Hello Michael and all fellow Listers !
I'm afraid that, by attempting to share some past experience, I spread a
little confusion to what my (and I really mean MY) position toward Patch/iX
and Stage/iX is.
First, let me say that I want these tools to continue to improve, and I am
highly confident they will. My very first experience dates back to the beta
test of 5.5, spring'96. At that time, about 20% of patches were not
Stage/iX-compatible, for some reason. From what I've seen more recently on
a few sites, this proportion has significantly decreased. Once again, just
my own perception. Patch/iX and Stage/iX may have a few bugs here and
there, but I do not remember ever killing a system or some other
catastrophe as a consequence of using these tools, even during the beta
test, when it was commonly accepted that these products could be less than
stable.
Last time I ordered patches, I received a tape with Autopat and
Autopat-format patches, i.e. patches installed perfectly, but they were
missing a few files that makes them Patch/iX-installable. RCEs on this list
must understand what I'm talking about. This is not, I repeat NOT, a
criticism toward Patch/iX, which I think is an excellent tool, but more a
remark toward local HPRC procedures, which have not been updated to take
advantage of these new tools. Once again, this did not happen in the USA.
Oh well, if some of my former teammates at the HPRC who sent me those
patches read this, it's certainly not a criticism toward you in person, but
more toward the lack of adaptability of the procedures you have to carry
out, whether you like them or not.
Now, about Patch/iX and Stage/iX, let me say a few things that most of you
probably already know. Patch/iX enables to install subsys, Powerpatches and
patches while users continue doing their business as usual. This is a net
reduction in planned downtime. Stage/iX enables to boot the system from a
staged area (a copy of MPE on disk that contains patched code), without the
need to create an SLT and boot from tape. This is another net reduction of
planned downtime. All in all, system uptime ratios will significantly
increase just by using today's standard tools. When trying to implement a
High Availability strategy, who would reject such tools ? Certainly not me,
hence my slight frustration when I received Autopat-format patches, and my
overwhelming support for Patch/iX and Stage/iX.
Oh, BTW, does anything similar to Patch/iX and Stage/iX exist on other OSs
? On Un*x ? NT ? OS/400 ? Anything else ? Not sure. Anyone has information
to share ? Once again, MPE might lead the world of cutting-edge technology.
Christian Lheureux
Consultant for Systems and Networks
APPIC R.H.
An HP3000 Systems Integrator
-----Message d'origine-----
De: Michael Dovano [SMTP:[log in to unmask]]
Date: mardi 13 avril 1999 22:59
A: [log in to unmask]
Objet: Re: TR: increasing the visibility of Patch/iX
Hello Christian and all:
Some clarification on my previous message about Patch/iX: I did not mean to
imply that things are perfectly OK with Patch/iX and Stage/iX at this point
and
we should all throw out AUTOPAT and be done with it. In fact, we have a
ways to
go before we in the MPE patch area can declare victory. What I DO mean to
say
is 1) I believe that the quality and usability of Patch/iX has improved
noticeably since it was first released, and 2) that we will continue to
make
improvements in Patch/iX (and Stage/iX) in the areas of reliability,
usability,
documentation, etc. I am personally committed to making this happen.
Really.
I'm not just saying that.
Anyway, as far as the specific issues of sending patches out without
Patch/iX
documentation and - in at least one case - without the correct files
(???!!!),
I'll be following up on this shortly.
Thanks for keeping me honest!
md
Christian Lheureux wrote:
> Mike,
>
> I strongly support your very encouraging perspective set forth for
> Patch/iX and Stage/iX. However, my own perception is that reality is not
> yet quite in sync with what you propose.
>
> I fact, I ordered a few patches about 3 weeks ago from an HP Response
> Center and, guess what, they sent me patches in Autopat format, missing a
> few files that I understandare necessary for Patch/iX. The install by
> itself worked fine, but with yesteryear's tool. Oh, well, the system I
was
> patching is a test box, so I did not care much about downtime, but, as an
> HP partner, I would appreciate to experiment with the most recent tools,
to
> be able to promote these tools.
>
> BTW, all of you on this list who know me probably remember that I have
been
> an HP Response Center Engineer for ... well ... many years, and I have
been
> involved, among others, in the beta test for 5.5 which, ironically
enough,
> introduced Patch/iX and Stage/iX.
>
> This is not ranting, it's simply my wish that reality ought to catch up
> with what I understand is the perspective set forth for Patch/iX and
> Stage/iX, Mike.
>
> Christian Lheureux
> Consultant for Systems and Networks
> APPIC R.H.
> An HP Systems Integrator
>
> -----Message d'origine-----
> De: Michael Dovano [SMTP:[log in to unmask]]
> Date: mardi 13 avril 1999 00:17
> A: [log in to unmask]
> Objet: Re: increasing the visibility of Patch/iX
>
> Just a couple of comments from the patch team here in the MPE/iX lab:
>
> Yes, Patch/iX has definitely gotten better, and I agree with Ted in
saying
> that if
> you haven't tried it lately, you should consider giving it a second
chance.
> Also, it
> is important to note that we consider Patch/iX and Stage/iX (and not
> AUTOPAT) to be
> the future of patch installation for MPE/iX. With this in mind, we are
> looking into
> some major improvements for both tools, such as a batch mode interface
for
> Patch/iX,
> much better reporting of patch information, and greatly expanding the
> number of
> patches Stage/iX can handle, to name a few.
>
> Also, we are currently working with the response centers on both sides of
> the
> Atlantic to increase the acceptance of Patch/iX. This should lead to
> changes in user
> documentation, as well as training for the support community.
>
> Other than that, we do value your feedback, so keep posting, complaining,
> and
> suggesting!
>
> md
>
> Ted Ashton wrote:
>
> > Thus it was written in the epistle of Mark Bixby,
> > > Perhaps more people would be using Patch/iX if it was actually
> mentioned in
> > > the standard "HP 3000 Patch Installation Instructions" that the North
> American
> > > Response Center sends along with snail-mail (well, FedEx) patches.
> >
> > I'm hoping that it will happen as the Response Center starts to trust
> Patch/iX
> > more. We're in a bit of a catch-22 here as the HPRC (at least when
I've
> > dealt with them) doesn't trust Patch/iX and so it doesn't get used,
> doesn't
> > get the kinks worked out and so isn't up to the level that the HPRC is
> > prepared to trust! I figure it's going to take a few of us who use it
> anyway
> > getting the ball rolling.
> >
> > To those out there who say, "Oh, I use AutoPat, we tried Patch/iX once
> and it
> > didn't work right so we don't use it now," give it a try again. It's a
> lot
> > better than it used to be. To the folks who are making it better, keep
> up the
> > good work, it ain't perfect yet (contact me if you want to know more
:-).
> >
> > Ted
> > --
> > Ted Ashton ([log in to unmask]), Info Serv, Southern Adventist
> University
> > ==========================================================
> > A man is like a fraction whose numerator is what he is and whose
> denominator
> > is what he thinks of himself. The larger the denominator the smaller
the
> > fraction.
> > -- Tolstoy, [Count] Lev Nikolgevich
(1828-1920)
>
> --
> Michael Dovano
> MPE/iX OS Patch Architect
> Commercial Systems Division, Hewlett-Packard
> 19447 Pruneridge Avenue
> Cupertino, CA 95014
>
> E-mail: [log in to unmask]
--
Michael Dovano
MPE/iX OS Patch Architect
Commercial Systems Division, Hewlett-Packard
19447 Pruneridge Avenue
Cupertino, CA 95014
Phone: (408) 447-5728
E-mail: [log in to unmask]
|
