Phone system mishaps in East Coast, L.A. questioned


WASHINGTON -- Striking similarities between nearly simultaneous computer malfunctions that disrupted local telephone service on the East Coast and in Los Angeles on Wednesday have raised questions among communications experts about the reliability of advanced networks that all the Bell telephone companies are now installing.

The problems experienced by both Pacific Bell and the Chesapeake & Potomac Co., which serves District of Columbia, Maryland, Virginia and parts of West Virginia, involved computer programs on advanced call-routing equipment, which uses the same new technology, one being adopted throughout the communications industry.

The problems, which were corrected in both areas by early evening Wednesday, made it impossible for about 9 million telephone customers to complete local telephone calls.

Although the origins of both malfunctions remained unclear yesterday, the difficulties at the two companies bore a strong resemblance toa brief but massive breakdown experienced by the American Telephone & Telegraph Co.'s long-distance lines in January 1990.

In all three cases, a problem at one switching center quickly corrupted other switches and paralyzed much of the system. Perhaps the biggest fear, federal regulators say, is that as telephone companies link their networks more closely, malfunctions at one company can infect systems at other companies and at long-distance carriers.

"What you want to avoid is the situation where one system contaminates another," said an investigator at the Federal Communications Commission who insisted on anonymity.

"I guess the ultimate concern is that software or hardware would be deployed in a way that the corruption could be processed through an entire network, and there would be no alternatives available."

As the telephone companies and government regulators tried to determine more precisely yesterday what went wrong, investigators at the communications commission said they also would look at several other questions:

Are there systemwide problems that have gone unnoticed until now? Can telephone companies reduce risks by reducing their dependence on one type of switching equipment? Were the disruptions caused by computer operators outside the telephone companies trying to sabotage the systems?

Officials at both companies discounted the possibility that a computer hacker might have caused the failures, and outside experts tended to agree.

"There's always that possibility, but most likely it was some kind of glitch or bug in the software," said A. Michael Noll, a professor at the Annenberg School of Communications at the University of Southern California and author of several textbooks on telecommunications technology.

Several independent communications experts said the problems reflected the difficulty of spotting all the hidden problems in complex software before putting it into commer

cial use.

"It's very hard to simulate all the possibilities in a laboratory," said Richard Jay Solomon, a telecommunications consultant and research associate at the Massachusetts Institute of Technology. "You have to go out in the field and keep your fingers crossed."

As more information became available yesterday, the two disruptions appeared to be almost identical.

Copyright © 2020, The Baltimore Sun, a Baltimore Sun Media Group publication | Place an Ad