News for hackers. Stuff that matters.
Re:Uptime 15,364 days – The Computers of Voyager [video] by segfaultbuserr on Friday October 18, 2019 @07:19AM (#21290344)
> Uptime 15,364 days

Is the uptime really technically true? Sure, the Voyager has been operating for 40+ years, but all embedded systems must have watchdog timers. And given how hostile the space environment is, I'll be surprised if the main system hasn't been reset by a watchdog timer for a couple of times to recover from fault conditions, thus the actual uptime must be much less than that.

Re:Uptime 15,364 days – The Computers of Voyager [video] by Zenst on Friday October 18, 2019 @09:20AM (#21291114)
https://www.wired.com/2013/09/vintage-voyager-probes/

"Practically all of Voyager's redundancy is gone now, either because something broke along the way or it was turned off to conserve power. Of the 11 original instruments on Voyager 1, only five remain"

Though I did wonder if they updated the firmware/code in all this time and als no definitive answer stood out: https://www.quora.com/Was-the-opportunity-to-update-the-Voya...

Re:Uptime 15,364 days – The Computers of Voyager [video] by coldnose on Friday October 18, 2019 @01:52PM (#21293540)
Voyager 2's computer was reset in 2010:

https://voyager.jpl.nasa.gov/news/details.php?article_id=16

It suddenly started returning corrupted frames. So engineers had it slowly transmit a core dump, and found a single flipped bit. To fix the problem, they "reset" the computer, which I assume means they rebooted it.

Re:Uptime 15,364 days – The Computers of Voyager [video] by gus_massa on Friday October 18, 2019 @07:40AM (#21290442)
From https://en.wikipedia.org/wiki/Command_Loss_Timer_Reset :

> Most spacecraft have more than one Command Loss Timer Reset for subsystem level safety reasons, with the Voyager craft using at least 7 of these timers.

Re:Uptime 15,364 days – The Computers of Voyager [video] by methehack on Friday October 18, 2019 @01:37PM (#21293400)
Isn't this splitting hairs? The real achievement it seems to me is continuous operation within spec. The watchdog timers (whatever they are) are part of the design that enabled this. I don't think it lessens the achievement at all.
Re:Uptime 15,364 days – The Computers of Voyager [video] by segfaultbuserr on Friday October 18, 2019 @03:13PM (#21294239)
> I don't think it lessens the achievement at all.

I was not implying that "the fact that the system has been rebooted lessens the achievement", I said none of it. I was just wondering about details of the system and technical accuracy of the statement, isn't it the point of posting on HN?

Running a probe for 15,364 days without even a single bit flip or poweroff would be an extraordinary miracle that exceeded all reasonable expectations, not simply the grestest accomplishment.

Please don't assume that every technical statements/questions imply undervaluation, criticism or attack, regardless of how common they are in tech.

Thanks.

Re:Uptime 15,364 days – The Computers of Voyager [video] by okcando on Friday October 18, 2019 @02:13PM (#21293721)
Sure, but it's not conventionally what is called "uptime", an uninterrupted period in a normal and responsive state. Voyager has had faults and been rebooted. In any other situation that I'm aware of, you'd have to say that uptime started over or that uptime is <100%.