47 years. It's been out there 47 years. It's awe inspiring that a) it's still operating, and b) some wicked smart folks in JPL can reconfigure its code remotely to get it back on its feet. A huge achievement for no other reason than it just is. Congratulations to the smarty's.
I was wondering if we had something to follow up the amazing Voyager I and II and so asked one of our A.I. friends. Unfortunately it listed spacecraft that I don't consider "deep space", so perhaps New Horizons is our only follow-on:
"Voyager 1 and 2: Both are still communicating with Earth as they journey through interstellar space, but their aging power sources mean they may go silent sometime in the 2030s.
New Horizons: Based on its plutonium power source, New Horizons should still be operating in 2034 as it studies other Kuiper Belt objects after its Pluto flyby.
Juno: NASA's Juno Jupiter orbiter should still be operational in 2034, with its mission currently extended into the 2030s.
Parker Solar Probe: With its solar electric propulsion and generous power supply, Parker should still be studying the Sun's corona in 2034 and perhaps even in the 2040s.
So of the deep space assets listed, the most likely to still be actively sending data back to Earth in around 2034 are New Horizons, Juno, Parker Solar Probe, and potentially the two active Voyager probes if their power holds out that long."
Are there any courses / tutorials / books on the engineering of this system? I'd love to know what kind of choices you make with software when your system needs to run for decades and be remote debuggable at the lowest levels.
Does anyone with more context know how are they rearranging the memory bits mentioned in the article. Are they patching this using malloc/realloc or is it something entirely different?
This is pure assembly code with absolute memory addressing. Probably non-contiguous memory regions originally used for other purposes with jump instructions.
Voyager must have it's program code written in memory banks. Each code session having a known start address. The code inside the failed memory chip must have been moved to free banks or then less important code was deleted to give room for it.
After that, it must have been a matter of rewriting every GOTO opcode that pointed to the failed chip.
According to DSN Now (https://eyes.nasa.gov/dsn/dsn.html) it is currently communicating with Voyager 1 at 40 bytes per second. You have to click the antenna that lists VGR1 and then click more detail in the side bar to get the signal strength (-160dBm) and data rate.
"Voyager 1 and 2: Both are still communicating with Earth as they journey through interstellar space, but their aging power sources mean they may go silent sometime in the 2030s.
New Horizons: Based on its plutonium power source, New Horizons should still be operating in 2034 as it studies other Kuiper Belt objects after its Pluto flyby.
Juno: NASA's Juno Jupiter orbiter should still be operational in 2034, with its mission currently extended into the 2030s.
Parker Solar Probe: With its solar electric propulsion and generous power supply, Parker should still be studying the Sun's corona in 2034 and perhaps even in the 2040s.
So of the deep space assets listed, the most likely to still be actively sending data back to Earth in around 2034 are New Horizons, Juno, Parker Solar Probe, and potentially the two active Voyager probes if their power holds out that long."
After that, it must have been a matter of rewriting every GOTO opcode that pointed to the failed chip.
Totally guessing.
Lots more discussion: https://news.ycombinator.com/item?id=40117599