Text 1671, 79 rader
Skriven 2004-12-31 09:04:36 av Ellen K. (1:379/45)
   Kommentar till text 1664 av Mike '/m' (1:379/45)
Ärende: Re: Airline flight + Luggage TangleUp
=============================================
From: Ellen K. <72322.enno.esspeayem.1016@compuserve.com>

The classic stupid programming mistake.

I got hit with one of those earlier this year -- I created a baby CRM database
in SQL Server that gets written to while the salespeople are also writing to
the legacy system...  So of course in my data model a lead number can only be
associated with one Person, while a Person can be associated with multiple lead
numbers...  Monitoring the data every night at the beginning, around the 3rd
day I suddenly saw a gazillion different names all with the same PersonID (i.e.
all but the latest one were in the PersonHistory table, but like every row was
a new name, IOW the system had the Person changing his/her name (and address
and phone number etc) a gazillion times)... blabla, turns out the legacy system
had code that asked for the next lead number (the legacy system is not a
relational database, so this had to happen programmatically) which made 1000
attempts, which if it didn't succeed after the 1000 attempts it passed back a
string of 10 spaces.  (The legacy system is 16-bit, so all numeric identifiers
are actually a 10-character string with leading spaces.)   Well, apparently on
the third day things got busy enough that
1000 attempts didn't get a new lead number, so it returned a 10-space string,
and then my system of course associated every successive lead with a 10-space
string for the lead number to the same Person as the first one.   I ended up
nuking the database and starting it over from
the initial population.

On Thu, 30 Dec 2004 10:30:33 -0500, Mike '/m' <mike@barkto.com> wrote in
message <tm78t0h7oi3t7jinu5l63v72mi6evar466@4ax.com>:

>
>http://www.cincypost.com/2004/12/28/comp12-28-2004.html
>
>===
>The computer software that crashed and grounded Comair's entire fleet on
>Christmas Day was an antiquated system due to be replaced in the coming
>months....
>
>Tom Carter, a computer consultant with Clover Link Systems of Los Angeles,
>said the application has a hard limit of 32,000 changes in a single month.
>
>"This probably seemed like plenty to the designers, but when the storms hit
>last week, they caused many, many crew reassignments, and the value of 32,000
>was exceeded," he said.
>
>Carter said SBS's newer system, called Maestro Crew, is a far more
>sophisticated system unlikely to run into the same problem that Comair
>faced....
>===
>
>  /m
>
>
>On Wed, 29 Dec 2004 19:42:42 -0800, Jeff Shultz <jeff@shultzinfosystems.com>
>wrote:
>
>>I figured that the Comair CIO was scheduled for exit counselling soon...
>>
>>On Tue, 28 Dec 2004 12:29:23 -0500, Robert Comer wrote:
>>
>>> All I could think of when I first saw the story was that I sure was glad I
>>> wasn't the programmer responsible. <g>
>>>
>>> - Bob Comer
>>>
>>>
>>> "Mike N." <mike@u-spam-u-die.net> wrote in message
>>> news:ig53t09oi557qoqjq41fle1t6r6r11knlv@4ax.com...
>>>>
>>>> http://news.com.com/U.S.+to+probe+Comair%2C+US+Airways+troubles/2100-1002_
3-5504993.html?tag=nefd.top
>>>>
>>>>  It'll be interesting to see how much of this is tech-related VS
>>>> human factors.  What kinds of DBs were running, and what was their
>>>> backup plan?
>>>>
>>>>
>>>>

--- BBBS/NT v4.01 Flag-5
 * Origin: Barktopia BBS Site http://HarborWebs.com:8081 (1:379/45)