Joseph Geiser wrote:
>
> Jeff Kell tells a great HP3000 Success Story (Alfredo --- did you take
> note?)
>
> > What is wrong with this picture?
>
> You told about how robust the HP3000 is, and how downright crummy the
> HP9000 can be, and when one adds Slow-racle, everyone run for cover!

Not just Slow-racle, but Slow-racle Forms/Designer, Slow-racle
Application Web Server, and all sorts of other nasty ickky thingys.  Bad
enough that you have to set your shared memory pool to ONE GIGABYTE to
make the instance start up in the first place, you get to tweak all
sorts of kernel parameters to make it happy (or as happy as it gets).
Richard can tell you all sorts of tales of woe, he has graciously picked
up the Oracle parts and run with it while I suffer with HPUX (when not
occupied with the network, but that's another story).

I just found it ironic that the 3000 swap/installation/production
rollout was uneventful at worst but spectacular at best, while the 9000
is still in pre-production, taking on hundreds of patches, doing little
useful work at this point.  And I'm blown away by the fact that we were
supporting 200+ users on the 969 without even breaking a sweat, and 90%
of those users coming in via host-based telnet (which was predicted to
be a resource hog, but I continue to be pleasantly surprised.  If I ever
run into Jeff Bandle, I owe him dinner :-) ).

> Great story Jeff!

Well thanks, but that's only true if you're on the listening end of the
conversation.  It's not so great being on this end, with management
pushing you toward the brink...

Jeff Kell <[log in to unmask]>