Linux Kernel 2.2.14

So everyone and their unkle wrote in to tell us that Linux 2.2.14 has officially been released. If your uptime isn't to sacred to ya, it may be worth upgrading. You know where to get the good stuff if ya need it.

Comments

Re:The importance (or lack thereof) of uptime

[Anonymous Coward]

Freud logic would suggest that guys with long uptimes are compensating for a certain "short coming" in other areas, if you know what I mean. So since you don't think a long uptime is a big deal you probably have a long penis and don't need to compensate.

Of course you probably know that and are trying to trick us in to thinking you don't care about long uptimes to make us think you have a long penis. What kind of excuse is that? script kiddies? puh-leeeze. you might as well have said that you don't want to ruin the environment by consuming the electricity needed for longer uptimes. like anyone would believe that.

So I'm calling your bluff. I bet you've got really long uptimes.

And for the record, my uptime is just the right size. Not too big, not too small. So don't even bother asking.

help a newbie out!

[Anonymous Coward]

I'm new to linux. What's the purpose of upgrading the kernel? is that something you need to do a lot? if your system is running ok, do you still need to upgrade?

I'm not actually planning on doing it, as I'm not proficient enough to recompile and replace a kernel yet, but I need to know how soon I have to try and take this step.

thanks.

Re:help a newbie out!

[mjuarez]

Well, the new kernel might just contain some security related fix, in which case it is recommended that you upgrade... in my experience, I've always upgraded kernels shortly after release, just so I can have the newest one, just in case. Have yet to see a bug in any of the kernels, though... I've been working with Linux from 2.0.29 up to now.

Re:help a newbie out!

[m3000]

You can read the replys on the other stuff, but as for upgrading your kernel, it's not too hard....if nothing goes wrong. I myself have only reconfigured it, still haven't gotten a full kernel upgrade to work for me. But it's different for everyone. But one article that helped me a lot (especially if you're using red hat or Mandrake (as they talk about RPM's)) can found found here.

Re:help a newbie out!

[Signail11]

One does not need to upgrade the kernel unless:
-Some major security hole has been found in a previous version of it
-The new kernel contains needed functionality
-The new kernel fixes some problem that affects you
If the system is running correctly and there is no pressing specific reason to upgrade the kernel, one would not have to upgrade it frequently. It is not a difficult task to patch/configure/compile/replace a kernel, but the sequence of tasks is rather exactly. Just read the documentation and make sure that you have a backup of your old kernel first.

Re:help a newbie out!

[J4]

It is not a difficult task to patch/configure/compile/replace a kernel, but the sequence of tasks is rather exactly.

What you state is indeed true, _however_...
Most newbies, in the US, at least, run RedHat.
RH, like the majority of distros, does not ship
stock kernel sources.
This can cause some parts of a patch to fail.
Not necessarily fatally, but certainly enough
to be disquieting to a newbie. I would suggest
either downloading the full sources _or_
waiting for RH to release an updated kernel rpm.
The former being the preferred method.
Patches are a good thing once you have
pristine sources to run them on.

Re:help a newbie out!

[Booker]

What's the purpose of upgrading the kernel?

To either fix bugs, or to gain features

is that something you need to do a lot?

not unless you need to fix a bug, or gain a feature. :)

if your system is running ok, do you still need to upgrade?

See above.

I'm not actually planning on doing it, as I'm not proficient enough to recompile and replace a kernel yet, but I need to know how soon I have to try and take this step.
You should look at the kernel HOWTO at www.linuxdoc.org/HOWTO - it's a bit daunting at first, but it's really not that difficult. The main reason to recompile your kernel is to tailor it exactly to your needs, removing all the cruft that doesn't apply to your system. Plus, as someone said, you'll never be a real Linux user until you do. :)
----

Why I switched to Net/OpenBSD...

[Anonymous Coward]

DISCLAIMER: This is _not_ a troll or flame - just one person's view of the situation with Linux (and FreeBSD for that matter...)

I recently switched from Linux to OpenBSD and NetBSD on my ix86 and Alpha machines after a year's evaluation of the free *BSDs on a couple of spare boxes; now I'm going to detail the reasons for the switch and elicit (hopefully constructive) feedback from /. readers.

Mind you, I did not switch for any political reasons (i.e. - GPL/FSF vs. BSD/MIT/X) - I actually prefer the 'restriction' of always-available source-code+improvements of the GPL and am philosophically aligned with the FSF.

Nor did I switch for preferring a 'Cathedral' development model over the 'Bazaar'.

Though I'm quite happy with most of the developers and contributors, I switched because I am becoming increasingly dissatisfied with the general, overall _attitude_ and _focus_ of the Linux community and, moreover, the practical and pragmatic potential pitfalls that I see Linux moving towards.

Here's a synopsis of my view:

1. Preoccupation with Microsoft

Why is the Linux community _so_ vocal about their dissatisfaction/hatred of Microsoft Corporation? Isn't this obsession somewhat absurd?

If proprietary and monopolistic are so terrible, why the soft spots (much less criticism at least) over Apple, Be, Sun or other Unix/hardware vendors? All of the above are really no better than MS and would *love* to be in Microsoft's place, reducing consumer choice and limiting their own liability.

How is this a liability to the Linux community? How about the proposed khttp? Four years ago, a kernel service like this would be unthinkable and considered an anathema to good Unix design; the post-Mindcraft attitude makes this permissible.

Is 'one-upping' NT4+IIS by introducing real dangers all that necessary to win corporate/popular mindshare? Why is this happening?

2. Lack of focus and the 'World-Domination' syndrome

Lack of focus for the general Linux community probably stems from the 'Bazaar' development model. While not necessarily bad (it is indeed the most democratic and fair development model, much more so over *BSDs), now that Linux has become more popular, the community now suffers from the 'World Domination' syndrome.

Does _every_ device need to be supported under Linux? Even devices whose quality and performance are questionable, when similarly-priced, higher-quality devices are available?

Personally, I use devices and only spec devices to others that I know are well-supported and whose vendors provide specs/development help to developers. For example, I only use Buslogic Multimaster and QLogic ISP-based SCSI controllers because they work, they don't time out, and their support does not vary (greatly) from one firmware revision to the next. The manufacturers have seen Linux as a viable market from the beginning and have provided specs/documentation to minimize the amount of reverse-engineering for developers. These are the vendors we should support, and perhaps the others will see free software as a viable market as well.

However, the Linux community wants *every* device supported, seemingly again to 'one-up' the other OSes (*BSDs, but particularly MS) and 'win over' more users. And as a result, the quality of support for many devices ranges from excellent to severely abysmal...

Can we decide on a range of devices to support (the level proportional to the amount of support from the vendors) and _really_ support those devices?

3. Lastly, the details and fine-tuning code.

For the most part, the kernel maintainers do an excellent job considering the volume and rapid rate of development on the kernel. Even so, is this rapid rate of development really conducive to a necessarily better kernel and/or OS? Also, are we ignoring potential pitfalls with the 'release early and often' methodology?

The details (an admittedly minor but poignant detail): disk device files and partitioning:

• Is the /dev/[s,h]da entry really satisfactory? Considering that Linux is moving towards enterprise computing with multiple disk controllers as a standard, shouldn't we be moving towards a controller-disk-partition naming scheme?
• Speaking of disk devices, shouldn't we drop the DOS disklabelling and go with BSD disklabels? They seem to make much more sense...
• Smaller details like NFS have always seemed to be ignored; in my situation, NFS seems more broken with 2.2.x than it was in 2.0.x. And we're not even at NFS v3.0 yet. Again, if we're going to pursue the enterprise computing realm (on heterogeneous networks), things like this must be addressed sooner than support for the latest 3dfx game card.
• Bugfixes in the Linux world are indeed rapid, but shouldn't we also consider how our fixes now can affect development in the future?

Let's compare this to the *BSD world... addressing the issues above.

1. The *BSDs are obviously not MS-friendly, but they don't seem to have many loud-mouths (i.e.- ESR) disparaging the Gates Empire at every opportunity. Though beset by lack of publicity (I believe they are waiting for the Linux hype to subside before pursuing the limelight), their users are growing, and increasingly including many ex-Linux users :(

2., 3. The *BSD teams are exceptionally well-focused, and that shows in the overall quality of code and distributions. Certainly the development is far less democratic, but it seems the users truly benefit from the focus on 'getting it right the first time.'

The hardware support (peripherals, not architectures since NetBSD supports just about arch) in *BSDs is limited in comparison to Linux but there is a marked difference - nearly all the supported hardware works well. For a new user (or a boss you're trying to impress), this is definitely preferable to wide-inclusion/spotty performance that we have under Linux.

Also, NFS works perfectly under *BSD. This is probably the number one reason why I switched.

Although the IP stack under Linux is good, it's even better on *BSD.

*BSD pays attention to the details. Bug fixes are fairly rapid (though not instantaneous as under Linux) but the maintainers always strive to make *correct* bug fixes - that is, fixes that won't become a potential problem six- ten- twelve- months from now...

*BSDs use proper disklabels, partitioning and device file naming schemes from the start; no need to worry about managing multiple host adapters, etc.

Sorry for the ranting... actually, I hope this will be considered constructive criticism.

The Linux community _does_ have an unusually high proportion of talented programmers, but, because of some of the more _social_ issues above, may lose mindshare and advantage to other free OSes...

~AC

Okay, I'll concede... Actually, I find that the Debian distribution, with its long release cycles, thorough and robust package management and exceptionally high-quality (in the Linux world) packages is the Linux distribution that approaches the quality of Net/OpenBSD. They obviously have extremely stringent standards (though not quite at *BSD level) and seem to attempt to 'get it right the first time'. Kudos to the Debian team!

probably not

[Anonymous Coward]

general rule (which doesn't only apply to kernels): if you don't need to upgrade, don't. reasons for upgrading might be:

1. you have some hardware, and a driver for it was added/fixed in the new kernel
2. there was some grevious security hole that was fixed (not the case with this release)
3. there was some grevious file corruption bug that was fixed (again, not the case)
4. for some reason, you upgrade a different piece of software, and it requires a newer kernel (far more common when upgrading from 2.0 to 2.2 than from 2.2.13 to 2.2.14, so it probably won't apply)
5. you feel the need for whiz-bang features, in which case you're probably already running the unstable kernels :), and consequently you wouldn't be asking this question

usually, reason #1 is the reason for upgrading. if you look at the changelogs for stable kernels, you'll find that the vast majority of changes are tiny (or occasionally not so tiny) fixes with obscure pieces of hardware that run on obscure systems (okay, maybe a slight exaggeration). anyway, unlike with the 2.3 kernels, you won't find any new whiz-bang features in the 2.2 kernels, so if you don't need to upgrade, don't!

I wish NcFTPd would put their server to real test

[Anonymous Coward]

NcFTP Software makes alot of great claims about the loads that their server can handle but I have yet to run into a heavily hit server that is running one. It would make a much cooler selling point if they could say it survives being a mirror of stable Linux kernel updates.

Ext3

[Kirth]

Applied immediately the 0.02c ext3-patch, patch
had no problems (except two obvious rejects of
2 lines of code in the kdb-patch, and further
2 lines in the ext3 patch), but it failed to
compile... I guess I'll have to wait until
Stephen Tweedie fixes this. And of course, the
missing ext3 is the sole reason that keeps me
from trying out 2.3.x-kernels.

Kirth

Re:Ext3

[LWolenczak]

Dude, you wouldn't happen to have the ext3 instlal instructions?
would you?
i lost mine, and linux.org.uk seems down

Re:Ext3

[whoop]

Was the error something about an "event" variable? They changed that in the fs code to like global_event.

I made up a patch based on 2.2.14pre18. It applies and compiles cleanly with this final 2.2.14. I took out some of the kdb stuff, because well, I don't care about debugging it. :) I guess the usual caveats when not dealing with the real author of something like this apply. But I've been up for almost three hours now with it (and several weeks in the 14pre kernels), and all filesystems are working just peachy.

Re:Ext3

[whoop]

mount /dev/whatever /mnt
dd if=/dev/zero of=/mnt/journal.dat
ls -i /mnt/journal.dat
umount /mnt
mount /dev/whatever /mnt -t ext3 -o journal=

The -o journal= is only needed the first time. Afterwards just edit /etc/fstab and put an appropriate entry for the mount point and set it to ext3. Hmm, I forget now what the kernel parameter was to do this for the root partition. Ah well, I guess you'll just have to wait for their site to come back, or find a mirror. Maybe Google has it cached...

Re:At last!

[whoop]

Ah, the great Slashdot Moderation Catch-22. Moderators only pay attention when a story first appears, using up their points early. After it gets over 100 posts, they only read the high-scoring ones other moderators have bumped up, sometimes bumping them up further. So you end up with a bunch of 4 or 5 point posts, and all the others remain at the bottom of the barrel. Only the truely brave dare go down there and read them.

Re:At last!

[whoop]

Or, given that I've not actually contributed to the kernel (yet),

Ah, let me tell you the joy one feels as a bonefied kernel contributor. I remember it like it was yesterday. Picture it, 1997 (or so), I guess something in the 2.1 line. A kernel is released, and some goofy driver that I used, a sound card or something, has a typo. Seizing the opportunity, I expertly craft a patch to handle this mishap, using all the resources at my disposal. Then, crossing my fingers, I wade through the compile to ensure everything works. 15 minutes later, it's compiled.

Then, the ever so daunting reboot and test it routine. I edit lilo.conf with a fury like I've never edited lilo.conf before. Run it, reboot, and closly watch the scrolling messages. Then, like the sun rising in the morning, I see it. Before my eyes is the startup lines for that piece of hardware, the IRQ, the IO address. It was a beauty I had only seen before when I actually paid attention to the bootup. Finally, after all was booted up, I did the final test. Playing a sound file, or whatever the thing was. And it worked! "Eureka!!" I shouted from my desk. The neighbor's dog starts barking. "Yes, boy. It is true. I have fixed the kernel," I reassure the young pup.

Hastily I booted up my PPP scripts. The modem fires up with it's random assortment of buzzes and bings. Then I see those magical characters from my ISP, "Your IP is now: 123.45.67.89." (IP changed to protect the innocent.) Without skipping a beat, I fire up pine and type one heck of an email to the kernel list. "Um, there was a typo in the xxx driver of 2.1.xx. Patch below," (an approximation, my memory isn't that good).

Anxiously I waited, day after day. "When is that crazy Fin gonna release a new kernel with my patch??" I asked myself. Each morning I awoke, checked the sunsite mirrors. With each passing day I only got more anxious to see my patch in all it's glory. And then, one day it happened. All my vigilant waiting had finally paid off. Linus released another revision of the kernel! I downloaded that patch like no other patch in the world, gunzipped it, and fired up less. Remembering which file it was I had to skillfully edit, I executed a search for it. Then a quiet peace fell upon the world. It was as though all the powers in the universe were converging to celebrate my patch. For there, upon that console screen, on that cold, wintry day (I'm guessing, it adds atmosphere to the story), I see the modification over which I toiled so diligently. I was now, a Linux kernel contributor.

That is my story, and that is all I have to say about that.

Re:AGP?

[demon]

No. It works as another PCI bus (which it really is - AGP is just a modified PCI bus), so AGP cards work fine, you just can't use the GART (graphics address remapping table), which is used for 3D texturing using texture data from system RAM. Support for this is in the 2.3.x kernel, but it doesn't seem to be very well-supported as yet...

Re:AGP?

[demon]

1) AGP has been in the kernel for quite some time.

As I said in another post, AGP is not explicitly supported in the current stable kernels, it's just used as another PCI bus (which is largely how the system treats/sees it). 2.3.x includes developing support for the special features of AGP (texture data in system RAM and (maybe?) DMA support).

How do you patch the kernel?

[V.]

I've used patch with other srcs before and not
had any problems but for some reason I can't
figure out how to get it to work with the Linux
src. I have the linux-2.2.12 src and the files
patch-2.2.14 and patch-2.2.13. What do I need to do?

lrwxrwxrwx 1 root root 12 Jan 4 17:58 linux -> linux-2.2.12/
drwxr-xr-x 18 root root 1024 Jan 4 17:40 linux-2.2.12/
drwxr-xr-x 2 root root 12288 Nov 14 18:28 lost+found/
-rw-rw-r-- 1 root root 3111595 Jan 4 17:37 patch-2.2.13
-rw-rw-r-- 1 root root 7269094 Jan 4 17:37 patch-2.2.14

Re:How do you patch the kernel?

[Centove]

linux/scripts/patch-kernel

Re:How do you patch the kernel?

[hensley]

Two ways:



cd into /usr/src/linux

patch -p1 (for both two patches)


or, easier and automated:

cd into /usr/src (or wherever your patches are)

/usr/src/linux/scripts/patch-kernel to apply all patches in the current directory)

Re:101 days on 2.2.10

[dangermouse]

yowch. better be careful there... ever filled /? It ain't pretty.

Re:The importance (or lack thereof) of uptime

[clasher]

Uptime is an important thing at times. On servers which thousands of users access regularly you had better not keep the machine down for long. Some admins who don't fully understand what they are doing are all too ready to install the newest software on busy servers occationally creating problems. Problems result in downtime which users don't normally like.
Sure if you are running a single user machine upgrade by all means but the only reasons to upgrade real servers are if the new kernels has security patches and possible imporovements in the drivers for your specific hardware. Otherwise keep the uptime high, the users appreciate it.

Re:What should I do to work with the new modules?

[Bake]

Why do you want to make a monolithic kernel (minus the scsi controller)?
Building a modular kernel isn't that hard, just a matter of putting " make modules ; make modules_install " after your " make dep ; make clean ; make zImage ; ".
There are several cons of building a monolithic kernel these days. In my experience the first think I have is that, for me, it's too big to be a monolithic one (even with make bzImage).
That's why I build everything I can as modules.
They're simple (IMO) and smart. If set up right (usually the case) they are only loaded when a program needs them and unloaded shortly after I turn off the program (my bttv modules do that anyway). That ensures that it only uses the RAM it needs and no more.
One question though, that SCSI controller isn't the one you hook up your scsi harddrive to (I'm assuming your harddrives are SCSI only..) is it? in which case you'll HAVE TO build support for it in the kernel itself (no modules here).

Re:The importance (or lack thereof) of uptime

[Hallow]

Yeah, you can tell this is just soooo accurate.
Their #1 uptime system is a HPUX box that's been up for over 100 years.

Re:uncle? what has he got to do with that?

[Luyseyal]

it's a figure of speech meaning essentially: infinity + 1.
i've seen the expression formulated as "everyone and their [dog|uncle|brother]".

-l

special king of bugfixes: security fixes

[hany]

you mentioned all reasons for/against upgrading the kernel very good.

i just want to point out security holes

with rule "not unless you need to fix a bug, or gain a feature." you can consider not to upgrade just because of for example tulip driver error because you are not experiencing troubles even when using tulip card.
BUT if this tulip bug is security hole you SHOULD update even when usual usage is OK. hackers are not producing usual thinkg on systems :)

disclaimer: take it as advice to newbies.

Re:security fixes fix bugs :)

[hany]

I figured security fixes were lumped in with bug fixes

exactly. i just "explained" special case of bug fix to newbies (i hope).

Re:Filesystem problems fixed?

[hany]

Any suggestions on what I could do to:
...
2) formulate a useful bug report?

take a look at /usr/src/linux/REPORTING-BUGS (2.2.X).

Re:Uptimes and kernel upgrades.

[madprof]

To be honest you shouldn't worry as the uptime for Linux boxes in general is as high as most other platforms already.
It is not uptiem alone that sells an OS to other people. If you can upgrade your kernel to take advantage of new functionality then that is more likely to make Linux appear robust, usable and so on.

Re:Link for the lazy

[flux]

Nice. People at kernel.org must be really happy now when you've shown THAT link instead of the mirrors-link. Of course, it might not be mirrored everywhere yet..

Re:Why?

[Sesse]

Not only IDE corruption -- filesystem corruption in general. Flushing all blocks would also mean flushing all _dirty_ blocks.

In addition, I've got some TCP/IP problems I hope 2.2.14 will fix, like returning fd 0 on accept() and generally breaking under high loads.

/* Steinar */

Re:AGP? -- Never will happen

[John Fulmer]

To be fair, AGP DMA transfers aren't supported by Linux, AFAIK, since Intel won't let go of the specs.

Also note that there are two (or more) AGP specifications that would have to be supported: Intel's, VIA's, and whoever else puts out an AGP chipset.

jf

Re:Framebuffer Devices

[Kesha]

I've used ATI MACH64 FB driver before (back in 2.1.13X I think). Do you have a different chipset?

security fixes fix bugs :)

[Booker]

I figured security fixes were lumped in with bug fixes, but I get your point. :)
----

Re:Uptimes and kernel upgrades.

[evonski]

VERY VERY difficult, if not impossible ("hot-swapable kernels"). Been discussed some time last year, check kernel archives. The HURD would more likely be possible to do this being a microkernel.

my $.02

Steve

Re:damn it...

[Ark]

My department moved to our new building about a year ago. In that transition I had to take down our "first linux production server" down to move it. I had a 112 day update.

It was plugged into a UPS and I was so tempted to move it without shutting down. Unfortunately the good side of me took over and I shut down the machine. I'd be at about 250 days or so of uptime after the move if it wasn't for the great-popcorn-burning-and-setting-off-the-fire-sup ression-system-power-outage of 1999. (Of course, myself and my fellow admins were late to the party as we were at lunch. The modem that is hooked up to the machine that is supposed to page us in the event of a power outage wasn't on the UPS. We maybe would have been back 5 minutes earlier. It wouldn't have really helped us, but it was a bit of egg on our face.)

I'm at 71 days on that machine, so I'm happy. I really wanted to shoot for a year....maybe I'll get it this time.

S/390s and VAXen

[Cato]

Yes, S/390s are very cool - I used to be sysadmin for a UNIX on Amdahl (IBM mainframe clone). It was so fast at I/O that grepping a big file (producing no output) was instantaneous - I thought the grep command wasn't working but it was just going blindingly fast, on a mid-1980s vintage mainframe at that...

As for Linux on VAX, see http://www.mssl.ucl.ac.uk/~atp/linux-vax/ - there is a port in progress, though right now NetBSD seems a better choice for everday use.

Re:AS/400s too...

[Cato]

Of course, somebody is hoping to port Linux to the AS/400 - see http://users.snip.net/~gbooker/as400.htm for details.

Given that the AS/400 has such an unusual architecture, it seems like a very difficult port, far more so than the S/390, but you never know...

Re:pppoe

[Griim]

PPPoE is in the 2.3 series, so I imagine it will be in the 2.4 series. Right now, building there are kernel patches to get it to work, which seem to give the least overhead, and there are a few userspace apps. You can find almost everything that's available here. (The one by David Skoll is apparently the best userspace app.)

Re:AGP?

[osjedi]

Wow, I must be some kind of kernel hacking God (kidding) because I have been using AGP with the stable kernel for about a year. Me thinkest thou art mistaken, my friend.

Re:What should I do to work with the new modules?

[Rozzin]

"... but I don't know how to make it so that I am using ONLY the latest module..."

When selecting what to include in or exclude from the kernel, choose "m" to include items as modules.

After you do "make zImage" or "make zlilo" or "make vmlinux" or whatever your choice for building the monolithic part of the kernel, issue "make modules" to build the modules, and then "make modules_install" to install all of the modules. If you don't already have this version of the kernel/modules installed, you've got nothing to worry about here--2.2.14's modules will all be in /lib/modules/2.2.14; when you boot/run with 2.2.14, the only modules that get loaded will be the ones from that directory.

Re:Uptime...

[Duckie01]

Which "all those smaller upgrades" are you talking about?

The only upgrades that cause me to bring a system down are kernel upgrades and hardware upgrades. I can do all other upgrades without rebooting the machine.

If my kernel supports my hardware, and there's not some major bug in it, what's the point of upgrading it? If my hardware is capable of doing the job, what's the point of upgrading it?

If it works, don't fix it...

Re:AGP?

[BJH]

Not particularly, asshole. I was correcting his acronym error, not indicating that USB support wasn't available yet.

BTW, USB support is included in the standard 2.2 kernel. Look under linux/drivers/usb.

PS: Go crawl back under your rock and die, you stupid piece of shit.

Re:could not be done

[Marco Schramp]

Theoretically is may be possible. It's not possible from a practical point of view.

Think about what you'd have to code if an internal structure changes (this happens a lot!). It would add to much code only going from one version to the next. I think they effort of implementing such a thing doesn't compare to the downtime of a few minutes that you would have.

If your uptime is really that sacred, that don't upgrade. Apparently your system is already stable enough that a reboot for a kernel upgrade is considered harmful. This implies that the current version of the kernel already suits your needs in terms of stability. Then there is no reason to upgrade anyway.

If security or stability is an issue, then you should take the time to reboot. Consider the time you lose because someone cracks your system or because unnecesary instability and you know why.

Regards,

Marco.

Does it support the parallelism?

[Amnesiak]

I know that some of the S/390's come in massively parallel configurations - like 12. Well, OK, it's massively parallel for me. Does the kernel have a limitation on the amount of processors? I thought that it was four per box.

Thanks and note

[Amnesiak]

Thanks. I just hope that everyone realizes that not everyone is a well established junkie yet - I knew where to get them but maybe some folks did not.

Re:Thanks and note

[Josho]

I'd like to enphasize what is said here. Be sure to provide links not just say "You know where to get the good stuff if ya need it." Given, it sounds cool, but not all of us are as knowledgable about where to find such things. Thanks.

Re:Uptime...

[Miguelito]

A few weeks ago I went into work to find the local paper with a front page article stating the utility company of this small town was going to upgrade the substations

You know, our local power company needed to "upgrade" our area again a few weeks ago. Nevermind that they'd done an all night (1am to 6am) outage less than 6 months before, but this time they decided it would be more convenient to shut power down at 5am on a Saturday! Allowing for up to 12 hours to do the upgrade of course! Surprise surprise, they started an hour late, and barely got the power back up before 6pm.. when it was already getting dark.

There's not a whole lot you can do at home on a Saturday with no freaking power. Needless to say noone in the neighborhood is happy with them right now. God forbid they might schedule an outage of this length for either a weekday when people will be at work, or better yet, during the night again when the impact to people will be minimal.

Re:That's valid, but one-sided.

[Helge Hafting]

we can only be offline for ten minutes every month"

You can have that, and keep up with new kernels also. Install every stable kernel that comes along, while keeping a copy of the running one. Just don't reboot.

Now, if the machine have to boot anyway (such as power failure, or the new kernel have something you must have) then you get the newest kernel. If that turns out bad, reboot to the old one that you kept.

2.3.36 just released too

[Mazzella!]

Just got the email from the Linux Kernel Distribution System ... 2.3.36 is out.
Here is the patch
. . .
Here is the full source

Re:AGP?

[kijiki]

the GART acts like a TLB for your AGP video card. It is chipset dependant. Basically, that huge texture you have allocated in your program from 0x4000 to 0x0A000 is, in reality spread all over physical memory in little page sized chunks (like most virtual memory). The GART makes all those little chunks look like a contiguous texture, by mapping the virtual address that the video card has to the various parts of physical memory on the motherboard. I'm not sure how it handles textures that have been paged out to disk, I assume it traps out for software to handle, just like the TLB does, or it may require you to MLOCK_ALL your textures into RAM.

Last I checked, VIA was being very open with the programming info required to operate the GART, but Intel was playing games. This may have changed lately.

Re:uncle? what has he got to do with that?

[blaine]

It's just a weird expression we have to imply a lot of people. An example:

"It used to be that only a few people had personal computers, but now everybody and their [uncle|brother|mother|cousin] has one!"

Hope that helps. English is a strange language indeed :) But then again, so is every other language.

Re:uncle? what has he got to do with that?

[DMDx86]

What's the fuss about Tulips? They're fast, stable, and light. Occassionaly, when I cannot find a Tulip, I try NE2000's, which just aren't quite the same.

Re:To the sys-admins at /.

[Ted Nitz]

uptime isn't the inverse of downtime, as odd as that may seem. It's quite possible to have very little down time but not years of uptime, it seems to me that in most cases 1 minute of downtime is quite acceptable when it allows you to fix problems. I think that the 45 seconds worth of reboot time to get important new features in the kernel is well worth it, even if it does ruin the whole uptime thing... I personaly think that the best way to measure uptime isn't in hours or years, but in percentage.
-Ted

Re:You guys EAT tulips?!?!?

[QuMa]

My cat absolutely loves tulip petals... As soon as he sees tulips, he charges and eats them away in no time... Weird animal...

Re:The importance (or lack thereof) of uptime

[Abigail-II]

Not that going down once every month will prevent you from having a day-long outage.

Well, if your database has been cranking for 17 hours to do the end of the month report, you crash, and you spend 7 hours running dbcc's untill your database server is back up, than you would count in my book as "having a day-long outage".

And that's excluding the time figuring out *why* the system crashed.

The only claim was that having an uptime ny longer than the most recent necessary security patch is stupid.

Not at all. That's why there are firewalls. Besides, for the majority of real computers out there (not counting the toys at peoples homes) net connectivity is just an extra, not a necessity.

-- Abigail

Re:The importance (or lack thereof) of uptime

[Abigail-II]

I would also hope that one performs the neccesary upgrades at off peak hours.

But you don't always have off-peak hours. Take a company with offices in in New York, London, Tokyo, Los Angeles and Singapore, and a central mail server. There are no off peak hours.

And even if there are off-peak hours, there's often so much to do, you cannot do it all in the maintainance window, and some things have to wait.

-- Abigail

Re:The importance (or lack thereof) of uptime

[Abigail-II]

The kernel is so out of date that any random script kiddie can grab an exploit or buffer overflow from bugtraq and root the system, obviously not a Good Thing if your computer is running any sort of critical task.

You got that all wrong. It should read:

The fact a script kiddie can send a single IP package to your machine is not a Good Thing if your computer is running any sort of task that has any importance.

If you really, really have to have (Inter)net connectivity on your box with a critical task (which seems unlikely) one should use a firewall.

-- Abigail

Re:MS SQL Server Sucks Rocks then coughs up blood

[Abigail-II]

dbcc's and 17hr queries - quess who is using Microsoft SQL Server.

I've no idea. I've never used Microsoft SQL Server.

Do you know that the upgrade to a full licence of SQL Server trashes any and all development / test databases you might have run using the Evaluation licence version?

Uhm, no. Why should I bother learning trivia like that? I've never run anything Microsoftish in my life. The closest I've ever got was having an unplugged NT machine under my desk.

-- Abigail

Re:Why?

[guacamole]

Stable is pretty good.

I beg to differ. The NFS is still quite broken.

Re:Changelist

[Nightcrawler]

Usually Alan Cox posts a very detailed and informative changelist on www.linux.org.uk which is much more useful than the stuff you can find on KernelNotes or Cutting Edge .

Re:rephrase of original post

[Rapier]

Hmmmm I've got a machine here running kernel 2.2.13 that's been up for 76 days now... but I guess I'll probably break down and reboot the system soon. Perhaps I will recompile the kernel, but hold out for an 80 day uptime before rebooting. :)

Re:Linux blows

[Telamon]

Oh please, get a clue. People don't break into systems because of the kernel, most break-ins are via the system daemons and other user-space programs. You want a secure system, run nmap on your box and then close down/firewall any service you aren't actually using. Common "insecure" services are sendmail and FTP.

Re:The importance (or lack thereof) of uptime

[Basje]

65535 seconds to be exacly (0xffff)

----------------------------------------------

pppoe

[jrs]

I know people are working on pppoe for linux now, but will it be in 2.4?

Re:pppoe

[john*mckown]

I haven't downloaded/installed it, but there is an entry for pppoe on freshmeat. It is from Dec 1999. Apparently it run in the user space, not kernel. Search on pppoe and you should find it.

Re:damn it...

[Binestar]

Try 426 days of uptime to move the server 10 feet around the corner. All 3 of them. I was fuming.

But what can ya do eh?

Re:damn it...

[hutchy]

My dear Chris, if those 400 and some days were so important to you , there was a solution. One could have plugged in a long extension chord, teased out the delivery end, matched polaritys with the existing (currently working chord) and tapped in. Thereby allowing you to "move around the corner". Just a thought for a possible next time.
Cheers,
hutchy

Re:Zap, ouch

[hutchy]

My appologies to those that don`t have the experience or the test equipment to accomplish this. Actually you don`t really need the test equipment. You only need a lightbulb socket with the bare pigtails. You plug the long extension cord into the same source or power strip as your computer. If no room, unplug a printer or other non-esential. You then open up the power cord to your computer, staggering your entry points. Bare the conductors and find the hot wire. You can find this with your lightbulb and pigtail setup. You also do this for your extension cord. Next step is to splice the coresponding wires together and insulate(tape). Phase and polarity will automaticaly align. Hope this helps. Your comment on leaving it alone if you don`t understand is valid. However, as I have plenty of experience I view this as really simple, and a simple explanation of what to do (as I hope I have elucidated) should not be beyond the capabilty of the magority of slashdot readers.

Cheers,
hutchy

Re:could not be done

[Score Whore]

And taaa-daaaa, I have an upgraded OS.

Does your kernel == your OS? It doesn't in my world.

Re:roll models

[Score Whore]

The point being made (if I read it correctly) is that Rob and Co. aren't particularily noteworthy, just infamous. Not the same thing, and certainly not to be emulated above other CS luminaries.

Re:Uptimes and kernel upgrades.

[Score Whore]

Theoretically it's possible to do that now. But if you aren't Terje Matheson or on of the L0pht guys, it might be difficult.

Re:The importance (or lack thereof) of uptime

[Score Whore]

If they can change your kernel, they can change the value reported by uptime.

Re:The importance (or lack thereof) of uptime

[zmooc]

Because it is an open protocol, cheating happens. Just like in Quake. But everybody with some brains in his head can separate the not-so-accurate-ones from the others. By the way...cheaters are usually no longer than about one hour in the list; by then the admin has disabled their account. Offcourse the statistics only make sense for OS's of which a lot have been registered so think of it what you like. For people that consider uptimes interesting, it is there. If you don't...well don't look.

Re:You guys EAT tulips?!?!?

[zmooc]

I do eat them. I usually take some fried tulips with mayonaise first, then I have some space-cake (with curry-ketchup and some pieces of tomato) followed by champignon-tea with LOTS of sugar (tastes REALLY fucked up) and then I have a nice Northern Lights/V42-joint and some Jagermeister.

Now you probably know the origin of my nick...

Re:The importance (or lack thereof) of uptime

[13th seer]

Superiority - You are correct. I've had Windows NT and even 95 boxes up for months at a time.

(pendantic) impossible. maybe you had 95 up for 54 days at a time, but any longer is not possible without system lockup. I think it was some sort of timing bug

Re:could not be done

[13th seer]

I think this idea has been thrown out as not being important enough. as far as coding difficulty versus usefulness goes.

Re:The importance (or lack thereof) of uptime

[miahrogers]

erm wouldn't that be a HUGE security risk? ie someone cracking into your system, replacing vmlinux with a new insecure vmlinux, rebooting, and changing your uptime to never show it?

that's all nice, but

[waddgodd]

Pardon me if I don't jump right up and download/compile it: I'm waiting for patch-int-2.2.14.X. I'm willing to bet that I'll never see hide nor hair of it anywhere on Andover, either :/ C'est la vie, pero por la tras ;)

Re:that's all nice, but

[G.A. Heath]

I will also wait for the international kernel patch to upgrade since there are fixes and new features that apply to me also, but I've already downloaded the patch so I have that much less downloading. And since you mentioned the International kernel patch we should point everybody to the place that tells them what it is and how they can setup an encypted filesystem with it.
The International Kernel Patch
and the encypted FS howto is here:
loopback encrypted filesystem HOWTO

consider yourself safe..

[niteq]

it's a good thing you're not in holland, Michigan. People have gone to jail for damaging tulips before, and I don't wanna know what kind of punishment munchin on one will bring ;)

Re:New platform introduced with v2.2.14 (S/390)

[galmeida]

I saw 3 of them (S/390), one is very old - water cooled.

Re:Oops

[Pppplahman]

Ah well, my arguments never hold up any good anyway. should have read My arguments never hold up any well anyway.

HuH?? Who are you kidding?? There is no such phrase as "hold up any well"...Unless you want to hold up any oil well in the couny, but that's beside the point. My point happens to be that the correct English for that phrase is:"My arguments never hold up VERY well..."

Okay, so I'm anal.:) Pppplahman:)

Re:uncle? what has he got to do with that?

[Simon Brooke]

Pardonez moi, ma petit choux de Pays-Bas, but tout le monde et sa tante a dit ca en la France pendant touts les annis.

[Or something. I don't really speak French].

What should I do to work with the new modules?

[Kit Lo]

I've just erased all that is Microsoft so I live with only Linux from now on. I have just installed RedHat 6.1 with the updates and whatnot from a few hours ago... I'm downloading the linux-2.2.14.tar.gz file (not the .patch file) from ftp.us.kernel.org, and I'm reading a page on upgrading the kernel from Linuxnewbie.org at the same time. However, I can't put my finger on how to properly install and use the new 2.2.14-bundled modules. I know the computer will "modularize" everything else when I follow the Linuxnewbie directions, but I don't know how to make it so that I am using ONLY the latest module for my Adaptec SCSI controller (the "aic7xxx" module) and nothing more. The last time I used Linux, I remembered something to the effect of "the aic7xxx module is already loaded (possibly the 2.2.12-20 version, from an initrd file)" during the bootup sequence. Obviously I screwed up something back then. Even more, I will have to take care of the 3dfx Voodoo3 stuff and the Sound Blaster Live! support for the latest kernel as well. What should I do?
--

Thanks!

[Kit Lo]

Thanks a lot for that advice.
--

Re:It is not hard

[m3000]

LIke I said, I don't think it's hard if nothing goes wrong. But the first few times I tried it, I kept getting compile errors. I'm not really an in-depth programer (yet) and so I had no idea what to do and just gave up. Then I read the article I posted above, installed some stuff that it suggested, recompiled it with some new configuring and it worked. So then I tried to install a new kernel, but when it booted up it said it couldn't mount root or something like that (been a while). Again, I tried a few things but never did get it to work. So if everything goes right, it's easy. But as I've experianced a lot with compiling, it doesn't always work right.

Re:damn it...

[m3000]

Well, I always turn off the computer at night. I don't like to pay for the extra electricity, and should we have a power surge or something, I don't feel like fixing the fscking file system cause it still doesn't have a journaling one yet. Damn I can't wait till that thing is implimented in the stable kernel, not having one has screwed me so many times. Anyway, right now I have 8 minutes of uptime (Winix box) and so I think I'll attempt to install the new kernal.

Re:damn it...

[m3000]

Well, I just dont' get to it. Sure, I guess it would be nice, but it's just a home computer, nothing important is on it. Even my mail is web-based so I don't even have to connect to my home computer to get it. There isn't any real reason to connect to it in the first place.

Re:damn it...

[m3000]

Lets' see, I boot up, start Netscape, start XMMS, and start XChat. Proceed to listen to music, surf the web, and chat in IRC for the next few hours while every once in a while writing some HTML (which I proceed to upload to a server so I can get it later from somewhere else should I need it). Now why exactly would i need to access anything away from home? Heck, I don't even think I've ever had the chance to even consider getting something from my computer at home from another computer. I hardly travel at all too. The only thing my computer does when I'm not sitting at it is crunch some SETI@home and distributed.net stuff. So no, I still don't see why I should leave it on all the time.

Marketing hype

[gadwale]

This would be a good time for somebody to post the email addresses for clueless reporters in the print media and zd net. We could all write in about what a great achievement this is and how it could crush windows 2000.

I would laugh my guts out if this showed up on the seven o'clock news right after "don't open email from strangers 'coz you might be vapourized". Something like this happened a couple of months ago about AOL ....

Re:The importance (or lack thereof) of uptime

[TheTomcat]

While this is true for the most part, I find it particularly annoying when I have to 'restart my computer' every time I change the smallest little thing. Be it my DNS servers, or my screen resolution (yea.. I know.. they fixed the resolution thing). Uptime also possibly shows how long since a CRASH. I can't count the times that netscape crashes on me at work, and then KEEPS crashing, crashing AFTER it is closed, and an OS that doesn't ALLOW me to open the 'task manager' if there's a critical error, but forces me to 'restart' is really annoying.

A large patch, BUT....

[BLiP2]

Well, someone commented that the patch was rather large, 1390 kb, the biggest to date for 2.2,
yet...

just out of curiosity, I loaded up "voices from the Hellmouth" from hof, and the grand total was 1783kb..

Re:That's valid, but one-sided.

[bigchris]

Some high-availability (am I using the right term there?) systems actually have uptime requirements (such as "we can only be offline for ten minutes every month") that make it risky to upgrade with every new kernel. Particularly since new kernels can introduce new bugs.

Er, shouldn't that be a dowtime limit, not an uptime limit? There is a subtle difference...

Here is the final changelog from Alan

[geirt]

It is to long to post here, so have a look at http://www.linux.org.uk/VERSION/ relnotes.2214.html

Have fun !

Re:AGP?

[spinkham]

1) AGP has been in the kernel for quite some time.
2) AGP has nothing to do with any webcam I have ever seen. Perhaps you mean USB?

Won't boot with empty IDE-Zip (non-Atapi!!)

[GlowStars]

Okay, first of all I have to admit that I did not yet try out either 2.2.14pre18 nor the final one but somewhere between 2.2.14pre12 and 2.2.14pre16 the kernel lost the ability to boot my PC if I do not insert a medium into my IDE-Zip!! I get lots of "Lost interrupt" messages during probing of that device. It's not a show-stopper but unnerving. Any ideas what caused this change in behaviour?

Cheers

Re:Filesystem problems fixed?

[Lae]

I've had similar errors on a Linux server, random filesystem corruption, damaged files and directories. The problem manifests itself only with the high disk activity, such as running pretty loaded squid cache. It seems what I was right suspecting bad motherboard (VIA 82C586-something, Celeron Slot 1). Turning off PIO, UDMA (it has fairly new IDE drive) and UDMA support in kernel seems to help. No problems so far, 20+ days uptime (it was no more than a week before). Squid cache directory was damaged every time. The kernel is 2.2.13, but I don't think filesystem corruption is kernel related in my case. Never, never buy cheap motherboard for a server and actively protest even if your boss tells you about limited budget. Bad hardware could be a whole can of worms. I hope I will get this bad motherboard replaced soon. Of course I can't reproduce the problem with Windows 95, high server load needed for the problem manifests itself. Sad story...

Re:The importance (or lack thereof) of uptime

[treke]

Why fix bugs that dont exist and add features you don't need? If security isnt an issue, and on an internal network it may not be as big of one, why take chances upgrading a critical component that is currently working fine. Now if security is an issue, or you want/need the new features by all means upgrade. But if it aint broke don't fix it.
treke

USB!!!

[ffatTony]

Your webcam is probably USB not AGP (unless it plugs into a slot inside of your case next to your video card : )

Re:When will linux have hot swap kernels?

[^switch]

When Linux is implemented on a micro-kernel :-)

Re:AGP? -- Never will happen

[bugg]

As the kernel needs not support for AGP.
Only your BIOS's chipset is concerned with the fact that the card is AGP..
That and your display manager (Your X server, for most of you) must support the chipset on the card.

Your X server (for most of you, the an XFree86 server) needs to be concerned. Not your kernel.

Re:Uptime...

[maarten_delft]

Of course it should be top priority to minimize downtime, but I think that the practice of "seeking uptimes" is pretty useless.

I have done my share of it, but now I think one should never ever hesitate to bring the system down when some essential upgrade is needed.

Otherwise the danger is that all those smaller upgrades effect build up to form one mega operation which is a lot more difficult to oversee and which might adversely affect total downtime.

Re:Linux blows

[rantenki]

It seems to me that it was true many moons ago that BSD (pick your favourite one) was the secure os of choice, but now, with linux development forging ahead at ungodly speeds, is there really much of a difference? Has anyone done a decent comparison lately?

Re:damn it...

[cnflctd]

It's a little known fact that .02% of the total US enery supply is squandered in the neurotic quest for longer and longer uptimes. Makes me think of that psychotic indian dude with the fingernails. But then again, look how much we spend on gourmet pet food...

Re:Uptime...

[Inoshiro]

I'm just trying to beat my 2.2.9 record. I don't upgrade kernels without good reason, and 2.2.14 has some good reasons. However, I also don't run kernels without devfs, so 2.2.14 must wait for Richard Gooch.
---

Re:ext3?

[DartX]

Sounds like a devel (2.3.X) sort of thing to me. Keep in mind this isnt the sort of thing we are likely to see in the "stable" 2.2.X series.

Re:AS/400s too...

[technos]

The first time I heard of the port to S/390, my response was something to the effect of 'You'd better start sharing those drugs, pal..'. Granted, I heard about it a month before the official announcement from an IBM employee, but still. S/390 was huge leap; AS/400 shouldn't be too far off.

Re:upgrade?

[technos]

should i upgrade to the new kernel, is there anything in it that i will notice and love

Probably not. Most of this release is just driver tweaks.

how hard is it to upgrade a kernel. Is it just a matter of downloading and compiling it, or is it a lot more complex?

There is an excellent document, the Kernel-HOWTO, that explains it in depth. Truthfully, there really isn't much more to it than that. (make clean && make menuconfig && make dep && make bzImage && make install) The only tough part is making it throught the kernel configuration; picking the right drivers, the filesystems, what to build as modules, etc.

Re:There is this thing called the Sun ...

[cheese63]

Brother, if kernel upgrades are your life then you need to get the hell out that house, sip some slimfast

I've heard from people that slimfast tastes fairly bad. Can you vouch for that?
By the way, I'm well aware that this post is offtopic, but I made it 2 levels underneath an offtopic post, as to not bother anyone.

Re:AGP?

[GooberToo]

I assume you mean USB????

Re:The importance (or lack thereof) of uptime

[goldfish]

*ahem*

Programming skill has a great deal to do with uptime if you happen to be writing kernel modules .... ;)

--
bje

Re:You guys EAT tulips?!?!?

[Drongo14]

Yep. We eat them as dressing on our sandwiches instead of iceberg salad ;-)
Actually, apparently they're horrible to eat, but people did eat tulip bulbs during the hunger winter in 1944/45. Basically all potatoes, onions and other foodstuffs had been sent out of urban parts of the country, and the only vaguely edible thing left was tulip bulbs.
Kind of gives a new perspective on the hunter/gatherer habits of man...

Oh yeah, and a happy 2000 to all of you.

Re:Whatever....

[Will the Chill]

The two major differences between us being that you posted anonymously, and were moderated at a 0. Just because I like to talk about my uptime doesn't mean I'm bragging. And CmdrTaco even said, IF your uptime isn't sacred to you, update the kernel. If Rob will acknowledge that uptimes are important, then that's good enough for me.

-Will the Chill

rephrase of original post

[Will the Chill]

Sorry, guys. What I MEANT to say was that this was a good uptime considering it's running kernel 2.2.13.

-Will the Chill

Re:rephrase of original post

[Airline_Sickness_Bag]

Well, I just installed the latest version of fetchmail on a friends' linux system. The uptime of that system is 95 days, which corresponds to the last big power outage here. Not really noteworthy, even if it is only running Linux 1.0.9 (redhat 2.0.2). -me

roll models

[Will the Chill]

Actually, I'm currently studying under one Dr. Cooke, who happens to be close friends w/ the likes of Ken Thompson and Vaughan Pratt. I've talked extensively w/ Dr. Cooke about such people, and have even showed interviews given by my prof w/ Thompson at one of the *nix group meetings I run at school.

Furthermore, I'm very good friends w/ a volunteer at GNU, and keep up w/ all of RMS's posts and such. Not to be outdone, I've also humbly requested the presence of ESR at a meeting of mine in April, and he's accepted the invitation.

I think it's safe to say I'm not giving into the "web hype-monkeys". I simply stated that I share many of the same view and opinions.

-Will the Chill

Link for the lazy

[ibanix]

The source code is available here:
http://www.kernel.o rg/pub/linux/kernel/v2.2/linux-2.2.14.tar.gz

Re:The importance (or lack thereof) of uptime

[TheKodiak]

That's fine. Percentage uptime is a good measure of that - not the length of each uptime. I would rather have a computer go down once a month for 1 minute than once every two years for a day. (Not that going down once every month will prevent you from having a day-long outage.)

No one is saying that Linux is bad. Well, ok, the guy you were replying to wasn't. The only claim was that having an uptime any longer than the most recent necessary security patch is stupid.

Re:To the sys-admins at /.

[TheKodiak]

No, no, I think we should use MTBF as a measure of system stability, so we can punish machines for restarting, only to decide that they need to restart again. (Hmm. Wonder what OS I could have in mind?)

Betraying The Promise Of Modularity

[tipjar administrator]

Yes. ptrace can provide a freeze/unfreeze, if
you can marshall the open pipes somehow; with
a "hibernate" system call that shut everything
down in a resumable state you could resume from
a new kernel with your uptime intact

Additionally it seems that the promise of modularity, which is that you will only need to
get the upgrades for your particular hardware, is
not being kept at all; I find myself downloading
millions of lines of code to fix problems with hardware I do not own when all I want is one particular feature.

I wonder how tricky it would be to make a database
that you could upload your .config file to and
download the patches pertaining to what you will
actually build... Comments?

Re:AGP?

[discore]

heh, for some reason i think you meant USB astrotek...
AGP is a slot usually used for video cards on your motherboard, USB is those flat 1/2 inch ports in the back of your computer, that your webcam/scanner/mouse/whatever is plugging into.

USB support for the linux kernel has been dreamed of for a while now. If i remember right if was in 2.2.13 but was Experimental. So I'd give it til 2.2.15 or .16.

tyler

Re:101 days on 2.2.10

[_GNU_]

twice, acctually...
as my system doesnt do anything critical, logs are the only thing that suffers from a full / , of course, I could just setup a logd on one of my macs or so.. but I don't have a need for that.

101 days on 2.2.10

[_GNU_]

I'm still on 2.2.10, compiled around 150 days ago (had a powersupply mishap :/)
otherwise running stable, some ppl tell me about problems with .10, but nothing for me.

Filesystem Size Used Avail Capacity Mounted on
/dev/hda1 39M 35M 1.2M 97% /
/dev/hda2 136M 91M 38M 70% /usr

hmm... haha. no space for the 2.2.14 source anyway ;)

Re:New platform introduced with v2.2.14 (S/390)

[hautis]

if just anyone could port Linux to VAX, things would be chilling.

There's a port in the works, latest patches were sent today (if it's 4.1. 2000 in where you're from).

Address is http://www.mssl.ucl.ac.uk/~atp/linux-vax/ and there's a link to all source and cross-tools and the patches.

However, there's always NetBSD.

Re:At last!

[abram_fettig]

I liked your story. Were I a moderator today I would give you a point.

To the sys-admins at /.

[SuperguyA1]

Don't listen to this please. The less downtime you guys have the better! Everyone else go ahead.

Re:The importance (or lack thereof) of uptime

[SuperguyA1]

How can you say that programming skill has nothing to do with uptime? It can't all be luck. Bugs cause crashes, crashes cause downtime.

Re:The importance (or lack thereof) of uptime

[mafried]

I believe he was refering to the programming skill of the user. That skill should (under most circumstances) have absoluty no effect on the overall uptime of the system.

If I write crappy code on Linux, and it crashes, I have a core file to look through, and a lot of work to do. If I write good code, then I have an application that works well. Either way, the uptime remains the same.

You can counter this by saying that bad (programming) skill would crash the OS if you were working on a system such as the Mac or Win9x, and thereby destroy the uptime. However, most of the time uptime does not consern the people who use these OSes for work, and the average user probably does not even know what uptime is.

Conclusion: uptime and programming skill are (unless you are doing kernel development) unrealated.

Re:AGP?

[tsphere]

it's not like the voodoo3 even _uses_ AGP features for anything. and everybody knows that's all that matters. ;-)

i'm just happy that the kernel can see that the slot's even there!

Bullshit

[smash_phase]

You don't upgrade a kernel for fun, only if there is an urgent need for it, to get something to work that is broken / not supported. This idea is really stupid.. What do you want to keep up?
A website? Bit difficult if you unload the kernel

Re:Bullshit

[PerlGeek]

I upgrade for fun - it gives me a warm feeling,
knowing I have the latest kernel out there. I
don't even have to run it, I just have to have it.
I run a 2.2.12 kernel, personally. I have a 2.2.13
patch, it's been sitting on my hard drive for a
couple months. Now that there's a new one, I must
have it. I'll wait till the mirrors have it to
download it, I think. Then I think I'll actually
compile a new kernel. It's been awhile, I'm out of
practice. Besides, it's fun (for me).

Re:When...?

[Mr. Hankey]

Sounds like a job for a Microkernel to me, I believe at one point this was one of the goals of the HURD. Anyone care to correct or clarify this?

HPT 366?

[Marcio Silva]

Anyone know if they'll ever be including the High Point ATA-66 controller drivers in the kernel? the patches have been around for a while.

Re:HPT 366?

[Doctor Wonky]

I believe that they have been in 2.3 since 2.3.13 Dr W

Noticed that...

[kcarnold]

...but figured that correcting my correction would be too degrading, not to mention utterly boring for anyone reading. Yeah, anal.

Amendments

[kcarnold]

Common interpretation seems to go against what I had intended to say. I really meant that, for every kernel release, the stable branch should be mostly static (and compared to the development it almost is) in general. I'm not a kernel insider so I don't know these things, but why IDE and [N]FS are broken should not be a concern today. After a few patchlevels to fix any bugs discovered by the increased number of systems running the stable kernels, the stable branch should stay static unless some very important bugfix or highly demanded feature comes along that can be integrated easily and without introducing [m]any bugs. The kernel developers should then concentrate fully on the development kernel, not worrying about the stable branch at all, and prefect the development branch. Then, when the development kernel runs out of things to fix and improve, it should be released as the stable and basically kept the same after that. Hmmm... that vaguely sounds like the original intent of the two-branched kernel.

Just curious (don't have enough time to search for myself), why is NFS, etc., "broken"? Was it "broken" in the older (2.2.2) patchlevels? And if so, why didn't the NFS maintainers fix it before releasing the kernel as stable? (See above)

If you think that this sounds like some pathetic, stupid, wrong, whining Linux newbie, then you are likely correct; I am not "in the know" as much as others.

Why?

[kcarnold]

Stable is pretty good. Concentrate on the devel branch and get us 2.4, guys. Any work on 2.2.x is going to be practically scrapped anyway come 2.4. I'm happy with 2.2.13.

Okay, maybe there are a few minor bugfixes. Ah well, my arguments never hold up any good anyway.

Re:Why?

[fsck]

There was IDE wierdness concerning geometry, I spent a few hours trying to get a 36gb ide drive to partition to no avail, on 2.2.13. Sure enough on some mailing lists it was said that 2.2.x has some bogus geometry problems so that big-ass disks are seen as ~2gb. 2.2.14pre? and up fix this. Thanks for fixing it too.

Re:Why?

[Inoshiro]

2.2.13 had odd IDE corruption issues. I like having to not worry about my IDE drive corruptioning itself...

On another note, they also made the memory buffer hash table more efficient. I also like speed :-)
---

A few questions and other stuff

[kcarnold]

What exactly do you mean by "old kernels"? (good grammar OFF) The current stable branch as opposed to devel? Early patchlevels of the current stable? Old stable (2.0.x)? (good grammar sort of ON)

I can't argue with "well tested." This is definately a necessity in a production, or even casual user's environment. However, I do contest your latter two points. First, the C libraries do not change, or should not change, from kernel to kernel. The only user-space C code that could change is the kernel include files, which are relatively simple (thus much less prone to error) and usually do not change much, if at all, from kernel to kernel. (Where there are changes, they usually do not involve the set of functions used by user programs, only kernel administration programs such as modprobe, ipchains, etc..) Thus there is also little or no concern about the programs that can run under that kernel.

I must also point out (the approprateness of the following point to the discussion is based on your definition of "old kernel", as questioned earlier) that an individual patchlevel is not itself updated when bugs are fixed. It is brought up to the next higher numbered patchlevel. The problem is that these new patchlevels can also introduce new features, which may (and usually does) involve questionable (that is, not guaranteed to be bug-free) code. Then, when these bugs are found, the patchlevel increases again, and more questionable code may be introduced. The developers wind up spending too much time on the stable branch and rarely coming out with something that is truely stable. This was my point (see my "Addendum" reply if the point seems unclear). Leave the stable branch the same if possible. No new features. Leave that for the new devel. Just make it and keep it bug-free.

That would be a suitable conclusion, but I have a few more issues. With no flamebait intended, I think that Microsoft's Win95 to Win95 upgrade was a good thing. They fixed a lot more than one bug (though they clearly added some), improved the UI, and added some handy features like taskbar toolbars that made the OS upgrade worthwhile. It isn't a very difficult upgrade in any case (though the cost may be prohibiting to some and annoying to most others). If uptime is critical and the system is any more than a small server, the administrators would use (I'll stick with Microsoft products here) Windows NT / Windows 2000 anyway.

I cannot argue with you or anyone else that there are systems which absolutely must remain up. In fact, this absolutely supports both my original and revised points. The IDE and NFS issues would not be here if the kernel developers only fixed bugs in the stable branch and made sure that the development branch was bug-free before releasing it to the masses. Keep in mind, however, that, at least in most cases, you are not required to upgrage your kernel. If a hospital or airline is using an older version of the kernel and it works fine for their purposes, then they do not have to upgrade to the latest kernel.

That's all.

Oops

[kcarnold]

Ah well, my arguments never hold up any good anyway. should have read My arguments never hold up any well anyway.

I often catch other people on that type of error; I should catch myself also.

Yes it is (Re:It is not hard)

[Punto]

I don't mean to scare anyone, but this are my experiences compiling kernels:

1) It took like 5 hours to compile 2.2.12 on a 486dx2 with 8MB of ram. It didn't work because I forgot to change some stupid option about the type of microprosessor (it was compiled for pentium). I used menuconfig, so I didn't go through _everything_.

2) For 2.2.13 it took 15 minutes (on a p233mmx with 40 mb), I went through every little option (in fact, I used config instead of menuconfig). When I reboot, the network doesn't work (I'm using the machine for masquerade).

So, I'm stuck with 2.0. that comes with debian.. Maybe I'm used to 'intelligent' and 'wizard' installations.. But I'll keep trying..

--

Slow Mirrors... and want crypto???

[pol-pot]

Well I just noticed that also has linux-2.2.14 and they have a lot of crypto in the kernel that isn't in the official kernel.

(That is why they doesn't put it on u.s. mirrors... (Crypto-regulations...))

Framebuffer Devices

[syn.ack]

Is the ATI framebuffer able to be used on PC's? Does it work in this version?

Re:Uptimes and kernel upgrades.

[stoev]

Then if you want to have hot swapable kernels, I think you have to pray for HURD.

In HURD it will be possible to replace almost all the kernel without reboot.

Re:Uptime...

[stoev]

I have one linux 2.2.13 15000 kilometers away from me. The stability of this box is very important for me, because my familly uses it to contact me.

What I am going to do now? I am compiling the new kernel 2.2.14 on this box 15000 km away and I am going to install it as default kernel. But I am not going to reboot it!!! The next reboot new kernel will start, but for now I will continue to enjoy uptime and clean kern.log

There are no important bugs and exploits in 2.2.13 and I don't see a reason to boot in 2.2.14 for now, but I will be ready with it.

It's called "humor."

[Xzzy]

Why can't people take what was said at face value and just get on with their lives?

There was no value judgements made in the small blurb that suggested those with long uptimes are more elite. It was merely a way to suggest that, if you've nothing important going on, get the patches and install them.

People react to small twists of phrases in these news entries like they're prophecies given by god, and feel that's an invitation to nitpick details far beyond the context they were used in.

Try not to be so frigging sensitive, folks. Read the news entry, glean what it means, and forgive the guy for putting a twist of levity into his news post. If you want a soapbox where you can rant againt people who brag about uptimes, go make yer own website.

Spare us from having to read over the top of you using each news post as a springboard to rant about your peeves.

Re:Stupid Question?

[ScutterBob]

hmm, underneith ie, from a shell it looks like unix, but as far as i can tell its setup as single user. The GUI is not exactly like anyhting else i know of, but not totally unlike X. It boots really fast, from POST its maybe 14 sec. on a cel.333 with 64MB. It uses a 64bit FS, is geared to media production or uhm at least using lots of multimedia stuff. I can't really find a whole lot i can mess with meaning, i can't find config files i'd like to. But I don't really use it much, just wanted to see what its like. It is very fast.

Bad idea

[gammatron]

Basically it's compiling the new kernel copying it over the old one and that's it.

Uhm, bad idea. You should always keep the old kernel. What if your new kernel doesn't work? Since you copied over the old one (which is known to work), you're screwed.
--

Woo hoo!

[KarMann]

Woo hoo! Looks like with this one, I can use my new joystick with Linux now! The Logitech Wingman Extreme Digital 3D wasn't in 2.2.13, although the same minus the 3D was, but I didn't notice the different name when I bought it. And just when I'd bought a book on device drivers so I could fix it myself, they go and pop this in!

Good... bad... I'm the one with the gun.

Re:it's not blind faith

[m.o]

What's wrong with "web hype-monkeys" ?!?!

Maestro audio driver

[G.A. Heath]

Zach Brown's Maestro audio driver is also included in kernel 2.2.14, for many of us this is a god send. I went to Zach Brown's webpage at redhat and retrieved the driver, but many users with the same notebook computer (toshiba Satellite 1555CDS) will be glad to see it in the kernel. If you have a notebook and wonder if your unit can utilize this module do lspci and look for something along the of "ESS Technology Maestro" in there. I know there are better ways to do it, but I don't have time to list them in a manner to avoid confusing the new users.

Re:The importance (or lack thereof) of uptime

[Signail11]

There is a difference between uptime as stored in /proc/uptime and system availability to users. Certainly, it would be nice if both could be kept up indefinitely, but real world issues prevent this from happening. If you have a server which thousands of users access, I would hope that you have some fault-tolerance mechanism is place. I would also hope that one performs the neccesary upgrades at off peak hours.

Re:AGP?

[jasno]

hmmm.. sounds like a confused windows user. flame me if i'm wrong, but I think the patch for win95B to use AGP required installing USB support...

FireWire??? (Re:USB!!!)

[yerricde]

Your webcam is probably USB

Or FireWire(TM) aka i.LINK(TM) aka IEEE-1394. High quality digital motion cameras (camcorders, etc.) tend to use the faster FireWire ports on Power Mac G3s, newer iMacs, and newer [LW]intel PCs.

Does any one out there really use Gnome or KDE?

I use GNOME.

WM is my favorite

To each his (or her) own.

I'm tired of start menu's (This also includes stylized "K"'s and "little feet")

The Mac "start"ed it all. It put its start menu in the upper left corner.

[OT] Your sig

[yerricde]

Tetris rules

No, Tetris sucks. The game of falling tetrominoes may rule, but Tetris sucks. Care to refresh your memory?

Re:Uptimes and kernel upgrades.

[yerricde]

Then if you want to have hot swapable kernels, I think you have to pray for HURD.

Or work on HURD yourself.

In HURD it will be possible
f0r 5kr1p+ k1dd135 (for script kiddies)
to replace almost all the kernel without reboot.

Security is always an issue. Maybe it's even a "known issue." Micro$oft Crapdows (any version after 3.11) has thousands of known "issues," and I'm not talking Korn CDs here.

Re:The importance (or lack thereof) of uptime

[jbarnett]

It is not the amount of uptime you have, it is how you use it. If you have an internal system that maintains a db that is used by the billing department between 8-6, it is alright if the system goes down after 6 oclock or upgrade/repairs.

Hell you could have a really bad percent of uptime, reboot at 6 and bring the system back up at 8 in the morning, the majority of the time, the system is down.

But when the system needs to be acessed (8-6) it is up running and fully funcational and secure due to the patches/upgrades that was done at 9 at night. When the system needs to do billing transactions, it is up and running great.

It is not the amount of uptime you have, but when and how you use the uptime. At least that is what my gf tells me.

Re:The importance (or lack thereof) of uptime

[jbarnett]

Just for the record I am not a small man, nor is my uptime.

Yes all Americans are...

[JammmGrrl]

...gullible, as I'm sure you've heard. We are all also arrogant and rebelious blokes, and, seeing as how I live in Washington state, all I drink is fancy coffee. :)

BTW, isn't it also true that Holland (or parts of Holland) are below sea level? I seem to remember something about dikes in my Americanized Hollish folklore memory banks....

You guys EAT tulips?!?!?

[JammmGrrl]

I knew you had them, but you eat them too?

Re:You guys EAT tulips?!?!?

[Wizard of OS]

Sure :)
Course not, I just typed it because I thougt you guys wouldn't know what Holland is if I didn't mention tulips and wooden shoos :)
If you ever visit the Netherlands, don't be surprised to see 0.000001% of the population with wooden shoos and 500 others types of flowers in the flower-shops :)
I've never really tasted flowers, but some people say they are great. I personally prefer steak with french fries ;-)
Thank you for your reply, it now is perfectly clear to me that Americans (or wherever you're from) just beleave anything a foreigner says ;-)

Re:Uptime...

[RuntimeError]

10:36pm up 20 days, 21:11, 2 users, load average: 0.30, 0.27, 0.29

I was hoping to do better than the average, which is supposed 59 days according to the uptime project.

The box is working perfectly, so I am wondering wether to upgrade or not. After all, "If it ain't broke, don't fix it"

Congratulation to all involved.

[Etam]

Going from 2.2.13 to 2.2.14 took unusually long but not unexpected; not only there are plentiful of changes, but bugs that came out were very closely tied to various hardware. All the hard work Alan and others who have put in finally pay off. Being a bystander that follow the development on the side, I can only watch in awe.

There are so many fixes in many areas and some significant re-design which took time to debug. Some of the new goodies that interest me include PCI parallel port handling (for your parallel port cdrom/cdrw), Matrox G400 frame buffer support and Network Telephone support.

I hope a lot of people can try out this release. And remember, if you find bugs, report them :)

Re:ext3?

[Etam]

Right now it isn't in 2.2.x nor 2.3.x kernel tree yet. There is a patch out there that can add this feature. My best guess is that it is not stable enough to be included yet.

Re:Uptime...

[firstmagic]

My UPS has worked wonderfully thusfar. It's kept my machine up through at least 4 power outages. One 3 second outage, one 20 minute outage, and two half hour outages (the last two were my doing -- had to cut the power to that part of the house to upgrade some wall power outlets).

10:16pm up 238 days, 11:41, 6 users, load average: 2.00, 2.00, 2.00

The load average of 2.00 is only because it's a dual proc running a pair of rc5 crackers for distributed.net -- they soak up idle time.

Still running 2.0.36

wooden shoes and tulips?

[GaiJin78]

i hafta confess im a dutchman myself too, but i havent seen a couple of wooden shoes for a long long time! and tulips... as far as i know, i have never TASTED one! :) Just for the heck of clarity, ok? i dun wanna the folks think we r just a weird race of ppl... :)

Re:AGP?

[Anonymous Coward]

No, in this case it makes sense. It's more of a motherboard chipset thing than a video card thing. AGP lets you transfer large amounts of texture memory (or something like that) over the AGP bus, using main memory as texture memory. You can't do that without changing the kernel (or writing a module). It's not a graphics library issue.

Re:AGP?

[Gleef]

As far as I can tell, AGP support has been in the kernel (as part of the PCI driver) since at least version 2.0.34, dated 1998-07-04. So it's been in there for at least a year and a half. Whatever problems you're having with your webcam, it should have nothing to do with "missing" AGP support.

----

How many webcams use GART?

[Gleef]

The original poster was complaining about AGP support because he wanted to use his Webcam. Digital cameras give simple 2D images, since when does this require kernel-level GART support?

----

Re:AGP?

[whoop]

For all of you that say AGP cards work in the kernel already, that's only half true. The Utah GLX has a kernel module that lets you use all the nifty fast memory transfers over the AGP interface. Sure the kernels can run AGP video cards, but this makes them much faster (or so I understand). The patch was already merged into the 2.3 kernels, but not the 2.2 ones. You can get it from the project's CVS server.

Supermount patch here

[Baddog]

Supermount isn't integrated into 2.2.14, however I have the patch mirrored here:

http://www.fargocity.co m/~ccondit/supermount-2.2.14-1.patch

This has been modified from the original supermount patch to patch cleanly on 2.2.14-final (md.c failed before).

Re:Uptime...

[dattaway]

The power outages seem to win...

A few weeks ago I went into work to find the local paper with a front page article stating the utility company of this small town was going to upgrade the substations. I had a spare car battery hooked to the UPS and thought it would ride through. Uh huh... looking at the logs, it looked like the battery was a half an hour shy of the four hours of no power.

Re:Filesystem problems fixed?

[drix]

They were a big problem around 2.2.10, were claimed to be fixed in 2.2.11, but happenened for me and a few others as late as 2.2.12. Hopefully they're all ironed out by now.

--

Filesystem problems fixed?

[Booker]

I saw on Kernel Traffic that there were some freaky filesystem corruption problems in the 2.2.x series... and I think I may have experienced this. Anyone know if this has been nailed down yet?
----

Re:Filesystem problems fixed?

[Booker]

I'm always a bit afraid to send in a bug report, because I'm not sure I know enough to put together a good one, and I don't want to increase the static on the list...

I did get a kernel oops that seemed ext2 related, and sent it to the ext2 maintaner, but got no reply.

Any suggestions on what I could do to:

1) be sure this really was a bug, and
2) formulate a useful bug report?

This last time, I got a whole slew of:

Jan 1 10:44:05 Lager kernel: EXT2-fs error (device ide0(3,7)): ext2_add_entry:
bad entry in directory #11873: rec_len is smaller than minimal - offset=0, inode
=11873, rec_len=6, name_len=1

entries in the log, and there were in fact several problems when I fsck'd. (This was with 2.2.13) Also, interesting (but probably unrelated) datapoint... it was the postfix files/directories that were damaged each time...
----

Re:Filesystem problems fixed?

[Booker]

Always send to linux-kernel@vger.rutgers.edu

Ok, ok, I get it! :) I promise to do that next time. As I said, though, I wanted to make sure that I didn't bother the list with something unimportant, and it's hard for me to tell sometimes...

Compiling 2.2.14 as we speak....
----

Re:Filesystem problems fixed?

[tao]

There has been lots of fixing of wicked corruption going on for v2.2.14; most of it were very special-cases, but some might have affected larger user-masses. I suggest you upgrade to v2.2.14; it's been tested for a loooong time now, and it seems really stable.

What you should remember is, that if you suspect file-corruption, please send a bug-report and a careful description (hardware, kernel-version, etc.) to linux-kernel@vger.rutgers.edu. More often than not, such reports can be of great help to us when we try to find bugs. This of course applies to other kernel-related bugs too (hangs, etc.)

Re:Filesystem problems fixed?

[tao]

Well, if this bug repeats itself for v2.2.14, you should submit another report. However, even if you send a report, always carbon-copy to LKML (Linux-Kernel Mailing List); not every bug that seem to relate to one thing have to, and sometimes the maintainer is away, misses one or two messages in his (her?) inbox or whatever.

Remember that the stream of mail to most kernel- hackers is huge, so if you just send them personal mail, simply procmail it away. You never know for sure. Always send to linux-kernel@vger.rutgers.edu (My, this must be the 3rd or 4th time I write that eddy here on Slashdot today...) as well. Others may be experiencing the same problems.

Remember, without bug-reports we don't know about the bugs...

Re:Filesystem problems fixed?

[spinkham]

From what I have read, 2.2.13 fixed all that pretty well.
I haven't followed up on that recently though, so I may be wrong..

Checklist was for 2.4

[Booker]

Checklist posted earlier was for 2.3.x -> 2.4

This article was about the 2.2.14 release. :)
----

Re:The importance (or lack thereof) of uptime

[Syberghost]

Let me get this straight; you think somebody will replace your kernel with whatever they want, reboot, and the uptime will be your tipoff?

And you think this hypothetical hacker will be savvy enough to code holes into the kernel (as opposed to into the utilities, such as login, which authenticate users), but won't be savvy enough to fake the uptime?

Uhm, sure, yeah, that'll happen. You'll be warned of it by the monkeys flying out of my butt.

Yes, in that extremely bizarre contrived case, it would be a security hole. However, if we're assuming the attacker can replace the kernel with one of this choosing, *EVERY* Unix-like OS has that hole. They have the hole now.

If your system is that owned, you shouldn't be trusting what any piece of software on it tells you, not even the BIOS if you're flash upgradeable.

Re:it compiled okay...

[tao]

Hmmm. That's bad. however, that description isn't a very good bug-report. I suggest you mail a careful description of what system you have and what your problems are. Compile-time related, missing/non-working drivers, file- corruption, hangs/oops:es/panic's, etc.

We do all we can to fix bugs, but we can't fix them unless people report them (ok, we do sometimes when we stumble over silly code by accident, when doing rehauling of code or when having little else to do, but those cases are not in majority.)

linux-kernel@vger.rutgers.edu is the place where you report your problems. Good luck!

Re:AGP?

[BJH]

AGP? Methinks you mean USB...

Boy, your face must be red right about now ;)

VM to boot new kernel

[SurfsUp]

Wouldn't it be possible to make a VM that boots the new kernel, then copies itself into the older kernels memory space?

Yes I guess that would work. The main problem would be getting hold of the other kernel's memory - you wouldn't have to do any copying at all, but you have to get the memory from somewhere. Presumeably you'd be running under VMWare, or better, an open source vm os, and there would be some kind of api for sharing memory among the various client os's. You'd use that to recover the memory from the halted os. Then figure out how to hot-swap your VM os and you're done. :-)

Re:could not be done

[um... Lucas]

Wouldn't it be possible to make a VM that boots the new kernel, then copies itself into the older kernels memory space? That way, supposing everything went smoothly, no reboot would be needed? Just curious if that's a possibility or not? it'd have to be handled in the kernel, so as to allow itself to be overwritten... Memory protection seems like it would be a hassel for anything but SW inside the kernel to accomplish that.

it compiled okay...

[kerouac]

but some stuff is still broke (for my set-up, anyhow).

I don't think that checklist posted earlier was completed.

*sigh* i guess that'll be worked out in the 'ac' patches...

Incase you DON'T know where to get the good stuff.

[Yakman]

List of Kernel Mirrors.

KernelNotes.org has changelists and things but hasn't been updated for 2.2.14 yet.

Re:damn it...

[SheldonYoung]

It's good for a home system, average for a server. It's almost painful to have to reboot a box with over 110 days of update just to put it on a UPS...

Re:The importance (or lack thereof) of uptime

[SEWilco]

Measure the uptime in terms of processor speed, so having a faster processor so you can reboot faster is less of an advantage...

Re:AGP?

[mircea]

AGP is supported in the stable kernel since many moons, my friend - why don't you check it again.

long uptime = proof of few resource leaks etc.

[coyote-san]

I agree that you should never perform a necessary upgrade just to preserve an uptime value, but I strongly disagree with your claim that a long uptime is meaningless.

A long uptime on an active system is proof that the system doesn't have significant resource leaks. The same logic applies to systems that don't require frequent disk "defragmentation", etc.

Resource leaks, by themselves, aren't dangerous other than the fact that they force you to reboot the system to recover the lost resources... but they are excellent indicators of the overall quality of the software. In my experience, all program with significant resource leaks have *always* had an unusually large number of other bugs... and the times that someone has tried to eliminate the leaks just to shut up their noisy boss they ended up fixing a large number of unrelated bugs.

Note I did not say that they "found" those bugs - many of the bugs were due to wild pointers that simply disappeared once the programmer took care to properly manage their resources. IMHO, the second most powerful bug-finding strategy, after fixing all warnings issued by the compiler, is elimination of resource leaks.

Re:Uptime...

[Inoshiro]

1 44 days, 04:36:04 | Linux 2.2.9 Sat Aug 21 18:58:10 1999
-> 2 39 days, 20:21:28 | Linux 2.2.13 Thu Nov 25 19:50:22 1999

So, you see, I am a mere 5 days from beating my 2.2.9 record (at which point I rebooted to use a upgraded kernel), when they go and release 2.2.14. Those bastards.

Ah well, since devfs for 2.2.14 isn't out yet (AFAIK), it'll be a while before I go and upgrade :-)
(Linus PLEASE put Devfs in 2.4)
---

Re:New platform introduced with v2.2.14 (S/390)

[technos]

Up that number by five or so. (We've got at least that many /. readers here) I've got two of them thirty feet from me. I won't get to play with them as intimatly as I'd like to, (I'm the 'low-end' guy, and we're only holding them for resale) but they won't leave without a 'Runs Linux' sticker over the shrink wrap.

I really wish someone would port Linux to one of the older rev AS/400's.. I've had a few chances to snag one off of a scrap deal, but w/ no OS they're useless to me..

MediaGX processor support?

[G27 Radio]

I've got a couple of boxes with MediaGX processors that I bought for $50 a piece...not as good of a deal as I'd originally thought considerring how f*cked up they are. I know Alan Cox has done some work with MediaGX processors to deal with some of the problems. One of the Linux Grrls gave me a pretty specific explanation of the problem (Thanks Kira :) Now I'm wondering if there will be better support in the 2.4 kernel.

The 2.2 kernel boots and runs fine until I try to do something radical like startx. 2.3 however, detects that it's a MediaGX at boot up, but locks up after 'checking the hlt instruction.'

numb

Re:Ken Thompson

[Will the Chill]

Wow, I can't believe that Tom Christiansen actually posted stuff about this.

If you'll follow the link that he gave to Computer, you'll see that Dr. Daniel Cooke is listed as one of the guys that wrote the article interviewing Ken. As I said in a previous post, Dr. Cooke received permission from Computer and Bell Labs to show SPUUG (South Plains Unix User Group; Texas Tech's little group that I coordinate) an unreleased video of him interviewing Ken in person. Alot of different material than that in the printed version.

Also, Dr. Cooke mentioned to me several times that he received QUITE a bit of flame from the Linux community when that stuff that Ken said was printed, even though he was just the one who wrote it up for Computer.

Again, I can't believe Tom actually posted about this stuff. Wow. Let me just say that backward: woW.

-Will the Chill

Telephony Support

[aunitt]

One very important thing in 2.2.14 is telephony support.

This could be an area were Linux could really shine. Telephony is all about reliability and can be very price sensitive, especially when you are trying to put together systems for small companies.

I should know, I work for a company producing exactly these kind of systems, unfortunately on NT :-(

For more details see this article on LinuxTelephony.

Re:AGP?

[nukem]

from the way i see the linux kernel, as long as your video card works good for u, the kernel doesn't need to mess with the video card alot, leave that to the grahics libraries to do, that's why u don't see much of anything in the kernel configuration about graphics, about the only thing i think the kernel might have to do with agp is provide access to that stuff to other progs, but isn't htat why there's graphics libraries?? i'm no kernel or linux expert tho, so this is just what i think, more than likely not the real facts.

Re:Ken Thompson

[atd3000]

What is sad is that sometimes, when our gurus age or get older, the shine on their accomplishments grows dull with their new, wrong headed ideas.

Look at Bobby Fischer's anti-semitism, or Einstein's belief in the solid state universe, etc.

Someday Linus will be 50 or 60 and he'll say that the hot new idea is a piece of crap.

It seems to me that people are first anonymous, then we find out about them from some great accomplishment(s), and then they become yesterday's news when the environment that created them changes.

If his comments about Linux being worse as a firewall than Windows are actually attributable to him, then it is obvious, at least on this topic, he is walking, breathing anachronism.

AGP?

[astrotek]

When does AGP get put into the stable kernel? and is the unstable one any good. I WANT MY WEBCAM =)

Re:AGP?

[tao]

I'd wager that *IF* AGP gets into v2.2.xx, it'll be for a fairly late one. The code isn't fully finished in v2.3.xx if I'm not all wrong.

When it comes to the experimental patches, everything depend on the hardware you have. I've been running v2.3.xx kernels on my trustworthy IBM PS/2 9556slc2 (SCSI) without troubles (apart from having to patch the ibmmca.c scsi-driver in an ugly way) whatsoever, others report lots of trouble.

v2.3 is good, no doubt about that; lots of new interesting stuff, but it isn't feature-complete; lots of stuff that works in v2.2 is broken for v2.3, such as some of the filesystems, etc (but as long as ext2 and fat works, I'm all happy), and the pre-patches are sometimes hard to get to compile.

Finally, more often than not, at least some platform is broken. Sparc seem to have had most problems, but they were fixed up just a while ago, if I'm not all wrong.

If you have important data on your disk or need 24/7 uptime, use v2.2.14. If you don't fall into any of those categories, and have any hardware not supported in v2.2.xx, try out v2.3; it'll be a nice experience, and we need all the bug- testers we can find.

Good luck!

Some limited mirrors...

[strredwolf]

us.kernel.org doesn't have it, but tux.org does. It's a 1.3 meg patch.

---
Another non-functioning site was "uncertainty.microsoft.com." The purpose of that site was not known. -- MSNBC 10-26-1999 on MS crack

Re:The importance (or lack thereof) of uptime

[Syberghost]

I've said it before, and I'll say it again; what we need in order to put a stop to this whole stupid argument for all time is a writeable /proc/uptime.

Let people fudge their damn uptime and all the BS will stop.

Changelist

[Rayban]

The changelist will be appearing here at some point in the future.

Hopefully soon :)

Re:The importance (or lack thereof) of uptime

[chrismcc@netus.com]

The kernel is so out of date that any random script kiddie can grab an exploit or buffer overflow from bugtraq and root the system, obviously not a Good Thing if your computer is running any sort of critical task.

I think you missed something important. remote and local exploits come from userland programs. bind, pop3d, etc. The kernel might have DOS problems, but AKAIK there are no remote root exploits for the linux kernel itself.

Re:Linux NFS

[Leos Bitto]

Yes, NFS is better in 2.2.14. That's why I am running 2.2.14pre14 on our production boxes - I rely on NFS much.

it's not blind faith

[Will the Chill]

It's called respect. I have a very high amount of respect for Rob, and others like him. The very simple fact that many geeks today aren't able to find suitable role-models in their everyday lives will lend this argument even more credibility. I will accept, to a certain extent, Rob's posts to be pretty authoratative. I've read /. for years, and am able to honestly say that I agree w/ pretty much everything the guy posts. Is it so bad that I happen to share roughly the same opinions w/ someone who is substantially more noteworthy than myself? It's not always about being a follower, you know...

-Will the Chill

Re:The importance (or lack thereof) of uptime

[billh]

Uptime is just that: a measure of how much time has elapsed since the last reboot of the system. It does not measure any of the following things:

-Superiority of an operating system

-Ability to administer a computer

-Programing skill

-"Eliteness/coolness"

Let us take this point by point:

Superiority - You are correct. I've had Windows NT and even 95 boxes up for months at a time.

Administration ability - depends on the circumstances. I have a colocated web server that I have been working on quite heavily since I installed it, and I haven't been within 30 miles of it since it was turned on 50 days ago. Uptime is currently at 50 days.

Programming skill - has nothing to do with uptime

Eliteness/coolness - while not quite the same thing, I am very close to closing a business deal with someone that I have been trying to get to sign on with me for months. In the end, it was the uptime that mattered. Or, more specifically, the fact that the machine didn't flinch during a live load test (real content, real users, no simulations) with this person present. The uptime is like a victory -- you can point to it frequently, and then show someone your logs to prove that your machine can do what you say.

Uptime == bragging rights in some circumstances.

uncle? what has he got to do with that?

[Wizard of OS]

Ok, I admit, I submitted. It could be the fact that I'm from Holland (nope, not Michigan, just that little country somewhere in europe where they wear wooden shoes and eat tulips ;) and I don't understand the expression.
AFAIK my uncle doesn't even know where the powerswich of his son's computer is, so I don't think he submitted a post about a new kernel ;)

That's valid, but one-sided.

[RainBrot]

How many Linux kernel bugs have there been that allowed users to gain root access? How many were fixed between 2.2.13 and 2.2.14?

Some high-availability (am I using the right term there?) systems actually have uptime requirements (such as "we can only be offline for ten minutes every month") that make it risky to upgrade with every new kernel. Particularly since new kernels can introduce new bugs.

My point is that it can be irresponsible to upgrade without knowing what the upgrade does, just as it can be irresponsible to not upgrade.

All that aside, not everyone is running mission critical servers. Some people use their computers for fun, and long uptimes can be a source of amusement.

I personally have two systems with long uptimes, and I will not be upgrading them. They're non-critical systems, and not worth messing with. Besides, I like to see how long it's been since the last power failure. :)

ext3?

[Millennium]

I remember Alan saying at one point that he was considering adding the current ext3 sources into the kernel. Anyone know if he's done this yet, or will that be going into the 2.3 tree?

Re:The importance (or lack thereof) of uptime

[JoeBuck]

For machines behind firewalls that are performing some task without any problems, it's best in many cases to just leave 'em alone and let them rack up the uptime.

On the other hand, for a visible machine with a static IP address, hosting web pages or other advertised services, you have to keep ahead of the script kiddies. But not all machines are in that category, far from it.

New platform introduced with v2.2.14 (S/390)

[tao]

Something that obviously passed many by is that v2.2.14 introduces a new platform; IBM's mainframe series of computers; S/370 and S/390. While it's hard improbably that more than maybe ten or fifteen readers of Slashdot even have seen one (I have; we got an offer to get one, but had no IPI-3 disks for it, and no OS; at that time the port didn't exist yet...), but they are basically very different and cool machines.

Have a look at IBM's homepage and search around for some information on them. They have BANDWIDTH.

This is at least cool, if nothing else. Now if just anyone could port Linux to VAX, things would be chilling.

Uptimes and kernel upgrades.

[delld]

Now that win2000 is supposedly comming out, and it supposedly needs fewer reboots, Linux uptime counters are going to have much more competition. Therefore, I call for hot-swapable kernels! I do not want to stop what I am doing, just to upgrade (or down-grade) my operating system! I want an uptime measured in decades!

Old kernels are still important

[coyote-san]

Old kernels are still important, for several reasons:

1) they are well tested
2) the C library for that kernel is well tested
3) the programs for that library are well tested

the importance of this can't be overemphasized. There are a lot of situations where it's much more important to work with a known quantity than to get the ultimate bit of performance or flexibility.

It's worth noting that one of the most damning complaints against Microsoft as an "enterprise class" OS & application suite is the fact that they have repeatedly demonstrated a cavalier attitude towards making big changes in a way that forces users to upgrade everything to fix a single bug in the kernel (e.g., Win95->Win98) or application (e.g., Office file formats).

That's why Linux, and all real enterprise-ready OSes, allow fairly independent maintenance paths for all major versions of the kernels/libraries/applications. It's a bit more work for the developer, but it's criticial when you're talking about systems which *must* remain up. (E.g., if a hospital's systems go down due to an unexpected bug in an upgraded OS, patients may die. If an airline's systems go down due to an unexpected bug, they can lose millions of dollars in lost bookings and contractual penalties for delays.)

Re:Ahh....

[Eimi Metamorphoumai]

Rob said so. If Rob told you to jump off a bridge would you do it?

I wouldn't be able to get anywhere near it. It would be /.'d to capacity. A total of maybe a foot difference between the height of the bridge and the pile of geeks next to it.

Changelog Info

[mwillis]

FYI - if you want the changelog for 2.2.14, just look at the last 2.2.pre14 kernel changelogs. Linuxtoday has a copy here:

http://linuxtoday.com/story.php3?sn=14481

It is a fairly long list of things. The S/390 port is there. Some nice-sounding bugfixes are there, so I'll probably recompile tonight. Also, supposedly it should now compile fine with gcc 2.95.

Re:The importance (or lack thereof) of uptime

[zmooc]

I disagree with you; my experience is that I only need to reboot if something goes terribly wrong or if I want to upgrade a `core' part of the system. Therefore one can say that operating systems with an average downtime that is rather low either are upgraded a lot, or crash a lot. I think the latter has the greatest influence still.

Off course not all systems run under the same conditions; windows computers are probably more often turned off at night than VMS systems, SunOS is usually used on high-end hardware while Linux often runs on crappy hardware and OpenBSD-systems probably have better admins than Linux-systems (no offense, but most unix-newbies tend to use Linux, not *BSD). But still I dare say that the uptime is a real good measurement for the stability of an operating system.

Apart from that I agree with the fact that one should not fail to upgrade because one wants to get the highest uptime possible. On the other hand, people shouldn't upgrade when there's no need to; if there are no new features/fixes in the new kernel which apply to your system, don't upgrade :)

Check http://www.uptimes.net for a list of uptimes per OS. There are about 500 hosts in the list, so it ought to give a rather clear view of the situation.

Re:Ken Thompson

[Tom Christiansen]

Isn't he the one who says that Linux is a piece of shit? Sounds like a great Slashdot role model to me!

Ken *invented* most of what you know as Unix and C. (It's fun to watch him and Dennis both disavow ownership and point at each other. :-) Without Ken, we wouldn't have Unix, and we probably wouldn't have C. And we most certainly wouldn't have Linux. If Ken said this, then I'm completely certain that he could have backed it up. But I don't recall having read anything by him that referred to Linux so scatologically. Please don't spread gossip and rumor, allowing idle speculation to blossom into bitter invective against a man hte likes of whose genius you seldom meet in one lifetime. Always get the exact quote and context.

[...time passes...]

Alright, here you go. Read this, which I got from IEEE Computer Magazine:

Computer: In a sense, Linux is following in this tradition. Any thoughts on this phenomenon?

Thompson: I view Linux as something that's not Microsoft-a backlash against Microsoft, no more and no less. I don't think it will be very successful in the long run. I've looked at the source and there are pieces that are good and pieces that are not. A whole bunch of random people have contributed to this source, and the quality varies drastically.

My experience and some of my friends' experience is that Linux is quite unreliable. Microsoft is really unreliable but Linux is worse. In a non-PC environment, it just won't hold up. If you're using it on a single box, that's one thing. But if you want to use Linux in firewalls, gateways, embedded systems, and so on, it has a long way to go.

Delving deeper, we have this article by Eric Raymond in Linux Today, in which he clarifies what Ken said, as follows:

The best news, I guess, is that Ken says he didn't intend to write off Linux itself as simply an anti-Microsoft backlash; what he was trying to say was that he believes the recent popularity of Linux in the press is an anything-but-Microsoft phenomenon. He adds ``i very much appreciate the chance to look at available code when i am faced with the task of interfacing to some nightmare piece of hardware'' and that ``i think the open software movement (and linux in particular) is laudable.''

Ken further adds ``i dont see eye-to-eye with microsoft's business practices.'' His original language was rather stronger and more entertaining, but he asked me not to quote that in order to avoid giving Lucent's lawyers heart failure.

The bad news is that Ken still thinks Linux is flaky. I offered to have VA Linux Labs ship him a machine so he could see what a properly tuned modern Linux looks like, but he said he couldn't accept. He adds ``i do believe that in a race, it is naive to think linux has a hope of making a dent against microsoft starting from way behind with a fraction of the resources and amateur labor. (i feel the same about unix.)''

I cited all the case studies and trend curves and statistics you'd expect me to. He didn't respond directly to those, but I hope I at least gave him some things to think about.

Ken did finish by saying ``i must say the linux community is a lot nicer than the unix community. a negative comment on unix would warrent death threats. with linux, it is like stirring up a nest of butterflies.'' (Hm. Butterfly T-shirts, anyone?)

The really bad news, of course, is that Ken was wrong about the volatile and irrational reaction by the members of the Linux community against those who cast aspersions on the current state of apotheosis of Linux--or of the FSF, for that matter. This kind of thing most certainly does happen, as all here can doubtless attest. So much for the good old days.

The importance (or lack thereof) of uptime

[Signail11]

"If your uptime isn't to sacred to ya, it may be worth upgrading."

Uptime should *never* be sacred to any computer user in the sense that preserving a high uptime should not preclude one from installing a neccessary software or hardward upgrade. What is important is that an operating system has the ability to run stably and for extended periods of time such that the use of the computer is not impaired. I've known quite a few users who claim ridiculously high uptimes (ie. > 1 year). The kernel is so out of date that any random script kiddie can grab an exploit or buffer overflow from bugtraq and root the system, obviously not a Good Thing if your computer is running any sort of critical task.

Uptime is just that: a measure of how much time has elapsed since the last reboot of the system. It does not measure any of the following things:
-Superiority of an operating system
-Ability to administer a computer
-Programing skill
-"Eliteness/coolness", whatever that is