Any Linux Gurus out there? Having trouble compiling Gentoo

[Apollyon]

Well, the title sort of explains it all. I'm trying to compile the latest Gentoo release on an old P2-266 machine I have, but I'm having some difficulty. Apparently, in the C code, somebody forgot to <include modules.h> or something to that effect, so that you guessed it, the modules don't compile. Now I don't want to have to wade through all of the code to ensure that that include statement is present, so if anybody can give me some help it would be greatly appreciated. At this point I am so frustrated and annoyed that I am thinking of abandoning Gentoo altogether for another brand.

[toml]

Did you ever solve this problem?

If not, I recommend going over to the forums at http://www.arstechnica.com/

[Apollyon]

thanks for your suggestion, I had trie d that previously but no luck. However there is a new release out so hopefully I will have better luck with that.

[inferno]

Well, the title sort of explains it all. I'm trying to compile the latest Gentoo release on an old P2-266 machine I have, but I'm having some difficulty. Apparently, in the C code, somebody forgot to <include modules.h> or something to that effect, so that you guessed it, the modules don't compile.
Well, good luck trying to compile Gentoo on an archaic machine. I once downloaded and compiled Gentoo on a P3 and it took me 3 days just to compile KDE. :idea:

One of the reasons why some people choose to give Gentoo a try is because of the the option to optimise it for a specific type of CPU. Ironically, this is also the reason why you might end up with an unstable system because the problem with optimisation (http://gentoo-wiki.com/CFLAGS) is that some of the GNU code is simply unsuitable for this kind of tweaking. Binutils (http://www.gnu.org/software/binutils/), for example, comes to mind.

It would be helpful if you could post the the exact error message that you received, by the way.

[Kuroyama]

Ap

By the time you read this I HOPE you have abandoned Gentoo for some RedHat7.3 or so. P266?????

Id recommend Suse 9, (the last distro I "enjoyed") but only with big ram installed. alas my P2 300 lapshit with only 64MB ram is an unhappy camper when X is loaded.

(I was utterly unable to compile 2.6 on Suse 8.)

[Apollyon]

Ahhhh noooo, I've actually got Gentoo R-5 up and running, but I'm having problems with the networking. I haven't tried installing X or kde or anything yet since I've got no network connection.

The whole idea was that since I've got this spare machine kicking around, I might as well use it for something. So I thought I'd set it up as a linux file server. But if it is going to run that horribly perhaps I should try another brand of linux. Have you got any good suggestions for a good distro?

Another thing I wanted to get into was embedded linux, so I'm hoping to fiddle around with the minimal configuration of hardware that will be acceptable.

So basically I really don't want to have to buy a new machine just to run linux. However, I might, if I really like it. :)

[Kuroyama]

Well

I started with Caldera (like 1.2 or so... I dont remember)
I cut my teeth on Red Hat from about 7.1 on. I havent done any work with Slackware, but I can tell you I DIDNT have fun with Gentoo. That doesnt mean its a bad distro, just that I couldnt get good use out of it in a short amount of time.

When in the states I worked as a Field Engineer for a storage company. RAID, tape, SAN, NAS, storage. Basically if I could get a distro up and flying (and stable) on a server with support for a fibre channel HBA, then it was a winner.

I used to be a big fan of Rhat, but then they got all greedy on me. When I went to linuxworld (linux trade show in N.Y.) in 03, peeps were all talking about jumping ship to Suse. Ive had very good success with that distro. (except putting a 2.6 kernel into suse server8... from what I found online...its just a headache, so I let it be and stepped up to 9)

When I went back in 04, peeps were talking about using Rhat advanced server and just paying the money. BUT Novell has bought Suse, AND Ximian (badass linux desktop) so in my mind they are poised to start bustin heads... but again, peeps are grabbing their ankles for red hat.

If you need to get up quickly, I recommend suse, (free download). Red hat has organization and mad documentation (though suse does too) but red hat wants that money. IMO Fedora Core (free red hat as of 03) is too cutting edge. The packages arent all stable. Its great for experimenting, but if you want to set it up and leave it... Id go suse.

sorry...maybe TMI?

[Apollyon]

No no no, it was great thank you. I wouldn't have known where to start without having to navigate through the zoo of different distros. You have been very helpful, thank you very kindly for taking the time to help inform me.

The reason I chose Gentoo is because a number of people I know have sworn by it. I am going to continue fiddling with it since I've already got it compiled, but I will remain open minded to trying some of the others.

One of the reasons why I really wanted to run linux is because of the amount of free open source software available (or so I am hoping) and particularly because I am interested in software applications from academic and research insitutions which do not exist in the Windows world.

Thanks again for your help :)

[Kuroyama]

Cool.

Like I said, just cause I couldnt get it running doesnt mean its a "bad" distro. I just means that Kuroyama couldnt get it up and running in the time he had, with the hardware he had.

Those kind of things can change overnight in the linux world with a new kernel, patch, or distro revision release. Especially for less "established" distros.

Personally, Im jonesing for an Apple powerbook so I can get my hands on some of that microkernel (ala BSD) action. (suffering from the "halo effect" from the iPods I bought for myself and the family...)

[inferno]

Ap

By the time you read this I HOPE you have abandoned Gentoo for some RedHat7.3 or so. P266?????
Well, there is no reason to abandon the former completely because distros like Gentoo and LFS (Linux From Scratch) do serve a purpose; by playing with them you get to learn a lot about GNU/Linux. Needless to say, this kind of hands-on experience is a great asset.

RedHat and SuSE have graphic system configuration tools, and although they can be handy at times the newbie admin does not really know what goes on underneath the hood by using them. People who rarely type in long command strings from the prompt will be in trouble if they do not have access to X on machine that they just logged into.

One of my gripes with SuSE is that YaST tends to overwrite configuration files that have been edited manually. There are ways to get around this, but it is very annoying. Other than that, I have to admit that it is one of the best commercial distros out there.

RedHat is quite expensive for the hobbyist, but this is why they started the Fedora project. Fedora Core 1 was unstable but the latest release performs surprisingly well. RedHat and SuSE are both aimed at a corporate environment.

[Kuroyama]

Inferno

Therin lies my problem. Being the corporate serf My mentality has been shaped likewise. While I find great value in tinkering with a product to fully understand it, I know that pointy-haired-bosses will have a hard time finding their ROI when their enigneer spends days getting ONE server up and running or has to spend more DAYS when we need to connect that server to a new fibre channel switch, FC to iSCSI bridge, or just direct iSCSI input...

support of "enterprise" peripherals is huge. It had been a loning of mine to quit my last job only to work a job that allowed me to focus on a much smaller set of products. instead of being jack of all (storage) trades, master of none.

But, then again, theres something to be said for having experience across such a widepsread list of products as well. Just cant have it both ways.

[inferno]

Inferno

Therin lies my problem. Being the corporate serf My mentality has been shaped likewise. While I find great value in tinkering with a product to fully understand it, I know that pointy-haired-bosses will have a hard time finding their ROI when their enigneer spends days getting ONE server up and running or has to spend more DAYS when we need to connect that server to a new fibre channel switch, FC to iSCSI bridge, or just direct iSCSI input...
Tinkering and actually getting things up and running in a production environment need not interfere with one another. Professionals who want to keep their skills up to date in order to stay ahead of the competition should spend time on getting to know a product as thoroughly as possible. Be it in their own spare time or whenever the opportunity arises at the "office". It is this kind of experience that will allow you to get things up and running in the least amount of time, which in turn makes the managers happy.

I do understand where you are coming from, however. The "trick" is to go through the manual and browse through solely the bits that you need in order to get your equipment online. The experience that you have acquired while working with other gear should come in handy you nicely as well. Kind of like a C programmer who needs to do some coding in Java whilst having no knowledge whatsoever of the latter.

support of "enterprise" peripherals is huge.
You mean in GNU/Linux? Absolutely. And to top it off, there are various support options available from companies like RedHat and SuSE. This certainly takes away those hurdles which prevent higher management from opting for an open sourse solution.

It had been a loning of mine to quit my last job only to work a job that allowed me to focus on a much smaller set of products. instead of being jack of all (storage) trades, master of none.

But, then again, theres something to be said for having experience across such a widepsread list of products as well. Just cant have it both ways.
Some would say that having a broad range of experience is a safer, since it allows some flexibility. Adaptability is the key to survival, should the tide turn. On the other hand, if you have specialised knowledge of a popular product and there are relatively few specialists in that particular field, you are set.

[Dialectic]

I know it's obvious, but I'll say it anyway: if you're going to become specialized, make certain to pick a specialty that you know will not be going away any time soon (mainframes and legacy systems, for example, may not be sexy, but they'll be around for a while yet). You pick carefully and you're good at what you do, the money won't stop.

[Kuroyama]

I should clarify:

When I say "specialize" I mean just doing the job of say a small or midhouse size systems admin. One where the set of hardware that I work with from day to day will only change 4-5%.

In the lab where I worked even the lab servers were subject to change based on who needed what parts, or what new server we were testing that week.

Compound that by the number of changing peripherals
The number of changing OS revisions
The number of changing kernels
Changing software revisions
patches
etc

You start to see what my life was like. The box you set up today may or may not exist tomorrow. Engineering was subject to the whims of sales/marketing, which made focus very difficult.

I agree that experimentation is huge. Indeed it was my forays into linux from 2 jobs back that set me up for eligibility of my last job.

However, from a "returns" standpoint. It made little sense to become familiar with a product that you will probably never have to work with in the field. (I NEVER once in my 5 years (at last gig) worked with a customer that had slackware running in the datacenter). Since you were expected to become familiar with all new enterprise storage products after a minimum of testing time... Id like to meet the guy that could effectively keep his sanity by running experimental linux boxes at home, being married, finding time to eat, excercise, breathe, and still be effective on the job by being up to the minute on all distros of UNIX, windows, linux, and their patches etc.