3
$\begingroup$

The Y2K scenario occurs, usually, with incremental counters (such as counting the seconds that pass with time). When you have a situation in which an incremental counter will infinitely increase, you have the potential to overflow depending on how much space you've allocated to storing this data.

enter image description here

See the following links for more information:

We've already lost a space craft dubbed Deep Impact to a Y2K scenario in which the craft counted the number of 1/10 seconds that had elapsed. It stored this number in a 32-bit unsigned integer. This means the maximum the number could reach is 2.147 billion. Once the number reached 2.147 billion it flipped to -2.147 billion and essentially crashed all software systems.

Self-contained wikipedia quote:

On September 10, 2013, a Deep Impact mission status report explained that mission controllers believe the computers on the spacecraft are continuously rebooting themselves and so are unable to issue any commands to the vehicle's thrusters. As a result of this problem, communication with the spacecraft was explained to be more difficult, as the orientation of the vehicle's antennas is unknown. Additionally, the solar panels on the vehicle may no longer be positioned correctly for generating power.[76]

On September 20, 2013, NASA abandoned further attempts to contact the craft.[77] According to chief scientist A'Hearn,[78] the reason for the software malfunction was a Y2K-like problem. August 11, 2013, 00:38:49, was 232 tenth-seconds from January 1, 2000, leading to speculation that a system on the craft tracked time in one-tenth second increments since January 1, 2000, and stored it in an unsigned 32-bit integer, which then overflowed at this time, similar to the Year 2038 problem.[79]


My question is:

Are there any other known cases of this, future, past or present, that you know of? Any computers running Unix with MET planned beyond 2038 may qualify. I would like to focus on currently at-risk in-flight missions, but would definitely not mind examples from the past! Honestly, even if the ISS has 1 computer on it that runs UNIX for testing purposes, even that would be interesting to know about!

$\endgroup$
7
  • 1
    $\begingroup$ "time overflow" rather than Y2K? $\endgroup$
    – user20636
    Mar 9, 2020 at 16:57
  • $\begingroup$ Do you ask for time counter overflows in future? $\endgroup$
    – Uwe
    Mar 9, 2020 at 17:53
  • $\begingroup$ The maximum of a 32 bit unsigned integer is not 2.147 billion and the minium is not -2.147 billion. Unsigned means there are positive values only, minimum is 0, maximum is 4.294 billion. There are 2^32 different values including zero, the maximum is 2^32 - 1. $\endgroup$
    – Uwe
    Mar 9, 2020 at 17:59
  • 1
    $\begingroup$ Shuttle had a similar but not exactly the same problem: the Year-End Rollover (YERO) problem. The basic problem was that the onboard computers' clocks rolled to day 366 and the shuttle's onboard time unit rolled to day 1 resulting in a loss of synchronization. (I may have gotten this backwards, but you get the idea) $\endgroup$ Mar 9, 2020 at 20:21
  • 1
    $\begingroup$ related and potentially(?) an answer: What is GPS' 19 year rollover and does it present a cybersecurity issue? $\endgroup$
    – uhoh
    Mar 10, 2020 at 5:56

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.