Put your RedHat 7.2 Rants here!
 

I've used Redhat 7.0 since it came out and 6.2 before that. If Emacs is the best piece of software I've ever seen, the RedHat 7.2 package (kernel 2.4.7-10) is the worst.

There are three problems:

1. Installer hangs.

If you 'check for bad blocks' when you format an ext3 partition, the installer will barf on you. Go on, try it :p . If you do the whole process quickly you can make it to the rpm installations before it crashes - but only if you wave a dead chicken over the computer while you install.
You'll still probably be dropped to a shell when you boot into your newly installed system.

2. Ext3 not implemented properly.

If the system unmounts cleanly when you call shutdown, why would it tell you there are errors on the filesystem when you reboot ? Init scripts reading a corrupted file flag might explain it ... ;-(

3. X Window freezes

Sooner or later this could happen, but the mean time to failure is about five minutes for RH7.2. If you're using a good distro, you might never have seen this.

4. FAT32 back in alpha!

Between RH7.1 and 7.2, someone at redhat.com appears to have lost code listings for FAT32 support (behind the filing cabinet?) as it's been relegated to alpha from full support in 7.0 and 7.1!! :p

5. Too much coupling between kernel modules

Recompiling the kernel in 7.0 or 7.1 is fine. For 7.2 I don't recommend it. On the second boot for one install, the system forgot where it put the keymap and dropped to a shell ?!?! OK, for something important like ext2fs support, I could understand this but for a keymap? ...

6. X server startup fails

On average after the second boot, usually because a combination of the above corrupts your file system making your system senile.

I am in the process of getting a refund and downgrading to 7.0. I'm doing this because previous distributions of redhat work. Which is always a bonus when you buy an OS.

Bert

edit:

That's more than three, I know. Once I started they just kept coming .. :rolleyes:

I haven't a clue what you are talking about. I have redhat 7.2 running on a POS e-machine. It runs for days just fine. I reboot and have no errors. I do know that the first time in installed it I had a bad burn and therefore it didn't work and rpesented many errors that you described. I have another friend that has it up and running on a 200mhz K6 with like 56mbs of RAM. He had the same problems with a set of CDs. However when he got a second set it all seemed to work great. He runs gnome and I run kde. So I think you should try a second set of CDs and see if that is the problem before you get high and mighty.
enjoy

It's a boxed distribution, and I was comparing it to previous distributions. Nothing high or mighty about that, they're simple observations.

Part of the problem is the kernel version, which I was careful to post as well. This is perhaps the wrong kernel version to be distributing with a redhat release, as (for example) the FAT32 support is not properly implemented.

Anyway, it was a rant! :D I've decided to post this sorry tale only after unsuccessfully reinstalling it for most of this week.

I'm getting round it by installing 2.4.8 from a free CD!

Bert

PS. If RedHat are distributing badly burned CDs, then they really are giving Linux a bad name.:tisk:

Redhat 7.2's setup and running problems are known entity too.

Bert

I believe that if you don't have an answer to the problem that you are bitching about, then you shouldn't bitch about it. Where's your distro at Bert?

Hey, don't start any controversies here. Bert is expressing his opinions on his experience with RH7.2.

-trickykid

I only have one minor rant about RH7.2:

You had to read right to the end of the docs given at install-time to find out about the append = "noathlon" bit. Seems a little unnecessary to me.

I had Bert's Problem #3 X-Win random freezes , never could figure out why. Therefore I switched to slack8, now everything is fine.

I've got a different complaint. 2.4.7 has basically the same NTFS support as 2.4.5. However, RH didn't hand out an NTFS pre-compiled module with 7.2 and yet Slackware 8.0 does.

Meanwhile RH makes ext3 an option, a filesystem that still has the word EXPERIMENTAL and a lot of bangs (!) after it in the menuconfig for 2.4.18!

I just don't understand all of this bleeding-edge-ware in a certain case, but none of it in others... Is this a shunning of windows compatability? or is this an issue of ext3 and vfat or NTFS being mounted at the same time so RH makes it a pain to do instead of slapping a big fat warning label on the box?

Also, and this is a personal gripe, ever since the switch to in-kernel pcmcia, its been the problem of the distros to properly build the contents of the /etc/pcmcia directory (*.opts, *.conf). RH seems to be using Hind's defaults from old pcmcia-cs, which are so behind on .conf structures that you have to download or create a .conf file for half the wireless cards on the market just to get cardmgr to bind modules that RedHat even compiled FOR YOU! Otherwise they just sit there in /lib/modules like dumb chickens. Admitedly Mandy 8.2 even has this problem, but at least they seem to have started tailoring /etc/pcmcia. Meanwhile I think (I haven't tried, but I've gotten the idea from postings and google), that Suse has their stuff sorted and will actually load the modules it has for the job, so its not like this is endemic to every distro, and the oversight is getting kinda glaring.

To approach the previous rants:

1. Had this happen once of 3, but on a laptop that didn't meet minimum specs, so I called it my fault.
2. Rh being Bleeding-ware alpha goobers, remember the gcc debacle of 7.0?
3. Don't be so quick as to blame X, Gnome is a much more likely candidate. Both of the 7.2 machines I have run KDE and haven't had this happen yet.
4. Never noticed, compiled my own kernel before untaring my archive of backups. I recomend doing the same. Actually, I was surprised at how easy it was on the usually grumpy RedHat.
5. See above, I never had a kernel compile right with 7.0. Fresh kernel source? or the mangled goop that ships with RH?
6. I boot to runlevel 3, always... keeps X from messing with my life if I don't want it to. I don't know anything about refunds... the only distro I've ever paid for was 7.0.

Solidarity,

Finegan

That's a valid point. The solution is to use the GUI (if you like Nautilus! ) and upgrade your kernel as soon as you can get in.

I really don't like this distribution now, and I feel it's irresponsible for redhat to be issuing with an unstable kernel - and don't get me wrong I was a die hard fan of redhat. The issue is that we know how to handle these problems but a proper noob won't.

Bert

Small rant. :-)

I buy RedHat.

That's right, I plop down my money at Redhat.com and buy their distro. I've bought 6.0, 7.0, and 7.2, so I'm not just downloading their stuff for free. We also use RedHat on many of our servers and even employee workstations where I work, so please don't think that I'm here to say "Redhat sucks" in any way.

After installing 7.2, I noticed that about every time updatedb ran via cron, the system would lock up. When I posted a support request for the issue, I was told, "Don't run updatedb." ~sigh~ I have installed every update so far, but I still get these lockups. It's maddening.

It would just be nice if there were a little more support available, at least for paying customers. Over USD 200 for a support ticket is way too costly for little old me. It just seems like there has to be some way (email support? Text chat? Newsgroups?) where Redhat could provide good support at a lower price which would still be cost effective for them and for the customers.

Don't get me wrong: I like Redhat. I'm just more and more disappointed with RedHat's support is all.

Redhat
 

I have used Redhat since 5.2 and believe me they have came a long and
positive way.
I have had problem with freezing and very slow loading of gnome or KDE
desktop , and I have found by going into /etc/sysconfig/harddisks and
removing the # sign in front of USE_DMA=1 it sure makes thing work a lot better.

J. Tate