Red Hat Linux 7.3 Released

qurob was the first of many readers to submit that Red Hat 7.3 has been released. Press release doesn't contain any surprises, just lists a bunch of stuff thats included with the dist. (Evolution, Mozilla, Apache). So go find a mirror if you're a Red Hat runner. Update: 05/06 14:05 GMT by T : christooley helpfully points out this list of mirrors.

Comments

Find a mirror?

[Anonymous Coward]

Other sites just get blindly /.'d without any regard for mirroring or performance concerns. I guess Redhat is special!

Re:Find a mirror?

[Anonymous Coward]

Yeah, perhaps Redhat takes the time to set up the mirrors BEFORE leaking the fact the release is out.

Just a thought, anyway.

What a coincidence

[PD]

I just bought a CD burner yesterday! Now I get to use it for something. I was actually hoping that Woody would be finalized, but this will do for now.

KDE3

[magicslax]

you'd think they would have touted kde3 a bit more instead of putting it at the bottom of the list. isn't this the first major distro to ship with the newest version of kde?

Re:KDE3

[GrenDel Fuego]

Suse 8.0 includes KDE 3.0, and was released last month.

Re:KDE3

[dmarien]

click through to their second page of new features. it's first on that list.

Re:KDE3

[1 (smarterThanYou)]

I thought that Suse 8.0 had already shipped in mid to late April. It was to include KDE 3 as well.

Re:KDE3

[Mandi Walls]

eh, they're a buncha crazy gnome freaks.

bastards like python, too.

may as well just put hot pokers in my eyes.

--mandi
(fan of purple linux...)

Re:KDE3

[nuOpus]

No .... SuSE comes with KDE3

Re:KDE3

[prmths]

I have a couple friends doing Debian packaging. I believe they'll be shipping with KDE3 on the next release...

But anyways, It's nice to see a few new things with out-of-the-box releases.

I've never really tried redhat - but I have tried Mandrake - I wasnt too thrilled about the autoconfig programs trying to tell me that the default out of the box NVidia driver is better than the one i had.. Ended up disabling all that stuff, and eventually switched back to my first distro, slackware...

I know redhat does a lot of auto-config stuff.. How well does it work?

Re:KDE3

[Diabolical]

Actually, SuSE 8.0 ships since a couple of weeks with KDE3. So, no, Red Hat isn't the first one to ship with KDE3. I don't know if there are other distro's shipping with KDE3...

Re:KDE3

[Anonymous Coward]

I'm pretty sure I was able to
cd /usr/ports/x11/kde3
make install clean
some time ago. But, I suppose you're special, too...

Re:KDE3

[Anonymous Coward]

KDE3 rocks, btw. And http://www.kde-look.org/ has some really neat eye-candy to make it even better.

Re:KDE3

[jonathan_ingram]

I believe they'll be shipping with KDE3 on the next release

It won't be in Woody, which will become Debian 3.0. It's not even in Sid yet.

Re:KDE3

[kpansky]

Gentoo has had KDE3 support for a while (almost since sources were released, I might add).

Re:KDE3

[Andreas(R)]

Actually, SuSE 8.0 was the very first distro to ship with KDE 3.0

I'm running SuSE 8.0 Pro now, and it's really great!

Re:KDE3

[Anonymous Coward]

Yes, if by "first major distro" you mean second

Re:KDE3

[odaiwai]

RH's autoconfig is pretty good. I can only recall one case of needing a driver because the supplied one didn't work, and that was something relatively unusual like a specific gigabit ethernet card. Otherwise, the stock kernels and installs have been fine.

dave

Re:KDE3

[gol64738]

hehe, wow, a non programmer...

Re:KDE3

[HiThere]

Yeah, but the python seems to be stuck at 1.5.

I know they include python2, but I assume there must be some reason for this. The first time it happened it was because of a licensing problem, but that's been solved (so now they *do* distribute python2).

Perhaps they're just being conservative, but the bug list on Python 2.2.1 (and 2.1.3) didn't list any bug that is obvious to me as a reason to stick with 1.5. I'd sure like to know what it is.

Re:KDE3

[tuffy]

Perhaps they're just being conservative, but the bug list on Python 2.2.1 (and 2.1.3) didn't list any bug that is obvious to me as a reason to stick with 1.5. I'd sure like to know what it is.

It's probably a compatibility issue, since plenty of Python 2.x code won't run under Python 1.x, and some 1.x code needs adjusting for 2.x. But 2.x should be the default and a python1 package left available for backwards compatibility. Maybe RedHat is saving that move for 8.0.

Re:KDE3

[johnnyb]

Simple - it's not backward-compatible. In order for backward-incompatible changes to go in, they need to release a new major number (i.e., 8.0).

Finnaly..

[Anonymous Coward]

wow, 7.2.97 was fun, but buggy... this better be better..

Missed it by that much.

[Chris Hiner]

Of course, I had to install 7.2 yesterday on a machine that couldn't wait. (I saw the closed 7.3 dir, and knew they'd release it the day after I installed 7.2.)

Maybe next time.

Re:Missed it by that much.

[Anonymous Coward]

Hey there Pondy!

X

Whats new Link

[bruceg]

HERE is a link to whats new in this release.

Re:Whats new Link

[christooley]

Thankfully their "New Features" page is wrong. It lists the new Mozilla build as 0.9.2 but the RPM for the installation is for 0.9.9. On the other hand I notice that the page title still reads "New Features for RedHat Linux 7.2" so it's probably not done yet. :)

Re:Whats new Link

[Zeio]

They removed linuxconf (good text console based configurator). You can get it here. RedHat still hasn't learned to use a real GCC release and instead they continue to use that lame 2.96. (so you will have to get GCC 2.95.3 for Linux 2.5 kernel compiles (per the directions in their docs) and get GCC 3.0.4 for everything else. This is getting ANNOYING, RedHat. ANNOYING.) "kgcc" (gcc 2.91.66-egcs.1.1.2-release)is still there for 2.4 builds.

KDE 3/ XFree 4.2 had some video corruption for me on a cheesy ATI card in a server, but no show stoppers. Seems like more goodness from the KDE team.

Good - they use glibc 2.2.5 - a standard GNU release, but they compiled it with the LAME 2.96 compiler. We shall SEE if they got the compiler It has been said that if a broken compiler compiles a library the library can be strangely broken and very difficult to debug. This goes to show RedHat why they shouldn't do this, and properly couple GLIBC 2.2.5 with GCC 3.0.4 as intended by GNU. Bero seems adamant about maintaining a 2.96 fork, which is costing time and resources and annoying users. I wouldn't care so much if 1.1.2, 2.95.3, 3.04 and RH-BROKEN.296-special were all included, but such is not the case. Lame.

Now that 2.2.5 can be compiled by GNU GCC, as well as KDE, RedHat is just being spiteful and not properly deprecating GCC 2.96X.

I give this a 5/5, because it's not a bad system, but it will require a man's touch. Mandrake and SuSE are starting to give this a run for its money, for sure.

NOTE: RedHat, please, just make it easier to play with the system and include the stuff that we all will go and download 20 seconds after install. Please. This compiling compilers like 2.95.3 and 3.0.4 is a waste of my time.

All in all, its good to upgrade at this point if you run a server to not have to do all the updates.

Note: About my insinuations about GCC 2.96 brokenness, I work side by side with a person who used to be on the GCC/GNU team, and has found strange bugs in certain version of the glibc that has been compiled by the 2.96 series. It went away when using release glibc compiled by release GCC. I personally have seen evidence that this is not FUD concerning GCC 2.96 - so please, all the flaming Bero zealots explain why is it now better to have a kluged compiler when the GCC team has far superceded it?

Re:Whats new Link

[Tet]

RedHat is just being spiteful and not properly deprecating GCC 2.96X.

No, Red Hat are just following their published policy of maintaining binary compatibility between releases with the same version number. Sure, they *could* have upped the version to 8.0, but then they'd have had to bump it up to 9.0 once gcc 3.x reaches sufficient stability, and that would have alienated a whole bunch of people a lot more than their current strategy. Whatever you may believe, gcc 3.x isn't yet ready for prime time. Remember that a company like Red Hat have to devote significant resources to support, and they will (as they should) follow whatever strategy leads to the most stable system, and hence the one that sucks least resoruces from the company. Red Hat aren't perfect, and have screwed up in various ways in the past. However, on this issue, I'm 100% behind them...

Re:Whats new Link

[Zeio]

I do not deprecate RedHat, and I use them - they are to me decidedly not evil.

However, why did gcc3 appear in 7.2 and not 7.3? All I ask is that yes, they can compile how they see fit, and so can I. My only request is to for them to provide the rest of the compilers for me that have been cleanly installed "their way," so that I don't have to go through the same shenanigan every time I upgrade or change a system or install a new one, etc.

On a side note, as far as GCC 3.X not being prime time, for C is surely is, I don't know about the rest, but for C its, as far as I can see, quite useable, stable and reliable with some interesting new optimizations. I also like ICC, from Intel, but they have very strange and frustrating licensing weirdness, and the kernel can't be compiled with it.

A lot of the GCC 3 is broke with regards to the C++, that's a crock. Both sides blame the other, but from what I have seen, most of the crap that doesn't compile right on GCC 3.x is the writer's fault, not the compiler. Think, what is harder, writing hello world or writing a compiler to compile hello world. I'm more inclined to believe the compiler guy that has to work on the project.

I see the reason to maintain binary compatibility to a point. For their manageability it makes sense, to some degree. So if its easier for them to put stuff out, go ahead.

I think that GNU has been a great force in the world, and to uselessly outpace them or point fingers at them is frustrating and bad for both sides of the camp.

One more note on RedHat, I am what would be the "customer," I do buy the media and get RH with new systems, etc. "Customers" who use this as a server don't like things being out of whack. I wish I could make it a requirement that the EGCS 1.1.2 release, 2.95.3 release, GCC 3.0.X release be included already to make things easier. It was there in 7.2, and then yanked out. I didn't hear the pissing and whining from the usual suspects about why this was done, but, I can only imagine they went off in some strange direction and have to dig themselves out quietly and slowly form this bastard fork, which, NO "readme," or "install" doc *EVER, EVER* requests. Face it. 2.96 is some RedHat only (Not Mandrake, Not SuSE) strange kluge. Programmers ignore in favor of GNU releases. Debian ignores it. It's a strange wart that needs wart removing acid, now. ;p

Re:Whats new Link

[Papineau]

A lot of the GCC 3 is broke with regards to the C++, that's a crock.

There you have it! The C ABI is pretty standard, and doesn't change between compiler releases. But the C++ side does. That's why you need a different libstdc++ for each compiler release. It's possible to upgrade the C compiler of a system easily (ICC, the Borland one, GCC). The C++ compiler is a bit more work, and RH prefered to not do it right now, but when GCC 3 will be ready for prime time (whether it's 3.0.4, 3.1, PI or 3.2 is not really relevant to the discussion).

Your suggestion of having 3 or 4 different releases of the same program is not a good idea: do you really expect users and developpers to choose between the different compilers dependant on the project they compile? "Do sdjy 0.58 needs EGCS 1.1.2, or is it safe to use 3.0.4? Oh right, it's 2.95.2." The goal of RH is to have only one: for the time being it's their own release, tagged 2.96-110, but it should evolve to a 3.?? based compiler for 8.0.

The other thing to keep in mind is that the gcc they ship now in 7.3 is tagged 2.96-110. Which means there's been a lot of patches applied since the beginning! When people grunt against gcc-2.96, they seldomly mention the RPM release. So yes, probably the first ones had a lot of problems (which explains why they shipped a kgcc in 7.0), but the current version works quite fine for me, even for kernel compilation (which explains why kgcc disappeared).

Re:Whats new Link

[Zeio]

RedHat develops for RedHat, no one out there develops OSS stuff for RedHat, more like GNU/Linux.

So, for RedHat internally, RedHat can use RedHat compilers.

For everyone else, there is GNU.

I have never used RH compiler for anything that I have done, and have compiler .SRC RPMS with the RELEASE compiler of my choice with no problems.

If I was actively developing, I would probably use Debian, or something far more conservative that what RedHat is right now.

This seems to be coming down to an interesting point in my mind; is this a server or a kiddie's workstation? Seems to me the motivation behind this deviance from release was to compile KDE, and arguably there were problems in KDE, because I never saw something like mozilla not compile, or any other convoluted and hairy C++ projects.

I think empirically coming up with opinions based on actually using a system is a valid way of doing things. Being "strategic" is not really imperative for RedHat's survival with regard to B2C. Most people who use RedHat still get driver RPMS in binary form from the vendor, don't compile anything and are GNU-impaired.. They do what MSFT has done, Dell is going to gleefully take and use your newest version. You don't really have to research your B2C commerce when you brokered large B2B deals. I would imagine that if RedHat was still more a B2C company, they would take this seriously.

About developers, they can read. If something calls for a compiler, you have to go install it anyway. You can compile with runtime specificity anyways, so more than one GNU compiler is easy. Case in point the 2.5 kernel. Boy, if I was a kernel hacker it would sure be nice to not have to go and install 2.95.3 RELEASE in usr/local.

The goal of RedHat is to make B2C customers happy, within reason. By the way, kgcc is still there in 7.3, and 90% of vendor supplied drivers, such as the e100 driver, check for kgcc and use it, because 2.96 is broken. I have all the updates, and the latest and greatest, and I can not for the life of my find valid reasons why RH avoids even putting in gcc3, like they did in 7.2 (that was a kluge by the way, it was a CVS checkout and not a release)

By the way, I don't think it's a good idea to comment about things, particularly kgcc, when its there in 7.3. That's like spreading FUD directly. I will proceed to show the audience that its still there, to date.


[root@rh73 root]# uname -a
Linux rh73 2.4.18-3 #1 Thu Apr 18 07:37:53 EDT 2002 i686 unknown
[root@rh73 root]# cat /etc/redhat-release
Red Hat Linux release 7.3 (Valhalla)
[root@rh73 root]# rpm -qa | grep cc
gcc-java-2.96-110
gcc-g77-2.96-110
gcc-2.96- 110
gcc-c++-2.96-110
gcc-chill-2.96-110
gcc-obj c-2.96-110

[root@rh73 root]# rpm -qa | grep compat
compat-glibc-6.2-2.1.3.2
XFree86-compat-l ibs-4.0.3-2
compat-egcs-c++-6.2-1.1.2.16
compat- egcs-objc-6.2-1.1.2.16
nss_db-compat-2.2-14
kde1 -compat-devel-1.1.2-11
kde2-compat-2.2.2-2
compa t-egcs-6.2-1.1.2.16
compat-libs-6.2-3
XFree86-co mpat-modules-3.3.6-44
compat-libstdc++-6.2-2.9.0. 16
compat-egcs-g77-6.2-1.1.2.16
kde1-compat-1.1. 2-11


And there you have it. So much for ONE compiler. EGCS 1.1.2 *AND* 2.96-110.

Like I said, I'm not RedHat's enemy, but it would be nice if they explained to people why (and did this with updated information and not the same old same old), and listened to my requirements, which are shared by everyone I know, particularly the OSS developers I know, who liked to use "matched sets" when possible.

Also, reader's note. In RedHat 7.2, Full Install, egcs 1.1.2, 2.96 and gcc 3.01(CVS) was included. That's 3 compilers.

You could have egcs 1.1.2 (for kernel-2.4 only), gcc 2.95.3(+CVS-stable if you are really picky) (this is what kernel 2.5 wants) and GCC 3.04 and *everything* I have ever used will compile, properly and cleanly. That's 3 release compilers, and it will all work.

To date there is a standing warning on using RedHat compilers by the kernel team. On several occasions I have seen posts to the kernel mailing lists with dmesg that indicated rh gcc was used, and the kernel hackers wont help them unless they use the proper compiler.

All in all this kind of argument makes it clear why SuSE and Mandrake and Debian and Slackware and TurboLinux have a place in the world today. They are all x86 primary, all backed by some major OEM in one way or another. RedHat could have wiped the floor, but technically, they have shortcoming that the others can address in one way or another.

It's times like these where I wish a big commercial entity put out their own dist from scratch, say IBM. Most of the Immunixes, HP-secure-linuxes ,etc, are rehashes of RedHat. Most notable what is different is the compiler ;p

Re:Whats new Link

[Papineau]

I see you have gcc-2.96 for C, java, C++, FORTRAN, objective-C and chill (although I don't know that last language). I also see you have egcs-1.1.2 for compiling C, C++, FORTRAN and obective-C code for a RH-6.2 target. You also have the libs to run those. If you check the 6.2 release, you can probably find some compat packages for targetting RH-5.2. And you can even install them if you want to target such an old release. You can probably even find some compat-libc5 rpms in RH-5.2. But in day to day use, or for compiling something for a RH-7.3 system, I don't see why you'd use egcs-1.1.2.

Back to kgcc... I don't see a kgcc in your rpm -a. By checking on a RH mirror, the only distro I can find with kgcc-1.1.2-40 is RH-7.0. After that, they always shipped only gcc-2.96-something (and it has been used extensively for kernels), except in 7.2 where gcc3 was also included. Unless you call kgcc something packaged as compat-egcs, there's no kgcc in 7.1, 7.2 nor 7.3.

Regarding the presence of gcc3 in RH-7.2, wasn't it meant as a technology preview only? Which explains both it's absence of 7.3 and the fact that it was a patched CVS checkout...

Of course, the other distributors can put whatever they want in their distros. The same way that some prefer CUPS and others LPRng. As long as they do the same thing...

Re:Whats new Link

[Zeio]

compat-egcs-6.2-1.1.2.16

So yes, there is a kgcc. Try it, kgcc -v.

Want to see the manifest for that?

The RPM is not called KGCC.

/usr/bin/egcs
/usr/bin/i386-glibc21-linux-gcc
/usr/bin/kgcc
/usr/i386-glibc21-linux
/usr/i386-gl ibc21-linux/bin
/usr/i386-glibc21-linux/bin/i386- g libc21-linux-env.sh
/usr/i386-glibc21-linux/inclu d e
/usr/i386-glibc21-linux/lib
/usr/i386-glibc21- li nux/lib/gcc-lib
/usr/i386-glibc21-linux/lib/gcc-l i b/i386-redhat-linux
/usr/i386-glibc21-linux/lib/g c c-lib/i386-redhat-linux/2.96
/usr/i386-glibc21-li n ux/lib/gcc-lib/i386-redhat-linux/2.96/SYSCALLS.c.X
/ usr/i386-glibc21-linux/lib/gcc-lib/i386-redhat-lin ux/2.96/cc1
/usr/i386-glibc21-linux/lib/gcc-lib/i 3 86-redhat-linux/2.96/collect2
/usr/i386-glibc21-l i nux/lib/gcc-lib/i386-redhat-linux/2.96/cpp
/usr/i 3 86-glibc21-linux/lib/gcc-lib/i386-redhat-linux/2.9 6/cpp0
/usr/i386-glibc21-linux/lib/gcc-lib/i386-r e dhat-linux/2.96/crtbegin.o
/usr/i386-glibc21-linu x / ib/gcc-lib/i386-redhat-linux/2.96/crtbeginS.o
/us r / 386-glibc21-linux/lib/gcc-lib/i386-redhat-linux/2. 96/crtend.o
/usr/i386-glibc21-linux/lib/gcc-lib/i 3 86-redhat-linux/2.96/crtendS.o
/usr/i386-glibc21- l inux/lib/gcc-lib/i386-redhat-linux/2.96/include
/ u sr/i386-glibc21-linux/lib/gcc-lib/i386-redhat-linu x/2.96/libgcc.a
/usr/i386-glibc21-linux/lib/gcc-l i b/i386-redhat-linux/2.96/libgcc.map
/usr/i386-gli b c21-linux/lib/gcc-lib/i386-redhat-linux/2.96/specs
/ usr/i386-glibc21-linux/lib/gcc-lib/i386-redhat-lin ux/2.96/tradcpp0
/usr/lib/gcc-lib
/usr/lib/gcc-l ib / 386-glibc21-linux
/usr/lib/gcc-lib/i386-glibc21-l i nux/egcs-2.91.66
/usr/lib/gcc-lib/i386-glibc21-li n ux/egcs-2.91.66/SYSCALLS.c.X
/usr/lib/gcc-lib/i38 6 -glibc21-linux/egcs-2.91.66/cc1
/usr/lib/gcc-lib/ i 386-glibc21-linux/egcs-2.91.66/collect2
/usr/lib/ g cc-lib/i386-glibc21-linux/egcs-2.91.66/cpp
/usr/l i b/gcc-lib/i386-glibc21-linux/egcs-2.91.66/cpp0
/u s r/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/crtb egin.o
/usr/lib/gcc-lib/i386-glibc21-linux/egcs-2 . 91.66/crtbeginS.o
/usr/lib/gcc-lib/i386-glibc21-l i nux/egcs-2.91.66/crtend.o
/usr/lib/gcc-lib/i386-g l ibc21-linux/egcs-2.91.66/crtendS.o
/usr/lib/gcc-l i b/i386-glibc21-linux/egcs-2.91.66/include
/usr/li b / cc-lib/i386-glibc21-linux/egcs-2.91.66/include/REA DME
/usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91 . 66/include/float.h
/usr/lib/gcc-lib/i386-glibc21- l inux/egcs-2.91.66/include/iso646.h
/usr/lib/gcc-l i b/i386-glibc21-linux/egcs-2.91.66/include/limits.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/in clude/proto.h
/usr/lib/gcc-lib/i386-glibc21-linux / egcs-2.91.66/include/stdarg.h
/usr/lib/gcc-lib/i3 8 6-glibc21-linux/egcs-2.91.66/include/stdbool.h
/u s r/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/incl ude/stddef.h
/usr/lib/gcc-lib/i386-glibc21-linux/ e gcs-2.91.66/include/syslimits.h
/usr/lib/gcc-lib/ i 386-glibc21-linux/egcs-2.91.66/include/va-alpha.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/in clude/va-arc.h
/usr/lib/gcc-lib/i386-glibc21-linu x / gcs-2.91.66/include/va-clipper.h
/usr/lib/gcc-lib / i386-glibc21-linux/egcs-2.91.66/include/va-h8300.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/in clude/va-i860.h
/usr/lib/gcc-lib/i386-glibc21-lin u x/egcs-2.91.66/include/va-i960.h
/usr/lib/gcc-lib / i386-glibc21-linux/egcs-2.91.66/include/va-m32r.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/in clude/va-m88k.h
/usr/lib/gcc-lib/i386-glibc21-lin u x/egcs-2.91.66/include/va-mips.h
/usr/lib/gcc-lib / i386-glibc21-linux/egcs-2.91.66/include/va-mn10200 .
/usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66 / include/va-mn10300.h
/usr/lib/gcc-lib/i386-glibc2 1 -linux/egcs-2.91.66/include/va-pa.h
/usr/lib/gcc- l ib/i386-glibc21-linux/egcs-2.91.66/include/va-ppc. h
/usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.6 6 / nclude/va-pyr.h
/usr/lib/gcc-lib/i386-glibc21-lin u x/egcs-2.91.66/include/va-sh.h
/usr/lib/gcc-lib/i 3 86-glibc21-linux/egcs-2.91.66/include/va-sparc.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/in clude/va-spur.h
/usr/lib/gcc-lib/i386-glibc21-lin u x/egcs-2.91.66/include/va-v850.h
/usr/lib/gcc-lib / i386-glibc21-linux/egcs-2.91.66/include/varargs.h
/ usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/li bgcc.a
/usr/lib/gcc-lib/i386-glibc21-linux/egcs-2 . 91.66/libgcc.map
/usr/lib/gcc-lib/i386-glibc21-li n ux/egcs-2.91.66/specs
/usr/lib/gcc-lib/i386-glibc 2 1-linux/egcs-2.91.66/tradcpp0
/usr/lib/gcc-lib/i3 8 6-redhat-linux
/usr/lib/gcc-lib/i386-redhat-linux / egcs-2.91.66
/usr/lib/gcc-lib/i386-redhat-linux/e g cs-2.91.66/SYSCALLS.c.X
/usr/lib/gcc-lib/i386-red h at-linux/egcs-2.91.66/cc1
/usr/lib/gcc-lib/i386-r e dhat-linux/egcs-2.91.66/collect2
/usr/lib/gcc-lib / i386-redhat-linux/egcs-2.91.66/cpp
/usr/lib/gcc-l i b/i386-redhat-linux/egcs-2.91.66/cpp0
/usr/lib/gc c -lib/i386-redhat-linux/egcs-2.91.66/crtbegin.o
/u s r/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/crtbe ginS.o
/usr/lib/gcc-lib/i386-redhat-linux/egcs-2. 9 1.66/crtend.o
/usr/lib/gcc-lib/i386-redhat-linux/ e gcs-2.91.66/crtendS.o
/usr/lib/gcc-lib/i386-redha t -linux/egcs-2.91.66/include
/usr/lib/gcc-lib/i386 - redhat-linux/egcs-2.91.66/include/README
/usr/lib / gcc-lib/i386-redhat-linux/egcs-2.91.66/include/flo at.h
/usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91 . 66/include/iso646.h
/usr/lib/gcc-lib/i386-redhat- l inux/egcs-2.91.66/include/limits.h
/usr/lib/gcc-l i b/i386-redhat-linux/egcs-2.91.66/include/proto.h
/ usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/inc lude/stdarg.h
/usr/lib/gcc-lib/i386-redhat-linux/ e gcs-2.91.66/include/stdbool.h
/usr/lib/gcc-lib/i3 8 6-redhat-linux/egcs-2.91.66/include/stddef.h
/usr / lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/include / yslimits.h
/usr/lib/gcc-lib/i386-redhat-linux/egc s -2.91.66/include/va-alpha.h
/usr/lib/gcc-lib/i386 - redhat-linux/egcs-2.91.66/include/va-arc.h
/usr/l i b/gcc-lib/i386-redhat-linux/egcs-2.91.66/include/v a-clipper.h
/usr/lib/gcc-lib/i386-redhat-linux/eg c s-2.91.66/include/va-h8300.h
/usr/lib/gcc-lib/i38 6 -redhat-linux/egcs-2.91.66/include/va-i860.h
/usr / lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/include / a-i960.h
/usr/lib/gcc-lib/i386-redhat-linux/egcs- 2 . 1.66/include/va-m32r.h
/usr/lib/gcc-lib/i386-redh a t-linux/egcs-2.91.66/include/va-m88k.h
/usr/lib/g c c-lib/i386-redhat-linux/egcs-2.91.66/include/va-mi ps.h
/usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91 . 66/include/va-mn10200.h
/usr/lib/gcc-lib/i386-red h at-linux/egcs-2.91.66/include/va-mn10300.h
/usr/l i b/gcc-lib/i386-redhat-linux/egcs-2.91.66/include/v a-pa.h
/usr/lib/gcc-lib/i386-redhat-linux/egcs-2. 9 1.66/include/va-ppc.h
/usr/lib/gcc-lib/i386-redha t -linux/egcs-2.91.66/include/va-pyr.h
/usr/lib/gcc - lib/i386-redhat-linux/egcs-2.91.66/include/va-sh.h
/ usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/inc lude/va-sparc.h
/usr/lib/gcc-lib/i386-redhat-linu x / gcs-2.91.66/include/va-spur.h
/usr/lib/gcc-lib/i3 8 6-redhat-linux/egcs-2.91.66/include/va-v850.h
/us r / ib/gcc-lib/i386-redhat-linux/egcs-2.91.66/include/ varargs.h
/usr/lib/gcc-lib/i386-redhat-linux/egcs - 2.91.66/libgcc.a
/usr/lib/gcc-lib/i386-redhat-lin u x/egcs-2.91.66/libgcc.map
/usr/lib/gcc-lib/i386-r e dhat-linux/egcs-2.91.66/specs
/usr/lib/gcc-lib/i3 8 6-redhat-linux/egcs-2.91.66/tradcpp0
/usr/share/d o c/compat-egcs-6.2
/usr/share/doc/compat-egcs-6.2/ C OPYING
/usr/share/doc/compat-egcs-6.2/COPYING.LIB
/ usr/share/doc/compat-egcs-6.2/README

Explain this, for example, in the e100 driver:
(e100-1.8.38)

# pick a compiler
ifneq (,$(findstring egcs-2.91.66, $(shell cat /proc/version)))
CC := kgcc gcc cc
else
CC := gcc cc
endif
test_cc = $(shell which $(cc) > /dev/null 2>&1 && echo $(cc))
CC := $(foreach cc, $(CC), $(test_cc))
CC := $(firstword $(CC))

Re:Whats new Link

[Zeio]

As a reply to myself, I'd like to point out to all the Zealots who support the broken 2.96, why does GENTOO completely compile from SOURCE, and support all the major packages RH-7.3 does using, the OH GOD, 2.95.3 COMPILER?!?!?!!?!!!!! I can't believe the FUD that gets spread about this compiler war that doesnt exist, only in RedHat's mind.

Gentoo Linux 1.1a

Gentoo Linux 1.1a features Linux 2.4.18+ and a modern GNU development environment (glibc-2.2.5, gcc 2.95.3), XFS, ReiserFS, ext3, LVM, ALSA, pcmcia-cs support, "vanilla" (stock) kernel compatibility for those who prefer unpatched kernels, Xfree86 4.2, OpenGL, KDE 3.0 and GNOME 1.4/2.0, tcp-wrappers, xinetd, iptables and Linux QoS tools, modern qmail (with optional mysql and LDAP support), postfix and exim MTAs, GRUB boot loader (LILO is still available if you need it), 1500+ up-to-date ebuild scripts of your favorite apps, an innovative dependency-based startup script design, and of course Portage, a completely open design and a great developer and user community.

Re:Whats new Link

[Zeio]

I think it might be Gentoo time.

As a reply to myself, I'd like to point out to all the Zealots who support the broken 2.96, why does GENTOO completely compile from SOURCE, and support all the major packages RH-7.3 does using, the OH GOD, 2.95.3 COMPILER?!?!?!!?!!!!! I can't believe the FUD that gets spread about this compiler war that doesnt exist, only in RedHat's mind.

Gentoo Linux 1.1a

Gentoo Linux 1.1a features Linux 2.4.18+ and a modern GNU development environment (glibc-2.2.5, gcc 2.95.3), XFS, ReiserFS, ext3, LVM, ALSA, pcmcia-cs support, "vanilla" (stock) kernel compatibility for those who prefer unpatched kernels, Xfree86 4.2, OpenGL, KDE 3.0 and GNOME 1.4/2.0, tcp-wrappers, xinetd, iptables and Linux QoS tools, modern qmail (with optional mysql and LDAP support), postfix and exim MTAs, GRUB boot loader (LILO is still available if you need it), 1500+ up-to-date ebuild scripts of your favorite apps, an innovative dependency-based startup script design, and of course Portage, a completely open design and a great developer and user community.

Sounds like RedHat minus the cruft and haphazard stuff. Funny how everything INCLUDING KDE 3.0 compiles.

Re:Whats new Link

[Menthos]

Gee, could you please stop the trolling? There's no reason to call a compiler broken just because you think it is (without proof). And this article was about Red Hat, not Gentoo.

Re:Whats new Link

[Zeio]

While I may not be able to say QED, I have personally seen evidence that leads me to believe there is something going on. I was merely pointing out that if Gentoo can compile essentially all of GNU with a "broken" and "heinous" compiler (with regards to 2.95.3), and producing a working dist (I have evidence, not proof that this is true as well), why RedHat chose to continue with his particular compiler.

I like RedHat, I aprecate them and wish them the best, but no one is without faults. Criticism can be constructive - and at this point, even if they are correct, the crap that has been said about 2.96 would have me running away from it to make the customers happy, or at least give them alternatives to suit their bias.

If a discussion of a feature in a distribution that's being announced isn't relevant to you and you refer to it as a troll, I'm sorry, Mr. Orwellian thought police brave new world prefect. You will not suppress me or my opinions or my experiences simply by trying to pass off an argument as a troll.

Re:Whats new Link

[Menthos]

While I may not be able to say QED, I have personally seen evidence that leads me to believe there is something going on.

So where is this "evidence"?

I was merely pointing out that if Gentoo can compile essentially all of GNU with a "broken" and "heinous" compiler (with regards to 2.95.3), and producing a working dist (I have evidence, not proof that this is true as well), why RedHat chose to continue with his particular compiler.

Ever heard about binary compatibility? The major version number of 7.x is there for a reason.

I like RedHat, I aprecate them and wish them the best, but no one is without faults. Criticism can be constructive

Well, your's isn't. You've been saying over and over in this thread that you think it's a broken compiler, but without ever mentioning why you think so. Thus it most certainly isn't "constructive criticism" because it lacks the fundamental part with the facts that would make it constructive.
Without it it is only seemingly pointless bad-mouthing.

and at this point, even if they are correct, the crap that has been said about 2.96 would have me running away from it to make the customers happy, or at least give them alternatives to suit their bias.

If you believe everything that people say, I wish you the best of luck.

If a discussion of a feature in a distribution that's being announced isn't relevant to you and you refer to it as a troll, I'm sorry, Mr. Orwellian thought police brave new world prefect. You will not suppress me or my opinions or my experiences simply by trying to pass off an argument as a troll.

Well, 1) you are not argumenting since you don't have any arguments besides "there are some other distributions that use some other compiler version" and 2) you aren't anything new that hasn't been carefully explained over and over on Slashdot and Bero's site for the last couple of years. Thus, unless you haven't been around for that long of a time and taken part in those discussions or read the explanations, you must be trolling.

Re:Whats new Link

[Zeio]

1) We compiled something here with 2.96 (C) it did not work. We compiled with 3.0.1 and it did. Since I cannot release our source code to the public as it isn't my property, you won't be seeing. We now use ICC because of the performance boost (that is considerable.) If RedHat took it upon itself to actually optimize the compiler, rather than what they have done, maybe I would use it.

2) I'm not inclined to believe what he says due to where he works - kind of simple.

3) Yes it is. As I am the one using the software, anything I say should be considered. Your snarky attitude will not help gain industry support for GNU. It's kind of amusing how large companies have to sugar coat this stuff because people like you won't even listen to someone's opinion. (Case in point: The fact Cobalt exists is a testament to customer requirements not being thoroughly evaluated by the likes of RedHat, there should have been NO room in the market for boxes like that.)

4) I believe what I see. I also want a choice. RedHat can compile anything it wants to with whatever compiler it wants to. I want to grab SRC RPMS or source packages and compile with what I want to.

5) Been around long enough, and Bero's site hasn't been updated in ages, that site was there before GCC 3 came along. He added a few uninspiring things about GCC 3 when it came along. Its funny, when that little tirade began it was RH-SECKR3T compiler vs. 2.91.66 and 2.95. It's like he did a replace with VI wherever there was 2.95 he put 3.01. I'm not inclined to believe the compiler hasn't been "fixed," because I don't consider the people working on GCC retarded, as you seem to.

While I don't claim to be an expert, or much of a contributor, lets just say this and a few other things makes me more interested in SuSE, Gentoo, Debian and Mandrake. Competition is good.

RedHat, prepare to lose market share. Shitting on people wit valid complaints and going against the grain doesn't win you friends or customers.

Thanks for trying to police my thought. As I had seen those pages before, and read them in their entirety, you have done nothing whatsoever to disarm my thinking about the RedHat compilers. I will continue to avoid them. You are way too evangelical, holding this Bero URL, a man with stock in the company being criticized, and holding it as gospel. It isn't.

Re:Whats new Link

[Menthos]

1) We compiled something here with 2.96 (C) it did not work. We compiled with 3.0.1 and it did. Since I cannot release our source code to the public as it isn't my property, you won't be seeing. We now use ICC because of the performance boost (that is considerable.) If RedHat took it upon itself to actually optimize the compiler, rather than what they have done, maybe I would use it.

So you are saying it doesn't work but you cannot reproduce it with any free sample code test case. Why on earth should I believe you? A test case shouldn't be that hard to produce, and you'd have to do that anyway if you wanted the alleged compiler error to get fixed. But there isn't a free test case. Only God knows why...

2) I'm not inclined to believe what he says due to where he works - kind of simple.

Understandable. He works at Red Hat and can be believed to have opinions in favor of Red Hat, so everything he says should be taken with a grain of salt. But you're not prepared to discuss anything he says. He brings up a lot of reasons for going with gcc-2.96-rh but all you say is "he works at Red Hat so I will completely ignore anything he says" instead of debating the individual points. Me thinks it's easier for you to dismiss him altogether rather than actually debating individual claims. Maybe because he actually brings up valid points?
Also, what you asked here was why Red Hat includes this compiler in their new version. I find it kind of natural that the full answer to this question you will only get from people working at Red Hat. But you won't accept an answer from a person working at Red Hat?!

3) Yes it is. As I am the one using the software, anything I say should be considered.

Maybe if you are paying for the software. I don't know if you or your company payed for your Red Hat boxes. But I can tell you that this "you should damn well listen to me when I say you suck" policy otherwise wont get you anywhere in the free software world. It's terribly rude and people (including developers) tend to avoid and ignore people that are rude, regardless of what they say have merits or not. There's only so much crap a single individual can take.

Your snarky attitude will not help gain industry support for GNU.

Who has the snarky attitude? In any case, I neither work for Red Hat nor speak for GNU.

It's kind of amusing how large companies have to sugar coat this stuff because people like you won't even listen to someone's opinion.

Not if they cannot provide valid criticisms in a polite manner, no.

(Case in point: The fact Cobalt exists is a testament to customer requirements not being thoroughly evaluated by the likes of RedHat, there should have been NO room in the market for boxes like that.)

This is getting terribly off-topic.

4) I believe what I see. I also want a choice. RedHat can compile anything it wants to with whatever compiler it wants to.

I'm happy that you give them that freedom.

I want to grab SRC RPMS or source packages and compile with what I want to.

Well, noone says you aren't allowed to (except for maybe support issues but I take it that you can live without that).

5) Been around long enough, and Bero's site hasn't been updated in ages, that site was there before GCC 3 came along. He added a few uninspiring things about GCC 3 when it came along. Its funny, when that little tirade began it was RH-SECKR3T compiler vs. 2.91.66 and 2.95. It's like he did a replace with VI wherever there was 2.95 he put 3.01. I'm not inclined to believe the compiler hasn't been "fixed," because I don't consider the people working on GCC retarded, as you seem to.

No I don't. And out of the core GCC developers, many of them are actually employed by Red Hat. Go figure. So Red Hat obviously knows something about compilers.

While I don't claim to be an expert, or much of a contributor, lets just say this and a few other things makes me more interested in SuSE, Gentoo, Debian and Mandrake. Competition is good.

Fully agreed. But your right to use anything else that you want to use doesn't give you the right to bad-mouth others without reason.

RedHat, prepare to lose market share. Shitting on people wit valid complaints and going against the grain doesn't win you friends or customers.

Now here is that trolling part again. You extract your discomfort with the Red Hat compiler into some holy war against Red Hat, and claim that a secret "Red Hat wants no customers" conspiracy exists. That's totally unfounded, irrelevant, and brings nothing to the discussion except for a lot of hot air. Trolling.

Thanks for trying to police my thought.

I'm not. I'm just trying to bring some sanity into your rants. It seems I have failed, because you are obviously still not prepared for having a serious discussion.

As I had seen those pages before, and read them in their entirety, you have done nothing whatsoever to disarm my thinking about the RedHat compilers. I will continue to avoid them.

You are free to.

You are way too evangelical, holding this Bero URL, a man with stock in the company being criticized, and holding it as gospel.

I'm not. I just want to have a serious discussion about compiler choices instead of one persons unfounded rant and bad-mouthing. Given this, it is only natural to start the discussion with debating some of the topics regarding Red Hat's compiler choice that Bero brings up. Debating individual claims and whether they are substantiated or possibly unfounded. But it seems that you are still not prepared to have that discussion; you are still continuing your rants and bringing them into more off-topic areas.

Also, I'm not sure Bero even has stock in Red Hat, or where you got that from. It was always my impression that he joined Red Hat rather late.

It isn't. Some say time is the fire in which we burn. My time is running out.

Well, go on with your life then instead of spending it ranting on Slashdot. But I must say that I'm quite disappointed that you are not prepared (and seemingly never will be) to have a serious discussion about the topic at hand.

Re:Whats new Link

[Anonymous Coward]

(Case in point: The fact Cobalt exists is a testament to customer requirements not being thoroughly evaluated by the likes of RedHat, there should have been NO room in the market for boxes like that.)

This is getting terribly off-topic.

4) I believe what I see. I also want a choice. RedHat can compile anything it wants to with whatever compiler it wants to.

I'm happy that you give them that freedom.

Wow. You sound like Marcello Tosatti:

Please compare and contrast Luther's 95 Theses with Marx's Das Kapital, including the later revolutions precipitated by each action, with particular attention paid to the rise and fall of 21st century Latin American oil prices and the ramifications of the Panama Canal. Um, five.

Re:Whats new Link

[Anonymous Coward]

"GCC" 2.96. THREE HUNDRED and EIGHTY NINE patches since its first release. I have the files here;

"gcc-2.96-20000731.tar.bz2" and 389 patch files. Sounds smart, you know, not having a current stable tree, and patching and hacking and slicing and dicing.

I also took notice, in the changelogs, the words tweak, hack, cure, FIX, over 500 times, FIX FIX FIX. If 2.96 was so great, why are there almost 2000 "fixes" since "release"?

Also, this changelog is BAD. Non descript, hacks, fixes and cures. Totally ridiculous.

Things like "partially revert sizeof overloading patch, it breaks other things"
"tweak Jim Wilson's patch so that we maintain binary compatibility"
"some more patches" "another cluster of patches."

After looking at this trash, I'd rather checkout of CVS -STABLE from GNU. This is just garbage.

Full log here. Its embarassing:

%changelog
* Fri Apr 12 2002 Jakub Jelinek 2.96-110
- 4 new patches - fix andhi abort (#63277) - some parallel makefile fixes (H.J.Lu, #62805) - fix loop unrolling (Franz Sirl, #63130) - don't move stack accesses accross prologue/epilogue stack
allocation/deallocation (Richard Henderson, PR opt/6165)

* Tue Mar 26 2002 Jakub Jelinek 2.96-109
- don't output .LPR* pic label into linkonce sections (#61931)

* Wed Mar 20 2002 Jakub Jelinek 2.96-108
- fix IA-64 packed unaligned structure initializers (PR c/5973)
- optimize visibility even if DECL_RTL has been already assigned
when seeing visibility attribute

* Wed Mar 13 2002 Jakub Jelinek 2.96-107
- make ia64 EH thread safe (#57443)
- fix ifcvt crash (Richard Henderson)

* Mon Mar 11 2002 Jakub Jelinek 2.96-106
- properly handle visibility attributes for duplicate decls

* Tue Mar 5 2002 Jakub Jelinek 2.96-105
- only emit visibility directives only with the definition
- fix for PR c++/1981 patch by Jason Merrill
- hardcode libc interface to libc6.2
- killed clear() hack, not needed anymore (#60097)

* Wed Feb 27 2002 Jakub Jelinek 2.96-104
- split gcc-ia64.patch into gcc-ia64.patch and
gcc-ia64-encode-section-info.patch
- 8 new patches - fix bitwise and/or/xor with constant operand in certain cases
(PR c/3444) - fix 64-bit pointer arithmetics in C++ (#58746, PR c++/4401) - fix regparm > 0 for functions returning struct (PR target/5755) - fix ICE in expand_and (PR c++/4574) - backport __attribute__((visibility("hidden"))) patch from 3.2
branch - fix floating point deficiencies on Atlas (Jan Hubicka) - using inside template functions (Nathan Sidwell, PR c++/1981) - fix for using in templates (Nathan Sidwell, PR c++/2184)
- 9 new patches from Mandrake gcc - fix array references with compound expr and ?: (Richard Henderson) - fix C++ flat initializers (Richard Sandiford) - handle C++ static member side-effects (Jason Merrill) - ensure reg-stack condition swap is valid (Jan Hubicka) - kill too tight check for i386 stack alignment (Richard Kenner) - fix check_dbra_loop (Jeff Law) - improve string literal length pedwarn (Anthony Green) - fix INT_MIN {/,%} -1 (Diego Novillo) - use -D,-U,-A option order as found on the command line (Neil Booth)

* Thu Feb 14 2002 Jakub Jelinek 2.96-103
- 3 new patches - fix access below sp on IA-32 (Richard Henderson, Jan Hubicka,
#55568, #57760) - fix destructor handling for template bases (Jason Merrill, #59766) - fix -fomit-frame-pointer (Brad Kaiser)

* Thu Dec 13 2001 Jakub Jelinek 2.96-102
- 3 new patches - initialize vt pointer in cin/cout/cerr/clog objects (#56119) - fix SPARC -m64 omitting neccessary zero/sign extends - don't allow insns reading from below frame pointer in
SPARC v9 return insn delay slot (Stephane Carrez)

* Tue Nov 20 2001 Jakub Jelinek 2.96-101
- 4 new patches - fix make -j issues - fix NaT issue with partial structure sets on IA-64 (Andrew MacLeod) - likewise when returning structure in registers on IA-64 (Andrew MacLeod) - fix SPARC ICE on LAPACK

* Mon Nov 12 2001 Jakub Jelinek 2.96-100
- 12 new patches - fix reload of insns without any reloads (#54571) - fix a flow ICE with nonlocal goto handlers (#55259) - fix C++ ICE on anonymous union on Alpha/Sparc (#51829) - fix Alpha ICE with asm("") (#46626) - fix Alpha inlining ICE (#54131) - frame unwind compatibility with 7.[12] binutils (#54724) - fix ICE on invalid C++ code (Nathan Sidwell, #55948) - fix CPP ICE on bogus -D arguments (Neil Booth, #54380, #55744) - support autoconf 2.52 (jfalk@netxpress.de) - fix SPARC typo in float.h (Alexandre Oliva) - fix SPARC ICE on pure function call (John Anglin) - fix SPARC ICE in change_address

* Mon Sep 10 2001 Jakub Jelinek 2.96-99
- ship /usr/bin/%%{gcc_target_platform}-c++/g++ instead of
/usr/bin/%%{_target_platform}-c++/g++ so we don't have hardlink
breakage (needed on sparc)

* Thu Sep 4 2001 Jakub Jelinek 2.96-98
- 7 new patches - fix strength reduction (#53025) - support seek*/tell* methods in classes (#53011) - fix ICE on tail recursion with user label on top (Jeff Law, #51477) - fix GCSE bug causing miscompilation with -Os (Diego Novillo, #52131) - backport GCC 3.0 fde lookup handling using glibc's _dl_iterate_phdr,
use proper eh sections for .gnu.linkonce functions to match
binutils (#49746) - fix libstdc++ stdin/stdout/stderr initialization, so that ftell works
on it even without any prior stdio calls (#52507) - fix objc class-ref handling, so that no relocations against .rodata
section are needed (John David Anglin, Rainer Orth, #52400)

* Mon Aug 20 2001 Jakub Jelinek 2.96-97
- 7 new patches - properly handle throw() exception specifiers in template decls (#51824) - fix dwarf-2 .debug_line section generation broken by patch added
in 2.96-91 - fix IA-64 varargs handling in presence of additional anonymous arguments
(#50757) - handle error conditions with friend classes gracefully
(Nathan Sidwell, #51533) - handle array initializers with side-effect elements properly
(Mark Mitchell, #51822) - fix a memory leak in cpp (Neil Booth) - fix ifcvt breaking eh regions (Richard Henderson, #48835, #51567)

* Thu Aug 9 2001 Jakub Jelinek 2.96-96
- 9 new patches - fix template processing in casts to array (Nathan Sidwell, #45824) - don't crash on := in C++ source (#47240) - issue declaration does not declare anything instead of ICE
(Nathan Sidwell, #48093) - fix SUBREG_BYTE related stabs bug (#50244) - ignore -fvtable-gc on alpha (#48933) - fix -ffunction-sections and -fdata-sections on alpha (H.J. Lu, #48581) - don't warn about if (&foo) if foo is weak (H.J. Lu, #50855) - set fail() if istream cannot extract a floating point variable
from the stream (#50722) - don't allow unresolved overloaded fns in both sides of ?: (Nathan Sidwell)

* Tue Jul 24 2001 Jakub Jelinek 2.96-95
- readd gcc-java subpackage, gcc3 will provide gcc3-java
- add build dependency on zlib-devel (#49286)
- 6 new patches - fix Fortran -ffixed-line-lengh-132 option (Mark Mitchell, #49326) - fix handling of C++ conversions to cv void * (Mark Mitchell, #49188) - avoid generating bogus .stabs (#49214) - avoid warning in (#49799) - fix for -pedantic (#49120) - fix make check in libio (#44657)

* Tue Jul 10 2001 Jakub Jelinek 2.96-94
- remove gcc-java subpackage, it is now provided from gcc3 src.rpm
- update libobjc to the one from gcc 3.0, so gcc3-objc and gcc-objc
can share the same libobjc.so
- don't run texconfig in texinfo configure - it can get stuck if
tetex is installed but tetex-xdvi is not.

* Fri Jun 22 2001 Jakub Jelinek 2.96-93
- use $RPM_BUILD_NCPUS in spec
- disable loop insn for -mcpu=k6 (Jan Hubicka, #23164)
- add SHF_MERGE support for Alpha
- disable jump threading patch temporarily

* Thu Jun 21 2001 Bill Nottingham 2.96-92
- ship /usr/bin/%%{_target_platform}-c++/g++ so we don't have hardlink
breakage

* Wed Jun 20 2001 Bill Nottingham 2.96-91
- add patch for dwarf2 file numbers (, )
- set in_section in the special case for "" in SHF_MERGE

* Wed Jun 20 2001 Elliot Lee 2.96-90
- Emit a newline after the .subsection pseudo-op.

* Thu Jun 14 2001 Jakub Jelinek 2.96-89
- SHF_MERGE support (together with binutils 2.11.90.0.8+)

* Wed Jun 13 2001 Jakub Jelinek 2.96-88
- prevent stack accesses below %sp in IA-32 (Richard Henderson)

* Fri Jun 8 2001 Jakub Jelinek 2.96-87
- fix friend handling in template class specializations (Nathan Sidwell)

* Wed Jun 6 2001 Jakub Jelinek 2.96-86
- 13 new patches - make jump threading work on ia-32, sparc (#39810) - optimize static constructors/destructors function - fix ICE on very questionable C++ code from JDK (#39858) - fix #ident handling if ident type was seen (#39929) - avoid regmove for unchanging pseudo (#40069) - fix diagnostic of invalid characters in identifiers (#41554) - fix another template mangling bug in the old ABI (#43071) - avoid moving CCmode regs around in loop and gcse on IA-64 (#42334) - fix ICE with complex float arguments to functions without
prototypes on IA-64 (#42339) - fix Fortran ICHAR() (Toon Moene, #42281) - fix IA-64 stop bit handling in presence of asm with multiple
outputs (Bernd Schmidt, #42560) - fix instantiation ICE (Nathan Sidwell, #43150) - fix -fsyntax-only (Nix, #43604) - fix default argument handling (Jason Merrill, #39466, #42123)

* Wed May 9 2001 Jakub Jelinek 2.96-85
- 4 new patches - pass -D__NO_INLINE__ to cpp if not inlining (#39824) - emit inline method in the object which references it (unless it
has been successfully inlined) in addition to the "exporting"
object (#39812) - kill trapping exprs in dsts of abnormal edges (Michael Matz) - fix nested expression statements in C++ (Mark Mitchell)

* Fri May 3 2001 Jakub Jelinek 2.96-84
- doh, 3 typos in the __frame_state_for patch

* Wed May 2 2001 Jakub Jelinek 2.96-83
- 4 new patches - fix __frame_state_for incompatibility between egcs 1.* and gcc-2.96-RH
(#37933) - fix a tradcpp ICE with -Dfoo= (#38551) - fix a sibcall and eh related ICE (#38466) - fix a template argument mangling bug (#31168)

* Fri Apr 27 2001 Jakub Jelinek 2.96-82
- 16 new patches - fix i386 if conversion optimization (#37054) - only warn about multi-line comments if -Wcomment or -Wall is given - fix a SUBREG_BYTE ICE with ia64 builtins - properly honour mode in cse lookups (Bernd Schmidt, #35122) - fix an ICE in dwarf-2 support (Jason Merrill) - fix an dwarf-2 output bug (Jason Merrill, #32801) - fix 2 bugs in ia64 exception handling - include ia64intrin.h header on ia64 - fix yet another alpha unaligned access (Richard Kenner, #37698) - fix gcc driver with -traditional -traditional-cpp
(Neil Booth, Chris Demetriou, #35963) - fix tree checking bug on bad arguments to inlined functions (#36986) - support -G option on ia64 (#33354) - fix -MD -o name_without_dot (#37697) - fix ia64 ICE seen in xfig (nested CONST rtl) - add a patch for m68k requested by Jes Sorensen (Jeff Law) - sync with ia64-000717-branch new patches
- fix functions using all local regs needing a frame pointer (Jim Wilson)

* Wed Apr 4 2001 Jakub Jelinek 2.96-81
- 7 new patches - fix constant folding of pointer comparisons where one pointer is
(void *)(-1UL >> 1) (#33690) - handle STRICT_LOW_PART in ifcvt (#33474) - throttle C++ inlining (Mark Mitchell, #29556) - define -D__EXCEPTIONS in g++ unless -fno-exceptions is passed - fix GCSE for function parameters whose addresses are taken (#34050) - fix null pointer check optimization (#23891) - fix dwarf-2 in presence of inline function redeclared locally
as extern (Jason Merrill, #25253)

* Mon Mar 26 2001 Jakub Jelinek 2.96-80
- 9 new patches - fix a SUBREG_BYTE related ICE (#32739) - update regrename.c to fix some -frename-registers bugs
(but -frename-registers is not considered supported) - fix Fortran lexer, so that it does not eat one whole line
after #line directive (#32230) - fix implicit operator= handling for classes with anonymous
aggregates (#32162) - fix libstdc++'s bastring.cc so that it does not warn with -Wshadow - update documentation about -Wshadow (#32160) - optimize C++ frontend by killing -fnew-abi support (it is not
ABI compatible with gcc 3.0 anyway) - store static variables from inline functions into linkonce sections
(to save space e.g. in mozilla)

* Fri Mar 16 2001 Jakub Jelinek 2.96-79
- 6 new patches - fix reg-stack return value handling (#20952, #21002, #31756) - fix glibc getdents64.c build on ia32 with --enable-kernel 2.4.1
(Bernd Schmidt, Jan Hubicka) - fix aspell miscompilation (#30297) - issue a clear error message about invalid ia32 floating point
asm constraints (#27137) - remove libc5 kludge from ia32 crtendS.o

* Fri Mar 9 2001 Jakub Jelinek 2.96-78
- 1 new patch, 1 fixed patch - fix Alpha combiner patch - fix bogus cpp warning (Neil Booth) - macro expand I in Objective C @I (Zack Weinberg)

* Thu Mar 1 2001 Jakub Jelinek
- libstdc++-devel requires libstdc++

* Tue Feb 27 2001 Jakub Jelinek 2.96-77
- 7 new patches - fix last cpp patch (so that -MD with -o works) - fix constant folding bug (#28864) - fix ADDRESSOF recognition (#29686) - avoid bogus -Wunused warnings in Fortran (Richard Henderson, #29559) - fix debugging of static local variables in stabs (Jason Merrill, #29548) - don't crash if fixup_var_refs_1 changed MEM into REG (Richard Kenner,
#29506) - add explicit barriers to flushrs and mov x = ar.bsp on ia64 (Jim Wilson,
#29155) - fix ia64 syscall_linkage security problems (David Mosberger)

* Thu Feb 22 2001 Jakub Jelinek 2.96-76
- 11 new patches - change some requires from %%{version} to %%{version}-%%{release} (#28200) - don't build java subpackage on ia64 - libgcj does not work there - fix C++ lookup if aggregate contains field with the same name as itself
(#21420) - clarify error message about taking address of an unqualified non-static
member (#24329) - better trim C++ inlined function graph cycles (#28385) - fix C++ static aggregate initialization (#20267) - fix Objective C garbage collection bug (#18827) - fix Objective C testsuite so that it does not use system libobjc - put only basename into -M goal (Nathan Sidwell) - fix Alpha reload bug (Richard Henderson, #26749) - fix Alpha combiner bug (Richard Henderson, #28509) - fix i386 reload bug with long long (Bernd Schmidt, #28209) - integrate patch.057 from gnupro-1117 tree

* Tue Feb 13 2001 Jakub Jelinek 2.96-75
- 10 new patches, 2 new sources - ia64 support - fix ia64 B-step errata handling code - fix a crash caused by lexing #line directives in some cases (#25851) - workaround ICE in dwarf2out (#24741) - avoid unaligned accesses on alpha and ia64 (Richard Henderson) - handle -foperator-names in the new cpp (#26457) - add nice diagnostic message about #if defined(xor) with -foperator-names
(Neil Booth) - fix paste avoidance after builtin (Neil Booth) - fix g++.jason/2371.C test so that new cpp does not complain about it
(Alexandre Oliva) - fix ICE on nested function with variable size array argument (#24506) - change gcc -v output to include minor release

* Sat Feb 3 2001 Jakub Jelinek 2.96-74
- 5 new patches - fix last cpp patch so that no whitespace is inserted at start of line
if last macro expansion resulted in no tokens (Neil Booth) - fix ICE during printing warning about overloading decisions (#23584) - honor no implicit extern "C" on linux in cpp (#26379) - fix layout of __attribute((packed)) enums in bitfields (showing up
in Linux DAC960 driver, #25440) - fix ICE when both prototype and function definition declares
variable size array argument (#24507)

* Wed Jan 31 2001 Jakub Jelinek 2.96-73
- 6 new patches - change new cpp so that it does not insert whitespace where no macro
expansion happens in C mode (Neil Booth, #25200) - don't ICE when inlining is replacing with constant a variable which is
only modified by assigning to itself (#24641) - mask constant argument to testl or andl if the operation should be done
in 8bit mode (#24921) - fix C++ anonymous union handling (#23682) - fix extract_bit_field so that it works even with XFmode (DJ Delorie)

* Fri Jan 26 2001 Jakub Jelinek 2.96-72
- 7 new patches, 1 new tarball - prevent inlining C++ functions containing static variables (#24253) - include headers sourced from -include or -imacros in -M output (#24317) - mangle overloaded operators declared in "C" linkage (#24685) - optimize out comparisons of two constants if at least one comes from
inline function arguments (visible e.g. on undefined __bad_udelay
symbols from some linux 2.4.0 kernel modules) - put in updated preprocessor, it seems stable enough and has tons of bugs
against both gcc 2.95.2 cccp and cpp used in 2.96-RH until now

* Thu Jan 18 2001 Jakub Jelinek 2.96-71
- 10 new patches - fix DECL_NEEDED_P (Mark Mitchell, #23401) - fix i386 assembly output for calls to symbol+offset (#23429) - fix C++ parser to cure GC crashes (#23489, #24060) - fix miscompilation due to not copied MEM/f flag (Alexander Oliva, #23502) - fix CC flags handling after sar on i386 (Richard Henderson, #24062) - only check inmode after checking type of argument in
fcmov_comparison_operator on i386 (#24115) - fix typos in i386 cmpqi_minus_1 (#24178) - fix thinko in alpha alpha_expand_block_move
(#23816, #24092, #24167, #24315) - fix fixup_var_refs (#24031) - fix my last fix for cpplib dependency handling (Graham Stott)

* Thu Jan 4 2001 Jakub Jelinek 2.96-70
- 11 new patches - Richard Henderson's debugging fix (#18707, #22671) - fix C++ #pragma implementation with -g3 (#21280) - avoid bogus warning whenever -fshort-wchar is used (#22786) - Richard Henderson's fix for alpha block move (#22972) - fix signedness of wchar_t in C++ (#22985) - Richard Henderson's alpha ze_and_ne fix (#23123) - simplify SUBREGs of VOIDmode CONST_DOUBLEs in addition to
CONST_INTs (#23211) - add -MD, -MMD and -MG support to -traditional cpp (#23254) - fix endless loops in cpp with some macros - fix patch for -fno-operator-names and -foperator-names - fix check for ISO C maximum string length (Geoff Keating)

* Mon Dec 18 2000 Jakub Jelinek 2.96-69
- honour -fno-operator-names in cpp (#22324)
- fix a bug in my do_compare_and_jump patch (#22418)
- register cp/input.c input stack with ggc

* Wed Dec 13 2000 Jakub Jelinek
- add Mark Mitchell's patch to fix ADDRESSOF patch

* Tue Dec 12 2000 Jakub Jelinek
- tweak Jim Wilson's patch so that we maintain binary compatibility

* Mon Dec 11 2000 Jakub Jelinek
- 4 new patches - partially revert sizeof overloading patch, it breaks other things
(but fix that bug in method.c) - if any errors were diagnosed already, don't ever ICE but issue
a user-friendly confused by earlier errors, bailing out message
and exit (#17103, #19609, #19612, #19613, #21861) - speed up C++ on Stepanov benchmark from abstraction penalty
3.88 down to 1.41 with -O2 -march=i686 (Jason Merrill, Jim Wilson)

* Wed Dec 06 2000 Jakub Jelinek
- 17 new patches - fix cpp handling of huge #error messages (#16965) - fix loop unrolling bug (#20580) - fix ICE with inlining compare on trees (#21125) - avoid ICE from integrate (#21167) - make sure volatile local variables don't loose their volatile status
(#21188) - apply Bernd Schmidt's local alloc fix (#21340) - fix loop MEM hoisting (#21376, #21423) - don't use tempnam in libf2c (Toon Moene, #21426) - fix declared with type qualifiers error message (Jeffrey Oldham, #21662) - fix truncxfsf2_2 pattern on i?86 - fix do_compare_and_jump in presence of inlining on trees - fix binding level handling for inlined returns - fix constant folding constant propagation, so that glibc string2.h
macros are properly optimized out (Bernd Schmidt) - fix pointer to member handling and related warnings (Nathan Sidwell) - fix diagnostic output of negative tree constants
(Nathan Sidwell, Jason Merill) - fix cleanup handling with gotos (Will Cohen) - Richard Henderson's 2 tail call fixes

* Thu Nov 16 2000 Jakub Jelinek
- 9 new patches - fix sizeof() overloading in C++ (#18369) - fix -fomit-frame-pointer with certain ia32 jumps on floating point
comparison (#20532, #20954) - handle more than 64k tokens on a line in preprocessor (#19290, #20443) - output #line directives even for headers generating no output (#20836) - fix __builtin_setjmp on sparc - fix miscompilation of glibc's e_sqrt.c on alpha (Richard Henderson)

* Mon Nov 6 2000 Jakub Jelinek
- 15 new patches - fix rtl mode of static class members (#18752) - fix ICE when using method undefined in its class (#19606, #19138, #20089) - fix ICE in place_field on bad input (#20268) - fix sparc ICE (#19884) - ressurect "confused by previous errors, bailing out" error
from g++ 2.95.x (#20104, #20105, #20508, #21322, #21323, #21639) - fix ICEs in binding level handling on bad input - fix named return values on sparc64 - fix vtable thunks on sparc64 - fix nested functions on sparc64

* Tue Oct 31 2000 Jakub Jelinek
- 7 new patches - fix postfix miscompilation (#18976) - fix ICE on NTL (#19092) - fix C++ ICE on bad input (#18764) - fix ICE on sidplay (#19462) - fix aliasing bug for testcase posted on lkml - add N constraint so that i386 in/out instructions
can be optimized by storing ports into instructions
again - fix cpp crash with invalid -A argument (#19974)

* Mon Oct 30 2000 Manfred Hollstein - add ability to build libg++-2.8.1.3 - add two libg++ related sources plus 4 libg++ related patches

* Mon Oct 23 2000 Jakub Jelinek
- 7 new patches - fix ICE in OpenOffice (Jason Merill, #20281) - fix sibcall miscompilation (#19543) - fix reload bug (Bernd Schmidt; #19313) - fix ICE with named return value (#18765, #21004 and GNATS bug 687)

* Wed Oct 18 2000 Jakub Jelinek
- 7 new patches
- since Aug 23 release, this should fix at least:
#12803, #17205, #17421, #17433, #18027, #18064, #18339, #18861, #19025,
#19182, #19607, #20011

* Wed Oct 11 2000 Jakub Jelinek
- 6 new patches

* Wed Oct 4 2000 Jakub Jelinek
- 4 new patches
- fix the bug reporting URL

* Fri Sep 15 2000 Jakub Jelinek
- fix cpp paste warnings
- fix alpha -mtune= patch
- Richard's fix for alpha aliasing bugs

* Wed Sep 13 2000 Jakub Jelinek
- add libio patch for glibc after Sep, 1st
- add -mtune= option for alpha
- fix libobjc install on sparc64

* Wed Sep 13 2000 Jakub Jelinek
- 15 new patches

* Wed Aug 23 2000 Jakub Jelinek
- 13 new patches

* Mon Aug 21 2000 Jakub Jelinek
- don't miscompile mawk

* Fri Aug 18 2000 Jakub Jelinek
- 16 new patches

* Thu Aug 3 2000 Jakub Jelinek
- take care of shared libobjc

* Wed Aug 2 2000 Jakub Jelinek
- some more patches

* Tue Aug 1 2000 Jakub Jelinek
- two loop patches from Bernd
- another loop fix
- two loop testcases

* Tue Aug 1 2000 Jakub Jelinek
- fix combine
- apply some more patches

* Mon Jul 31 2000 Jakub Jelinek
- update to 20000731 to fix sane reload breakage and other things
- fix 20000728-1 testcase
- some more patches

* Mon Jul 31 2000 Jakub Jelinek
- don't segfault in cpp on Alpha when input is a directory
- fix soft-fp preprocessing bugs

* Tue Jul 25 2000 Jakub Jelinek
- put the tradcpp segfault patch actually in, ho hum

* Mon Jul 24 2000 Jakub Jelinek
- don't merge stack adjustements accross asm(), should
fix kupdate oops in 2.4 kernel

* Mon Jul 24 2000 Jakub Jelinek
- update to 20000724
- fix an ICE seen in 586 kernel
- fix tradcpp segfaults

* Fri Jul 21 2000 Bill Nottingham
- yet another cpp fix

* Thu Jul 20 2000 Jakub Jelinek
- another tradcpp fix from Zack

* Wed Jul 19 2000 Jakub Jelinek
- update to 20000719
- tradcpp fix from Zack
- kill deprecated GNU ## extension used warning

* Mon Jul 17 2000 Jakub Jelinek
- Another sibcall fix (sibcall4)
- cpp fix for -include

* Sat Jul 15 2000 Jakub Jelinek
- Put Richard Henderson's mysql compilation fix instead to play safe
- Move libstdc++-compat stuff into compat-egcs package

* Fri Jul 14 2000 Jakub Jelinek
- Zack Weinberg's cpp fixes
- Fix mysql miscompilation

* Thu Jul 13 2000 Jakub Jelinek
- update to 20000713
- fix i386 -fpic miscompilations
- new version of sibcall3 patch

* Tue Jul 11 2000 Jakub Jelinek
- update to 20000711
- added spec file option to select: - fnew-abi by default and libstdc++-v3
- remove cccp hack, use tradcpp

* Fri Jul 07 2000 Jakub Jelinek
- Richard Henderson's alpha ICE patch
- better check for make -j count (Matt Wilson)
- 2 sibcall miscompilations and one ICE fixed

* Mon Jul 03 2000 Jakub Jelinek
- fix alpha ldconfig miscompilation due to aliasing
- fix i386 ICE/misoptimization introduced in 20000702

* Sun Jul 02 2000 Jakub Jelinek
- update to 20000702
- fix up streambuf.h for glibc 2.2

* Fri Jun 30 2000 Jakub Jelinek
- avoid putting cccp as /usr/bin/cpp, instead hack xcpp
so that it looks at $FULLPATH/cccp/cpp for cpp binary

* Thu Jun 29 2000 Jakub Jelinek
- fix the cpp fix from yesterday
- fix strength reduction ICE with -g -O2

* Wed Jun 28 2000 Jakub Jelinek
- fix cpp to put correct #line commands in ORBit
- rth's alpha summarize_insn fix

* Tue Jun 27 2000 Jakub Jelinek
- fix sparc32 vaarg patch vs. aliasing
- don't include fixincluded headers
- move gcc file list back to files section

* Fri Jun 23 2000 Jakub Jelinek
- fix interface vs. implementation C++ bug

* Thu Jun 22 2000 Jakub Jelinek
- update to 20000622
- fix sibcall alias related miscompilation
- --disable-checking
- don't warn about trigraphs unless -ansi for now

* Wed Jun 21 2000 Jakub Jelinek
- hopefully last step towards sparc rpm -m64 support
- Jason Merill's fix for the typename ICE
- Jan Hubicka's fix for wine ICE

* Wed Jun 21 2000 Jakub Jelinek
- next step towards sparc rpm -m64 support
- don't abort in C++ when a function has statement expression in its
arguments

* Mon Jun 19 2000 Jakub Jelinek
- first step towards even sparc rpm being able to build 64bit stuff
- fix one C++ ICE and one SPARC-only ICE

* Sat Jun 17 2000 Jakub Jelinek
- avoid throwing needed eh code away

* Tue Jun 13 2000 Jakub Jelinek
- update to 20000613
- fix two more ICEs

* Thu Jun 8 2000 Jakub Jelinek
- copy requires/provides from egcs, so that e.g. gcc-c++ requires libstdc++
- fix libstdc++-compat packages
- some ICE fixes

* Wed Jun 7 2000 Jakub Jelinek
- update to 20000607
- use as/ld -relax options if available on sparc
- use mandir, infodir and prefix macros

* Tue Jun 6 2000 Jakub Jelinek
- fix x86 ABI breakage with unions
- some other fixes

* Mon Jun 5 2000 Jakub Jelinek
- put libgcj into separate source package

* Fri Jun 2 2000 Jakub Jelinek
- update to 20000602
- fix libgcj bootstrap ICE

* Fri Jun 2 2000 Jakub Jelinek
- update to 20000601

* Thu May 25 2000 Jakub Jelinek
- don't use %f32-%f62 for -m32 on SPARC - proper fix
is to fix up reload

* Wed May 24 2000 Jakub Jelinek
- include old cccp for imake
- make libjava compile with new iconv prototype

* Tue May 23 2000 Jakub Jelinek
- fix ifcvt ICE
- fix SPARC -m32 -mvis ICE

* Sun May 21 2000 Jakub Jelinek
- update to 20000521

* Tue May 16 2000 Jakub Jelinek
- update to 20000516
- fix jc1 ICE

* Mon May 15 2000 Jakub Jelinek
- update to 20000515
- update libgcj to current CVS
- fix C++ sibcall related ICE
- avoid endless loop in javac

* Thu May 4 2000 Jakub Jelinek
- update to 20000504
- fix ICE when compiling kernel on Sparc64
- update libgcj to current CVS, so that it actually compiles
- fix ICE in libgcj

* Wed May 3 2000 Jakub Jelinek
- update to 20000503
- fix C++ breakage in Qt
- fix libstdc++.{so,a} symlinks for glibc 2.1.x

* Sat Apr 29 2000 Jakub Jelinek
- use uint64 for jword on 64bit arches

* Sat Apr 29 2000 Jakub Jelinek
- use -n for libgcj* subpackages.
- avoid removing %{GCC_PREFIX}/lib* from /etc/ld.so.conf on package
uninstalation (some other package might still use it).

* Fri Apr 28 2000 Jakub Jelinek
- include libgcj again (current CVS snapshot)
- convert from sparc64 only into all arch package
- split libgcj into libgcj and libgcj-devel

* Wed Apr 20 2000 Jakub Jelinek
- build all the languages and libraries with the exception of libgcj

* Mon Apr 17 2000 Jakub Jelinek
- two new patches

* Fri Apr 14 2000 Jakub Jelinek
- update to 20000414

* Sat Apr 8 2000 Jakub Jelinek
- update to 20000408
- remove a bunch of commited patches, add some new and updated ones

* Thu Mar 23 2000 Bernhard Rosenkraenzer
- Fix up cpp

* Wed Mar 15 2000 Bernhard Rosenkraenzer
- Deal with the build system compressing man pages

* Mon Jan 31 2000 Jakub Jelinek
- another cluster of patches

* Tue Jan 25 2000 Jakub Jelinek
- Some more patches

* Wed Jan 19 2000 Jakub Jelinek
- Initial hack for a 64bit sparc compiler

* Sat Dec 11 1999 Bernhard Rosenkraenzer
- Obsolete egcs*, g77
- Add egcs 1.1.x'ish libstdc++ versions to libstdc++-compat

* Wed Dec 8 1999 Bernhard Rosenkraenzer
- Fix build on sparc

* Tue Dec 7 1999 Bernhard Rosenkraenzer
- Add -warn patch (adapted from egcs-1.1.2 RPM)
- drop release number to 1 for 7.0 tree

* Tue Oct 26 1999 Bernhard Rosenkraenzer
- 2.95.2 release

* Sun Oct 24 1999 Bernhard Rosenkraenzer
- initial RPM

Oh brother...

[Linux Freak]

Just great, now my LAME Guide is even _MORE_ out of date. ;^)

Re:Oh brother...

[magicslax]

so take it down! please! one of the biggest obstacles i faced when trying to 'learn' linux was the vast ammount of horribly out of date liturature online. does a newbie really want to see this? you're doing more harm than good, i say.

Re:Oh brother...

[Linux Freak]

Well sue me for not having had the time (YET) to update it. I do have a pre-release of the next version. Now I'll need to update it to 7.3, but one of these days...

It's considered bad form to complain about volunteer efforts, btw. Maybe if you pay me the salary I am earning at my full-time job I could quit it and devote more time on things like LAME.

By the way, a newbie guide written for Red Hat is considered "Off topic"? What moderator and what drug?

Re:Oh brother...

[Anonymous Coward]

Don't criticize this guy for trying to help your lamer ass out. I checked out his site. It's not THAT out of date. If you encoutered this site (even in your newbie days), it would still be considerably helpful. If however, people like you continue to criticize the quality of this FREE site, he may cease to produce content all together, leaving you up the proverbial fecal estuary sans a method of locomotion. In the end, it could be I that is doing more harm than good... By feeding a troll...

Re:Oh brother...

[einer]

You ass. He's doing this for free. Not to mention, that his site is not 'horribly out of date,' it's quite navigable, and obviously took a great deal of time. When someone does the community a favour like this, it's best not to throw about unfounded criticisms and flames. Ungrateful troll (I suppose that IS the only variety).

hopefully it's better than the latest Mandrake

[Anonymous Coward]

http://www.theregus.com/content/4/24861.html for the registers first look.

In related news....

[Anonymous Coward]

Adore 7.2 patches out now.

Nice spoiler, jerk!

[PD]

Press release doesn't contain any surprises, just lists a bunch of stuff thats included with the dist. (Evolution, Mozilla, Apache)

Well thanks a lot, jerk. Some of us in California haven't even had a chance to read it yet, and you've given away the ending. Didn't the negative feedback from the Lone Gunmen snafu teach you anything?

Sheesh!

and is not nor will be downloadable for a while

[Lumpy]

It's not available on any mirrors or the ftp.redhat.com site.

Hmmm. I wonder if they are going to pull a Suse and not offer iso's anymore.

Re:and is not nor will be downloadable for a while

[Lumpy]

Ack ,call me stupid and beat me over the head...

Where I looked I didn't see it but it's there in my face... Valhalla is the name of it..

Yes it's there on msu.edu so the redhat ftp server it must be lurking somewhere...

Re:and is not nor will be downloadable for a while

[havardw]

I'll tell you which mirror I'm getting 500 KB/s from after I finish my download. ;)

Re:and is not nor will be downloadable for a while

[NickisGod.com]

It's not available on any mirrors or the ftp.redhat.com site.

Downloading it right now. Check the .edu's.

Re:and is not nor will be downloadable for a while

[dstanley]

You must have looked too soon.
I saw the iso's sitting there. I just hope my usual mirrors have them so I can get them after work.

Re:and is not nor will be downloadable for a while

[brodiedreamyou.ca]

No actually it is avaliable right now, i'm downloading it from the official ftp right now. Most of the mirrors arn't updated yet though.

What, no Red Hat 8.0?

[bytor4232]

Isn't Red Hat supposed to go x.0, x.1, then x.2? Whats with x.3? A real break from their traditional release model.

Re:What, no Red Hat 8.0?

[Nighttime]

Redhat only increment the major version number if the release will break binaries compatability.

Re:What, no Red Hat 8.0?

[BrokenHalo]

Means they've got nothing to say that they hadn't already said in 7.2; ie nothing to justify any more than a point release...

Re:What, no Red Hat 8.0?

[Burnon]

The story (as I've heard it, mostly just speculation) is that 8.0 will come when there are enough changes to justify some major breakage in compatibility. I guess the new version of gcc isn't where it needs to be for Red Hat to switch to it, nor is GNOME 2 ready. No kernel 2.6.x, no new glibc, etc... Everything else (even KDE3 I guess) is considered to be more evolutionary, rather than revolutionary.

Re:What, no Red Hat 8.0?

[MrJones]

I think 7.3 is a good numbering strategy. Because it is basically a kernel & programs update. Is not a big change like going from kernel 2.2 to 2.4.
Also, I wonder what SuSE and Mandrake will do when they release version 10 en 1,5 years :-)
Should they start from 1.0 again? Or go for the HPUX way?

Re:What, no Red Hat 8.0?

[Anonymous Coward]

and what is hpux way?

Re:What, no Red Hat 8.0?

[guacamole]

Uh, come on. I agree about most that you have said but waiting for 2.6.x kernel is a bit silly. I am pretty certain RH isn't waiting for a 2.6.x kernel to release the 8.0 (just like they didn't wait for 2.4.x in 7.0). The work on 2.5.x just started. That
probably means that we'll probably see Linux 2.6.x a couple of years from now ..

Re:What, no Red Hat 8.0?

[Burnon]

Hey - I was just running down the list of usual suspects that warrant an increment in the major version number. A shift to a new kernel series would fill that role, but obviously, there's no new stable kernel.

Re:What, no Red Hat 8.0?

[Anonymous Coward]

I hated hpux 10.x, 11/11i seem cool. Red Hat 8i?

A Question

[Jouster]

With up2date, how much longer can RedHat release CDs?

It has been our policy at work for some time now to grab whatever the latest release is, run up2date on it, and modify a CD image of the old CD so it has the new RPMs.

Is this prevalent? Will it become more so?

Jouster

Re:A Question

[tuffy]

With up2date, how much longer can RedHat release CDs?

As long as the majority of RedHat users don't have high speed internet access, CDs are still a viable method of distribution. The bandwidth of a box full of RedHat CDs in the trunk of my car is a helluva lot more than anything a measly 56k modem can provide, that's for sure...

Re:A Question

[Anonymous Coward]

It has been our policy at work for some time now to grab whatever the latest release is, run up2date on it, and modify a CD image of the old CD so it has the new RPMs.

Is this prevalent? Will it become more so?

Only for Debian users. apt-get dist-upgrade. Duh. I was a Red Hat noob for years until I found nirvana in Debian. It's just so much easier to maintain and keep updated than Redshit.

Re:A Question

[byran lei]

>>With up2date, how much longer can RedHat release CDs?
>
>As long as the majority of RedHat users don't have high speed internet
>access, CDs are still a viable method of distribution. The bandwidth
>of a box full of RedHat CDs in the trunk of my car is a helluva lot
>more than anything a measly 56k modem can provide, that's for sure...
>
>
CD distrubtion isn't going anywhere. Only a moron would try to claim otherwise. With high speed internet operators putting more and more speed and upload/download restrictions on their services,downloading multi-disk distros will never become a attractive option. The same applies to the BSD port system. In most cases, it'll still be cheaper to buy the cdset rather than wasting time and money actually downloading the files.

Re:A Question

[stilwebm]

I remember in 1992, even as late as 1994, many people said "individual telephone lines will never allow data transfers faster than 28800bps".

Of course DVD-ROMs (or another large capacity, inexpensive medium) will be more popular by the time bandwith makes a 650MB download seem trivial.

Re:A Question

[richie2000]

The bandwidth of a box full of RedHat CDs in the trunk of my car is a helluva lot more than anything a measly 56k modem can provide, that's for sure...

But the latency, the latency... You need a bike. :-)

Re:A Question

[Anonymous Coward]

Your so l33t and cool, can I join your club?

Jackass.

Re:A Question

[rbook]

And as late as 1989, many people said "individual telephone lines will never allow data transfers faster than 9600 bps."

And before that, it was thoguht that 300 bps was the theoretical limit of paired copper wires.

FYI, DSL runs on those same wires... how fast is DSL up to now? ;-)

Re:A Question

[Anonymous Coward]

But the high speed internet providers could also add local caches of things like new Redhat releases (and movie trailer sites, and other bandwidth hogs). If it doesn't leave their fiber, the cost is practically zero, and if they can coax users from other providers to use their caches or mirrors, they might even make a bit on the deal.

Re:A Question

[HydroCarbon10]

I remember in 1992, even as late as 1994, many people said "individual telephone lines will never allow data transfers faster than 28800bps".

In many places, those people are still correct.

Thank you Sprint...I wish you did.

Re:A Question

[MattBurke]

One word: support.

When you buy a pre-packaged linux distro, you get official support on it whereas when you download, you have Google...

Re:A Question

[Bios_Hakr]

There are certian packages you will never be able to just upgrade. If I am not mistaken, gcc is linked, at compile time, agianst your installed kernel. For more information on what it takes to build a linux system, go to linuxfromscratch.org. Basicly, even us guys building linux from the ground up have to rebuild our systems on occasion. Not mutch of a prob when you just need to pop in a mandrake disk, but a real bitch when you spend 10+ hours compiling linux/x11/kde/mozilla/other assorted things. Not to mention configuring...

Re:A Question

[rapidweather]

I would expect them to release CD's with Book as a source of income. I bought Redhat 7.1 from http://www.chguy.net/ for a few dollars, and that came with a nice sheet of instructions that I found very helpful. He downloads using broadband, and burns copies for you. They just have the disk number in magic marker, no label, etc. I got Redhat 6.1 with the book "Red Hat Linux Secrets" by Naba Barkakati, and that is very complete, and was $40.00 new. When I can come up with bread to get a new HDD, I'll also ask the genie for enought to visit chguy.net and have him send me RHL 7.3. That'll install on a slave HDD, right under the master I have all my goodies on now.

Re:A Question

[gkatsi]

gcc linked at compiled time to the kernel...
this is hilarious...

Excellent

[Anonymous Coward]

For all my mission critical systems I always deploy RedHat Linux, coupled with the excellent EXT2 filesystem and MySQL.

Beat my uptime!

Re:Excellent

[Apostata]

...and when my man comes home from a tough day at the office, he wants Stove Top Stuffing. Stove Top; no other stuffing will do.

Re:Excellent

[gatesh8r]

Beat my uptime!

Wow; 10 days on Windows XP!? You mean you either didn't have to reboot or have to put in a security patch in a 10-day span?

My longest uptime on my personal box was 4 months. Then my uni cut the power -- damn them, and I didn't even have a good UPS.

Re:Excellent

[Anonymous Coward]

Well, I have seen Novell boxes up for over 2 years doing file and print. Granted they only had around 60 people hitting on them a day.

I have also had a Mandrake 7.x box running at a church for over 6 months with no problem.

Please not that I don't agree with keeping a system running that long (2+years) without a reboot. WHEN it does finally crash you will find all sorts of problems...

Skewed Information

[Qwerpafw]

Just because slasdhot's readers mainly use linux doesn't mean that we need to almost facetiously notify readers of all the patches et etc.

I mean, most people would complain, wouldn't they, if slashdot ran a headline "Microsoft releases Windows update--adds minro functionality change and stability" or "Apple release new MacOS X change. Nothing much new". Just because its "news for nerds" doesn't mean the editors should ignore the "stuff that matters" part.

The vast majority of people who read slashdot are probably more interested in things like NASA, copyright law developments, et etc. It might be nice if there were a slashdot page specifically for linux kernel, distro, et etc updates. ala apple.slashdot.org, perhaps.

Just a thought (not a flame.)

Oh, and BTW, I'm going to download 7.3 now ;)

What? No 8.0

[graystar]

When was the last time there was a x.3 release? I am disappointed, I was expecting version 8 by now...it musnt be as good as they thought!

Re:What? No 8.0

[essdodson]

Yes, everyone knows that version numbers are all that matter. I'm sure if FreeBSD bumped to 11.0 you guys would swarm to it.

conflicted information

[sebol]

http://www.redhat.com/software/linux/rhl_new_featu res.html mentioned that, mopzilla is 0.9.2
but
http://www.redhat.com/software/linux/pl_rhl.html said it's 0.9.9

and with galeon 1.2.0 too :)
Yes!

Re:conflicted information

[Burnon]

Yeah - I noticed that too. I sent a message to the webadmin - I wonder if they'll fix it.

thanks (not)

[Anonymous Coward]

I am watching the throughput on my mirror download drop from the 162.94 kB/s I got for disk one to less than 60 since this announcement on /. Now I'll have to wait till this afternoon to install! I guess I should be happy I got this much so far, though.

What's new?

[Delirium Tremens]

Mozilla 0.9.2 !?

Isn't that a little conservative?

Re:What's new?

[Delirium Tremens]

Alright, it is probably a typo in their release notes. The full package list says Mozilla 0.9.9 . Way better.

Re:What's new?

[*xpenguin*]

They also screwed up their on that page.

Re:What's new?

[Anonymous Coward]

Conservative perhaps. It's a good thing. I would rather have slightly older software that has been thorougly tested! Quality control is important to me. Most Linux developers don't care about the quality of their products - they just slap shit together, hope it compiles, then sends it out.This leads to major corruption bugs in the kernel and other programs. Which is why redhat is so important to the Linux community - it increases the quality and value of linux.

Re:What's new?

[Anonymous Coward]

I would rather have slightly older software that has been thorougly tested!

..Yes, ...0.9.2, ...It was thoroughly tested, many many bugs were found... and the fixes were in 0.9.3. ...And the cycle continued...

It's beta software. Unlike release quality software, when comparing beta versions, older is more buggy, not more stable. (Testing doesn't make the software more stable, fixing it after the tests does, and 0.9.9 is more fixed than 0.9.2)

Not that it matters, it was a typo, RH7.3 has Mozilla 0.9.9

Re:What's new?

[Anonymous Coward]

Interesting.... It contains three versions of the same kernel!

"kernel 2.4.18 The Linux kernel (the core of the Linux operating system)."

Is This a Stable Release?

[ras_b]

I thought RedHat's stable, production level releases all were *.2 (i don't remember a 6.3, and 6.2 was the release for a while). Is this considered a stable release that can be used in production systems?

Re:Is This a Stable Release?

[ispel]

I thought RedHat's stable, production level releases all were *.2 (i don't remember a 6.3, and 6.2 was the release for a while). Is this considered a stable release that can be used in production systems?

The 7.3 version number indicates that the release is a incremental upgrade from (the excellent Redhat 7.2), w/o major feature changes.

It's my understanding that Redhat considers all their numbered non-beta releases to be stable and production ready. Their 7.0 release had some major component upgrades which gave their x.0 releases a bad wrap for some people, however, the issues (with GCC, security fixes) were fixed timely in the form of downloadable upgrades. The 7.x series has been great and rock-solid on the desktop (I've been using 7.1 and 7.2 as my desktop at home), and I'm looking forward to trying 7.3.

Re:Is This a Stable Release?

[BJH]

It's my understanding that Redhat considers all their numbered non-beta releases to be stable and production ready.

Yeah, right... as someone who's used (for various purposes and on various architectures) RH4.2, 5.0, 5.1, 5.2, 6.0, 6.2, 7.0 and 7.2, I think I can say that an RHx.0 release is generally just a kind of "let's throw it out there and hope that we can put out the bugfixes fast enough to avoid anyone noticing that large parts of it don't work", an RHx.1 release is "this *should* fix all the problems with x.0", and an RHx.2 release is "OK, we're done, onto the next major release, guys".

Re:Is This a Stable Release?

[Cro Magnon]

My guess is that it will be more stable than 7.2 (that shouldn't be too hard).

Re:Is This a Stable Release?

[WeirdKid]

That would be "half are below the median" -- not necessary the average (a.k.a. the mean).

Re:Is This a Stable Release?

[dananderson]

Yes it (RH 7.3) is a stable release intended for production systems. RH 7.3 has been through beta.

That said, I always wait at least a month (sometimes several if there's nothing in it I really need) for the "users" to find the more obscure bugs.

Also it gives time for the early adopters to post any problems and workarounds on the various mailing lists for me to search in http://google.com/ and http://groups.google.com/ (USENET news)

Mirrors are found :-)

[French Thias]

I've put up a list of mirror servers that are known to be fully synced with the release here :

http://freshrpms.net/mirrors/valhalla.html

I've also already rebuilt a pre-configured apt and its reposiroty for use with Valhalla, as well as many custom packages (lame, gkrellm, glimmer, nessus, xmame...)

Having already tested it a bit, I must say this release looks darn good and stable so far! Maybe it's because there are fewer changes than usual (which explains this being 7.3 and not 8.0).

Matthias

Anyone used MrProject?

[nagora]

Project management is one of those "Linux still doesn't have ..." items that MS'er come out with and its something our growing company is going to need soon, so has anyone used it "for real"?

TWW

Re:Anyone used MrProject?

[prmths]

I have the power management functions i need up and running.. I never trusted the HD shutdown thing in *any* OS..also never trusted the full sleep function on *any* os.. that includes windows which claims to be king of power management. Besides that, I have the basics like screen blanking, auto-shutoff, battery monitoring, etc on my C Series Fujitsu lifebook (Slackware 8, 2.4.18-pre9) via APM .. It works great... The problem is when you have issues with power management under windows (like i always did) you just shrug it off as a hardware incompatibility.. if people have trouble with it under linux.. then they assume something's wrong with the OS. Power management has always been problematic for me... and I've just learned to avoid certain aspects of it... regardless of OS.

Re:Anyone used MrProject?

[essdodson]

Ick! We all know that this would be determental to Linux. Productivity is something we choose to leave for Microsoft to handle. We may not actually get anything done, but damnit, I've got the source! That's what matters, right?

Re:Anyone used MrProject?

[nagora]

I agree with and understand everything you said, but I was asking about the project management software in 7.3.

TWW

Re:Anyone used MrProject?

[Burnon]

I've found it to be awfully alpha in quality. I last used it several releases ago, so I may be out of date. I hope it improves in the future, but to be honest, a linux-based application without a MS-Windows client wouldn't be very helpful in many heterogenous computing environments.

You might have luck looking at one of the web-based project management servers. One of the projects at my company picked one that seems to work well with your typical browsers on all types of operating systems.

Re:Anyone used MrProject?

[Anonymous Coward]

I have heavily used MS Project for a while during a previous job and MrProject (v 0.5.0) can't compare with it. For example, there only seems to be one relationship type between tasks (finish to start), I couldn't find a way to freeze a baseline and keep on tracking my project with the same file. It only seems to be able to keep track of task completion. If you want to use it in a group environment, there doesn't seem to be any functionality to allow each resources to update their schedule automatically.

MrProject is really useful to build a super to-do list for 1 project with no more than 3 people but it's not ready to track multiple projects with a common resource pool, there's just too much functionality still missing.

Re:Anyone used MrProject?

[Anonymous Coward]

Looks really nice. Uses simple XML file format. Easy to use. It does not read or write MS files, and lacks a lot of the routing/approval features of the product it tries to clone. Very nice in a Linux shop, but does not really work as a work-with product.

Re:Anyone used MrProject?

[FuzzyBad-Mofo]

Are you serious? My company tried to inflict MS Project on us. All involved grew to dislike the entire thing immensely

It lasted for all of two months, until the manager who suggested it got thrown out..

Fuzzy

openoffice?

[guacamolefoo]

I didn't see open office listed - is this going to be included with the 7.3 standard package?

Guac-foo.

Re:openoffice?

[Anonymous Coward]

StarOffice 5.2 is included with RH 7.3

No, OpenOffice at all.

versions questions....

[jeffy124]

What version of xfree comes with the new distro? Specifically, can it support a SiS 6326 card?

Re:versions questions....

[BrokenHalo]

You'll be lucky (or clever) if it does... I used to have a 6326 and nothing supported it; I ended up having to get a new mobo... (btw, I can recommend nvidia's Geforce cards).

Re:versions questions....

[jeffy124]

the most recent version of xfree has support for it, but it is extremely buggy and slow. worse, SiS seems to have no interest in developing a driver for it, so it's up to the community to write one. i havent checked on the status of the driver in a while, maybe i should do that. i'd but a new card, but money is a bit hard to come by for me recently. I'd also help xfree out, but i dont have experience or anything dealing with hardware interactions.

Something's broken...

[Voidhobo]

[red@hat]# apt-get dist-upgrade
bash: apt-get: command not found

Anybody else experiencing this problem?

Re:Something's broken...

[Enry]

Depending on how Debian works at times, that's not exactly a problem...

-Enry
Running Debian unstable and RH 7.1 servers

Re:Something's broken...

[Ankou]

Hey I have no problem running apt-get on RH. If you cared to search enough you would have found you can get apt-get for Red Hat 7.X from:

http://freshrpms.net/apt/

Re:Something's broken...

[Menthos]

http://apt-rpm.codefactory.se/ is also a useful link if you like to setup apt-rpm on Red Hat 7.2. Will possibly be updated for 7.3 soon.

Re:Something's broken...

[Daengbo]

Take a look at freshrpms.net for the apt package and source list for 7.2

Re:Something's broken...

[Anonymous Coward]

Just curious, why does your grandfather ask to suck your dick so much? Is he just teething his gums or something?

Re:Something's broken...

[Anonymous Coward]

Jesus, you debian trolls should apt-get a life. If you love your little tool soo much, you'd realise you can use it on any system that uses RPMs for package managment. Just install it. And don't go complaining that you "shouldn't have to install it", because, as a debian user, after an install, you have to upgrade EVERYTHING because it's so out of date.

Re:Something's broken...

[198348726583297634]

Also, slackware is very nice.

I am chicago-style. Mama mia! I have-a no pants!

Re:Something's broken...

[arkane1234]

You've got your facts wrong there... it's the fact that debian has bug fixes for individual packages. They *do* have version-r(number) that they update every so often ya know.

Don't think I'll go that way again...

[BrokenHalo]

After a brief fling with Slackware, I used RH from 5.2 to 7.1, learning to live with and work around their broken packages.

I switched to Mandrake 8.1 and liked it, but IMHO 8.2 was rushed out the door too quickly, and I love it no more. :-(

Now I'm just hanging out for Woody to be released...

Re:Don't think I'll go that way again...

[spongman]

why wait?

seriously, Woody is more stable than most other distros even before release. Yes, it's still got a few critical bugs, but they're mostly for non-x86 platforms. it's not like getting the final bits is going to be any harder than typing 41 characters.

Re:Don't think I'll go that way again...

[jdavidb]

I use woody on my blueberry ibook, with good results. I'm a little peeved that I'm still running a 2.2.x kernel, though when I tried to compile my own 2.4.x I trashed everything. (Should have used the 2.4.x maintained by the PowerPC Linux porting guys.)

Re:Don't think I'll go that way again...

[Anonymous Coward]

Debian users are a bunch of know-it-all pricks for the most part. All they do is troll the internet and look for people talking about other distros and then try to shit on them. What a bunch of insecure faggots.
All you every hear from debian users is how great apt is. Whatever, get over yourselves. You have like 3% of the linux market and while the rest of use go about our jobs using our distros of choice in silence, all of you debian pricks insist on telling us about how our distro sucks so much.

Well a big fuck you to the debians who troll /. and the rest of the net you all should just shut the fuck up and cram apt where the sun dosen't shine. There is more to a distro then just updating the fucking packages.

Re:Don't think I'll go that way again...

[Anonymous Coward]

I will second that:

`up2date` works just fine for me.

Differences from Beta?

[Bilbo]

I just finished downloading the recent RH Beta release and burning to CD, though I haven't had a chance to install anywhere yet. Anyone know of differences between the Beta and the official release?

Re:Differences from Beta?

[Lethyos]

Bug fixes?

Re:Differences from Beta?

[Anonymous Coward]

Check the MD5Sums for 7.3 iso against the beta iso's you downloaded. If they are the same then the last beta was the official release if not, better get the 7.3 iso's.

Re:Differences from Beta?

[Bilbo]

> Bug fixes?

Well, I sort of assumed that...

What I mean is, which ones? Which last minute changes managed to sneak into the official release between the time the Skipjack release came out, and when this was officially blessed?

(Or, to put it another way, which RPM's am I going to see immediate updates for on the Up2date or red-carpet services?)

Re:Differences from Beta?

[GrenDel Fuego]

Here's a big difference between the beta and official releases:

Official support.

I find up2date a very useful tool. I don't know how cleanly it'll support the beta if at all. Also, you can't really know for sure that all bug fixes that come up later will apply cleanly.

Best plan is to just get the final release.

Re:Differences from Beta?

[Menthos]

Umm, that method wouldn't work even if the code was all the same, simply because the version number has changed. The betas were not called 7.3, and the version number is mentioned on quite a few places in the distro. So even if all code was the same md5sums wouldn't match.

However, I strongly suspect there are more changes then just the version number change.
And as other people have pointed out, betas are unsupported and not to be used on production systems. That means that there will never more be updates or any kind of even the slightest security fix for the past betas in the future, so you're really out on your own if you are running the beta.

Re:Differences from Beta?

[Bilbo]

> That means that there will never more be updates or any kind of even the slightest security fix ...

Good point! I was assuming I could install the Beta, poke around a bit, and then just use red-carpet or up2date to get all my RPM's up to current releases. However, since RH specifically avoided calling the Beta "7.3", I don't know if even the "free" update services would know what to do with the release. (Red Carpet has specific "channels" for each release, probably because of dependency issues.)

Any way you look at it, you're probably going to have to do a clean install of any system you install the Beta on, once you get past the poking around stage and want to put it to real use.

Sigh... Good thing CDR's re still cheap... ;-)

Re:Differences from Beta?

[Papineau]

Once you burned the images to CDRs, you can use rpm -Fvh * on each CD. It will upgrade only the packages that are present on your system.

Or you can loopback mount all the isos, then symlink all the packages in a single directory, and run a single rpm -Fvh *. That way, if a devel package is on another CD, it won't matter.

Another thing I always do before an upgrade is use the same command with --test. If there's a dependancy problem, I can see it before anything bad happens.

Good luck with up2date: I never used it myself (prefer ftp+rpm).

x.3 release

[Goronguer]

This may be a minor point, but Red Hat deserves credit for calling this release "7.3" instead of "8.0". Especially when their pattern for years now has been x.0, x.1, x.2, x+1.0..., it shows admirable restraint for them to break the pattern and resist the temptation to call this a major-version release, when it is in fact an update release. Let's hear it for truth in advertising!

Re:x.3 release

[anthony_dipierro]

They just realized that most people hear ".0" and think "unstable".

Re:x.3 release

[newbiescum]

Doesn't Red Hat only increment the major version number when there is a "major" compiler version change? If so, then they're just following their normal practices which I suppose is a good thing.

Re:x.3 release

[Darren Winsper]

Red Hat change the major version number when they break binary compatibility. Since 7.3 is binary compatible with 7.0, they didn't go to 8.0

Re:x.3 release

[r6144]

Because they didn't include bloody things like gcc 3.1 that will break binary compatibility (at least for C++)

Re:x.3 release

[randombit]

They just realized that most people hear ".0" and think "unstable".

Which, in the case of Red Hat, is correct. Someone at securityfocus did a graph that showed the number of security bugs found in RH releases, covering 5.x-7.0 or so (this was maybe last year sometime). The x.0 releases had two or three times the number of security problems that the .2 releases had.

I still use Red Hat on my home machine, but I always make a point of avoiding x.0 completely, and only installing x.1 about 3-5 months after it's been released (and then grabbing all the new updates right away). I've been doing this successfully since 5.2, and am often glad for it, for example people I knew who installed 6.0 got screwed by all kinds of nastly problems.

Install apt4rpm

[gatesh8r]

and watch it go away

Re:Install apt4rpm

[Anonymous Coward]

yeah, and then watch apt get uninstalled by another fucking RPM depencancy.

fact: if you install apt-rpm, make an apt repository of the current redhat 7.2 updates and apt-get dist-upgrade, this will quite cutely "upgrade" apt off your box. good eh?

another fine production by redhat. sometimes i think they dont *want* you to upgrade over the nwtwork, what with the up2date "mysteriously" breaking for all but their latest releases, annoying subscription service, and their complete pig-headed stance of using a package manager that does not sort out its own dependancies for you, leaving you to recurse the dependancies for yourself.

So...

[NickisGod.com]

So, do I need to download all three discs, or will they let me get away with one, like Mandrake and Debian do?

This isn't a troll. I'd truly like to know...

Re:So...

[Jouster]

If you do just a "standard" (workstation) install, you only need Disc 1.

If I remember correctly, server installs require more than that.

Personally, I just get Disc 1, and use rpmfind.net whenever I need a given server.

Jouster

Re:So...

[yobbo]

1 cd only for mandrake? you have to be kidding me.

I tried getting away with 1 disc on the latest release - then i realised packages such as half of the freakin gnome control center were located on disc 2.

Re:So...

[Menthos]

Two CD:s are minimum, if you want any -server package to be installed at least. And even if you go for two CD:s it may be possible that packages you choose are on the third CD (some binary rpm packages are on the third CD).
So my advise is to get all three of them; that usually saves a lot of trouble later on.

That review is ridiculous

[Apostata]

Regarding that review, the guy may have some salient points (XFS), but he goes about it like an absolutely elitist asshole. Half the time he doesn't explain what kind of errors he's experiencing, nor does he attempt to troubleshoot. Lastly, and most clearly, his level of professional journalism is revealed in his childish anti-Gnome rant which - to this - reader dropped any respect I had for his article down 10 places.
Mandrake 8.2 is not the Grail, but to call it a f'ing disaster is just juvenile.

Old version of Mozilla?

[Fatal]

Redhat 7.3 only comes packaged with Mozilla 0.9.2 according to the release notes. Isn't that a bit old? I would have expected 0.9.9 at least!

Re:Old version of Mozilla?

[bero-rh]

Error in the announcement. It's actually 0.9.9.

Re:Old version of Mozilla?

[schmidt]

Has it been patched for the recently discovered security hole?

http://sec.greymagic.com/adv/gm001-ns/

Re:Old version of Mozilla?

[PrimeEnd]

Is the StarOffice 5.2 also an error? If not, why such an old version? Why not OpenOffice 1.0?

Re:Old version of Mozilla?

[bero-rh]

It's not.
OpenOffice 1.0 was released way too late to get through the QA process (can't reveal the schedule of course, but take a look at the changelogs in packages to get an idea about when the release had to be deep-frozen ;) ).

There are a couple of other things that prevent it from getting into Rawhide at the moment.

Off the top of my head (there are probably some more):

• It doesn't build without Sun JDK. We're looking into porting to gcj, but it's quite a way to go. Since gcj in any gcc prior to 3.1 is rather sucky, this was not even possible for a 7.x release.
• The UNO stuff requires a specific version of gcc, and it's not the "right" one.
• The installation process is not suitable for packaging. (Try building an RPM of something requiring GUI input during installation...)

These are all fixable because it's Open Source, but they require a considerable amount of time.

Also, the database application is missing (because it couldn't be relicensed), and some people depend on it.

I'm expecting OpenOffice in the base distribution in the next release... But this is not an official statement and much less a promise.

Re:Old version of Mozilla?

[Rakarra]

* It doesn't build without Sun JDK. We're looking into porting to gcj, but it's quite a way to go. Since gcj in any gcc prior to 3.1 is rather sucky, this was not even possible for a 7.x release.

Ah yes, Java. "Write once, run anywhere(1)." *snicker*

(1) With (non-trivial) restrictions.

From looking at the release notes

[wiredog]

It seems that if you have RH7.2 and you have run up2date weekly you have everything except for Moz, kde3, Evolution, and some gnome collab app. Since I've been running kde 3 since the release and have Moz 1.0-RC1, I see no need to download the ISO's.

Re:From looking at the release notes

[Burnon]

I've noticed a serious improvment in kernel interactivity. GUI applications on the 2.4.9-ish kernels that Red Hat was shipping for 7.2 ran like a dog whenever I tried to run a build or two of a large software project in the background.

I had been applying the preemptive kernel patches, which improved things a little, but still left GUI response jerky (completely subjective on my part). The 2.4.18-ish kernels in the 7.3 beta didn't have this responsiveness problem at all.

So, if you aren't into rolling your own kernels, and you frequently find your CPU load higher than 1, the upgrade may be worth it for you for that alone.

Re:From looking at the release notes

[bracher]

Mostly true. The first exception that comes to mind is X. 7.2 (even with the updates) runs XFree86 4.1.0; 7.3 includes 4.2.0.

- mark

Re:From looking at the release notes

[ralzod]

7.3 is apparently still packaged with the vulnerable zlib 1.1.3 version.

Re:From looking at the release notes

[Menthos]

Umm, you can't just look at version numbers of packages like that and say "that's vulnerable". Have a look at the list of patches applied in the source rpm:s, I'm fairly certain you will find the necessary security patches if you do.

Re:From looking at the release notes

[antdude]

You can use red-carpet if you are GNOME user to get Mozilla, Evolution, and other updates.

Re:From looking at the release notes

[Kynde]

It seems that if you have RH7.2 and you have run up2date weekly you have everything except for Moz, kde3, Evolution, and some gnome collab app.

That's not entirely true...
The 7.2 is aka "Enigma" where as the 7.3 is a lot cooler "Valhalla". If that's not a reason to upgrade then nothing is.

(I'm still sad that I had to update away from the "Guinness")

Re:From looking at the release notes

[Kynde]

... for the record and for those of you that dont trust those cd upgrades: You'll go far (and also get the "Valhalla" name :)) by merely downloading the redhat-release noarch.rpm somewhere and let up2date -u take over. I've been upgrading one these boxes here since up2date first appeared all the way to 7.2 beta by updating the redhat-release rpm first and then letting the up2date -u do the rest.

It's not always all that pretty, atleast with 7.1beta there were some problems and some of the packages had to be manually updated before up2date could be allowed to take over.

Nonetheless the point is that I'm not keen on upgrading a fully running box from cds, besides you'll loose all the information about which files had to be moved/renamed/created-as-new.

I know some ./configure && make && make install fans piss on all these rpms, but RedHat has indeed put a lot of effort into producing a well functional dependency tree and they're nearly there. You still can build an embedded box with their set (given that practical bare minimum lies somewhere around 90 rpms installed), but a full distrinbution upgrade by a mere 'rpm -F rh-latest/*.rpm' after a few mandatory rpm -uhv first is close to what it should be.

Re:From looking at the release notes

[ralzod]

> you can't just look at version numbers of
> packages like that and say "that's vulnerable".

My mistake. I actually thought that in this case the "patch" was to upgrade to v 1.1.4. Why wouldn't they just include the newer version of zlib in this release anyway?

Re:From looking at the release notes

[Anonymous Coward]

however, if you use red-carpet and ximian gnome, you'll break redhat upgrades, unless you remove ximian first, (which you definatly don't want to do in the future).

Remember folks!

[gatesh8r]

If you can bet better than bitrate on a mirror; it's not there!

Who cares?

[glrotate]

They could call it 7.2345346 or 56.8. I think most Gnu/Linux users try to be a bit more rational than to worry about whatever version number they decide to slap on it.

I think Red Hat should be given credit based on the quality of the release, not the version number.

Re:Who cares?

[Arandir]

It may not make a difference to current GnuLiX users, but it does to those who aren't.

I remember thinking Patrick Volkerding was a schmutz for coming out with Slackware 7.0, until several months ago when I heard at Fry's: "No dude! Get this Manhat 7.1, it's newer than that Redrake 7.0 you've got in your hands."

Linux is dying!

[essdodson]

It is official; Netcraft confirms: Linux is dying

One more crippling bombshell hit the already beleaguered Linux community when IDC confirmed that Linux market share has dropped yet again, now down to less than a fraction of 1 percent of all servers> Coming on the heels of a recent Netcraft survey which plainly states that Linux has lost more market share, this news serves to reinforce what we've known all along. Linux is collapsing in complete disarray, as fittingly exemplified by failing dead last [samag.com] in the recent Sys Admin comprehensive networking test.

You don't need to be a Kreskin [amdest.com] to predict Linux's future. The hand writing is on the wall: Linux faces a bleak future. In fact there won't be any future at all for Linux because Linux is dying. Things are looking very bad for Linux. As many of us are already aware, Linux continues to lose market share. Red ink flows like a river of blood. FreeLinux is the most endangered of them all, having lost 93% of its core developers.

Let's keep to the facts and look at the numbers.

OpenLinux leader Theo states that there are 7000 users of OpenLinux. How many users of NetLinux are there? Let's see. The number of OpenLinux versus NetLinux posts on Usenet is roughly in ratio of 5 to 1. Therefore there are about 7000/5 = 1400 NetLinux users. Linux/OS posts on Usenet are about half of the volume of NetLinux posts. Therefore there are about 700 users of Linux/OS. A recent article put FreeLinux at about 80 percent of the Linux market. Therefore there are (7000+1400+700)*4 = 36400 FreeLinux users. This is consistent with the number of FreeLinux Usenet posts.

Due to the troubles of Walnut Creek, abysmal sales and so on, FreeLinux went out of business and was taken over by LinuxI who sell another troubled OS. Now LinuxI is also dead, its corpse turned over to yet another charnel house.
All major surveys show that Linux has steadily declined in market share. Linux is very sick and its long term survival prospects are very dim. If Linux is to survive at all it will be among OS dilettante dabblers. Linux continues to decay. Nothing short of a miracle could save it at this point in time. For all practical purposes, Linux is dead.

Fact: Linux is dying

Oh wait, nevermind

Re:Linux is dying!

[charnerd]

What on Earth are you talking about?

Re:Linux is dying!

[arkane1234]

That'd be all well and good if there wasn't one true factor in the whole equation:

Linux isn't a commercial Operating System

:)

Unfair

[Anonymous Coward]

Hey! How come they can release it with Mozilla? Isn't that anti competative? I think they should make a modular version of Red Hat so that OEMs can include IE if they choose (and IF MS would write a version of IE for such a buggy OS).

Re:Unfair

[Anonymous Coward]

Surely you're a troll, but I'll bite anyway...

How come they can release it with Mozilla?
Mozilla is freely (re)distributable.

Isn't that anti competative?
You mean, Red Hat using their monopoly power in desktop systems to gain a monopoly for their browser in the browser market? Red Hat is missing two components that would make that anti-competitive. Guess what they are. (hints in bold)

I think they should make a modular version of Red Hat so that OEMs can include IE if they choose
If MS didn't threaten OEMs against providing other OSes in the first place, an OEM could easily take Mozilla out of the distro, replace it with another browser, and not break the whole system. It is modular.

Gnome programs

[daserver]

What strikes me about the release notes is that allthough kde 3.0 is a big update there is no kde programs mentioned.
Instead they show us all the great gnome office programs, gnomemeeting evolution.... And yes gnome 1.4.1 is also mentioned so it's not to compensate.. :)

Re:Gnome programs

[Enry]

Red Hat effectively started GNOME, or at least was heavily involved with the beginning of it. There were a few releases where they did not distribute KDE because it was not GPL'd. Red Hat has a very long history with GNOME, and so it doesn't suprise me that they would emphasize their GNOME apps over the KDE ones.

Re:Gnome programs

[prmths]

I've never really run KDE much.. and i honestly couldnt knowingly name more than a few 'KDE' programs off the top of my head, although the ones i could name are excellent pieces of software even though they may be slow on non-modern machines. (which is true with most software written these days) Either way, both KDE apps and gnome apps run in the same gui simultaneously without having either of the WM's loaded (although the libs need to be there) The only thing I see having an effect is the general theming of the widgets. It'd be nice for X to have a single widget theming setup and have all window managers/programs use that -- and let the window managers and desktop environment programs do their main job -- create the general underlying interface for running programs
I love having a choice in how my interface behaves. Although i would definitely like to see all apps have a common look to 'em if i wanted them to.. regardless if they're gnome, kde, or the more classic X widget sets. (I also like having some select few programs have a totally different look -- and sawfish does an excellent job doing that under gnome)

Re:Gnome programs

[Anonymous Coward]

RHAT also employs Owen Taylor and Havoc Pennington, two of the main developers of GTK and Gnome.

Re:Gnome programs

[Cheesy Fool]

I've just installed openbsd and kde3 on a amd k6-2 266 with 96mb ram and although the compiling took a while, kde3 runs great.

Re:Gnome programs

[Anonymous Coward]

KDE has always been a combination of GPL, LGPL, Artistic and BSD like licensed code.

Re:Gnome programs

[Anonymous Coward]

"ones i could name are excellent pieces of software even though they may be slow on non-modern machines"

Try another distro like gentoo then. I can run KDE 3 pretty fast on my p-pro 200 (a bit faster than KDE 2.2.2), and about the same speed as GNOME/gtk apps.

Re:Gnome programs

[Anonymous Coward]

Well, they did on the next page, where all the KDE stuff is listed first. I think finally, redhat is actually listening to USERS about KDE now, instead of satifying their own ego.

Re:Gnome programs

[arkane1234]

Try another distro like gentoo [gentoo.org] then. I can run KDE 3 pretty fast on my p-pro 200 (a bit faster than KDE 2.2.2), and about the same speed as GNOME/gtk apps.

Awww yeah... Gentoo is da bomb.

I use it on my 450Mhz which I have ran Redhat, Mandrake, Debian, and various other distributions on. I have to say, the optimizations of Gentoo are astounding. (namely because it's compiled specifically for your processor)

Try it in Test Drive

[Test Drive]

I have our Red Hat Test Drive system updated now to Red Hat 7.3. As always, accounts with us are free, and you get access to a number of different systems. Try out the latest releases of operating systems on our hardware before you commit it to your own!

Re:Try it in Test Drive

[Sits]

Can you tell us a little more about who Test Drive is for? Things aren't very clear from the linked web page...

Re:Try it in Test Drive

[Test Drive]

Sure. Test Drive is mostly intended for developers who want to port or test their code on platforms to which they might not otherwise have access. However, it is open to anyone. When you sign up, you get a free shell account on our systems which, so long as you obey the rules, you should be able to use indefinitely. We try to keep all our offerings up to date, and we're generally pretty responsive if you let us know about something you'd like to see in the program. You may also want to check our FAQ, or get in touch with us if you have more questions.

Features....

[DeadBugs]

I'm still waiting on Mozilla 1.0, KDE 3.1, GCC 3.0 but here's the list anyway...
Here's the full package list

KDE 3.0 Open Source graphical desktop environment

Quick access to programs, utilities, and tools

Customizable desktop for added convenience

Manage settings easily with KDE Control Center

Applications include Konquerer file manager and internet browser, KInternet automatic Internet dialer, KOffice Office Suite, and Kooka multimedia viewer.

GNOME 1.4 with Nautilus File Manager

Seamlessly view content locally or on the internet via URL-based naming.

Choose the right user interface for your level of expertise, from novice to expert.

Evolution comprehensive personal management system integrates e-mail, calendar, and contact information functionality:

Evolution Mailer -- e-mail environment with features enabling you to easily create, send, forward, save, and print e-mail using standard POP3 and IMAP protocols.

Evolution Calendar -- keep track of appointments, deadlines, and task lists.

Evolution Contact Manager --maintain lists of people and their phone numbers, addresses, and other identifying information.

XFree86 4.2.0 for improved graphic display and hardware support

GNOME Meeting video conferencing solution

MrProject open source project management software

Updated Mozilla Web browser

Printing Configuration Tool

Scalable printing configuration architecture that supports over 500 printers

Automatically upgrades older systems configured with printtool

Redesigned boot program speeds system startup.

USB 2.0 support

Improved support for digital cameras with gPhoto2, gtKam, and Kamera Server Features

Complete Web, mail, ftp, file and print servers included.

Apache Configuration Tool

Quick configuration of Apache options

Easily create and manage multiple virtual hosts

BIND Configuration Tool

Simple setup of forward and backward name resolutions and DNS lookups.

LPRng and CUPS print daemons

Sendmail and Postfix message transfer agents (MTAs) with switching tool

LBA48/ATA133 support for drives > 137GB

Updated PostgreSQL open source database System Security and Administration Features

Firewall Configuration during installation allows you to select from three predefined levels of security or customize your own security level.

Graphical Kickstart Configuration

Red Hat Network maintains system security Upgraded Core System Components

The following major system components are included in Red Hat Linux 7.3:

2.4.18 kernel

gcc 2.96-RH

glibc 2.2.4

XFree86 4.2.0

Mozilla 0.9.2

OpenMotif 2.1.30

Perl 5.6.1

Apache 1.3.23-9

Re:Features....

[teg]

Should be mozilla 0.9.9, which is what is included.

crap!

[0x20]

I just finished downloading the skipjack ISOs about 30 minutes ago.

Does anyone know the differences between skipjack and valhalla?

Re:crap!

[Anonymous Coward]

Name allready implied it: Skip it jack**

Re:crap!

[Anonymous Coward]

I have skipjack. Overall it's nice, but it's a little too buggy. For example, whenever I click on preferences in Galeon, it crashes.

Re:crap!

[Anonymous Coward]

One is a beta and the other is the real release, duh

Re:crap!

[0x20]

Thanks, Sherlock.

Did anyone else notice

[Anonymous Coward]

that the features link shows "Red Hat 7.2"?

no Enlightenment

[Anonymous Coward]

Where is Enlightenment in this release?

Valhalla?

[MJArrison]

So what does Valhalla mean? I found a lot of google hits, but what is RedHat specifically referring to?

Re:Valhalla?

[JanneM]

The hall of the Valar, or the home of the gods in nordic mythology.

/Janne

Re:Valhalla?

[Anonymous Coward]

See there ...

http://www.dictionary.com/search?q=Valhalla

See the "Linux is Dying" post plays on this meaning...

Valhalla

\Val*hal"la\, n. [Icel. valh["o]ll, literally, hall of the slain; valr the slain (akin to AS. w[ae]l, OHG. wal battlefield, wuol defeat, slaughter, AS. w[=o]l pestilence) + h["o]ll a royal hall. See Hall, and cf. Walhalla.] [Written also walhalla.] 1. (Scand. Myth.) The palace of immortality, inhabited by the souls of heroes slain in battle.

It's also a place in the state of New York in the United States of America although I wouldn't want to be caught dead there :-)

Re:Valhalla?

[Lardmonster]

The hall in which Odin received the souls of slain heroes.

Re:Valhalla?

[maraklov]

the 'linux is dying' post vanished! or is it just me?

Re:Valhalla?

[Anonymous Coward]

valhalla is where vikings go when they die

Re:Valhalla?

[Anonymous Coward]

It is also the name of a non profit LAN gaming association in Adelaide, South Australia :)

Re:Valhalla?

[Russ Steffen]

More imporantly, what is the relationship between skipjack and valhalla? Red Hat codenames have been linked by double meanings since the 3.0.3 release (Picasso).

Re:Valhalla?

[myz24]

I prefer to think they are talking about a steak house near me which serves the tastiest prime rib you can imagine. MMM, red meat...er hat

Re:Valhalla?

[Descartes]

Ok, this is the best I can come up with:
According to dictionary.com a skipjack is a type of sailboat.
Valhalla is the viking afterlife for warriors who died in battle. Viking warriors were often either buried in sailboats or cremated in them.

Re:Real Moron

[Glanz]

"Real" morons love XP, the "Real Morons' OS".. . Now if you think this is flame bait, you are mistaken. You are living proof of my statement, ergo, it is merely a fact.

In other news

[Anti-Microsoft Troll]

Microsfot released Windows XP 1.1.

The press relsease contains no surprises, just a list of additional bugs and security holes that have been purposefully added to frustrate and antagonize the user.

A spokesman for Microsoft is quoted as saying, "So? What are you going to do about it?"

Errata??

[Anonymous Coward]

So where are the errata?

I have 7.3 installed (no, not the pre-release versions). But several packages should have newer versions because of errata.
Currently the errata page does not list anything newer then 7.2

Ximian just updated their installer to support the version of RPM (4.04) shipped with RH 7.3, so I might try to install the Ximian desktop this afternoon as the RH desktop is rather boring.

Re:Errata??

[Anonymous Coward]

And here is the errata page, although the general support page does not link to it yet.

http://rhn.redhat.com/errata/rh73-errata.html

10 errata available, classified as:

6 security updates
3 bug fixes
1 feature enhancement

Aiight!

[Ixe]

Time to go sit at the ftp sites trying to get the ISO's all day.... oh well

What a bummer to have broadband and be stuck at under 10k/s during the /. effect.
Thankfully RedHat has pretty big pipes if I recall...

And including the new Emacs!!

[xiox]

emacs-21.2

You may have heard about Windows XP...

[NOT-2-QUICK]

Actually, I believe Slashdot did run a story on the release of a minor revision to a Windows operating system... You may have seen it - it was the announcement for the release of Windows XP!!!

Perhaps you have heard of it otherwise referenced as Windows NT 5.1!

In fact, IIRC, Microsoft - not to mention the majority of tech the world - made quite a huge commotion about the release of this new OS. And despite the XP operating system being touted as a revolutionary new step in computing, it still only garnished enough pull within the closed ranks of the windows developer community to qualify for an increment of the minor revision to the Windows version...

But hey, that's just Microsoft...why would we - a technically knowledgeable, primarily Linux community - get excited over a minor revision to the Red Hat Linux. After all, it is only one of the most stable - not to mention commercial - distros currently in existence! :-)

Re:You may have heard about Windows XP...

[j-turkey]

Why get excited about the RH 7.3 release if its not called Linux XP?


-Turkey

Re:You may have heard about Windows XP...

[arkane1234]

After all, it is only one of the most stable - not to mention commercial - distros currently in existence! :-)

You might want to retract your statement about it being one of the most stable....

Sure, it's commercialized to hell and back, but it's not one of the most stable. One of the most beginner user-friendly... (mandrake and redhat rank up there)

Re:Whats new

[L1nUx h4x0r]

glibpr0n-1.0: better/faster rendering of skin tones, including the ability to mathematically extrapolate other angles.
gpr0n-sync-0.9: Automatically download the latest pics and movies with this great piece of Free software.
gpr0n-view-1.1: By far the greatest viewer for graphics, both still and movies. Using glibpr0n's full API set, this program can do it all.
m$-hack-3.11: Tired of typing M$ all the time or trying to come up with snappy new insults for the company everyone loves to hate? M$ hack connects to a distributed network via a completely free software P2P network and can automatically be used to correct your typing in any application. You type "Microsoft" and it replaces it with a real-time generated mock. Essential for all the l33t h4x0rz out there.
b4sh3r-1.0: The newest shell for Linux, released under the GPL, this is quite possibly the least user friendly shell around. Providing a direct link to kernel APIs, all commands are given as 16 bit common command codes, or 32 bit addresses to the actual kernel code. If you don't want your mom/girlfriend/wife (I know, I know, I include girlfriend and wife because I'm an optimist) to mess with your l33t setup.
sl4sh3r-9-1.0: Slasher-9 is the world's most sophisticated trollbot. Drawing upon the same P2P network (Did I mention all the traffic is seperated over FTP, HTTP, HTTPS, and SMTP ports, with 4096 bit encryption?) that m$-hack built, this bot ensures any /. user (or even AC) of first post each and every time. Or, start this bad boy with the parameter --karmawhore and it will automatically retrive articles and post them in the thread, and summon all of the other bots (via the p2p network) to mod it up as "Interesting").
lk4@$$ez-0.1: This nifty little utility can make anyone feel like a kernel hacker. By scouring the net, and subscribing to the lkml, this bot sends you messages containing all the stuff that Alan, Linus, and the rest of the hacker gang, are saying, but in English! That's right. Now, when you go to linux get together, you too can sound like you know what the hells going on. Proposed expansion ideas include showing you the code, corretly commented, so that anyone can create and submit a patch.
fsckd4man-2.0: This utility, created for network defense, can be used to find exploits to your own computers connected to a network. Using advanced random-cypher-encrpyted-stenenographic-pseudo-frac tal-metasculptic-herristics-reivew tacktics, this bad boy can make it into, and grant you root permissions, on any box that has any type of security flaw. Remember, this is only intended for use on your own systems for protection.
Db-comp-0.4: This piece of GPL software is an alternative to M$'s C# complier. Db, C#, they sound the same to me.
All this and more!

Question from a Winlubber...

[lmckayjo]

Is Redhat 7.3 going to make it easy to install X on my Geforce4? If not, is there ANY current distro that I can actually install by myself (Doesn't hold my hand but won't require me to remember VI commands to boot into X) that will *automatically* set up my X server? My concern isn't driver support, but just having something better than the standard XF86Config that does nothing more than confuse the configuration file into having NO idea how to boot X on my machine, and confuse me beyond my simple understanding of Linux config'ing. Anybody have Hed Rat 7.3 on a Geforce machine? Anything to make the process real easy? I kinda hope not, since I just downloaded 7.2 yesterday...

Re:Question from a Winlubber...

[Anonymous Coward]

if you want a driver thats gonna work right then you need to goto nvidia's site and download it. none of the distros will ship the driver

Re:Question from a Winlubber...

[MrPoopyPants]

The directions on nVidia's site for installing the drivers are pretty good. There's also a program called Xconfigurator, which is a quick and easy way to configure your X server.

Since you seem to understand what X is I'd say you're way ahead of a lot of Winlubbers I have tried to help in the past...

Re:Question from a Winlubber...

[Menthos]

Get rid of the Nvidia card and replace it with a Matrox or ATI one, or card from any other manufacturer that allows free drivers to be written. I'm serious. That change alone will make your experienses with graphics configuration in distros to be so much better.

Re:Question from a Winlubber...

[Anonymous Coward]

Get rid of the Nvidia card and replace it with a Matrox or ATI one

Replace his Geforce4 with something from Matrox or ATI? Are you serious or just trolling?

The most high end Matrox or ATI doesn't do dri-OpenGL as fast as glx-OpenGL on a low end Geforce3.
And this guy has a GEFORCE 4!

If this guy doesn't want to use vi, then simply install redhat using the default "nv" drivers, download the "nvidia" drivers from Nvidia's sight and edit his X config with Kedit or Gedit or Open Office or something. Just restart X and load up Return to Castle Wolfenstien :)

Re:Question from a Winlubber...

[mohaine]

Nvidia cards do have open source 2d drivers for XFree, although I would be suprised if the GForce4 is supported yet. On my Geforce2, it just installs and works.

Nvidia is the only hardware Mfg that I know of that activily supports their hardware in Linux. Their Linux OpenGL drivers rock, even if they are closed source. I for one am very happy to see Nvidia put effort toward Linux.
ATI and Maxtor may have OS GL drivers, but they suck.

Why should we have to write drivers to run their cards?

Re:Question from a Winlubber...

[Anonymous Coward]

I just started using SuSE 8.0; it comes with dummy drivers for GeForce that allow you to use non HW-accelerated stuff out the box, but the autoupdater (YaST2) will download the NVidia .rpm's from their site (license restrictions or whatever) and install them. Easiest I've used - don't even need to edit xf86config manually - Sax2 does the rest.

Re:Question from a Winlubber...

[Anonymous Coward]

Okay, hold up hoss.

Nvidia drivers are great. The closed source drivers are better than the open source drivers, but both are quite good. So far we're in agreement.

ATI drivers suck. They're totally open source but they're big chunky afterthoughts of the Windows drivers. So far we're in agreement.

Matrox (who I'm assuming you meant when you wrote Maxtor) makes drivers just as good as Nvidia. AND they're totally open source, which makes them better. Matrox may not have any 3d performance worth considering, but you know what? A LOT of people DON'T CARE! The cards kick ass in 2d, and the drivers are rock solid.

Just because Matrox makes cards for a market sector that doesn't include YOU doesn't mean they don't include Linux. Their Linux support is phenomenal.

Re:Question from a Winlubber...

[flez]

This is what worked for me (granted I have a (much) older card (TNT2)):

Get Mandrake 8.2.
Go here and grab the perl script. Run it according to the directions.

Enjoy your 3D goodness.

.f

Re:Question from a Winlubber...

[Menthos]

Replace his Geforce4 with something from Matrox or ATI? Are you serious or just trolling?

I'm as serious as can be.

The most high end Matrox or ATI doesn't do dri-OpenGL as fast as glx-OpenGL on a low end Geforce3. And this guy has a GEFORCE 4!

This thread was about out-of-the-box driver support and configuration. There's nothing beating free drivers there.

If this guy doesn't want to use vi, then simply install redhat using the default "nv" drivers, download the "nvidia" drivers from Nvidia's sight and edit his X config with Kedit or Gedit or Open Office or something. Just restart X and load up Return to Castle Wolfenstien :)

And repeat every time there's a new kernel or XFree86 or new release...

Southern California Mirror

[schmutze]

If someone could help me check out my bandwidth that would be great:

http://toughguy.caltech.edu/pub/linux/redhat/lin ux / .3/

thanks,
chad

Re:Southern California Mirror

[schmutze]

oops sorry about that, it's too early for me:

http://toughguy.caltech.edu/pub/redhat/linux/7.3

chad

Re:Southern California Mirror

[Anonymous Coward]

I'm on dual T1s connected via Sprint. I get 30-35 Kb/sec at around 12:30 CST.

Re:Southern California Mirror

[Anonymous Coward]

I'm pulling down the ISOsat 6.5 MB/sec. This is an an Athlon box runnig RH 7.2. Very cool!

It probably doesn't hurt that I'm across town at JPL, though.

Re:Southern California Mirror

[Anonymous Coward]

120-130 KB/s from Santa Monica!!! Thanks for the ISOs.

shame

[Anonymous Coward]

1) Why do they include apache 1.3 instead of 2.0.35?

2) Yes there is a kind of BIND Configuration Tool. Buti t is so basic, that after trying it once i returned using vi....

Re:shame

[halfelven]

Apache2 is kind of early. It has problems with PHP, among others.
But yes, just like you, i can't wait for it to be included in the distro.

Patience is the keyword... :-)

Don't Give Into The Darkside

[rizzo]

Gentoo is the one, true light.

Re:Don't Give Into The Darkside

[Anonymous Coward]

the force is strong with this one ;)

Re:Don't Give Into The Darkside

[gatesh8r]

If we can him to the darkside... ;-)

Re:Don't Give Into The Darkside

[kpansky]

I have been using Gentoo for over a month now. The experience has been quite good. If you enjoyed slackware's DIY/simple attitude or Debian's simple apt-get system or BSD's ports then you will enjoy this distro.

Installation is pretty easy (maybe not for a complete noob) except for the lack of vi and the presence of nano (hint: developers).

Re:Don't Give Into The Darkside

[Anonymous Coward]

I'm your father... I'm your father...

I'm your father... I'm your father....

FTP download help

[Anonymous Coward]

I'm currecntly at school, I want to download the isos'. I have telnet access with my computer at home, how do i start the download without it canceling when I disconnect from telnet?

Re:FTP download help

[dorix]

You could do something like:

nohup (wget ftp://.../.../..disc1.iso ; wget ftp://.../.../..disc2.iso) &

Then you could exit and the subshell with the FTP's should continue.

Re:FTP download help

[gergi]

run it in the background with &

wget [f,ht]tp://path_to_iso_1/disk1.iso &
wget [f,ht]tp://path_to_iso_1/disk2.iso &
etc
logout

Re:FTP download help

[MCZapf]

wget -b ftp://whatever/blah

works for me.

Re:FTP download help

[Papineau]

wget [f,ht]tp://path_to_iso_1/disk1.iso & wget [f,ht]tp://path_to_iso_1/disk2.iso & etc logout You may want to use nohup wget ... rather than wget directly if you intend to logout while downloading.
Or wget -b ..., which sends it to the backgroud. Along with -o file, it enables you to have it in the backgroud and still have a log of what happens.

Don't forget that some FTP servers will limit the number of concurrent connections a single host can sustain at a time. And if you download more than one image at the same time, you'll (likely) prevent somebody else from getting his.

Re:FTP download help

[El Prebso]

screen is your friend.

Re:Prez Sez:

[Anonymous Coward]

Linux is for losers. Take the advice of the president and stop using crappy Linux operating systems. I find it highly amusing that Lusers have decided to construct this totally haphazardly functioning OS from scratch, when the Berkely license code was already very mature out in the wild. Morons.

If GW Bush slammed an OS like that, I'd switch to it in a heartbeat, even if it crashed my system and made it catch fire. I figure if that moron didn't like it, they must be doing something right.

Does the distribution still include Netscape?

[pomakis]

I know they've switched over to Mozilla (a move I wholeheartedly agree with), but I was wondering if they still ship with Netscape as well (at least for this release).

Re:Does the distribution still include Netscape?

[_|()|\|]

Yes:

$ rpm -qa |grep -i netscape
netscape-common-4.79-1
netscape-communicator-4.79-1

Re:Does the distribution still include Netscape?

[pomakis]

Good, because until Mozilla bug 58554 is fixed, I think I'm going to stick with Netscape 4.x!

Re:Does the distribution still include Netscape?

[Ioldanach]

Good, because until Mozilla bug 58554 [mozilla.org] is fixed, I think I'm going to stick with Netscape 4.x!

Which would be a worthy note if that was a real bug. Looks to me like an enhancement. Just because a piece of software can't do something another similar piece can doesn't make it a bug. An enhancement is something that adds useful functionality.

Re:Does the distribution still include Netscape?

[pomakis]

Which would be a worthy note if that was a real bug. Looks to me like an enhancement. Just because a piece of software can't do something another similar piece can doesn't make it a bug. An enhancement is something that adds useful functionality.

I consider this a bug. If I tell Mozilla to pass documents of type "image/gif" to the application "xv" and it doesn't, that's a bug. It's unfortunate that this was marked as an "enhancement request", because it means that it's less likely that anybody is going to look into it anytime soon.

Re:Does the distribution still include Netscape?

[bero-rh]

It's still included.

Both Konqueror and Mozilla are better for most stuff by now, but unfortunately, Netscape 4.x is still the only browser that does Java without the need of shipping a not legally redistributable JDK.

Re:Does the distribution still include Netscape?

[kryps]

Hi!

What do you mean by not legally redistributable?

The license for the Sun J2SDK 1.4.0 seems to suggest that it is perfectly fine to include it in a Linux distribution such as Redhat Linux.

Please enlighten me!

-- kryps

Re:Does the distribution still include Netscape?

[bero-rh]

There are several problems in the license.
The part I'm referring to is this:

2. License to Distribute Software. Subject to the terms and conditions of this Agreement, including, but not limited to Section 4 (Java Technology Restrictions) of these Supplemental Terms, Sun grants you a non-exclusive, non-transferable, limited license to reproduce and distribute the Software, provided that (i) you distribute the Software complete and unmodified (unless otherwise specified in the applicable README file) and only bundled as part of, and for the sole purpose of running, your Programs, (ii) the Programs add significant and primary functionality to the Software, (iii) you do not distribute additional software intended to replace any component(s) of the Software (unless otherwise specified in the applicable README file), (iv) you do not remove or alter any proprietary legends or notices contained in the Software, (v) you only distribute the Software subject to a license agreement that protects Sun's interests consistent with the terms contained in this Agreement, and (vi) you agree to defend and indemnify Sun and its licensors from and against any damages, costs, liabilities, settlement amounts and/or expenses (including attorneys' fees) incurred in connection with any claim, lawsuit or action by any third party that arises or results from the use or distribution of any and all Programs and/or Software. (vi) include the following statement as part of product documentation (whether hard copy or electronic), as a part of a copyright page or proprietary rights notice page, in an "About" box or in any other form reasonably designed to make the statement visible to users of the Software: "This product includes code licensed from RSA Security, Inc.", and (vii) include the statement, "Some portions licensed from IBM are available at http://oss.software.ibm.com/icu4j/".

IANAL, but for me, this implies:

• non-transferable -- we can't allow anyone else to copy our CDs
• bundled as part of, and for the sole purpose of running, your Programs -- we don't write anything in Java, so we'd be shipping it for a different purpose, e.g. to view someone else's Java applets -> we'd violate the license.
• You do not distribute additional software intended to replace any component(s) of the Software -- while this is probably meant to say you can't require someone to install JDK and then remove javac to replace it with something else, it can be interpreted as "If you ship JDK, you may not ship any replacements for parts of it [such as GCJ, Jikes or Kaffe]". We ship gcj.

Re:Does the distribution still include Netscape?

[FredGray]

So is RedHat not in a position to negotiate a special license with Sun for the JDK? If I recall correctly, they were willing to grant Debian explicit permission to distribute it in non-free, but the sticking point was a requirement that Debian "indeminfy and hold harmless" Sun against any claims. As a nonprofit, Debian (i.e. Software in the Public Interest) wasn't willing to accept that level of liability.

Re:Does the distribution still include Netscape?

[kryps]

Section 3 allows you distribute redistributables:

3. License to Distribute Redistributables. Subject to the terms and conditions of this Agreement, including but not limited to Section 4 (Java Technology Restrictions) of these Supplemental Terms, Sun grants you a non-exclusive, non-transferable, limited license to reproduce and distribute those files specifically identified as redistributable in the Software "README" file ("Redistributables") provided that: (i) you distribute the Redistributables complete and unmodified (unless otherwise specified in the applicable README file), and only bundled as part of Programs, (ii) you do not distribute additional software intended to supersede any component(s) of the Redistributables (unless otherwise specified in the applicable README file), (iii) you do not remove or alter any proprietary legends or notices contained in or on the Redistributables, (iv) you only distribute the Redistributables pursuant to a license agreement that protects Sun's interests consistent with the terms contained in the Agreement, (v) you agree to defend and indemnify Sun and its licensors from and against any damages, costs, liabilities, settlement amounts and/or expenses (including attorneys' fees) incurred in connection with any claim, lawsuit or action by any third party that arises or results from the use or distribution of any and all Programs and/or Software, (vi) include the following statement as part of product documentation (whether hard copy or electronic), as a part of a copyright page or proprietary rights notice page, in an "About" box or in any other form reasonably designed to make the statement visible to users of the Software: "This product includes code licensed from RSA Security, Inc.", and (vii) include the statement, "Some portions licensed from IBM are available at http://oss.software.ibm.com/icu4j/".


Regarding the other objections: I am sure that Sun will allow you to bundle it if you ask them to.

With all due respect for opensource Java implementations such as kaffe and gcj: They are nowhere near offering production quality. Thus having a recent JDK included in Redhat Linux would be really great.

BTW: I am not sure but if I recall correctly SuSE Linux ships with a Sun JDK.

-- kryps

Re:Does the distribution still include Netscape?

[Kynde]

(ii) you do not distribute additional software intended to supersede any component(s) of the Redistributables (unless otherwise specified in the applicable README file),

Like Bero said, the above would appear to state that because RedHat ships gcj they cannot ship JDK.

Nonetheless, I totally agree with you on Sun most likely being willing to allow RedHat to ship JDK.

BTW: I am not sure but if I recall correctly SuSE Linux ships with a Sun JDK.

May be so, I'm not a SuSE user, but in that case I'm guessing SuSE doesnt ship gcj, Jikes nor Kaffe...

Re:Does the distribution still include Netscape?

[m_evanchik]

Hilarious.

Bugzilla is blocking links from slashdot!

Re:Does the distribution still include Netscape?

[subsolar2]

Could Redhat do what the people at CodeWeavers do with their CrossOffice product? Microsoft does not allow redistribution of the free web fonts, but what they do is allow the user after installation to automatically download and install the free web fonts so the system has the "standard windows" fonts available for windows applications.

I would think you could do the same thing with JDK and just ship a simple downloader and installer program to automatically download and install Sun's JDK.

Re:Does the distribution still include Netscape?

[archen]

Think i'd be more worried about zlib in Netscape than any mozilla bugs. Did netscape issue a fix for zlib?

Re:Does the distribution still include Netscape?

[kryps]

Like Bero said, the above would appear to state that because RedHat ships gcj they cannot ship JDK.

I am quite sure that the paragraph in question is not aimed at other packages such gcj or kaffe but rather disallows that a vendor adds classes intended to replace core classes of Sun's Java implementation thus creating an incompatible version of Java.

May be so, I'm not a SuSE user, but in that case I'm guessing SuSE doesnt ship gcj, Jikes nor Kaffe...

Enough guessing ;-)

SuSE 8.0 ships with Sun J2SE 1.3.1, IBM JDK 1.3.0, jikes and GCJ according to their package list at: http://www.suse.de/de/products/suse_linux/i386/pac kages_professional/index_all.html.

-- kryps

Re:Real Moron

[steveeq2]

Look guys, am I the only one who thinks that XP is a good oeprating system? It's pretty damn stable from my opinion and it's a very good GENERAL PURPOSE operating system. Note what I said: GENERAL PURPOSE. Meaning, you don't have to spend hours and hours configuring XF86Config, you don't have to go searching for drivers for less-mainstream hardware, and installing softare si relatively easy (as opposed to Linux). Remember guys, most people out there aren't Slashdot geeks. Most people out there (from my opinion) just want to browse the web, type a word processing document here and there, and answer their email. THIS DOES NOT MAKE PEOPLE MORONS, just typical users.

And before you go on calling me a moron, I'm using Linux to write this message. I prefer Linux for development because it's more flexible.

up2date from 7.2 to 7.3?

[ceswiedler]

Is is possible (and easy) to use up2date to upgrade from 7.2 to 7.3, at least for certain packages like KDE?

I've been meaning to upgrade to KDE3, even have the RPMs, but up2date works so much better.

Re:up2date from 7.2 to 7.3?

[Menthos]

Some say it's possible to use up2date for upgrading by tweaking /etc/redhat-release to the new value, thus tricking up2date in the next run to upgrade the whole distro. I haven't tried it myself.

The easiest and supported way of upgrading from one release to another is of course using the installer. Just get the cd:s, pop them in and select "upgrade existing install". This is supported and will also take care of interrelease changes (like boot loader change, ext3 migration etc.), which most hacks for upgrading that only updates packages won't. Granted, there seems to be few of those changes this time, but I'd recommend the CD upgrade method any day.

Re:up2date from 7.2 to 7.3?

[Kynde]

Some say it's possible to use up2date for upgrading by tweaking /etc/redhat-release to the new value, thus tricking up2date in the next run to upgrade the whole distro. I haven't tried it myself.

I've upgraded an abundance of boxes that way, although no tweaking necessary. I've just fetched the redhat-release .noarch.rpm from some ftp site and the let up2date take over.

There have been minor issues here and there, which have forced me to download and upgrade some of the new rpms manually, but it's been really nonissue. I've done this since 7.0beta and as soon as the up2date is open for non subscribers loads of boxes here will go from "Enigma" or "Skipjack" (7.3beta) to "Valhalla".

That way you'll avoid letting their installer do it's trickery. I've had few pretty fscked-up experiences with the cd upgrades.

Re:up2date from 7.2 to 7.3?

[GreyLightning]

Yes, you can update your redhat-release number and do updates against the new redhat major version. My method of doing this is to update the redhat-release RPM which is the stub marking the version number. There are some major caveats, though. There may be a lot of packages you simply don't have, and you'll have to find these one at a time and merge them in. Some RPMS get repackaged into separate parts. e.g. the chunks of glibc that got moved into glibc-common with the 7.1 release. You'll have to manually search for these chunks. They usually have a similar name, but occasionally you'll find something like a library that's be moved to a totally different bundle. I think this is probably a lot of work. I can do it, but it's time-consuming and I don't know a lot of other people willing to sort through RPM conflicts for a few hours. There are also occasionally weird incompatibilities with updating your redhat major version. e.g. the Oracle client installer won't run on redhat 7.1. It's a java-based program, and it just won't execute. It comes packaged with its own jre, works fine on 7.0. For machines that aren't your desktop or test playgrounds, I would stick to updates off the installed base version.

Re:up2date from 7.2 to 7.3?

[edgarde]

Tried this (on a non-critical box), but up2date isn't working for non-subscribing cheapskates like myself

Error message: Free service limited due to high load, please try again later.
Error Class Code: 51
Error Class Info: Due to extremely high traffic, access to Red Hat Network is currently limited to subscription customers. Please try again later. If you would like to become a subscription customer, go to https://rhn.redhat.com/preview/priority_service.px t for more information.

I'll try again later. I'm DL-ing ISO images for when this fails utterly.

Easy way to find mirrors

[0x0d0a]

check filesearching.com for redhat 7.2 mirrors, and try 'em until you find one with 7.3

Marketing this

[Anonymous Coward]

Sure it may look good on paper, but I wonder if these guys have thought about the opinion of the general public of Open Source/GNU/Linux etc.

I have been involved in the marketing (dirty word I know!) of software and hardware to non-technical people for a number of years. The consultancy group I work for numbers many of America's top blue-chip electronics and software corporations among its clients, I have over 11 years experience of marketing, and 4 years experience of software development (VB) and systems administration (NT 3.51), in addition to a marketing science qualification from one of America's top business schools - so it's safe to say that I know what I am talking about when it comes to computers and marketing.

I have been keeping an eye this forum for quite some time now, as part of my daily intelligence gathering, I find the robust exchange of views, and technical arguments make an interesting diversion from some of the other corporate bullshit I have to deal with in my working day. I also read corporate intelligence reports from the Gartner group, Forrester, the Meta group, and Olsen Online Business Intelligence Services. Slashdot has often proved to be far more accurate when it comes to the technical details,and I am often amazed at the incredible levels of intelligence and insight shown by its readership, some of whom demonstrate a knowledge of Linux and Operating systems far in advance of anyone I have ever met, even in the IS department of major corporations. For this reason, I feel I should contribute my 2c to the debate about the future direction of Linux and the whole Open Source movement in general.

I feel I can do my bit for the Open Source community by offering (free of charge) some of my hard-earned knowledge straight from the bloody trenches at the front-line of tech-Marketing. Normally I would be paid over $4000/day for my perspective, but Slashdot - this one's on me. You people can think of it as my small and unworthy attempt to "give something back" to the Community.

Why Linux/Open Source has an image problem in major US Corporations and what the community can do about it. Like any movment, political or religious, Open Source/Linux has its Leaders, High priests and Gurus. These high profile individuals represent the public face of the organization. Like it or not, these people are associated with the product in the eyes of the buying public. One of the first things the Linux movement must do in order to gain acceptence by middle-America and Joe-and-Jean Sixpack and their 2.4 kids, is to develop what we in the Marketing profession call a "Happy Face".

When Joe Sixpack drives past a McDonald's, he associates it with the smiling face of Ronald McDonald the clown,and quality food served quickly. When he is choosing a collect-call company, the smiling face of Al Bundy (of TV's Married with Children) springs to mind, and when he thinks of fried chicken in large capacity bucket-like containers, it is the image of the happy-go-lucky avuncular Colonel with his associations of good old Southern hospitality that sticks in his memory. (In marketing terms this is known as a "positive association". Because the image puts the consumer into a "buying-receptive" mental state).

Linux/Open Source lacks any kind of "Happy Face". Now this in itself is not a problem, were it not for the fact that Linux has several extremely high-profile advocates who are the exact opposite of "Happy Faces" in that they invite negative associations into the consumers head and put him/ her into a state known by Marketers as "passive-aggressive sales-message rejection" (In layman's terms they don't want to buy the product).

Now, I will not lower the tone of the debate by naming names. I will give a few brief profiles and community members will know who I am talking about.

In reverse order of harmfullness we have the laconic, dour nothern European. Not known for his sense of hunor, and with far too many nights spent coding when he should have been out partying he creates an image of Linux as the OS of choice for "friendless geeks who never got laid". (note - I do not subscribe to this viewpoint, but trust me some of my focus group members do).

Then we have the good old gun-toting libertarian self-proclaimed open source guru. Although M.R. studies show that 78% of PC owners show right-wing bias this person is too wacko and off-the scale for them. He alienates them, and in the worst case scares them that they risk being physically harmed if they don't agree with his fundamentalist libertarian "philosophy".

Finally we have a bearded Communist hippy. Do I need to say any more ?

So the normal consumer associates Linux with a sucicidal friendless nerd from some godforsaken corner of Northern Europe, a plainly insane right wing lunatic, and an "alternative lifestyle" Communist throwback to Woodstock with a facial hair problem. Is it any wonder that time after time, the message comes back from my focus groups that Linux is for wierdos ?

Here are a few example comments from a focus group session from Q3 1999 in response to a question about their attitudes to Linux and open source software, you'll get the general idea.

"Linux - that's that geek system right ?"
"I tried Linux but it was too hard for me to install, then that guy flamed me on the newsgroups"
"I don't want any Open Source software because it is written by communists and I am concerned about security"
"My boss says Linux was written by Communists and Gun-Nuts"
"Linux is used by Communists who hate capitalism and Microsoft"
"Open source software cannot be any good because it is written by college students and hackers."
"Linux is not compatible with my USB peripherals"
"I would like to try Linux but my buddies would think I was a Commie"

I could go on and on with these genuine responses, but I think I've illustrated my point well enough. Linux has a serious image problem.

What to do about it is more problematic. Open Source proponents and Linux advocates are fiercely independent and proud of their alternative stance. They see any form of marketing as "selling out to da man" or "not groking it" or becoming a "suit" Any mention of money or financial rewards is derided, and developers are supposed to be content with "Kudos" from the community. Whilst this might be ok at college, or if you are tremendously wealthy, it cuts no ice with Joe Sixpack who was raised on Microsoft and associates Bill's millions with the quality of the software his company puts out. From the focus group again:

"If Bill (Gates) is worth that much money he must make the best software in the world."
"Microsoft must know what they are doing - the whole world uses their software."
"The best programmers work for Microsoft - they have the most tech-savvy hackers there."
"Microsoft spend millions on their software I think it is the best in the world. (referring to IE5)
Again the message is clear: Microsoft is winning the hearts and minds not only of Joe Sixpack, but also Juan Sixpack in South America, Jean-Paul Sixpack in France, Jeroen Van der Sixpack in the Netherlands, Nkwele-Olamu Sixpack in West Africa, Mohammed-Al-Sixpack in Iran, Kulwant Chandrasekhera Sixpack in India, and Boris Sixpack in the Russian Federation.

Their message is powerful, international, and presented relentlessly with no internal bickering and bitching.

What can be done ?

There are no easy answers. The Linux/Open Source community has proved unwilling or unable to accept critisim (even constructive criticism such as this) gracefully, preferring to mount foul-languaged assaults on the personal integrity of anyone who steps out from the party line.

I offer no easy solutions, however here are a few pointers:

1) As a damage limitation exercise Linux/GNU should appoint itself a "Marketing Spokesperson". This person would be the "official face of Linux/ GNU/Open Source". First and Foremost, they would wear an expensive suit, especially when talking to the press or when dealing with high-profile major corporation with deep pockets and $$$s to spend. I realise this is ridiculous from a technical perspective, but with my blend of tech-savvy and marketing exprience, I realize the importance of presentation over technical merit. It goes against the grain of the community, but if we are to become the next Microsoft (and why else would we be in this game if not to win it at all costs), we must fight them on our battleground, but with the same weapons they use against us.

2) The Penguin logo MUST go ASAP. Although it seemed "cute" and funny at the time, in the eyes of the corporate MIS department it just looks juvenile. Linux needs a new logo, preferably one of those kind of eliptical ones with a swoosh that in the eyes of the public can mean one thing: Hip and cool DOTCOM Corporation. The logo should be bland, yet robust, non-controversial yet ahead of the curve, and toned in serious businesslike colors such as gray, silver, and white. It should transcend culture and religion to be internationally recognized like the Coca-Cola image is all over the world.

3) Downplay RMS, Linus, ESR, etc. They are technicians with zero understanding of the general public, or of software consumers in general. Indeed many of them only write their program for themselves to "scratch an itch". This is hardly the way to gain public acceptance.

4) Direct X - A MAJOR stumbling block on Linux's road to world domination is the lack of Direct X support for Linux. This trivial omission means that most games will not run on Linux. Linux could gain 1000's of new games by simply implementing the DirectX api. This is a no-brainer. Kernel support for XML would be a big performance booster too in the B2B and B2C application area, and would make Linux buzzword compliant for XML.

5) Finally FOCUS GROUPS. Before you think about starting that new open-source project, (be it a new web browser like Mazola, or simply a new front-end for the cdplayer application) Get a focus group together. Use a few minutes of your non-tech-savvy friend's time. If you don't have any friends like that, try your folks, or your grandparents. Ask them what they would like to see in your new program. This way, you will gain "market perspective" on the likely acceptance of your product by the "normal people" of the world.

thank you for your time

Re:Marketing this

[Anonymous Coward]

something needs to be said here:

a) you actually spent more time talking about just how impressed you are with your own credentials than you did talking about the topic at hand.

b) "Lets just make DirectX run on Linux"

gee, why didn't I think of that?

Which VM?

[tholti]

Does anybody know which VM (AA's or RvR's) is used in the accompaning 2.4.18 kernel? Alan Cox is using the RvR-VM in his ac-branch, so Red Hat Linux 7.3?

Re:Which VM?

[Anonymous Coward]

they have rmap and o(1) sheduler. and a bunch of other stuff too :) and, most importantly, the obviously very effective stresstesting.

How is KDE3 running?

[ChrisWong]

I have been underwhelmed by Red Hat's packaging of KDE in the past. For example, in a boxed release (either 7.1 or 7.2), kdehelp's "back" and "forward" buttons didn't work. When KDE 2.2.2 RPMs were released, they helpfully included (and required) a version of Qt that froze the desktop: I had to disable klipper. The current KDE3 RPMs for RH 7.2 from Red Hat have their own glitches: ksplash goes kblooie at startup, and konqueror seems to have this big memory leak that bloats its footprint over time. I wonder if anyone at Red Hat even tries to use KDE.

How is KDE3 running on RH 7.3? Does Konq still have that memory leak?

Re:How is KDE3 running?

[Anonymous Coward]

Bill Gates will be ice skating with the Devil the day Redhat decides to give KDE some respect and ship RPM's that haven't been deliberately broken.

Re:How is KDE3 running?

[HiThere]

Bill Gates will be ice skating with the Devil the day Redhat decides to give KDE some respect and ship RPM's that haven't been deliberately broken.

Ah! These must be good then.

Re:How is KDE3 running?

[bero-rh]

I have been underwhelmed by Red Hat's packaging of KDE in the past. For example, in a boxed release (either 7.1 or 7.2), kdehelp's "back" and "forward" buttons didn't work.

This must be related to your setup. It doesn't happen here, and it doesn't happen to anyone else, at least not to anyone who cares about it enough to report it (as always, I can't fix problems I'm not aware of).

Even if a problem seems obvious to you (say, a crash on startup), go ahead and report it because chances are it happens only on your setup or your hardware. If it were really as obvious as it seems to you, it would have been fixed.

When KDE 2.2.2 RPMs were released, they helpfully included (and required) a version of Qt that froze the desktop

Also, for your setup only. Worked perfectly here.

The current KDE3 RPMs for RH 7.2 from Red Hat have their own glitches: ksplash goes kblooie at startup

This is a known problem. It only occurs on first startup though, which is why I didn't notice it before uploading the packages to kde.org.
It has been fixed since, and is fixed in 7.3, along with several other problems we've noticed.

and konqueror seems to have this big memory leak that bloats its footprint over time.

Not reproducable here. Report details here or here.

I wonder if anyone at Red Hat even tries to use KDE.

Yes. Plenty of us do. I haven't seen any other desktop running on any machine in our .de office for quite some time.

The only two desktops I ever use are KDE and text mode. Konqueror and lynx are my favorite browsers.

enlightenment

[koekepeer]

hey! they dumped enlightenment! i know they don't use it in gnome anymore, but to throw it out completely is not so nice IMHO. hoep it's not because they hold a little grudge towards rasterman...

Re:enlightenment

[Anonymous Coward]

What is not "nice" about it? No one uses it. They are dumping a lot of packages that are old and no one uses like E. They are also dropping things like X3.3.6 and things like mars_nwe. If you want E download it yourself. Redhat can't keep dragging along every package no matter what.

It always funny how users will complain about something that is free.

Re:enlightenment

[Anonymous Coward]

E has been dead for a long time...
everyone I know who had be big supporters of E now are switch to either Afterstep or sawmill. They really lost themselves...once Rasterman and Mandrake were the toast of the town. There is no reason to keep a package that no one uses except for the E developers. Btw who do not even listen to their own people...

Re:enlightenment

[Anonymous Coward]

"Btw who do not even listen to their own people..."

Gee like you? ....

I agree those that still use Enlightenment is barely on the radar scope...this project and the people that develop it have gone into the where are they now files.

Re:enlightenment

[Anonymous Coward]

Just looked on their site...the last entry about the next version of E was mar 28, one wonders..if the E developers...finally realized how buggy, memory leaked E16 and stop developing E17

Re:enlightenment

[Hitokage_Nishino]

I'd like to stamp a big WHATEVER on your post.

First of all, E is not dead. The E devs are very busy preparing DR17 of Enlightenment.

Second, E DR16 still works wonderfully for me and it will take a LOT of improvement on Sawfish's part for me to switch over... and I'm no dev myself.

Re:enlightenment

[Anonymous Coward]

Yeah buts been what two years of development time for E17? You think it would have been ready by now. They have nothing to show that time and investment has taken place except for a few screenshots and few libraries. Even the E devs admit it will be at least year before E 17 will be out. The world pass them by long time ago.

Re:enlightenment

[Anonymous Coward]

I have heard this as well. I know from reading ck and gilbert sites, there is no effort to develop E17. CK points out that no one has put any effort into helping him build E17 right. Oh well...It was great while it lasted...

Re:enlightenment

[analog_line]

Maybe no one you personally know uses Enlightenment, but I know a whole lot of people who do, myself included. Just because you don't like it doesn't mean everyone doesn't like it. Blah blah no man is an island blah blah.

Re:enlightenment

[Anonymous Coward]

I think your taking the universe out of context. Its not enlightenment was cool...its just that there is better stuff that has been developed since E16, and nothing has been changed since 16.5, there is no use for it anymore...

Re:enlightenment

[Anonymous Coward]

Speculation is that Raster has totally lost to code , while the rest of the E developers are clueless on what to do next.

Re:enlightenment

[Anonymous Coward]

I heard Mandrake is now writting programs for xp and has totally ditched E

Re:enlightenment

[koekepeer]

crap. go check out cvs, and follow the e-devel mailing list.

wait for version 8 if you want gcc 3.0

[gimpboy]

since redhat increments the major version number when they break binary compatability, i would expect we will have gcc 2.96 until version 8.

Re:wait for version 8 if you want gcc 3.0

[Anonymous Coward]

> since redhat increments the major version number when they break binary compatability,

Nope, not really. For example, they included KDE 3 which broke binary compatability with KDE 2. When they include GNOME 2, which will break binary compatability with GNOME 1.x, they won't release Redhat 8 either. They will only release Redhat 8 when there is a new version of glibc.

redhat 4.x - libc
redhat 5.x - glibc 2.0
redhat 6.x - glibc 2.1
redhat 7.x - glibc 2.2
redhat 8.x - glibc 2.3 (when it's ready).

rsync

[gimpboy]

try finding an rsync server. then you can get the diff's possibly.

Instant ISOs available

[neuroid]

If you shell out some cash for RHN, they have 'Instant ISOs' available.

Much faster than the mirrors I've tried - check it out.

Re:Instant ISOs available

[jmorris42]

So did it actually work for you? I tried twice with Skipjack and would get a hundred or so megs into it and it would fail. I'm guessing Moz is loading it into ram before saving. Wish they would offer up their ISOs via ftp with my RHN userid/password.

Re:Instant ISOs available

[Anonymous Coward]

I had this same problem with mozilla (0.9.8?), ran out of memory. I ended up using Konqueror which saves it directly to disk.

Of course it's off-topic

[Anonymous Coward]

The news story was about the new release of redhat. Plugging your out-of-date redhat newbie guide is off-topic...dot sure what part of that you don't understand. A new redhat release does not mean that any redhat-related post is on-topic.

Re:Of course it's off-topic

[alt.sex.fetish.jesus]

You've got a point, but do you think the guy deserved to lose 2 points on a marginally off-topic post? (Besides, a new release of Linux _does_ mean his guy gets a little older, so it kinda sorta _is_ on topic). Seeing the load of crap postings on here that either get ignored, or modded up as "Funny" I'm not sure why he got punished.

Besides, don't the moderator guidelines encourage moderating UP rather than moderating DOWN? I'll be watching for this one in meta-moderation.

(Now watch this posting get moderated down as well.)

pattern also seems to follow kernel versions

[GroundBounce]

Might just be coincidence with the compiler/binary compatability thing, but it seems that major RH releases also track kernel versions:

RH 5.x - kernel 2.0.x
RH 6.x - kernel 2.2.x
RH 7.x - kernel 2.4.x

Re:pattern also seems to follow kernel versions

[Fabian Kroenner]

Both of you got it wrong!
Red Hat release numbers indicate the version of the C library (for binary-compatibility) - like so:

• 4.x - libc5
• 5.x - glibc2.0
• 6.x - glibc2.1
• 7.x - glibx2.2
• 8.x - go figure!...
  

Re:pattern also seems to follow kernel versions

[Papineau]

Don't forget the version of the C++ compiler. I don't have the different version numbers shipped with the different distributions, but from one version to the other the ABI (application binary compatibility) is not guaranteed, if not plain absent.

IE, you can't link some part of a C++ program compiled with gcc-2.95.3 with some other compiled with gcc 3.0, although you can do the same thing with a C program.

I think the compiler thing will be one of the major compatibility changes for RH 8.0.

(And damn them, I left my dorm room last week!)

Re:pattern also seems to follow kernel versions

[Cro Magnon]

I believe RH 7.0 still used kernel 2.2. 7.1 was the first one with 2.4

Japanese version?

[alt.sex.fetish.jesus]

When RH 7.2 came out, I was surprised (and impressed) to see that they made the Japanese version available at the same time. Not so, this time around.

Anybody know how long before 7.3J is ready?

Re:Japanese version?

[Anonymous Coward]

The Japanese version is the same as English, German, Spanish, Italian, French, Chinese, Korean, etc. It's all integrated into the same distro now.

Did they patch Mozilla 0.9.9?

[Anonymous Coward]

Mozilla 0.9.9 has a terrible bug:

http://sec.greymagic.com/adv/gm001-ns/

(also present in RC1, but fixed in the latest
nightly). Does 7.3 ship with a patched 0.9.9?

Re:Did they patch Mozilla 0.9.9?

[Anonymous Coward]

Doesn't anyone care about this stuff anymore?
I mean, it *is* a pretty serious bug. "News
for Nerds. Stuff that Matters" my ass :)

your sig

[Anonymous Coward]

run your google again... it's grown:

Results 1 - 10 of about 186,000,000. Search took 0.09 seconds

well

[Anonymous Coward]

I'm running a Geforce4 on Redhat 7.2 just fine. Before screwing with your X settings, download the NVIDIA kernel and GLX drivers from nVidia.

So, all the official mirrors are slow..

[Anonymous Coward]

Try this super fast unofficial one:
http://www.sheriff.martin.fl.us/rh/ or anon ftp to the same box.

To late...

[CptNoSkill]

I made a youthful indescration.. and am now one with the dark side.

Re:To late...

[layingMantis]

what the fuck is an "indescration" ???

the Dark Side has clouded your mind and spelling abilities, Jedi.

Seems fast from here

[Wee]

I'm on a Road Runner cable modem in San Diego. I'm getting ~80Kb/sec pulling down all three simultaneously.

Not bad. Thanks for the mirror!

-B

Can you upgrade and leave the Kernel alone?

[cs668]

When I have upgraded my Redhat system in the past it has always installed the kernel from the cd. If I would try to deselect the kernel package it would still force its installation.

Since I am usually building my own kernels it would be nice if I could upgrade to 7.3 without installing the kernel packages.

Is there a way to do this?

Re:Can you upgrade and leave the Kernel alone?

[guacamole]

Consider not building your own kernels. RedHat's kernel packages are very flexible and there should be very few users who really need to compile a custom kernel.

Re:Can you upgrade and leave the Kernel alone?

[charnerd]

I'd just rebuild the kernel if I were you after you install 7.3. But you could also just save off the kernel file and use it again as your kernel after you install 7.3. It's also important though to have your kernel source tree available in /usr/src/. Good luck!

Re:KDE3 -- and may I add (spoilers)

[Laplace]

I installed SuSE 8.0 on my PII Celeron 333. KDE3 rocks on it. I installed Mosfet's Liquid theme (http://www.mosfet.org/liquid) and I just love it. Snappy, fun, all that. Like having Aqua without the need for expensive hardware. Yummy.

Re:Real Moron

[Glanz]

The only word I objected to in your first intervention is "real".. when describing XP. If topheavy code crashes, "call-homes", embedded insecurity, telly-tubbyesque graphics, a fat price tag, and being treated like a complete idiot by an OS with an embedded Gates EGO, is your idea of a "real" OS, then I guess you are right

More Project Management under Linux

[GroundBounce]

Although it's not open source, another program to check out is Intellisys Project Desktop. It is Java-based, so it is cross-platform, which can be useful in some environments. It is also more mature than MrProject or Toutdoux (sp?).

Interesting

[flanker]

Not to be a Red Hat basher (I run various flavors of RH on all my machines) but do they usually not mention the GPL-required downloadability in their press releases?

Re:Interesting

[Isaac-Lew]

Not all of the software that ships with RH 7.3 is released under the GPL (XFree86, apache, perl, mozilla & openssl/ssh are the examples I can think of right off the top of my head).

Re:Interesting

[Blrfl]

Nothing in the GPL (for those packages that are GPLd) requires that download information be included in press releases.

diff 7.2 7.3

[wormbin]

Here is a full diff between 7.2 and last night's rawhide which should be pretty close to 7.3.

diff-72-73.txt.bz2

Mandrake too...

[EvilAlien]

Mandrake unleased their crazy KDE lovin' upon an unsuspecting public on April 23rd. Red Hat is just following the trend.

Its not that big of a feature to the GNOMEish, but I 'll give KDE 3 a shot when I rebuild my dev box with RH7.3 today. I'm not a big fan of KDE, and each time I try it, I fail to be suitably impressed. Maybe 3 will change all that =)

Re:Mandrake too...

[Cro Magnon]

Didn't Mandrake ship with KDE 2.2.2, saying that KDE 3 was too unstable? And if Mandrake thought it was unstable, do you really want to use it?

Re:Mandrake too...

[EvilAlien]

Yes, 8.2 didn't ship with KDE 3. They have recently released packages for KDE 3 separately. Personally, I don't want to use it for a production workstation, and I don't intend to. I tend to stick to GNOME/Sawfish or GNOME/Enlightenment.

Re:Mandrake too...

[optikSmoke]

Didn't Mandrake ship with KDE 2.2.2, saying that KDE 3 was too unstable?

Which is because they released 8.2 before KDE 3 came out (I'm pretty sure).

And if Mandrake thought it was unstable, do you really want to use it?

Ignoring the obvious pot-shot, it wouldn't have made sense for Mandrake to release a distribution with a version of KDE that would have been from CVS (because, like I said, KDE wasn't ready at the time). Unless you like it when distributers include unfinished software?

Re:Mandrake too...

[Anonymous Coward]

I tend to stick to GNOME/Sawfish or GNOME/Enlightenment.

ewwwwwwwwwwww... kde/fluxbox or (usually) just fluxbox for me, thx.

D'oh!

[buford_tannen]

And *of course* RedHat releases 7.3 _one_ day after I get home from college and my 10Mbps and arrive back at good ole' 56K. It appears that Red Hat has conspired against me, to force me to go and buy a boxed set. I will surrender this time, but only this time.

gcc-2.96

[Jacek Poplawski]

When RedHat will switch from gcc-2.96 to something else? That's the only reason for me to install Slackware instead RH/MDK.

Re:gcc-2.96

[Sits]

When GCC 3.0 compiles everything perfectly? I hear there can be problems using GCC3.0 to compile the kernel and Mozilla...

Re:gcc-2.96

[Papineau]

Probably in their next revision (8.0). Actually, they ship it right now, as gcc3 packages rather than gcc-3. So if you want to use it, you can. Their testing must have shown that gcc 2.96-RH is the one they want to ship right now.

If they change their default compiler to gcc 3.x, then they'll up the version to 8.0 because that new compiler (mostly the C++ compiler, but maybe the Java one also) won't be binary compatible with the previous one.

And if that's the only reason for you, why don't you give it a go yourself? It's pretty easy to do, especially if your second choice of distro is Slackware rather than say Suse or even Debian.

Re:gcc-2.96

[bero-rh]

There's not much of a problem with 2.96.

Earlier versions than 2.96 are not an option because they don't do real C++ (see http://www.bero.org/gcc296.html).
3.0.x releases are rather broken and don't have any real advantages over the current builds of 2.96.

gcc 3.1 will be a very good release, even better than 2.96. It is what we're likely to use in the next major release (unless, of course, gcc 3.2 comes first and is good).

Re:gcc-2.96

[Anonymous Coward]

Actually GCC 2.96 *is* a problem.

Our company ships two products that run on Linux -- one ships in binary (it's a COMMERCIAL application) and it's built on GCC 2.95.3. It's a major PITA to make people build 2.95.3 to run apps, but that's what we have to do. They're unhappy.

Our other product ships as source, and the supported compiler is (again) GCC 2.95.3. To officially support them, we require it. But now with GCC 2.96-98 and later, at least 2.96 works.

But why fork another version of GCC? Why should we need to support an extra compiler version (amongst all the other OS and compiler versions we support) because you continue to use a proprietary compiler version?

Sucks, man.

Re:gcc-2.96

[tempest303]

except their 'proprietary' compiler is the only one that's not broken.

With RH 8, (the next major release, no doubt) they'll surely use gcc 3.x and be in line with the rest of the world.

Hopefully it'll include Apache 2, Gnome 2, OpenOffice 1.x, and Mozilla 1.x! That'd be a lot of milestones in one release... It'll be an interesting Fall/Winter when 8.0 ships.

Re:gcc-2.96

[Compenguin]

In the most recent issue of Dr. Dobb's it says that 2.96 is only marginally superior to 2.95.3 and 3.0.4 is far superior to both in the c++ arena

Re:gcc-2.96

[Kynde]

But why fork another version of GCC? Why should we need to support an extra compiler version (amongst all the other OS and compiler versions we support) because you continue to use a proprietary compiler version?

Why not really? 2.95.3 is buggy when it comes to C++. Neither does the 2.96 support C++ like 3.xs do, but atleast it's not that buggy. It's a big hop from 2.9x to any 3.x, and as I see it having an intermediate compiler is not that bad at all, sort of a bugfix release of 2.95 if you will.

Actually at my work I've been forced to back port code to avoid some of the C++ flaws present in 2.95.3. (Actually nevermind that, but the countless hours I've spent with gdb hunting them down and finally accepting them as compiler errors was the bitching thing. Over the years I have learned that the flaw is always between the keyboard and the chair (speaking about real operating systems here :)). It can be quite annoying when it's not.)

(and yes, I admit that at the time RH first shipped 2.96 it wasnt pretty, but they've come a long way since)

Re:gcc-2.96

[Kynde]

In the most recent issue of Dr. Dobb's it says that 2.96 is only marginally superior to 2.95.3

Marginally superior, well yes, but when that marginal holds bugfixes, I'm ok with it.

2.95.3 is for the kernel (and bloody good for that, too) and 3.x is where we'll be (and yes, it will be fantastic), but right now the 2.96 feels right.

Re:gcc-2.96

[MSG]

one ships in binary

I was about to suggest that you include a copy of libstdc++ with your application, but I just noticed that the RPM package and serveral files in the CVS repository all bear the GPL. That would make it illegal to link proprietary applications against libstdc++, wouldn't it? :( I think I'm going to bring this up on the redhat-list and see if anyone has anything to say about this...

because you continue to use a proprietary compiler version?

It's not "proprietary". All of the code is available. It's Free Software.

Re:gcc-2.96

[Anonymous Coward]

no, 8.0 will not ship until glibc 2.3 comes out.
the next release will probably be 7.4.

i've said it once, and I've said it again, RH only releases on new glibc releases.

rh 4.x=libc5
rh 5.x=glibc2.0
rh 6.x=glibc2.1
rh 7.x=glibc2.2
rh 8.x=glibc2.3

Re:gcc-2.96

[Anonymous Coward]

Yeah but as I understand that wasn't
the reason anyway.

A major reason for going to 2.96 was
for the compaq alpha processor support.

Re:gcc-2.96

[PrismaticBooger]

There's not much of a problem with 2.96.

Except, of course, it's not gcc3, and isn't compatible with gcc3's ABI.

It doesn't bother me so much that gcc 2.96 is the primary compiler in Red Hat 7.3; but the availability of gcc3 RPMs for concurrent installation with 2.96 was extremely useful for those of us who need to compile modern C++ code that just won't work with 2.96. Those RPMs are sorely missed in Red Hat 7.3.

Is there any chance they'll make it back in an update?

Re:gcc-2.96

[Anonymous Coward]

> I was about to suggest that you include a copy of libstdc++ with your application, but I just noticed that the RPM package and serveral files in the CVS repository all bear the GPL. That would make it illegal to link proprietary applications against libstdc++, wouldn't it?

Not illegal unless they don't make the required offer to provide sources in the same way/cost as their binaries.

"proprietary" != "closed source", which is what you were implying.

Is it worth getting 7.3 from 7.1 and 7.2?

[antdude]

I am happy with my 7.1 and 7.2 installations. I do a lot of updating with programs I use like OpenOffice, GKrellM, Lopster, GNOME, KDE (KDE3 now) etc.

Are there anything in 7.3 worth that will want me to upgrade? From what I saw, nothing is really new except updated components. I will install 7.3 from scratch if I ever had to reinstall due to a HDD failure or something.

Thank you in advance. :)

Re:Is it worth getting 7.3 from 7.1 and 7.2?

[halfelven]

From what i've heard, the kernel is significantly better.

Re:Is it worth getting 7.3 from 7.1 and 7.2?

[slagazor]

I'm just wondering has anyone ever used the upgrade feature built into the install CD ? I'm wondering how smooth it is. Rob.

Re:Is it worth getting 7.3 from 7.1 and 7.2?

[Micah]

I haven't upgraded to 7.3 yet (will do today if I can manage to get the ISOs) but for all previous versions, the upgrade has been nearly perfect. No worries.

yeah redhat network...

[Polo]

So I guess joining the redhat network worked out pretty well. I'm downloading the iso images in parallel at the capacity of my cable modem.

Re:yeah redhat network...

[Papineau]

I don't know if that's from an FTP or HTTP server, but in doing so you're probably limiting the number of people who can download the isos from RHN, since a lot of FTP servers have a limit on the number of concurrent connections (that's especially true for anonymous connections, but also to a lesser extent for authenticated connections). Those people have paid the same thing as you: why would you prevent them from getting it? One at the time would be better for everybody, and it might not even take you more time (since with all of them, you're maxing out your cable-modem, with one you should be able to also)!

(Saying this from a DSL line with 3 concurrent downloads from a single mirror... bad me! Bad)

Re:yeah redhat network...

[jaju]

I too managed to do it without the subscription.
Depends on how you choose your mirrors.

which VM?

[Anonymous Coward]

Linux is dying. It has set back the state of computing for 10 years with its unstable VM and incompatible networking that can only talk to itself. What a bunch of wasted effort reimplementing what's already in the more stable higher performance FreeBSD.

that's a very lame review

[halfelven]

It surely doesn't fit into a "thegurus.com" website.

my mirror of the iso's

[gimpboy]

hey..

i just got mine. if you need to feel free to suck them down here.

the md5sums all check out.

have fun.

Ext2 is not "mission critical". Use XFS instead

[halfelven]

You're buying yourself a ticket to disaster if you use Ext2 on "mission critical" servers.
Better use XFS. From the link below you can download a modified Red Hat distribution (just XFS was added to the Red Hat kernel, there's no other change) which allows you to install directly on XFS:

http://oss.sgi.com/projects/xfs/

XFS is the filesystem used by SGI (Silicon Graphics) on their 1024-CPU video servers and supercomputers.

Re:Ext2 is not "mission critical". Use XFS instead

[arkane1234]

You're buying yourself a ticket to disaster if you use Ext2 on "mission critical" servers.
Better use XFS. From the link below you can download a modified Red Hat distribution (just XFS was added to the Red Hat kernel, there's no other change) which allows you to install directly on XFS:

Problem is, in a mission critical situation you want to be able to split, grow, and other things with your partitions on the fly. XFS doesn't provide the shrinking ability.

EXT3 and ReiserFS do... unfortunately it's best to take ReiserFS offline when you shrink a partition.

This is simply amazing

[rosewood]

I have a strange track record. I don't *need* a linux box but I like to keep one running for web serving, ftp, a router, etc. This means that some times I have to pull hardware from the box for other necesities. Well, yesterday I finally got the box back up and running (and much more powerful then it was in its last conception). I installed and have throughly updated Redhat 7.2. Now, 7.3 comes out.

This has happened to me for every major . release since 6.0! I sware, if you want the next version of redhat, I just need to install it and update it. Pfft!

Re:This is simply amazing

[ejasons]

And the string of betas (tirelessly reported on Slashdot) over the last month didn't give you a hint that there was a new version coming out?

Re:This is simply amazing

[Pierre]

Think of the power you have over Redhat. Simply by installing a new version you FORCE them to announce a major upgrade.

Hmmmmmm, If you install 7.3 on 3 or 4 boxes maybe we could 8.0 out imeadiately!

Re:This is simply amazing

[moosesocks]

odd... the same thing seems to happen to me with every distro execpt for debian (maybe if I try now, a new version will come out. the last time I tried was about a year ago).

Or it could simply be because of the absurdly large amount of time it takes a 56k user to download 1.2 gigabytes of ISO images

no, it's patched

[halfelven]

It's based on 1.1.3 allright, but it's patched against that bug.
They just ported the patch from 1.1.4 to 1.1.3

SiS6326 with XFree (was Re:versions questions....)

[Urchlay]

My workstation at work had an AGP SiS 6326 when I first started my current job. It worked more or less OK with XFree86 4.1.0 and later 4.2.0, but the SiS driver doesn't offer much 2D acceleration (not sure how much of this is a deficiency of the card, and how much the driver). It's slightly faster than VESA framebuffer, and unlike the framebuffer driver, you can crank up the refresh rate and resolution (I think I was running 1152x864 at around 80Hz). There's even an experimental DRI (3D accel) module for the SiS cards, though it's useless as the driver only allows use of 4M of video RAM (not even glxgears can run with only 4M, and quake wouldn't even try to use it)

If you're stuck with a SiS card, your best bet is XFree86 4.2.0 (which Redhat 7.3 has, so it should be OK)

The *one* nice feature about the SiS is that its XFree86 4.1.0 and 4.2.0 drivers support rotation.. so you can flip your monitor on its side and run it in `portrait' mode (though you can't flip back to landscape mode without editing XF86Config and restarting X). This is unbelievably cool for web browsing and reading PDF docs that have the aspect ratio of standard printer paper.

To stay slightly on topic here.. The 6.x versions of Redhat would detect the card and configure X for you, but until you edited XF86Config by hand and added a line like this:

Option "xaa_no_color_exp"

...the card was unusable (e.g. fonts came out as solid black blocks without this line)

Hm, that's still not all that on-topic.. but since RH7.3 ships with XFree86 4.2.0, you won't need to do this (only applies in 3.3.x). Like I said above, the card is usable, but not great, in 4.2.0. Get a Real Card as soon as possible.

Re:Prez Sez:

[halfelven]

Take the advice of the president

When did he said that?

Halls of Valhalla

[Anonymous Coward]

Where Ragnar is hanging out these days.

Oh.. wait a minute..nobody bought any of Loki's games... nevermind.

Great for geeks

[horza]

I have to admit I'm quite excited about installing Gentoo (just waiting for my new gfx card to come through the post) and will install it instead of Mandrake 8.2. Red Hat was my first exposure to Linux and serves well as an eye opener. I tend to stick with what I know but on a spare machine I tried a couple of other distros. Mandrake 8.1 instantly converted me. Now I want a fast desktop, no worries about keeping my software up to date, and want to get rid of the bloat. The two most attractive distros are Scorcery and Gentoo, with the latter appearing much better supported. This is for my desktop machine. I'll keep Red Hat on my production server as I like the fall-back of business support, even though I never have and probably never will use it.

Phillip.

I'm new, please don't hurt me.

[SimplexO]

I am new to the linux world, and my first distrobution was RedHat 7.2, however I really wanted to wait for the distrobution with KDE3 in it. You know how being young and bored in a dorm room can get, so I installed 7.2 using 2 CDs (although my 2nd CD was corrupt; always check your md5sums). 7.3 has 3 CDs. Can anyone explain why 7.3 is now on 3 CDs?

Re:I'm new, please don't hurt me.

[Anonymous Coward]

You forgot to said young and stupid >:)

Re:I'm new, please don't hurt me.

[Anonymous Coward]

I think that in previous distributions, the Powertools cd contained extra, and sometimes useful, packages. RH discontinued the powertools cd, and bundled the most useful packages in with the main distribution.

Re:I'm new, please don't hurt me.

[SimplexO]

You forgot to said young and stupid >:)

Thank you. I'm not young (maybe compared to some) and I'm definitely not stupid. I just know how harsh the nerds are on this page when someone asks a question that just possibly might have already been answered. I figured a little ego-stroking might do the trick to get the information I needed.

Also, congratulations on your use of the English language.

Better Kernel?

[antdude]

How is that? I usually use Linux as a workstation and play Q3A and RTCW.

Re:Better Kernel?

[halfelven]

It has the low-latency patch included, plus other goodies.

Getting slower....

[Wee]

I was at ~80KB/sec, but it seems to be leveling out at around 40KB/sec. Which is still not too bad.

-B

Mirror in Europe

[Yenya]

My mirror still has some 30 Mbps of free bandwidth, so if you are in Europe, you can try to download from it.

Re:Mirror in Europe

[JanneM]

Thanks! I'm getting around 120KB/s from you.

/Janne

Re:Mirror in Europe

[Anonymous Coward]

Gracis, I'm In Texas and still getting great speed. In fact, the first two ISO's are already done...

Re:Mirror in Europe

[jrwyant]

Thanks again! I'm at my DSL max (~180kiB/s.) :)

MD5SUM

[Anonymous Coward]

Can somebody please post the correct MD5SUMs of the ISO-images?
Thanks
Russell

Re:MD5SUM

[repetty]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

c9a4d963a49e384e10dec9c2bd49ad73 valhalla-SRPMS-disc1.iso
41b03d068e84d2a17147aa27 e704f79b valhalla-SRPMS-disc2.iso
cb91810ce8173039fed24420 407e4c59 valhalla-i386-disc1.iso
ec1b813d32ffdc8edc2be2617 35d17de valhalla-i386-disc2.iso
5dc81ce523cfddf99b4d4d63e 91bcaa7 valhalla-i386-disc3.iso
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE8z/oCIZGAzdtCpg4RAsMvAJ9+xOn4Pw1T0mp8zVT6 4c EDWuqqKwCfblTd
4Lw0SvJC+v/6JbGIxJWL7aA=
=0xs+
- ----END PGP SIGNATURE-----

Re:MD5SUM

[Anonymous Coward]

Thanks. Are you sure those are the right isos? Are they really called valhalla? I thought the codename was skipjack. Strange ...
Thanks anyway.

My experience with Dead Rat Linux 7.2

[layingMantis]

This is some info for any geeks considering this bloatware concerning my personal exp. with it:

1. LILO does not work. This forced me to rely on a boot floppy.(see next item)

2. The option under the KDE menu to make a boot floppy doesn't work!! This is pathetic.

3. It's slow. Yes, that's right folks, slower than Windows 98, with which is shares a partition.

4. The bottom menu bar, Kicker or whatever the hell its called, sucks. You can easily add an applet, for example, but you can't 'right click' and take the damn thing off. This is stupid.

5. The documentation manual is a FUCKING JOKE, just like with Windows.

6. The driver for my Soundblaster Live! card is straight up BROKEN. There is an open source driver out there, but it entails that the kernel not be compiled with any other drivers for the card. In other words, you must re-compile the kernel. Yep, user friendliness at its best.

7. There's a little peice of shit prog called "lokkit" that is supposed to configure your firewall. It is BROKEN as well. I used it once and couldn't get on the 'Net no matter how hard I tried to fix it. I had to re-install.

sound good to you? admittedly, I'm no Linux veteran. So if you are and enjoy recompiling kernels and poring over thousands of configuration files, then have at it. But please, don't pay for this shit like I did.

$60 bucks for a fucking Red Hat sticker is not a good deal.

Re:My experience with Dead Rat Linux 7.2

[Sj0]

1)Use Grub, or rather, don't change the setting in the first place, because lilo isn't default in RedHat 7.2. It was set as default when you installed the thing, don't complain when you mess with settings you don't understand and you break the OS.

2) No comment -- I didn't even know this existed.

3)So is Windows XP. It's a grim fact of life that if you want to run a modern OS, it will be slow. Ask anyone using Windows XP on 64 MB of ram about that. On the other hand, Linux can be made to run fairly lean, and I've had it running on a p133 without too much trouble.

4)Hit ctrl-esc to bring up the KDE equivilant to the task manager, click on "tree view", and kill the applet. It's sitting right under the "kicker" item. No matter which OS you are running, there will sometimes be applications which have lousy interfaces. Linux has quite a bit better fidelity when you want to kill these applications.(ie. you don't need to hunt around for a utility on the internet to uninstall it, like some Windows programs)

5)No comment. I may have a comment the day I find a manual which actually helps me do something. So far, intuition gets me further than that, so I usually need to surf to find my answers.

6)You'd be suprised how easy it is to exist without ever needing to re-compile. I had the same sort of problem (sis AC97 sound on a sis 735 mobo), but downloading the latest kernel from RedHat fixed it. Downloading 7.3 would likely fix both our problems.

7) There are two possibilities in this case, and I'm inclined to go with the first: Odds are, your networking never worked. Since you were booting off the floppy, it's possible that your interfaces never came up. The other possibility is that you were mucking around with something you didn't understand -- even at high security, the firewall doesn't block outgoing communications. It blocks incoming communications(like hackers trying to hack your box).

One tip: When you paid that 60 dollars for RedHat, you weren't paying to have to hack around to get things working; Calling tech support was an option, and they likely would have talked you through an installation which wouldn't have had the same problems.

I would have agreed with you if we were talking about RH 5, because it's UI was an embarassing attempt at cloning the windows gui. I would have agreed with you if we were talking about RH 7.1 even, because the threat of data loss from a single power outage was too great to recommend it. RedHat 7.2, on the other hand, is the version I believe is ready for prime time, and I am quite excited about trying out RedHat 7.3, with even more hardware support, an even better UI, and more graphical tools to make the world an easier place to be for a mouse user.

Re:My experience with Dead Rat Linux 7.2

[layingMantis]

Well, i knew i'd be flamebait on that one (what with criticizing Linux and all) but anyway....

#1 I actually did use the default one (I forgot whether it was GRUB or LILO).........but when I'd try to log in.........no Windows (or DOS) option. Only Linux.
#4 -- o.k., i'll try your advice on that one, but I still think it is dumb that you can add an applet so easily, but can't take it away via the same method.
#5 -- you are lucky to have your intuition, because the little two books they give seriously aren't worth a damn. It seems thrown together by like one guy. I resorted to buying a book (Running Linux, o'reilly) which has helped.
#6 -- well, I hope you are right, and the 7.3 kernel will support my (very common) soundcard. But won't I still need to re-compile, and answer lots of obscure hardware questions? Not for the faint of heart.
#7 -- with the "lokkit" thing, there are only like 3 options, low-medium-high (and then a short custom list), so I still don't think it was my fault there. (also found a couple of other similar testimonies on the Net) My networking was working okay till I messed with that damn thing.

so all in all, not the best experience. I'm leaning toward trying Mandrake now, but I damn sure won't pay for it. I'm a fool for having done so with Red Hat.

oh, one more aside before I bore you to tears -- when I went to register for my "support" (which only covers basic installation questions - no soundcard stuff allowed) - their site said my product ID did not exist. Imagine my anger; already I was pissed off and frustrated! To their credit, they responded promptly to my indignant e-mail with a working number. Alas, by then my mind had been made up.

there is one thing all this has made me realize, though. "Free" software like Linux still has an amateurish, non-integrated feel to it. Windows 98 seems fairly tight by comparison; it even plays my mp3's! (I still despise it of course.) Maybe OS-X?

~mantis

Re:My experience with Dead Rat Linux 7.2

[georgeb]

#4 -- o.k., i'll try your advice on that one, but I still think it is dumb that you can add an applet so easily, but can't take it away via the same method.

I expect some major improvements of audio drivers in post-2.5 kernels. ALSA seems to be a bit more versatile than OSS. OTOH, I haven't tried an ALSA install on 2.4 for some time. If you don't mind about non-RPM installations, you may be fine (and have some minor headache when upgrading kernels).

I think it's an AC97 onboard soundcard that's not supported by OSS (the 2.4 kernels). And it will take some time until 2.6 that's for sure... ;)

Re:My experience with Dead Rat Linux 7.2

[Sj0]

#1 I actually did use the default one (I forgot whether it was GRUB or LILO).........but when I'd try to log in.........no Windows (or DOS) option. Only Linux.
That's wierd.

#4 -- o.k., i'll try your advice on that one, but I still think it is dumb that you can add an applet so easily, but can't take it away via the same method.
I just installed 7.3 last night, and now that feature exists in both KDE and Gnome...

#5 -- you are lucky to have your intuition, because the little two books they give seriously aren't worth a damn. It seems thrown together by like one guy. I resorted to buying a book (Running Linux, o'reilly) which has helped.

To be honest, for something as big as Linux, it's probably best that you just go out and buy a book on the subject anyway.

#6 -- well, I hope you are right, and the 7.3 kernel will support my (very common) soundcard. But won't I still need to re-compile, and answer lots of obscure hardware questions? Not for the faint of heart.

Nope, no re-compile, no questions, nothing. The installer will take care of it, and if you upgrade the kernel seperately, Kudzu will take care of it. Driver support in Linux (and more importantly, the ability of the end user to install drivers without knowing about the underlying OS) is getting really good.

#7 -- with the "lokkit" thing, there are only like 3 options, low-medium-high (and then a short custom list), so I still don't think it was my fault there. (also found a couple of other similar testimonies on the Net) My networking was working okay till I messed with that damn thing.

Again, I'll have to take your word for it.

so all in all, not the best experience. I'm leaning toward trying Mandrake now, but I damn sure won't pay for it. I'm a fool for having done so with Red Hat.
Live and learn, I guess...

oh, one more aside before I bore you to tears -- when I went to register for my "support" (which only covers basic installation questions - no soundcard stuff allowed) - their site said my product ID did not exist. Imagine my anger; already I was pissed off and frustrated! To their credit, they responded promptly to my indignant e-mail with a working number. Alas, by then my mind had been made up.

My bad. I never used their tech support, so I didn't know how it was.

there is one thing all this has made me realize, though. "Free" software like Linux still has an amateurish, non-integrated feel to it. Windows 98 seems fairly tight by comparison; it even plays my mp3's! (I still despise it of course.) Maybe OS-X?

Er...We are in disagreement there. I am finding that over time, Linux as a whole becomes more professional, and at this point suprises me sometimes by what it can do. Even the utilities it comes with are getting better, to the point where I can use a machine without having a terminal open 24/7.

P.S. I'm not flaming you, I'm trying to help -- but after a hard day as a computer tech, my diplomacy is all gone. :)

Kernel hacks, kjournald

[jovlinger]

I have had exactly the opposite experience. RH 7.2 (2.5.7-10 IIRC) with ext3 is snappy and responsive, even under the heavy cpu and i/o load of a background kernel compile.

But I wanted ALSA, so I grabbed 2.4.18, and installed that. It is absoultely HORRIBLE. With any sort of i/o in the background, the mouse is laggy, GUI latency can be measured in large fractions (and sometimes numbers of) seconds. top This is for exactly the same setup otherwise. Low latency patches don't help much. maybe a little, but it still is unacceptably laggy. Forget xmms + pan (one program to feed the other :-)) at the same time.

So there are two possibilities:
1) fsked up my 2.4.18 config, and thus ended up compiling a really crappy kernel. But I've been compiling kernels since 1.2.13, and have yet to have one behave anywhere NEAR this badly.
2) RH have significantly hacked 2.4.7 to make it useful. Does anyone know whether the same hacks have happened for the 7.3 kernel?

Thoughts?

Re:Kernel hacks, kjournald

[jovlinger]

Oh just to add:

It appears to be orthogonal to memory pressure, but related to i/o. Running top suggests that kjournald is the culprit, as it both sucks up a (comparatively) large fraction of cpu time and % (low single digits under i/o load), in addition to bing near the top of the runq when UI lag sucks most.

Re:Kernel hacks, kjournald

[greenrd]

Another reason to use reiserfs rather than the backward-compatible, conservative hack ext3fs.

Re:Kernel hacks, kjournald

[bero-rh]

First of all, don't use 2.4.7-anything.
It has some major problems including a remote root exploit. Please upgrade to either the 7.2 errata kernel, 2.4.9-something, which fixes all known security problems, or the 7.3 kernel.

So there are two possibilities:
1) fsked up my 2.4.18 config, and thus ended up compiling a really crappy kernel. But I've been compiling kernels since 1.2.13, and have yet to have one behave anywhere NEAR this badly.
2) RH have significantly hacked 2.4.7 to make it useful. Does anyone know whether the same hacks have happened for the 7.3 kernel?

2, and possibly 1 as well.

Red Hat kernels are always patched quite a bit to make them more stable/usable, but 2.4.18 doesn't look THAT bad for me (maybe related to different hardware or different setups).

Since kjournald appears to be the culprit, the Red Hat version of 2.4.18 is likely to fix the problem because it uses a newer version of ext3 and everything related to it.

Re:Kernel hacks, kjournald

[jovlinger]

Thanks.

A net search indicates that I may be the only person to get crappy ext3 / i/o performace under 2.4.18, which suggests I'm incompetent to compile my own kernel, and should just use someone-else's.

So I guess I'll be experiencing the excitement of a semi-backed-up (my mp3s are living on borrowed time) upgrade over the weekend. woo hoo!

Re:Kernel hacks, kjournald

[Anonymous Coward]

A net search indicates that I may be the only person to get crappy ext3 / i/o performace under 2.4.18, which suggests I'm incompetent to compile my own kernel, and should just use someone-else's.


not true, happens to me too. 2.4.17, same symptoms - low latency patches don't help much, thought it was my own fault, etc. in fact the performance is worse than win2k!! I guess everyone else is too embarassed to say anything.

speaking of embarassing, I couldn't get rpm to install KDE 3.0. so I will have to try RH 7.3

-- Joe Luser

Because it's bigger! :D

[N0Nick]

More software is added to distributions, and software is generally getting larger.

Re:Because it's bigger! :D

[SimplexO]

I realize that in general, new stuff == larger files, but I how does a minor update constitue an extra 500MB CD?

ah, sigh...

[darkonc]

I've been trying for the last hour to connect to my favorite mirror for a couple of RPMs that I need, and it keeps coming back full... but that's never happened in recent memory. My favorite server's almost always completely unloaded (that's why it's my favorite!). Then I think: "This feels like a slashdot. I wonder if the 7.3 release has just been announced?".

Sure enough.......
Maybe I'll get my RPMs tomorrow (or tonight at 4am).

RedHat and licenses

[Lac]

This is really weird. Browsing the left navbar on the release notes, I just discovered that Red Hat Linux comes with a fancy EULA. Yes, the type that you are assumed to agree to by installing the product. No, I am not making this up. Read for yourself.

http://www.redhat.com/licenses/rhl_standard_us.htm l

Now, I have not read the EULA. Perhaps there is nothing sinister here. They probably have a very valid reason for doing this. I suppose the lawyers insist on it. But I had never heard of this practice and I doubt many here have. To me, it seems inconsistent with RedHat's reasonable, pro-open source, transparent attitude.

I mean, the document says that this applied to 7.2 as well. I installed 7.2 on quite a few machines and never once did the installation program warn me that I was simultaneously "signing" a legal document. Even if that EULA really is harmless, I should be told about it.

PS: I love Red Hat, bot for their product and their attitude. It is an amazing company, but this comes as a shock to me. And the more I think about it, the weirder I feel about it. Which probably explains why I have been editing this post for 15 minutes now.

Re:RedHat and licenses

[Lac]

Sorry for replying to myself this quickly. But you really have to look at this document. Look at the first part, that talks about the auto-signing "feature" of the EULA.

By installing any or all of the software included with this product, you agree to the following terms and conditions.

Now, I really love and respect Red Hat. I haven't lost any respect for them since they started making it big and I am sure that this is just a big, big mistake. Really./P>

But think about what they are saying! If I use the SRPM to compile and install the source code for kernel, or Emacs, or any GNU tools, I am agreeing to this? Has someone gone insane at RedHat? That software is covered under the GNU GPL. Slapping a EULA on it isn't even remotely legal!

Putting a EULA on a distribution is one thing. Putting one on "any or all of the software included" with it is another entirely. I hope this is just a big mistake, or that I just misunderstand this whole issue and am blowing it all out of proportion. Anyone at RedHat care to comment?

Re:RedHat and licenses

[Lac]

Replying to myself again. This is really a weird document. They have this little bit of sanity in there, which voids part of what I have said.

Most of the Linux Programs are licensed pursuant to a Linux EULA that permits you to copy, modify, and redistribute the software, in both source code and binary code forms. However, you must review the on-line documentation that accompanies each of the Linux Programs included in this product for the applicable Linux EULA. Review these Linux EULAs carefully, in order to understand your rights under them and to realize the maximum benefits available to you with Red Hat Linux. Nothing in this license agreement limits your rights under, or grants you rights that supercede, the terms of any applicable Linux EULA.

But then they have this part again.

CAREFULLY READ THE FOLLOWING TERMS AND CONDITIONS BEFORE INSTALLING ANY OF THE SOFTWARE PROGRAMS. INSTALLING THE SOFTWARE PROGRAMS INDICATES YOUR ACCEPTANCE TO THE TERMS AND CONDITIONS SET FORTH IN THIS DOCUMENT AND OF THE END USER LICENSE AGREEMENT ASSOCIATED WITH THE SOFTWARE PROGRAM. IF YOU DO NOT AGREE WITH THESE TERMS AND CONDITIONS, DO NOT INSTALL THE SOFTWARE PROGRAMS.

So it is not clear to me that this EULA is not adding restrictions on top of all software packages includet with RHL, including the GPL ones. One particular problem is that you have to agree to this license before you can install the software and review individual licenses to see which rights you have.

Perhaps I'm freaking out needlessly here. I just find it _so_ weird that Red Hat has a EULA. Oh well. I'll get over it. It's not like I'm a license freak or anything.

Re:RedHat and licenses

[ainsoph]

I like Redhat. A bunch of years back I got paranoid about that stuff, similar stuff with the company and stopped using it. I went back to Redhat for 7.2, and I actually feel really good about them. You know why? Cos I trust Alan Cox. He works for them and as soon as they did some shit that was questionable, he would be the first to jump ship. I trust that.

Apache

[smittyoneeach]

Curious why they haven't gone to 2.0.x for httpd...

Re:Apache

[Garfunkel]

likely because things like PHP still don't work with Apache 2.0.x

ed2k?

[Anonymous Coward]

ed2k links anybody?

Can I upgrade from Beta 1?

[KidSock]

I installed Beta 1. Can I safely run the installer over that? Or perhaps I can just to the uptodate business?

Works with Ximian? (wasRe:Old version of Mozilla?)

[morris57]

Bero, hopefully you will see this and be able to answer...(I always appreciate seeing your answers whenever a new Red Hat release comes around.)

Can users of Ximian Gnome desktop upgrade to RH7.3? I've been keeping up to date with Red Carpet, so I didn't upgrade from 7.1 to 7.2, but I'd like to try Valhalla. Any known problems between Red Hat's packages and Ximian's?

Thanks!

KRUD is the answer.

[Duke]

If you do not have the bandwidth, maybe a couple of filled-to-the-brim CDs every month is the answer.
http://www.tummy.com/krud
The April CDs contain about 1.4GB.

Re:KRUD is the answer.

[Anonymous Coward]

Is KRUD just a RedHat for modem users, or does it have any other advantages besides saving bandwidth?

Re:Works with Ximian? (wasRe:Old version of Mozill

[bero-rh]

Remove Ximian first. They're playing the rpm Epoch game, so installing their packages breaks updates unless they are removed prior to updating.

Happy to see x.3 (a success story)

[Yort]

I, for one, am quite happy they decided to do 7.3 instead of 8.0. I work for SGI, and while we have plenty of things we'd rather not brag about, one that is almost universally touted as an advantage by customers is that we've consistently come out with updates to our 6.5 OS (on a quarterly basis) for the last *four* *years*. We've introduced tons of new features and support for brand new architectures, but always kept binary compatibility.

Of course, you do have the "Thisdistro 8.1 must be better than Thatdistro 7.3" problem, but I think Red Hat is probably big enough now that the backwards compatibility for their current customer base is a bigger plus for them.

Re:Works with Ximian? (wasRe:Old version of Mozill

[morris57]

Thanks for the quick answer!

Alternate gcc

[Compenguin]

I don't see the alternate gcc3 packages, 7.2 shipped with gcc-2.96 as the pramiry compiler and gcc3 packages as a secondary compiler, why did they kill gcc3?

Irrelevant ISOs?

[Anonymous Coward]

Am I the only one who feels ISO releases are becoming irrelevant? Or maybe I've learned Linux enough to build my own packages?

Anyway, looking at the list of 7.3's features I realize most of my 7.2 packages are newer than than the ones that come with 7.3:
Evolution - 1.0.3, I've golatest
OpenOffice 1.03, the latest (that's on my Deb notebook, true)
Mozilla - 1.0 RC1
.. and so on.

One reason why I like that this ISO is out is that I want to build my own version of 7.3 (get rid of useless packages, change some .conf files, change stupid Mozilla-search-Netscape.com-in-sidebar prefs and like) so that I can have a good and consistent installs without using Kickstart.

DumbFuk

dead already...

[bani]

nicely 404-compliant

ATI Radeon 8500 All in Wonder Support?

[Edward Teach]

Downloading it now. I hope that it supports my AIW 8500. Haven't had any luck gettin X up and running with 7.2.

Junkbuster deprecated?

[deepestblue]

According to the release-notes, junkbuster is deprecated. What is the alternative filtering proxy that RedHat supports?

Re:Junkbuster deprecated?

[ChrisWong]

Junkbuster does not support HTTP/1.1, so a lot of browsers have problems with it. There is no ad-blocking proxy "supported" by RH as far as I know. The apparent successor to Junkbuster is Privoxy: it's not yet officially supported but it seems to usable now.

Whats NOT new

[JeffMings]

A unified CLI-based administration tool seems to be missing. People bitch and complain about LinuxConf, but it really does make administration via SSH easy, and it is MISSING from 7.3, with no apparent replacement! Bero, if you are reading this, perhaps you could explain the intended replacement or alternative? A LOT of admins will find a shortage of CLI tools to be a big problem.
Thanks for listening to my rant/question.

Re:Whats NOT new

[referee]

It's called VIM.

Re:Whats NOT new

[qurk]

Hmm. I'd say the removal of linux conf is a good thing. In fact I think that is the clincher to get me to turn all my 7.2 puters into 7.3 on that fact alone. Linuxconf sucks. O.K. I concede it could be a good program - IF IT WORKED. You change stuff in linux conf and then you go edit the text files anyways :P Also the interface for linux conf is so ungodly ugly and awful, I mean if it's interface for changing stuff was anything like mainstream or even logical then maybe my opinion would be different, but as it is I think linux conf bites the big one and it annoyed me when it got installed simply by choosing everthing in the redhat installer. I definately applaud the decision to get rid of that peice of crap.

Now I wouldn't be against a unifed configuration program that worked or made sense but linux conf just sucks ass.

Use the upgrade feature

[Wee]

I installed and have throughly updated Redhat 7.2. Now, 7.3 comes out.

Sounds like a great time to see if the upgrade install feature works. Since you only recently installed, there can't be much custom (non-RPM I suppose) stuff to break. If the upgrade goes well, nothing more to do except remember that when RH 8.x comes out, you can sdafely upgrade instead of re-install. If the upgrade fails, then you can submit a bug report and install freshly. Either way you have a 7.3 install.

-B

Re:Use the upgrade feature

[rosewood]

Since this does not have Gnome 2.0 or anything big, and I just got done installing Ximian - Im just gonna stick with 7.2

Also, Ive always been early adopters of the . releases and stuff is always broken

Re:Use the upgrade feature

[Anonymous Coward]

ffia, gnome 2.0 will be included in redhat 7.4, unless glib 2.3 comes out before gnome 2, in which case it'll be in redhat 8.0. it wasn't included in 7.3 because it's not ready yet to be shipped in a commercial distro.

This release does have something big tho: kde 3.

i'm not dead yet ; )

[gimpboy]

i just had to shut it down for a bit to set the max connections limit for that directory.

Question about 7.2 support

[guacamole]

My understanding is that redhat's current support statement looks like this "We will support all minor releases of the current major RedHat Linux release and the latest version of the previous major release." That means that right now 6.2, 7.0, 7.1, 7.2, and 7.3 are supported but once 8.0 is released, only 8.0 and 7.3 will be supported. If this is true I will be scrwed big time because I just finished upgrading some 50 6.2 boxen to 7.2 and I was thinking that since 7.2 is the latest (and the last) minor release of 7.x series, I can keep on running those machines for a couple of years without upgrading. Now, if RH stops providing updates for 7.2 after 8.0, we'll be screwed big time. Can someone confirm or deny my fears?

Re:Question about 7.2 support

[charnerd]

Why would you be screwed? Just migrate to 7.3. It's not that hard. You can even hire me to help you. Good luck!

Wishlist for 8.0

[pyrotic]

7.3 has a few things missing, and I'm still dreaming of the day RH ship a distro which does everything I need out of the box.

This release comes without linuxconf. While linuxconf did have its faults, and apacheconf, bindconf et al look very nice under xwindows, I need something I can admin remotely with a modem over ssh. While I do apache and bind with a text editor, it's nice to have a gui for admining user accounts, rc scripts etc.

Apache 2.0 came too late to QC in this release. I'm half tempted to build my own, with PHP, mod_bandwidth etc, but I'd rather stay in sync with RH. Multithreaded webserver looks really handy for persistent pooled SQL connections.

Analog is something my customers always like for looking at http logs. It used to be on the powertools disk. Remember powertools?

At least we now have a system with posfix available in the default install. Guess I will install 7.3, and do a kickstart script to grab stuff that's missing.

imlib and enlightenment

[AstroJetson]

Careful anybody out there running enlightenment. I upgraded (via up2date) my imlib a few weeks ago and suddenly enlightenment would not start up. When e started up it would get to 91% and just die. Backing up to the previous imlib rpm fixed the problem. Don't know if this problem still exists in 7.3, but I'd be careful if I were you.

Re:imlib and enlightenment

[Anonymous Coward]

e16 is not officially supported by redhat anymore

Well, just like it always has

[Nailer]

The current KDE3 RPMs for RH 7.2 from Red Hat have their own glitches: ksplash goes kblooie at startup, and konqueror seems to have this big memory leak that bloats its footprint over time.

I've installed them on six different red Hat 7.2 machines and not encountered this, and so have most Red Hat users I've spoke to who installed KDE on Red Hat 7.2. Personally, I just duped Red Hat's packages in a local apt repository and then apt-get installed kdebase. I had to uninstall some minor stuff (like switchdesk-kde) that wasn't yet ported to KDE 3, but other than that, everything went AOK. There appears to be something amiss with your system - try visiting irc.openprojects.net #redhat for a hand.

Re:Prez Sez:

[Anonymous Coward]

Yes, that moron is President of the United States, while you are an obese hippy dork sitting in your mother's basement and grovelling for Hot-seet-tits-21-xxx@aol.com to pay attention to you in your chat room.

Yea!

[orpheus2000]

I get to keep my RHCE current for six more months! :-)

reiserfs ??

[Anonymous Coward]

When will anaconda include this fs? Is RH waiting for reiserfs ver 4 ??

Low latency kernel ?

[bicatu]

Hi, I was wondering if the kernel included already has the low latency patch applied. If not can we get it from rawhide ?

what, no NonStop Kernel systems?

[Anonymous Coward]

disappointing there are no NSK systems offered!

happy HPQ to you!

Not exactly a minor update

[N0Nick]

Skipjack (7.3) is not exactly a minor update;
Except for newer versions of software, it comes with many new programs that weren't in Enigma (7.2): KDE 3.0, Evolution, Gnome Video Conference, etc., as you can see in the "New Features" docs.

Why no Tramp?

[wray]

Perhaps Tramp is just not used that much, but I noticed that for RedHat 7.3 XEmacs has the Tramp, package, but GNU Emacs 21, does not. Does anyone know why this would be, or is this just an oversight?

I personally would really like to have this capability there by default, not that it is /that/ hard to install yourself, but why shouldn't I expect consistency :-)

As an aside, perhaps someone could post a link to the definitive URL for posting these types of suggestions to RedHat -- is it just bugzilla?

Thanks.

Red Hat release names Enigma Skipjack Valhalla ...

[gnugnugnu]

Valhalla may be the home of the Norse Gods, where warriors slain in battle go when they die ...

... but what i really want to know is what is the connection between SkipJack and Valhalla.

Enigma (7.1) and Skipjack(7.2) were both encryption algorithms

For explanations of the older release names:
http://freshrpms.net/redhat.html?links=on
http://www.smoogespace.com/documents/behind_the_n a mes.html

(why cant slashcode automatically convert all valid URLs into clickable hyperlinks eh?)

Bero returns another volley.

[Queuetue]

Like most days with a major RedHat announcement, Bero is here in force, defudding like a peruvian congressman. Good to see it!

Bero, is /. ambassador an official role for you at RedHat, or do you just do it because it needs to be done?

CLI remote administration

[LinuxHam]

Webmin still works very hard to be very usable under Lynx. Just ssh into the box and point Lynx to localhost:10000. You can firewall off port 10000 if you don't want to admin from a remote browser. Works great for me.

LVM

[DJPenguin]

Anyone know if LVM is in the kernel? This would be the only reason i would have to cook my own kernel...

Oh yeah and what about devfs?

Cheers!

You need more sleep

[ChrisWong]

I think you need more sleep. Do you even realized what you typed?

• Kdehelp back button not working: you reported this bug! You filed it as bugs.kde.org: "Bug#31123: Back button in khelpcenter has no effect". There's no denying the bug reported by someone as dependable as your honorable self.
  
• Bad version of Qt: the KDE 2.2.2 release page explicitly tells you not to use Qt 2.3.2. Guess which version of Qt you decided to use with the Red Hat RPMs? Yup: qt 2.3.2 and it's an RPM dependency. The frozen desktop bug is documented at bugs.kde.org as bug 34949 ("system freeze on KDE startup").
  

So: get some sleep and write again when your fingers are resynced with reality. I want my KDE to work. "Not reproducible here" does not sound like a good excuse at this point.

Redhat 7.3 trouble

[Tersevs]

Okay, I totally expect that this post will be considered trolling. But anyway...

In the 5.0 days i ran RedHat, but then i migrated to Mandrake. I thought that it was time to try RH again, so i neglected upgrading to Mandrake 8.2 and waited for RH 7.3... Now... Im kinda disappointed. Its seems darn buggy.

First: Whenever i close a KDE application (Konsole or Konqueror or whatyever), The KDE Crash Handeller pops up.

Second: Mouse scrollwheel support? I've tried all kind of drivers (and even migrated my old X-windows configuration file.

Third: NVidia driver. The src.rpm does compile but there seem to be some issued with the mesa library that causes it to crash.

Am I the only one with a Geforce256MX card and a logitech wheelmouse running KDE?

Wonder why there is no redhat 7.3 faq on redhat.com yet... Isnt the mandrake site more informative towards the community?

Now i havnt giving up on RH7.3 yet, but i AM downloading Mdk 8.2 ISO:s while writing this.

Is FTP or HTTP better?

[LinuxHam]

Speaking of which..

I always thought HTTP traffic was much easier on servers than FTP traffic, yet 95% of the mirrors are FTP servers. Am I correct in that thinking? Maybe the thought comes from the relative apparent ease of load balancing web traffic vs FTP traffic.

And as far as maxing out the connection on just one download, many mirror-oriented servers limit transfer rates. I often find that multiple downloads each get the same rate as just one download from the same server.

Re:Kernel hacks, kjournald (use hdparm)

[draziw]

Take a look at how your disk/interface is setup.
hdparm -Tt /dev/hd[a-z] will do a read test and give you a base line. then try:
hdparm /dev/hd[a-z] to check your settings.
You are probably not running the fastest settings by default. Here is what I'm using on most of my systems: multicount=16, I/O support=3 (32-bit w/sync), unmaskirq=1, using_dma=1, keepsettings=1 (just keeps the other settings if there is a reset, etc on the drive), readahead=8. After changing those, and checking for stability, try the hdparm -Tt again, and see if you get better numbers. "man hdparm" has a lot more info, and /etc/sysconfig/harddisks has a little. You can then configure to use those settings on each startup by editing /etc/sysconfig/harddisks. You can also run diff settings for each drive in your box.
Regards,
Ryan

Re:Kernel hacks, kjournald (use hdparm)

[jovlinger]

Thanks for the info, but I get much better performance with the RH kernel than my own -- ALL else being equal, including hdparms.

Or am I misunderstanding something here? Are you suggesting that hdparm tweaks are part of the kernel?

Re:what, no Neue Slowenische Kunst systems?

[Anonymous Coward]

Leben Heißt Leben!

hehe... Appologies...

[Lethyos]

Couldn't resist. :)

OpenOffice 1.0

[Anonymous Coward]

First of all thanks for your info, but Openoffice 1.0 it is still available into Mandrake, so lets go for it!!!