Re: Kernel OOPS with 1.2.11

lilo (TaRDiS@mail.utexas.edu)
Sun, 30 Jul 1995 21:32:36 -0500 (CDT)


On Sun, 30 Jul 1995, Chris Woods wrote:

> On Sun, 30 Jul 1995, lilo wrote:
>
> > You should investigate 1.3.13. It looks very solid. YMMV, of course. Um,
> > if you're getting rid of all your NE2k's, maybe you could send them to me?
> > Mine work fine.... :) :) :)
>
> I'm in a position (unfortunate as it may be, I used to enjoy having the
> latest kernel *first* ;-) ) of having to be over-cautious. 1.2.10 works
> fine, is rock-solid for me in all 11 machines. If/when I see a need to try
> a 1.3.x kernel (probably not until 1.4.x is released, then I use those) I
> will definitely investigate.

*Nod*. But I suggested 1.3.13 realizing you needed something very solid.
Remember that, no matter how much we try, a production kernel is not always
inherently more solid that a test one--especially since Linux test kernels
that don't involve radical changes are often rock-solid. I always evaluate
kernel releases on a case-by-case basis; I put up 1.3.12 because it looked
*very* stable (and still seems so) and because I needed source route packets
dropped and a few other features that appear to work quite well at this
release (such as masquerade).

Anyway, just as you wouldn't put up a new production release without being
fairly sure of its stability, I wouldn't rule out test releases without
getting feedback on *their* stability. Anyway, as always, YMMV....

lilo :)