[plug] Kernelnn

[John Summerfield]

On Wed, 31 Mar 1999, Greg Mildenhall wrote:

> On Wed, 31 Mar 1999, John Summerfield wrote:
> > The 2.2 kernels are rather new and unproven. 
> I'd say for most hardware they are as stable or better than 2.0 these
> days. While I'd be inclined to wait before putting it on a server, I see
> no reason to settle for 2.0 on a desktop machine.
> 
> > I'm running 2.2.x because I need one of the new features, 
> > but the system was more trouble-free on 2.0.3x.
> What has 2.2 been doing to you?

Worst, on the system I needed it, I had three EIDE drives plus a CD drive.
hda: QUANTUM FIREBALL ST3.2A, ATA DISK drive
hdb: QUANTUM FIREBALL_TM2110A, ATA DISK drive
hdc: Maxtor 88400D8, ATA DISK drive
hdd: ATAPI 24X CD-ROM drive, 128kB Cache

On the two or three occasions I accessed hdd (except for playing music -
that was fine) the system locked up cold.

CAD was ineffective.
The LRB sort-of rebooted, but the system immediately hung with timeouts on
hdc. Gave me a nasty fright wrt the nice newish drive.

The system had to have its power cycled to restore control.

I think the problem's fixed: - I'm not keen to try, and as another machine
on the LAN also has a CDD I don't feel the need.

Since then problem though, I've shuffled drives and the other machine now
has the 8 Gb drive and my workstation has a 4.3 Gb Fujitsu drive.
Additionally, the drives have been shuffled to this:
hda: QUANTUM FIREBALL ST3.2A, 3079MB w/81kB Cache, CHS=782/128/63, UDMA
hdb: FUJITSU MPB3043ATU E, 4125MB w/0kB Cache, CHS=525/255/63, UDMA
hdc: QUANTUM FIREBALL_TM2110A, 2014MB w/76kB Cache, CHS=4092/16/63
hdd: ATAPI 24X CD-ROM drive, 128kB Cache


Apart from that, I've found a couple of errors in Documention/Changes
(corrected as at 2.2.4) and the yp-series seems to have got temperamental.
I plan to recompile them with the 2.2.x kernels in place and see if that
makes a difference.

The new support was a bit patchy too. Had problems swapping between 230 Mb
(512-byte sectors) and 640 M (2048-byte sectors) on my 
scsi : 1 host.
  Vendor: FUJITSU   Model: M2513A            Rev: 1200

It transpired that it works if I press the button, but not if I use Jeff
Tranter's eject 2.0. Alan Cox has a M2513A too, so I left him 7 Jeff to
sort it out.

It seems though that the kernel (or at least that bit of it) wasn't
reporting all errors.

Oh, I had to get a newer fdisk and mkfs than mentioned in Documention/Changes


Oh, I wasn't amused with the ipchains rpm either. It removed ipfwadm which
made it harder to revert to 2.0.3x if I wanted.

It would have been better to rename the ipfwadm command and adjust that
nice wrapper to call the real ipfwadm for a 2.0 kernel, do it the new way
for 2.2.


However, I have put the 2.2.4 kernel on a system where it wasn't needed.
That's when I found the problem with ipchains.




-- 
Cheers
John Summerfield
http://os2.ami.com.au/os2/ for OS/2 support.
Configuration, networking, combined IBM ftpsites index.