Mars Rover Curiosity had an apparent major glitch on Wednesday, February 27, in the "A" side computer, the primary computer running the rover, and part of a twinned pair. After a short evaluation, NASA scientists put the "A" side computer into safe mode, and booted the "B" side computer up. But all seems well. The "A" side is back on line.
On February 27, just as Curiosity seemed to be within a day of beginning analysis of the sample of rock interior she had drilled from a Mars rock called by NASA "John Klein" -- disaster! The on board computer operating the rover didn't send data and failed to enter sleep mode when ordered to. Despite having a duplicate computer (the "B" side computer) that had been used for some of the approach and landing, this looked to be a major problem. However, NASA engineers soon discovered that things were perhaps less dramatically wrong than first thought. "Disaster" amounted to a small segment of the rover's programming being corrupted; most likely by the high energy cosmic ray radiation that strikes the surface of a planet with as little atmosphere as Mars has. The rover is hardened against such high energy strikes, but one hit the computer's directory (table of contents) anyway, sending the program into an endless loop. Much relieved, NASA scientists determined that if they sent the "A" side computer into Safe Mode and woke up the "B" side computer to keep Curiosity's necessary functions going, they could repair the damage fairly quickly. The "B" side computer, identical to the "A" side except for data updates, was given command of the rover, and the NASA computer scientists set about doing what computer scientists do... they began repairing the "A" side computer.
By Saturday, the "A" side computer was back on line, but some extensive and significant testing and reestablishing is definitely in order. The idea now is to make the "A" side computer the backup, at least temporarily. So the "B" side computer is being taught where everything and every part of the rover is, using data from the "A" side that was collected and calibrated during the weeks of testing that so badly frustrated many followers of the rover's adventures... followers who were expecting a great deal more travel and science early in the program.
NASA expects to take at least this week to get the rover ready to recommence its rock dust analysis and its travels. But followers of the rover who have been awaiting analysis of the rock dust will have to endure at least a month more of waiting. The sun will come partially between Mars and the Earth, and the risk of corruption of data in commands sent during that period is too great. If another serious glitch occurred, it might not be possible to gain the necessary control of the rover until a month had passed and the sun's interference was no longer a concern. The rover uncontrolled for a month doesn't bear contemplating.
Mars Rover Curiosity has been on Mars about seven months, and unlike its predecessors who each had several severe computer problems in that same amount of time, this is the first serious issue it has suffered. If this is typical of the strength of the machine, there may be many years of fascinating data collection and publication ahead. The rover's two year mission was extended indefinitely several weeks ago.
Meantime, even while Curiosity is down, NASA continues to produce data and photographs taken earlier and downloaded to the Jet Propulsion Laboratory's computers for interpreting and making presentable for the world to view. Among the hundreds of photographs is the self portrait below taken February 3 by Curiosity's MAHLI (Mars Hand Lens Imager). The photo itself is a mosaic of sixty-six separate shots melded together.