It was every federal worker's nightmare. Malcolm Tindall, a 53-year-old air traffic controller, was shaving when his supervisor called and said the results of his drug test had come back positive.

"Yeah, sure," Tindall recalls laughing. "Somebody call in sick and you want me to come in early?"

"This is serious, Mel," said a shaken Walter S. Smith, the air traffic manager at the Santa Rosa Tower in northern California. Smith couldn't believe it himself. Tindall, an 18-year federal controller who never took sick leave and climbed the 96 stairs each day to the tower, had tested positive for marijuana, PCP and cocaine.

It is not true, Tindall gasped. "You're taking away my livelihood. I've got two years to go before retirement, and I'm not going to be able to get a job at a 7-Eleven with this on my record."

Five hours later, the official nightmare ended. Tindall was found to be the victim of an administrative error at the drug-testing laboratory, was taken off administrative leave, had his draft letter of termination destroyed and took the rest of the day off.

Tindall's case, an isolated example, assumes new significance as the federal government prepares to begin widespread random testing of thousands of workers for illegal drug use.

It shows that the strictest scientific guidelines cannot prevent human error, that good working relationships between supervisors and employees are important, that a small amount of illegal drug use has gone undetected, and how terrifying even one wrong result can be to the victim and coworkers.

President Reagan proposed government-wide random drug testing for workers in sensitive positions more than a year ago. After a series of delays, the program appears likely to go into effect as early as May.

Strict scientific guidelines for the tests are ready for publication in the Federal Register. And Congress is expecting notification soon of the costs and extent of the detailed program.

As a result, the spotlight has focused on the pioneering program of the Department of Transportation, which has tested 890 workers at random and gotten seven positive tests, not including Tindall's, which was first reported in the Federal Times.

"The important thing to remember here is that, in general, the system worked," said Melissa J. Allen, deputy assistant secretary for administration of the Department of Transportation. "Within five hours, we had verified that the test was not positive for that employee {Tindall}, and the employee is back on the job and nothing is in his record."

"We regret the inconvenience," said Michael Terretti, vice president for marketing and sales for CompuChem Laboratories, which conducted the test, "but we must stress that the system did work, the mistake was caught, and in two hours the error was rectified."

". . . A whole line of people . . . believed this fellow," said Janice LaChance, the spokesman for the American Federation of Government Employees. "DOT did get right on it and did it all quickly and got it resolved, but it really showed the potential for disaster and for careers being ruined in this whole program."

Dan Jenkins, the facility representative for the National Air Traffic Controllers Association, said, "I thank God that of all people in the FAA they decided to foul up a test on, it was Mel, not somebody younger or more active in the union. . . . Mel is the least likely guy in the entire FAA to be on drugs," he said.

CompuChem Laboratories is a $25 million-a-year, publicly held North Carolina company that has one of the best reputations in the country for forensic drug testing -- testing that will stand up in court.

The company uses highly technical and expensive processes to detect the presence of drugs in urine, which is collected under elaborate security procedures by employees of Upjohn Co.

CompuChem says it has performed 8 million to 10 million tests on 1 million samples for various clients and "this is the first time {an administrative error similar to Tindall's} has ever happened" despite more than 500 legal and administrative challenges, Terretti said.

Terretti said it occurred as the samples were coming into the laboratory and taken out of their mailing containers in groups of three. Tindall's sample -- taped shut, labeled with identifying information and sealed with security tape initiated by Tindall -- was placed on the top of the paper work accompanying another sample.

When the test, which was conducted on a small sample of urine while the rest is retained in the original bottle, had a positive result, CompuChem reported the results to Tindall's supervisor without double-checking against the identifying material remaining on the bottle.

Territti said the employees involved have been given further training and "made to realize the seriousness of this work." No one was fired.

In addition, he said, new procedures have been instituted: Containers are opened one at a time, a new computer check ensures that every sample is double-checked with the bottle it came from and, before positive results are reported to employers, a new layer of human review of all results against all evidence has been added to make sure the result is accurate.

"If an employee stands up and says 'This must be a mistake,' we will either ask for a second test of the sample from the same lab or an interlab transfer and second test," the Transportation Department's Allen said.

She said that the seven employees with verified positive tests have all entered or are about to enter rehabilitation programs without protest.

The results reported to Tindall were traced to another employee, Allen said, who has been notified, and is "in counseling." She said the employee "was functioning" and that the worker's supervisor had not suspected drug use. She did not say whether the employee was an air traffic controller.

By all accounts, Smith's support for Tindall was important in resolving the matter expeditiously. "I'm an ex-cop," said Smith, who was Ronald Reagan's bodyguard briefly when Reagan was governor of California. "If this were true, there would have been indications."

"I stuck my neck out." Smith said. The flight surgeon "claimed their tests were flawless. But we finally convinced him to let us talk to Washington, where they agreed to ask the lab to double-check."

Smith "stuck by me," Tindall said. "I'd be in deep {expletive} if it weren't for him. But that's not the case everywhere. The FAA has a lot of combat zones where it is us against them."

The Santa Rosa Tower has only nine controllers, and Smith knows all of them well. He once was a controller himself, working side-by-side with Tindall.

"Walt knew Mel," said Jenkins, the union representative. "If it had been a big, center facility where the chief doesn't see his employees very often, I don't know that the guy would have fought quite as hard. What would have happened if he had been younger, worn his hair longer, been more active in the union?

"We were also lucky that the lab was able to find the mistake. What happens when they make an absolutely honest mistake that they can't find? It's bound to happen," Jenkins said.

Smith said that FAA Administrator T. Allan McArtor called Tindall to apologize. "I think it should have come from the White House, President Reagan himself. It was a terrible thing to do to a guy," Smith said.