Human error causes Reuters Mexico Mkt data glitchgreenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread
Monday January 3, 3:10 pm Eastern Time
Human error causes Reuters Mexico mkt data glitch
MEXICO CITY, Jan 3 (Reuters) - Due to human error, Reuters screens carried incorrect data from the Mexican Stock Exchange Monday morning, Reuters said.
``The cause of this was traced to a computer which, in recent months, had participated in test data sessions with the Mexican Stock Exchange. Due to an operational oversight the resulting test files were not deleted from the computer in question,'' Reuters said in a letter to the Mexican Stock Exchange.
``This situation was not caused by any date-related software or hardware failure,'' Nancy Bobrowitz, a Reuters spokeswoman, said.
Reuters said that when the Mexican bourse sent out pre-opening messages Monday morning, it set off distribution of the data contained in the Reuters computer test files.
``Mexican Stock Exchange instruments were affected,'' Reuters said.
Reuters said the problem was identified and a back-up computer put into operation prior to the market's opening.
Reuters also said that by the time the Mexican Stock Exchange opened all real-time updates from the exchange were being processed correctly, but that problems persisted in the closing prices for the last trading day.
-- Homer Beanfang (Bats@inbellfry.com), January 03, 2000
This must be a computer that was just put into operation today.
There are two types, fully remediated systems that have old but fixed code on them and that worked over the CDC and replacement ones that have just been brought on line. This seems to be a replacement one.
These are the ones that have the greatest chance of producing silent but undetected errors that are noticed later.
Wonder how many transactions today produced bogus results today by a replacement computer and will show up in bills,etc by next month?
Of course by then it will be declared a human error because the trail is too cold.
-- William R. Sullivan (email@example.com), January 03, 2000.
Y2K problem was a human error, wasn't it?
-- gomer (firstname.lastname@example.org), January 03, 2000.