I can tell you that, absolutely positively, you are WRONG. My manager
and I did *EVERYTHING* we were supposed to. The disk vendor lied to us
(ok, partially our fault, we went with non IBM disk).
I just want to make sure that you understand me correctly. I don't want to imply that you or your manager did your job wrong. You checked with every vendor, you got greenlights from everywhere, it's okay to assume that everything will work.
However, without a FULL compatibility check and no fallback plan, you can run into problems. If you do the math, 8 days of downtime are probably cheaper than maintaining an exact replica of the production system for QA purposes. That doesn't mean it's impossible, though.
A technician or developer should never gamble with systems. That's for management to decide.
And that solved the problem how? My friends was a consultant, brought
It won't happen again. The Windows Admin made a mistake, you tell him, and the problem will be fixed the next time you need to roll out a new software.
I've observed that the main thing that causes problems in stable systems
is change. If an application is running fine in V5R1, and IBM doesn't
Support it anymore, and there isn't any need for change, the main thing
that WILL cause problems is change.
I prefer planned changes with planned problems over unplanned changes with unplanned problems. YMMV.
But they should hit them with the cluestick whenever they have
the opportunity.
That's a very easy way to loose customers.
Naah. You'll just need sales guys to wrap the cluestick in enough cotton candy.
It's fine with me to close this thread with your realistic sight of the situation, and my naïve "that's how it should be world" (that wasn't sarcasm.).
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of David Gibbs
Sent: Tuesday, July 24, 2007 8:20 PM
To: Midrange Systems Technical Discussion
Subject: Customers vs. ISV (was: New redpaper)
Lukas Beeler wrote:
Then how do you explain the 8 days of downtime the company I used to
work for experienced when trying to upgrade our CISC system to a RISC
system?
I don't know. But what I do know is that it is entirely possible to
avoid those 8 days of downtime.
I can tell you that, absolutely positively, you are WRONG. My manager
and I did *EVERYTHING* we were supposed to. The disk vendor lied to us
(ok, partially our fault, we went with non IBM disk).
The cheap variant: Roll back if it doesn't work within half the allotted
upgrade timeframe.
Sorry, Lukas, you really can't speak with authority on this ... you have
no idea what our situation was. There was no cheap or simple variant.
Once we started the upgrade we were committed.
Luckily we were able to get IBM involved and do very low level traces on
the system to identify the drive, platter, cylinder, and head, that was
generating the error. It was only then we got the disk vendor to fess
up and give us a free disk upgrade (same capacity, newer hardware).
You should kick the Windows Admin. 50 Desktops and no software
deployment solution?
And that solved the problem how? My friends was a consultant, brought
in to upgrade this package ... and the customer only had a part time
system administrator (backups, change paper in the printer, do routine
maintenance on PC's when required).
If you system is running fine, your application isn't dependent on new
functionality in later OS400 versions, and you hardware support is
being
handled by a 3rd party vendor, what's the driving need to upgrade?
The _POSSIBILITY_ of getting unforeseen consequences if you don't.
I've observed that the main thing that causes problems in stable systems
is change. If an application is running fine in V5R1, and IBM doesn't
support it anymore, and there isn't any need for change, the main thing
that WILL cause problems is change.
But they should hit them with the cluestick whenever they have
the opportunity.
That's a very easy way to loose customers.
As I said before ... I work for an tools vendor ... requirements of a
development tool does not (and should not) dictate what OS400 version is
used (within reason). Especially when there are downstream
ramifications (i.e., the customer is a ISV themselves ... and has their
own release level requirements).
david
p.s. I think this horse pulling this thread has gotten enough beating
... perhaps we let it go back to barn for some hay?
midrange.com
