[SIM] 'Infrastructure Year 2000 Problems
Posts or replies since May. 16 2:40 AM 1998
[go back one day] [go back one week]
90. Author: David C. Hall ( dhall )
Date: May. 20 9:03 AM 1998
There is a 50% to 60% chance that each major
telecom carrier will suffer at least one
failure of a mission critical system,
according to Gartner Group. Small, midsize,
and foreign-based carriers will be affected
the most. US fire and police department
dispatch systems are vulnerable based on FCC
notes. Failures could range from billing
problems to a complete lack of service. It is
interensting to note that most words being read
are reissues of stuff the Yr2K community has
been saying are possibilites for several
years. We really have not been able to confirm or
deny these predictions by getting real live
test data from ANYBODY. I was invited to a state
House Committee meeting in May and listened to
the rep from US West tell us that they can
guarantee dial tone in their system, period.
That was an interesting statement since they
have not yet conducted ANY system-wide tests to
determine if the Yr2K fixes work properly when
stressed.
I hope that they are correct, but I do tend to
discount ANY statement made about
computerized equipment that does not have test
results to back it up. My experience with
computerized stuff says that it tends to not
work right the first time you do anything, even
with isolated items of equipment. Since we are
working with very interconnected systems
consisting of several tiers of complexity made
by several different vendors, I am more
inclined to doubt sweeping statements not
covered by tests. Based on this experience I am
going to discount ANY statements about how much
ANY system will work until there are test
results confirming that statement. This
basically means that we can now ignore any
statements made by our basic infrastructure
companies (electric, telecom, etc.) until
test results are made known or until late 1999,
since most tests are scheduled to run then. I
would appreciate any opinions anyone has in
this area.
Dave Hall
dhall@enteract.com
91. Author: David C. Hall ( dhall )
Date: May. 22 8:11 AM 1998
Embedded systems are finally getting press
coverage. Janet Reno is now talking about
embedded systems! At her weekly press
conference she briefed reporters on the
problems of date-sensitive hardware in places
like medical devices and electric power
plants. The URL for this is
cgi.pathfinder.com.
(really!)
Dave Hall
dhall@enteract.com
92. Author: David C. Hall ( dhall )
Date: May. 25 10:25 PM 1998
Some notes from a correspondent:
We have found varying levels of problems with
the following types of systems: 1) Medical -
defibrilators, ultrasound units, and medical
imaging devices; 2) Mining - DCS, SCADA,
trucks, PLCs, slope monitoring; 3) Oil and Gas -
SCADA, custody transfer metering, gas
chromatographs; 4) Power Generation - DCS,
SCADA, re-closers; 5) Airports - flight
information, baggage handling, UPS; 6)
Petrochemical - DCS, lab equipment, PLCs,
SCADA; 7) Buildings - PABX, voicemail,
elevators, HVAC, fire controls, surveillance
systems; 8) Food and Beverage - product date
stamping, production reporting; 9) Printed
Media - ink systems, bundling and distribution
systems, paper handling; 10) Heavy Industry -
PLCs, system interfaces; 11) Water Utilities -
SCADA, chlorination systems, billing
systems.
Please do not send a request to "identify the
exact equipment and manufacturers" since I do
not know them and am not going to ask. It does not
matter what specific item of equipment was
tested, since there is no such thing as
identical items of equipment or installation
of systems. The only reason I include types on
this conversation is to inform you that here are
some types that have tested non-compliant. You
should therefore test YOUR equivalent systems
and equipment to determine its compliancy.
Many things are going to malfunction and if you
do not know which ones, you are probably not
going to be able to recover from the failures.
Organizations need to , at the very least,
determine WHAT will malfunction and what the
malfunction will mean to the operation of their
enterprises. Testing of key operational
processes must be done, or you run the risk of
unknown failures. Proving that you do not have a
problem will be harder than trying to find out if
you do have a problem.
I am insisting that anything that could affect
MY health and safety (electricity, water,
food, pharmeceuticals, health care, etc.)
PROVE that they have no Year 2000 problems or
impacts. I do not want to wake up anytime in 2000
to blackness, cold, no water, inadequate
health care, etc. because no one asked
questions and everyone just ASSUMED that they
had no problem.
Dave Hall
dhall@enteract.com
year2000.unt.edu |
