Linux Kernel 2.6.8 Released. Oh wait, it's 2.6.8.1

Gleng writes "The latest Linux Kernel, 2.6.8 has been released. The changelog is here. Don't download that though! A follow up to patch it to version 2.6.8.1 has been released, which fixes an NFS client bug in 2.6.8. The mini-changelog is here."

Comments

Feline Poop!

[Anonymous Coward]

Fuck you, all you motherfucking LambdaMOOers you! That's right, fuck y'all!

First kernel...

[Ianoo]

Is this the first production kernel with a w.x.y.z numbering scheme instead of the more usual w.x.y? I can't help but wonder how many poorly written version checking Perl scripts this will break...

Entirely untested, but Obviously Correct(TM)

[tod_miller]

"Entirely untested, but Obviously Correct(TM)"

Nice changelogs! *runs and installs*. I tend to wait until a major release to a nice looking distro before I upgrade. That is me though, just a user...

Pats on backs all round though!

Re:Entirely untested, but Obviously Correct(TM)

[photon317]

The user way is apparently the way of the future, which sucks for a lot of us. Gone are the days of the vanilla stable kernel branch actually being stable. :(

Quote from Linus (Re:First kernel...)

[blwrd]

On Sat, 14 Aug 2004, Christoph Hellwig wrote:

>>
>> Cane we make this 2.6.9 to avoid breaking all kinds of scripts
>> expecting
>> three-digit kernel versions?

Well, we've been discussing the 2.6.x.y format for a while, so I see
this
as an opportunity to actually do it... Will it break automated scripts?
Maybe. But on the other hand, we'll never even find out unless we try it
some time.

Linus

Re:Quote from Linus (Re:First kernel...)

[spitefulcrow]

What about people using RC versions? Or kernel sources patched by distribution vendors that have a dozen things in the extra version (linux-2.6.7-gentoo-r11, etc.)

Re:Quote from Linus (Re:First kernel...)

[irc.goatse.cx troll]

Renaming it Linux-2003-Professional would show if it breaks automated scripts to, but that doesnt mean its a Good Idea(tm)

Wow, again?

[Anonymous Coward]

I could have sworn the exact same thing happened the other day.

You know, I can understand some duplicates sneaking in - some things aren't so easy to search for I guess. A word like "Linux" isn't exactly tricky though, and you'd think "2.6.8" appearing in a story two days ago would clinch it...

So... the editors don't read the site, and can't code a duplicate checker to save their life. Oh dear.

Is this ...

[flex941]

... dupe? Yes it is ...

Re:Is this ...

[desau]

No -- this is not a dupe. Read the version numbers carefully.

Re:Is this ...

[flex941]

Since thread starts with 'Linux Kernel 2.6.8 Released.' + info about 2.6.8.1. it's still a dupe.

from the i-fooled-you dept indeed!

[Inoshiro]

Topic: Linux:

"
Linux Kernel 2.6.8 Released. Oh wait, it's 2.6.8.1
On August 16th, 2004 with 6 comments
Gleng writes "The latest Linux Kernel, 2.6.8 has been released. The changelog is here. Don't download that though! A follow up to patch it to version 2.6.8.1...
Linux > Operating Systems, Linux, IT

Linux Kernel 2.6.8 Released
On August 14th, 2004 with 203 comments
J ROC writes "According to The Linux Kernel Archives kernel 2.6.8 is now out. It includes some fixes from 2.6.7. Happy upgrading." You may want to read this...
Linux > Upgrades, Linux, IT "

If you'd read the comments in the original 2.6.8 release story (rather than this dupe), you'd see they mentioned the NFS fix -- several times in +5 comments.

Dupe checking? What's that?

NFS should be declared illegal anyways

[fluor2]

Anyways, people running NFS should be reported for crime anyways. It's hilarious that companies still run this VERY unsecure filesharing method. It's so easy to hack it, that it should be declared illegal.

Re:NFS should be declared illegal anyways

[Anonymous Coward]

easy to hack if you don't setup the rest of your system wisely. I've seen an entire university running NFS without a single problem for years.

Go checkout the Unix Admin Handbook... it was written by some very sharp IT *nix admins at the University of Colorado and other places. One of the best IT books I've ever read. CU is running NFS in all of it's computer labs, it's also running SSH, keycoded doors and remote admining (pushing config files across multiple platforms and OSes) without a single hitch. If you want to see how to do *nix admining right, to the point where the securities in NFS become irrelevant, go check out those guys...

p.s., off-topic this if you want.

Re:NFS should be declared illegal anyways

[fluor2]

i bet there is some kind of ip you can steal that make you auto-root on any of these boxes (by getting the key for ssh auth).

Re:NFS should be declared illegal anyways

[Anonymous Coward]

Not if they've set it up properly. It's really easy to set up a completely secure NFS server, and even if your NFS is set up to allow TCP connections, which by default it is NOT.

The default behavior of NFS is to "squash_root" meaning any request sent as root is changed to regular user priv.

If you did get root on a box connected to the network that already has NFS rights, you _might_ be able to spoof your UID and hack into someone's homedir, but that also would only occur if you didn't set your network properly... and is still worst case scenario for a completely default NFS install.

Where did you get your crazy ideas about NFS, have you ever tried it? Oh wait you probably ran it once two versions back when you were still "l33t." Come on man get with it!

Irresponsible

[vigilology]

If you didn't change the format, nothing would break.

Mandrake lockup.

[Anonymous Coward]

"Linux Kernel 2.6.8 Released. Oh wait, it's 2.6.8.1"

Mandrake cooker. Kernel-2.6.8.0.rc2.2mdk was the latest good one. The following one's apparently either don't like 'X', or at least the "nv" driver. A solid lockup.

Oh where is my VIA VT8237 RAID drivers for 2.6.x.

[boy_afraid]

Cold, freezing, dark, and lonely. My room is quiet. No sound of fans or hard drive spinning. No high pitch whining of the monitors, clickity-clack of the keyboard, rushing of the mouse, or blaring of the speakers. It... is... all... quiet. Sitting on the floor, staring up into the cieling, hoping, waiting, wondering.... is this all for nothing? My eyelids are being tugged down, shadow people are laughing at me. The cool breeze makes me shiver with a rash of goose-bumps having thier way with me. Death is knawing at my soul, no reason to live, to go on. I yearn to slumber forever in this darkness.

Only one thing can bring me from the event horizon of death... and that is...
VIA GETTING OFF THIER F*CKING ASS AND RELEASING VT8237 RAID DRIVERS FOR LINUX KERNEL 2.6.x FOR THE AMD64 PLATFORM!!!!

I SWEAR I'LL NEVER USE VIA AGAIN. THIS WAS THE LAST STRAW!!!!!

Kernel Release Levels

[GeekBird]

Is it me, or are kernels just not really ready for production prime time until the "y" level (of a w.x.y.z or w.x.y kernel) is 10 or more? I'm really starting to get gun shy with some of the weirdnesses that have happened in the 2.4 and now 2.6 release cycle. Maybe I'm just conservative...?

If Linus says it, he must be right

[millermj]

Okay, I'm not quite as naiive as the subject line says, but now that we've gotten rid of odd-numbered kernels, 2.6.y.z format is really not that unexpected. I much prefer that over 2.6.8-pre1-am2. By comparison, 2.6.y.z should be an easy problem to fix.

Not alone.

[SphericalCrusher]

I don't see why everyone is making a big deal out of it. Sure, it's embarrishing, but everyone makes mistakes -- especially open source programmers. It's the ones who get right down to it and fix their problems first that really makes a difference in the world. Look at Microsoft and all of the things they have had to fix over... and over... and over... and they still have problems with it today.