Tom Brennan wrote:
>When I'm trying to get to an IBM service and it's not
>available, it's highly likely the problem is on the IBM side.
>Why? because if it was a public/local internet problem I wouldn't
>be trying to get to an IBM service - I'd be trying to solve our
>internet issue so my wife can post pictures of her lunch on Facebook.

First of all, we're not talking about whatever Internet connections you use
for posting to Facebook. We're (typically) talking about corporate and
government networks with highly controlled public Internet access,
sometimes coping with "great firewalls" and other external interference.
There are *plenty* of factors beyond any single vendor's control that can
and do go wrong. It could be as simple as a DNS misconfiguration, a bad
routing table entry, a hosts file intercepting something it probably
shouldn't, a firewall that isn't configured correctly.... The list is
practically endless, and the fact some other site might work doesn't
necessarily mean it's an IBM problem.

I don't know why this concept seems so hard to understand, but let me try
again. YES, in my view, IBM ought to improve what it can improve. But you
MUST plan for the possibility of not being able to reach IBM, or any other
vendor, at every moment in time if you're running a mission critical
operation. Even if IBM executes perfectly, there are myriad reasons why the
path to IBM might be closed off to you at a particular moment in time.

Clark Morris wrote:
>If you need to get a fix, the Internet had better be working and most
>data centers are probably in deep trouble if it isn't working.

"Deep trouble" might be true, but the fact remains that those with mission
critical needs, who take them seriously, plan for exactly this class of
problems and have contingencies.

"The Internet" in this case is a multifaceted set of services and
configurations. It's entirely possible, even common, for your site to have
narrow "Internet" problems that nobody else is experiencing that are,
nonetheless, highly disruptive. Maybe somebody figured out how to clip your
wings selectively using router-targeted malware, to pick a random example.
Many problems are possible, and many happen.

>Microsoft understands this.

You've mentioned Microsoft and Amazon. They've both had widespread service
outages, many publicly known and well documented.

Again, to repeat, I think all vendors, including IBM, ought to improve what
they can. That's a necessary condition, but it's not sufficient and never
can be. You MUST have a "Plan B" that does not rely on your ability to
reach your vendor(s) continuously if you have mission critical needs and
take them seriously. In the case of Knowledge Center, IBM is equipping you
with an excellent "Plan B" option called the IBM Knowledge Center for z/OS,
included at no additional charge with your base z/OS license. You may or
may not have other contingencies, but that's an excellent one and is
available at no additional charge.

--------------------------------------------------------------------------------------------------------
Timothy Sipples
IT Architect Executive, Industry Solutions, IBM z Systems, AP/GCG/MEA
E-Mail: [email protected]

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to