, it just doesn't look great. But you can change it. Dpkg-reconfigure console-setup and switch to something more appropriate.
* Is that vim-tiny ?! what's wrong with vim ? Or is your 1.44 floppy not big enough ?!
Adds too many dependencies for the base if I remember correctly.
* Systemd shouldn't be an option, it should be the default. It's not just inane "modern" features. There's serious performance and design benefits.
That would be great, but unfortunately Systemd depends very heavily on Linux and Debian wants to be compatible with other kernels as well.
* where is my tmux ? why do I need to apt-get it ? I want my tmux !
Just install your tmux. It is in the repos.
* stop asking me about locales during install; it's en us utf8. I don't live in US and English isn't my native tongue and it's still en us utf8. ask me after the first login to set the locales but leave the default, the default.
Some people prefer otherwise. You can still change your per-user locale at any time.
* don't ask me about what to install. install the base system and ask me after the first login about installing the rest. it's a waste of time, space and resources doing anything else.
Waste of time? You must be stressed. Take a look at FAI, it sounds like you should automate this thing.
* I need sudo, I want sudo. why waste my time during install on something that can be changed with two commands after install ? just set it as default and add it to the post install.
Earlier you said that you wanted just the base, but it turns out that you actually want something else too. Have you thought about that other people might want something else too in their install, and that the base install is a good fit for most of them?
* mail? ha? is that a posix thing ? just put something minimal by default that doesn't bother me. just, go away...
Overall, my biggest problem is that lack of acknowledgement about the post install process. It's like the debian team lives under some strange illusion people don't go through a long post install to setup their system with the software they like. The default are not stable of time tested. They're just old. The entire installation process seems antiquated.
Just, clean this up please.
You seem like a very happy person.
Debian is hardly aimed at casual users. You may be familiar with its far more popular offspring, Ubuntu and Mint, which are built especially for that market. Debian remains firmly in the domain of professionals and enthusiasts; and that is a group who need catering for too
Try "configure-debian".
Who in 2013 still doesn't "apt-get remove nano" first thing after a fresh Debian install?
Exim in Debian is shockingly poorly maintained. Not only does the maintainer think that a system user named 'Debian-Exim' (yes, really capitals and really with the Debian- prefix) is perfectly reasonable, he splits the relatively reasonable exim.conf into thousands of tiny, impossible to debug fragments stuffed full of so many comments that the combined configuration is vast. No wonder the upstream exim mailing list has washed its hands of bug reports from Debian exim users. What a mess.
I don't know who uses Debian, out of all the Linux users I know 0 of them do. But you looking for a system / upgrade that doesn't change habits then welcome to Gentoo!
Looks like packages.debian.org isn't updating properly. Currently wheezy has a 3.2.41 based kernel. My understanding is they intend to follow the 3.2 kernel series in wheezy.
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
LDAP can be used for long complex information but it's not hard to feed it with shorter, simpler, information and present a friendly UI that assumes it only has this shorter, simpler, information - which is certainly possible if you're building an Active Directory compatible server into a router (which thanks to Andrew Tridgell and his fellow SAMBA developers you can) with all administration done via the router's UI.
DHCP configuration files can be very complicated too, but routers tend to have very friendly webmin UIs that hide most of the complexity, combining information known to the router, with the basics that a user is actually likely to want to change (ie "This is the default range of IPs to give out, give static IPs to the machines on this list, and the router can figure out the rest because it already provides DNS, the default route, etc, etc.")
You are not alone. This is not normal. None of this is normal.
There is the GNU Hurd that GNU is working on, and then there is Debian, Arch and one more distro working on different distros. Is there anything that stops people who won't get accepted in the GNU Hurd team from working instead w/ the Debian or the Arch projects?
Thanks idunham. I appreciate the tip. I'll probably try the path of least resistance with squeeze-backports first.
friends don't let friends teleport drunk
sudo-ldap is kind of an out-lier; I've set up NSS LDAP I don't know how many times, against both OpenLDAP and Active Directory, and I've never bothered with sudo-ldap. I can see why people would but it is entirely possible (and IMHO just as easy) to not use it.
One thing that does bug me is that nslcd doesn't understand nested Groups in AD.
How do you authenticate to do sudo then?
The only think I can think is that you are authenticating locally instead of against ldap.
True. Debian is mostly about rock solid reliability and reasonably long support, not about end users. Though debian testing is a pretty neat bleeding edge distro.
Stupidity is the root of all evil.
I have to admit that I got halfway through an LDAP book and decided not to bother for the less than a dozen MS Windows users on site. While it scales up beyond the 32k users that the old domain method hands out and has a pile of other goodies it's a bit much for a site that's mostly NFS and only a little bit of SMB. I should look at it again since there's bound to be some sort of template that will get the implementation time down below what looked like a week :)
No, since those ar far, far less of a closed shop. The Debian committee was weird for quite a while but those just in it for the politics have probably got bored and wandered away by now (leaving nothing to show they were ever there apart from the "iceweasel" name as an insult to firefox).
Conversely, one of the reasons GNU+Linux on the desktop has not been a hit is because of the abhorrent support for commonly used multimedia file formats. The common user may not understand the technical names of software codecs and the multimedia formats they encode to and decode from, but they sure feel it when they are shipped a default installation that can read or create any possible format out there.
It's an immense achievement that an operating system with a Social Contract dedicated to only including Libre Software in its main repositories can be automagically more versatile at handing more media formats than proprietary OSes like Windows or Mac OS X. "Casual users" want their audio players, video players and editors, etc. to read and write anything, and the news here is that the complex libraries and codecs are included to be used by any software that needs to call them.